[layout] change the source directory layout

The logilab.common package now lives in a logilab/common directory to make it pip compliant.

Related to #294479.

47 files changed, 14291 insertions, 0 deletions

diff --git a/logilab/__init__.py b/logilab/__init__.py
new file mode 100644
index 0000000..de40ea7
--- /dev/null
+++ b/logilab/__init__.py
@@ -0,0 +1 @@
+__import__('pkg_resources').declare_namespace(__name__)
diff --git a/logilab/common/__init__.py b/logilab/common/__init__.py
new file mode 100644
index 0000000..6a7c8b4
--- /dev/null
+++ b/logilab/common/__init__.py
@@ -0,0 +1,175 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Logilab common library (aka Logilab's extension to the standard library).
+
+:type STD_BLACKLIST: tuple
+:var STD_BLACKLIST: directories ignored by default by the functions in
+  this package which have to recurse into directories
+
+:type IGNORED_EXTENSIONS: tuple
+:var IGNORED_EXTENSIONS: file extensions that may usually be ignored
+"""
+__docformat__ = "restructuredtext en"
+
+
+from logilab.common.__pkginfo__ import version as __version__
+
+STD_BLACKLIST = ('CVS', '.svn', '.hg', 'debian', 'dist', 'build')
+
+IGNORED_EXTENSIONS = ('.pyc', '.pyo', '.elc', '~', '.swp', '.orig')
+
+# set this to False if you've mx DateTime installed but you don't want your db
+# adapter to use it (should be set before you got a connection)
+USE_MX_DATETIME = True
+
+
+class attrdict(dict):
+    """A dictionary for which keys are also accessible as attributes."""
+    def __getattr__(self, attr):
+        try:
+            return self[attr]
+        except KeyError:
+            raise AttributeError(attr)
+
+class dictattr(dict):
+    def __init__(self, proxy):
+        self.__proxy = proxy
+
+    def __getitem__(self, attr):
+        try:
+            return getattr(self.__proxy, attr)
+        except AttributeError:
+            raise KeyError(attr)
+
+class nullobject(object):
+    def __repr__(self):
+        return '<nullobject>'
+    def __bool__(self):
+        return False
+    __nonzero__ = __bool__
+
+class tempattr(object):
+    def __init__(self, obj, attr, value):
+        self.obj = obj
+        self.attr = attr
+        self.value = value
+
+    def __enter__(self):
+        self.oldvalue = getattr(self.obj, self.attr)
+        setattr(self.obj, self.attr, self.value)
+        return self.obj
+
+    def __exit__(self, exctype, value, traceback):
+        setattr(self.obj, self.attr, self.oldvalue)
+
+
+
+# flatten -----
+# XXX move in a specific module and use yield instead
+# do not mix flatten and translate
+#
+# def iterable(obj):
+#    try: iter(obj)
+#    except: return False
+#    return True
+#
+# def is_string_like(obj):
+#    try: obj +''
+#    except (TypeError, ValueError): return False
+#    return True
+#
+#def is_scalar(obj):
+#    return is_string_like(obj) or not iterable(obj)
+#
+#def flatten(seq):
+#    for item in seq:
+#        if is_scalar(item):
+#            yield item
+#        else:
+#            for subitem in flatten(item):
+#               yield subitem
+
+def flatten(iterable, tr_func=None, results=None):
+    """Flatten a list of list with any level.
+
+    If tr_func is not None, it should be a one argument function that'll be called
+    on each final element.
+
+    :rtype: list
+
+    >>> flatten([1, [2, 3]])
+    [1, 2, 3]
+    """
+    if results is None:
+        results = []
+    for val in iterable:
+        if isinstance(val, (list, tuple)):
+            flatten(val, tr_func, results)
+        elif tr_func is None:
+            results.append(val)
+        else:
+            results.append(tr_func(val))
+    return results
+
+
+# XXX is function below still used ?
+
+def make_domains(lists):
+    """
+    Given a list of lists, return a list of domain for each list to produce all
+    combinations of possibles values.
+
+    :rtype: list
+
+    Example:
+
+    >>> make_domains(['a', 'b'], ['c','d', 'e'])
+    [['a', 'b', 'a', 'b', 'a', 'b'], ['c', 'c', 'd', 'd', 'e', 'e']]
+    """
+    from six.moves import range
+    domains = []
+    for iterable in lists:
+        new_domain = iterable[:]
+        for i in range(len(domains)):
+            domains[i] = domains[i]*len(iterable)
+        if domains:
+            missing = (len(domains[0]) - len(iterable)) / len(iterable)
+            i = 0
+            for j in range(len(iterable)):
+                value = iterable[j]
+                for dummy in range(missing):
+                    new_domain.insert(i, value)
+                    i += 1
+                i += 1
+        domains.append(new_domain)
+    return domains
+
+
+# private stuff ################################################################
+
+def _handle_blacklist(blacklist, dirnames, filenames):
+    """remove files/directories in the black list
+
+    dirnames/filenames are usually from os.walk
+    """
+    for norecurs in blacklist:
+        if norecurs in dirnames:
+            dirnames.remove(norecurs)
+        elif norecurs in filenames:
+            filenames.remove(norecurs)
+
diff --git a/logilab/common/__pkginfo__.py b/logilab/common/__pkginfo__.py
new file mode 100644
index 0000000..55a2cc3
--- /dev/null
+++ b/logilab/common/__pkginfo__.py
@@ -0,0 +1,57 @@
+# copyright 2003-2014 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""logilab.common packaging information"""
+__docformat__ = "restructuredtext en"
+import sys
+import os
+
+distname = 'logilab-common'
+modname = 'common'
+subpackage_of = 'logilab'
+subpackage_master = True
+
+numversion = (0, 63, 2)
+version = '.'.join([str(num) for num in numversion])
+
+license = 'LGPL' # 2.1 or later
+description = "collection of low-level Python packages and modules used by Logilab projects"
+web = "http://www.logilab.org/project/%s" % distname
+mailinglist = "mailto://python-projects@lists.logilab.org"
+author = "Logilab"
+author_email = "contact@logilab.fr"
+
+
+from os.path import join
+scripts = [join('bin', 'pytest')]
+include_dirs = [join('test', 'data')]
+
+install_requires = [
+        'six >= 1.4.0',
+        ]
+test_require = ['pytz']
+
+if sys.version_info < (2, 7):
+    install_requires.append('unittest2 >= 0.5.1')
+if os.name == 'nt':
+    install_requires.append('colorama')
+
+classifiers = ["Topic :: Utilities",
+               "Programming Language :: Python",
+               "Programming Language :: Python :: 2",
+               "Programming Language :: Python :: 3",
+               ]
diff --git a/logilab/common/cache.py b/logilab/common/cache.py
new file mode 100644
index 0000000..11ed137
--- /dev/null
+++ b/logilab/common/cache.py
@@ -0,0 +1,114 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Cache module, with a least recently used algorithm for the management of the
+deletion of entries.
+
+
+
+
+"""
+__docformat__ = "restructuredtext en"
+
+from threading import Lock
+
+from logilab.common.decorators import locked
+
+_marker = object()
+
+class Cache(dict):
+    """A dictionary like cache.
+
+    inv:
+        len(self._usage) <= self.size
+        len(self.data) <= self.size
+    """
+
+    def __init__(self, size=100):
+        """ Warning : Cache.__init__() != dict.__init__().
+        Constructor does not take any arguments beside size.
+        """
+        assert size >= 0, 'cache size must be >= 0 (0 meaning no caching)'
+        self.size = size
+        self._usage = []
+        self._lock = Lock()
+        super(Cache, self).__init__()
+
+    def _acquire(self):
+        self._lock.acquire()
+
+    def _release(self):
+        self._lock.release()
+
+    def _update_usage(self, key):
+        if not self._usage:
+            self._usage.append(key)
+        elif self._usage[-1] != key:
+            try:
+                self._usage.remove(key)
+            except ValueError:
+                # we are inserting a new key
+                # check the size of the dictionary
+                # and remove the oldest item in the cache
+                if self.size and len(self._usage) >= self.size:
+                    super(Cache, self).__delitem__(self._usage[0])
+                    del self._usage[0]
+            self._usage.append(key)
+        else:
+            pass # key is already the most recently used key
+
+    def __getitem__(self, key):
+        value = super(Cache, self).__getitem__(key)
+        self._update_usage(key)
+        return value
+    __getitem__ = locked(_acquire, _release)(__getitem__)
+
+    def __setitem__(self, key, item):
+        # Just make sure that size > 0 before inserting a new item in the cache
+        if self.size > 0:
+            super(Cache, self).__setitem__(key, item)
+            self._update_usage(key)
+    __setitem__ = locked(_acquire, _release)(__setitem__)
+
+    def __delitem__(self, key):
+        super(Cache, self).__delitem__(key)
+        self._usage.remove(key)
+    __delitem__ = locked(_acquire, _release)(__delitem__)
+
+    def clear(self):
+        super(Cache, self).clear()
+        self._usage = []
+    clear = locked(_acquire, _release)(clear)
+
+    def pop(self, key, default=_marker):
+        if key in self:
+            self._usage.remove(key)
+        #if default is _marker:
+        #    return super(Cache, self).pop(key)
+        return super(Cache, self).pop(key, default)
+    pop = locked(_acquire, _release)(pop)
+
+    def popitem(self):
+        raise NotImplementedError()
+
+    def setdefault(self, key, default=None):
+        raise NotImplementedError()
+
+    def update(self, other):
+        raise NotImplementedError()
+
+
diff --git a/logilab/common/changelog.py b/logilab/common/changelog.py
new file mode 100644
index 0000000..2fff2ed
--- /dev/null
+++ b/logilab/common/changelog.py
@@ -0,0 +1,238 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Manipulation of upstream change log files.
+
+The upstream change log files format handled is simpler than the one
+often used such as those generated by the default Emacs changelog mode.
+
+Sample ChangeLog format::
+
+  Change log for project Yoo
+  ==========================
+
+   --
+      * add a new functionality
+
+  2002-02-01 -- 0.1.1
+      * fix bug #435454
+      * fix bug #434356
+
+  2002-01-01 -- 0.1
+      * initial release
+
+
+There is 3 entries in this change log, one for each released version and one
+for the next version (i.e. the current entry).
+Each entry contains a set of messages corresponding to changes done in this
+release.
+All the non empty lines before the first entry are considered as the change
+log title.
+"""
+
+__docformat__ = "restructuredtext en"
+
+import sys
+from stat import S_IWRITE
+
+from six import string_types
+
+BULLET = '*'
+SUBBULLET = '-'
+INDENT = ' ' * 4
+
+class NoEntry(Exception):
+    """raised when we are unable to find an entry"""
+
+class EntryNotFound(Exception):
+    """raised when we are unable to find a given entry"""
+
+class Version(tuple):
+    """simple class to handle soft version number has a tuple while
+    correctly printing it as X.Y.Z
+    """
+    def __new__(cls, versionstr):
+        if isinstance(versionstr, string_types):
+            versionstr = versionstr.strip(' :') # XXX (syt) duh?
+            parsed = cls.parse(versionstr)
+        else:
+            parsed = versionstr
+        return tuple.__new__(cls, parsed)
+
+    @classmethod
+    def parse(cls, versionstr):
+        versionstr = versionstr.strip(' :')
+        try:
+            return [int(i) for i in versionstr.split('.')]
+        except ValueError as ex:
+            raise ValueError("invalid literal for version '%s' (%s)"%(versionstr, ex))
+
+    def __str__(self):
+        return '.'.join([str(i) for i in self])
+
+# upstream change log #########################################################
+
+class ChangeLogEntry(object):
+    """a change log entry, i.e. a set of messages associated to a version and
+    its release date
+    """
+    version_class = Version
+
+    def __init__(self, date=None, version=None, **kwargs):
+        self.__dict__.update(kwargs)
+        if version:
+            self.version = self.version_class(version)
+        else:
+            self.version = None
+        self.date = date
+        self.messages = []
+
+    def add_message(self, msg):
+        """add a new message"""
+        self.messages.append(([msg], []))
+
+    def complete_latest_message(self, msg_suite):
+        """complete the latest added message
+        """
+        if not self.messages:
+            raise ValueError('unable to complete last message as there is no previous message)')
+        if self.messages[-1][1]: # sub messages
+            self.messages[-1][1][-1].append(msg_suite)
+        else: # message
+            self.messages[-1][0].append(msg_suite)
+
+    def add_sub_message(self, sub_msg, key=None):
+        if not self.messages:
+            raise ValueError('unable to complete last message as there is no previous message)')
+        if key is None:
+            self.messages[-1][1].append([sub_msg])
+        else:
+            raise NotImplementedError("sub message to specific key are not implemented yet")
+
+    def write(self, stream=sys.stdout):
+        """write the entry to file """
+        stream.write('%s  --  %s\n' % (self.date or '', self.version or ''))
+        for msg, sub_msgs in self.messages:
+            stream.write('%s%s %s\n' % (INDENT, BULLET, msg[0]))
+            stream.write(''.join(msg[1:]))
+            if sub_msgs:
+                stream.write('\n')
+            for sub_msg in sub_msgs:
+                stream.write('%s%s %s\n' % (INDENT * 2, SUBBULLET, sub_msg[0]))
+                stream.write(''.join(sub_msg[1:]))
+            stream.write('\n')
+
+        stream.write('\n\n')
+
+class ChangeLog(object):
+    """object representation of a whole ChangeLog file"""
+
+    entry_class = ChangeLogEntry
+
+    def __init__(self, changelog_file, title=''):
+        self.file = changelog_file
+        self.title = title
+        self.additional_content = ''
+        self.entries = []
+        self.load()
+
+    def __repr__(self):
+        return '<ChangeLog %s at %s (%s entries)>' % (self.file, id(self),
+                                                      len(self.entries))
+
+    def add_entry(self, entry):
+        """add a new entry to the change log"""
+        self.entries.append(entry)
+
+    def get_entry(self, version='', create=None):
+        """ return a given changelog entry
+        if version is omitted, return the current entry
+        """
+        if not self.entries:
+            if version or not create:
+                raise NoEntry()
+            self.entries.append(self.entry_class())
+        if not version:
+            if self.entries[0].version and create is not None:
+                self.entries.insert(0, self.entry_class())
+            return self.entries[0]
+        version = self.version_class(version)
+        for entry in self.entries:
+            if entry.version == version:
+                return entry
+        raise EntryNotFound()
+
+    def add(self, msg, create=None):
+        """add a new message to the latest opened entry"""
+        entry = self.get_entry(create=create)
+        entry.add_message(msg)
+
+    def load(self):
+        """ read a logilab's ChangeLog from file """
+        try:
+            stream = open(self.file)
+        except IOError:
+            return
+        last = None
+        expect_sub = False
+        for line in stream.readlines():
+            sline = line.strip()
+            words = sline.split()
+            # if new entry
+            if len(words) == 1 and words[0] == '--':
+                expect_sub = False
+                last = self.entry_class()
+                self.add_entry(last)
+            # if old entry
+            elif len(words) == 3 and words[1] == '--':
+                expect_sub = False
+                last = self.entry_class(words[0], words[2])
+                self.add_entry(last)
+            # if title
+            elif sline and last is None:
+                self.title = '%s%s' % (self.title, line)
+            # if new entry
+            elif sline and sline[0] == BULLET:
+                expect_sub = False
+                last.add_message(sline[1:].strip())
+            # if new sub_entry
+            elif expect_sub and sline and sline[0] == SUBBULLET:
+                last.add_sub_message(sline[1:].strip())
+            # if new line for current entry
+            elif sline and last.messages:
+                last.complete_latest_message(line)
+            else:
+                expect_sub = True
+                self.additional_content += line
+        stream.close()
+
+    def format_title(self):
+        return '%s\n\n' % self.title.strip()
+
+    def save(self):
+        """write back change log"""
+        # filetutils isn't importable in appengine, so import locally
+        from logilab.common.fileutils import ensure_fs_mode
+        ensure_fs_mode(self.file, S_IWRITE)
+        self.write(open(self.file, 'w'))
+
+    def write(self, stream=sys.stdout):
+        """write changelog to stream"""
+        stream.write(self.format_title())
+        for entry in self.entries:
+            entry.write(stream)
+
diff --git a/logilab/common/clcommands.py b/logilab/common/clcommands.py
new file mode 100644
index 0000000..4778b99
--- /dev/null
+++ b/logilab/common/clcommands.py
@@ -0,0 +1,334 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Helper functions to support command line tools providing more than
+one command.
+
+e.g called as "tool command [options] args..." where <options> and <args> are
+command'specific
+"""
+
+from __future__ import print_function
+
+__docformat__ = "restructuredtext en"
+
+import sys
+import logging
+from os.path import basename
+
+from logilab.common.configuration import Configuration
+from logilab.common.logging_ext import init_log, get_threshold
+from logilab.common.deprecation import deprecated
+
+
+class BadCommandUsage(Exception):
+    """Raised when an unknown command is used or when a command is not
+    correctly used (bad options, too much / missing arguments...).
+
+    Trigger display of command usage.
+    """
+
+class CommandError(Exception):
+    """Raised when a command can't be processed and we want to display it and
+    exit, without traceback nor usage displayed.
+    """
+
+
+# command line access point ####################################################
+
+class CommandLine(dict):
+    """Usage:
+
+    >>> LDI = cli.CommandLine('ldi', doc='Logilab debian installer',
+                              version=version, rcfile=RCFILE)
+    >>> LDI.register(MyCommandClass)
+    >>> LDI.register(MyOtherCommandClass)
+    >>> LDI.run(sys.argv[1:])
+
+    Arguments:
+
+    * `pgm`, the program name, default to `basename(sys.argv[0])`
+
+    * `doc`, a short description of the command line tool
+
+    * `copyright`, additional doc string that will be appended to the generated
+      doc
+
+    * `version`, version number of string of the tool. If specified, global
+      --version option will be available.
+
+    * `rcfile`, path to a configuration file. If specified, global --C/--rc-file
+      option will be available?  self.rcfile = rcfile
+
+    * `logger`, logger to propagate to commands, default to
+      `logging.getLogger(self.pgm))`
+    """
+    def __init__(self, pgm=None, doc=None, copyright=None, version=None,
+                 rcfile=None, logthreshold=logging.ERROR,
+                 check_duplicated_command=True):
+        if pgm is None:
+            pgm = basename(sys.argv[0])
+        self.pgm = pgm
+        self.doc = doc
+        self.copyright = copyright
+        self.version = version
+        self.rcfile = rcfile
+        self.logger = None
+        self.logthreshold = logthreshold
+        self.check_duplicated_command = check_duplicated_command
+
+    def register(self, cls, force=False):
+        """register the given :class:`Command` subclass"""
+        assert not self.check_duplicated_command or force or not cls.name in self, \
+               'a command %s is already defined' % cls.name
+        self[cls.name] = cls
+        return cls
+
+    def run(self, args):
+        """main command line access point:
+        * init logging
+        * handle global options (-h/--help, --version, -C/--rc-file)
+        * check command
+        * run command
+
+        Terminate by :exc:`SystemExit`
+        """
+        init_log(debug=True, # so that we use StreamHandler
+                 logthreshold=self.logthreshold,
+                 logformat='%(levelname)s: %(message)s')
+        try:
+            arg = args.pop(0)
+        except IndexError:
+            self.usage_and_exit(1)
+        if arg in ('-h', '--help'):
+            self.usage_and_exit(0)
+        if self.version is not None and arg in ('--version'):
+            print(self.version)
+            sys.exit(0)
+        rcfile = self.rcfile
+        if rcfile is not None and arg in ('-C', '--rc-file'):
+            try:
+                rcfile = args.pop(0)
+                arg = args.pop(0)
+            except IndexError:
+                self.usage_and_exit(1)
+        try:
+            command = self.get_command(arg)
+        except KeyError:
+            print('ERROR: no %s command' % arg)
+            print()
+            self.usage_and_exit(1)
+        try:
+            sys.exit(command.main_run(args, rcfile))
+        except KeyboardInterrupt as exc:
+            print('Interrupted', end=' ')
+            if str(exc):
+                print(': %s' % exc, end=' ')
+            print()
+            sys.exit(4)
+        except BadCommandUsage as err:
+            print('ERROR:', err)
+            print()
+            print(command.help())
+            sys.exit(1)
+
+    def create_logger(self, handler, logthreshold=None):
+        logger = logging.Logger(self.pgm)
+        logger.handlers = [handler]
+        if logthreshold is None:
+            logthreshold = get_threshold(self.logthreshold)
+        logger.setLevel(logthreshold)
+        return logger
+
+    def get_command(self, cmd, logger=None):
+        if logger is None:
+            logger = self.logger
+        if logger is None:
+            logger = self.logger = logging.getLogger(self.pgm)
+            logger.setLevel(get_threshold(self.logthreshold))
+        return self[cmd](logger)
+
+    def usage(self):
+        """display usage for the main program (i.e. when no command supplied)
+        and exit
+        """
+        print('usage:', self.pgm, end=' ')
+        if self.rcfile:
+            print('[--rc-file=<configuration file>]', end=' ')
+        print('<command> [options] <command argument>...')
+        if self.doc:
+            print('\n%s' % self.doc)
+        print('''
+Type "%(pgm)s <command> --help" for more information about a specific
+command. Available commands are :\n''' % self.__dict__)
+        max_len = max([len(cmd) for cmd in self])
+        padding = ' ' * max_len
+        for cmdname, cmd in sorted(self.items()):
+            if not cmd.hidden:
+                print(' ', (cmdname + padding)[:max_len], cmd.short_description())
+        if self.rcfile:
+            print('''
+Use --rc-file=<configuration file> / -C <configuration file> before the command
+to specify a configuration file. Default to %s.
+''' % self.rcfile)
+        print('''%(pgm)s -h/--help
+      display this usage information and exit''' % self.__dict__)
+        if self.version:
+            print('''%(pgm)s -v/--version
+      display version configuration and exit''' % self.__dict__)
+        if self.copyright:
+            print('\n', self.copyright)
+
+    def usage_and_exit(self, status):
+        self.usage()
+        sys.exit(status)
+
+
+# base command classes #########################################################
+
+class Command(Configuration):
+    """Base class for command line commands.
+
+    Class attributes:
+
+    * `name`, the name of the command
+
+    * `min_args`, minimum number of arguments, None if unspecified
+
+    * `max_args`, maximum number of arguments, None if unspecified
+
+    * `arguments`, string describing arguments, used in command usage
+
+    * `hidden`, boolean flag telling if the command should be hidden, e.g. does
+      not appear in help's commands list
+
+    * `options`, options list, as allowed by :mod:configuration
+    """
+
+    arguments = ''
+    name = ''
+    # hidden from help ?
+    hidden = False
+    # max/min args, None meaning unspecified
+    min_args = None
+    max_args = None
+
+    @classmethod
+    def description(cls):
+        return cls.__doc__.replace('    ', '')
+
+    @classmethod
+    def short_description(cls):
+        return cls.description().split('.')[0]
+
+    def __init__(self, logger):
+        usage = '%%prog %s %s\n\n%s' % (self.name, self.arguments,
+                                        self.description())
+        Configuration.__init__(self, usage=usage)
+        self.logger = logger
+
+    def check_args(self, args):
+        """check command's arguments are provided"""
+        if self.min_args is not None and len(args) < self.min_args:
+            raise BadCommandUsage('missing argument')
+        if self.max_args is not None and len(args) > self.max_args:
+            raise BadCommandUsage('too many arguments')
+
+    def main_run(self, args, rcfile=None):
+        """Run the command and return status 0 if everything went fine.
+
+        If :exc:`CommandError` is raised by the underlying command, simply log
+        the error and return status 2.
+
+        Any other exceptions, including :exc:`BadCommandUsage` will be
+        propagated.
+        """
+        if rcfile:
+            self.load_file_configuration(rcfile)
+        args = self.load_command_line_configuration(args)
+        try:
+            self.check_args(args)
+            self.run(args)
+        except CommandError as err:
+            self.logger.error(err)
+            return 2
+        return 0
+
+    def run(self, args):
+        """run the command with its specific arguments"""
+        raise NotImplementedError()
+
+
+class ListCommandsCommand(Command):
+    """list available commands, useful for bash completion."""
+    name = 'listcommands'
+    arguments = '[command]'
+    hidden = True
+
+    def run(self, args):
+        """run the command with its specific arguments"""
+        if args:
+            command = args.pop()
+            cmd = _COMMANDS[command]
+            for optname, optdict in cmd.options:
+                print('--help')
+                print('--' + optname)
+        else:
+            commands = sorted(_COMMANDS.keys())
+            for command in commands:
+                cmd = _COMMANDS[command]
+                if not cmd.hidden:
+                    print(command)
+
+
+# deprecated stuff #############################################################
+
+_COMMANDS = CommandLine()
+
+DEFAULT_COPYRIGHT = '''\
+Copyright (c) 2004-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+http://www.logilab.fr/ -- mailto:contact@logilab.fr'''
+
+@deprecated('use cls.register(cli)')
+def register_commands(commands):
+    """register existing commands"""
+    for command_klass in commands:
+        _COMMANDS.register(command_klass)
+
+@deprecated('use args.pop(0)')
+def main_run(args, doc=None, copyright=None, version=None):
+    """command line tool: run command specified by argument list (without the
+    program name). Raise SystemExit with status 0 if everything went fine.
+
+    >>> main_run(sys.argv[1:])
+    """
+    _COMMANDS.doc = doc
+    _COMMANDS.copyright = copyright
+    _COMMANDS.version = version
+    _COMMANDS.run(args)
+
+@deprecated('use args.pop(0)')
+def pop_arg(args_list, expected_size_after=None, msg="Missing argument"):
+    """helper function to get and check command line arguments"""
+    try:
+        value = args_list.pop(0)
+    except IndexError:
+        raise BadCommandUsage(msg)
+    if expected_size_after is not None and len(args_list) > expected_size_after:
+        raise BadCommandUsage('too many arguments')
+    return value
+
diff --git a/logilab/common/cli.py b/logilab/common/cli.py
new file mode 100644
index 0000000..cdeef97
--- /dev/null
+++ b/logilab/common/cli.py
@@ -0,0 +1,211 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Command line interface helper classes.
+
+It provides some default commands, a help system, a default readline
+configuration with completion and persistent history.
+
+Example::
+
+    class BookShell(CLIHelper):
+
+        def __init__(self):
+            # quit and help are builtins
+            # CMD_MAP keys are commands, values are topics
+            self.CMD_MAP['pionce'] = _("Sommeil")
+            self.CMD_MAP['ronfle'] = _("Sommeil")
+            CLIHelper.__init__(self)
+
+        help_do_pionce = ("pionce", "pionce duree", _("met ton corps en veille"))
+        def do_pionce(self):
+            print('nap is good')
+
+        help_do_ronfle = ("ronfle", "ronfle volume", _("met les autres en veille"))
+        def do_ronfle(self):
+            print('fuuuuuuuuuuuu rhhhhhrhrhrrh')
+
+    cl = BookShell()
+"""
+
+from __future__ import print_function
+
+__docformat__ = "restructuredtext en"
+
+from six.moves import builtins, input
+
+if not hasattr(builtins, '_'):
+    builtins._ = str
+
+
+def init_readline(complete_method, histfile=None):
+    """Init the readline library if available."""
+    try:
+        import readline
+        readline.parse_and_bind("tab: complete")
+        readline.set_completer(complete_method)
+        string = readline.get_completer_delims().replace(':', '')
+        readline.set_completer_delims(string)
+        if histfile is not None:
+            try:
+                readline.read_history_file(histfile)
+            except IOError:
+                pass
+            import atexit
+            atexit.register(readline.write_history_file, histfile)
+    except:
+        print('readline is not available :-(')
+
+
+class Completer :
+    """Readline completer."""
+
+    def __init__(self, commands):
+        self.list = commands
+
+    def complete(self, text, state):
+        """Hook called by readline when <tab> is pressed."""
+        n = len(text)
+        matches = []
+        for cmd in self.list :
+            if cmd[:n] == text :
+                matches.append(cmd)
+        try:
+            return matches[state]
+        except IndexError:
+            return None
+
+
+class CLIHelper:
+    """An abstract command line interface client which recognize commands
+    and provide an help system.
+    """
+
+    CMD_MAP = {'help': _("Others"),
+               'quit': _("Others"),
+               }
+    CMD_PREFIX = ''
+
+    def __init__(self, histfile=None) :
+        self._topics = {}
+        self.commands = None
+        self._completer = Completer(self._register_commands())
+        init_readline(self._completer.complete, histfile)
+
+    def run(self):
+        """loop on user input, exit on EOF"""
+        while True:
+            try:
+                line = input('>>> ')
+            except EOFError:
+                print
+                break
+            s_line = line.strip()
+            if not s_line:
+                continue
+            args = s_line.split()
+            if args[0] in self.commands:
+                try:
+                    cmd = 'do_%s' % self.commands[args[0]]
+                    getattr(self, cmd)(*args[1:])
+                except EOFError:
+                    break
+                except:
+                    import traceback
+                    traceback.print_exc()
+            else:
+                try:
+                    self.handle_line(s_line)
+                except:
+                    import traceback
+                    traceback.print_exc()
+
+    def handle_line(self, stripped_line):
+        """Method to overload in the concrete class (should handle
+        lines which are not commands).
+        """
+        raise NotImplementedError()
+
+
+    # private methods #########################################################
+
+    def _register_commands(self):
+        """ register available commands method and return the list of
+        commands name
+        """
+        self.commands = {}
+        self._command_help = {}
+        commands = [attr[3:] for attr in dir(self) if attr[:3] == 'do_']
+        for command in commands:
+            topic = self.CMD_MAP[command]
+            help_method = getattr(self, 'help_do_%s' % command)
+            self._topics.setdefault(topic, []).append(help_method)
+            self.commands[self.CMD_PREFIX + command] = command
+            self._command_help[command] = help_method
+        return self.commands.keys()
+
+    def _print_help(self, cmd, syntax, explanation):
+        print(_('Command %s') % cmd)
+        print(_('Syntax: %s') % syntax)
+        print('\t', explanation)
+        print()
+
+
+    # predefined commands #####################################################
+
+    def do_help(self, command=None) :
+        """base input of the help system"""
+        if command in self._command_help:
+            self._print_help(*self._command_help[command])
+        elif command is None or command not in self._topics:
+            print(_("Use help <topic> or help <command>."))
+            print(_("Available topics are:"))
+            topics = sorted(self._topics.keys())
+            for topic in topics:
+                print('\t', topic)
+            print()
+            print(_("Available commands are:"))
+            commands = self.commands.keys()
+            commands.sort()
+            for command in commands:
+                print('\t', command[len(self.CMD_PREFIX):])
+
+        else:
+            print(_('Available commands about %s:') % command)
+            print
+            for command_help_method in self._topics[command]:
+                try:
+                    if callable(command_help_method):
+                        self._print_help(*command_help_method())
+                    else:
+                        self._print_help(*command_help_method)
+                except:
+                    import traceback
+                    traceback.print_exc()
+                    print('ERROR in help method %s'% (
+                        command_help_method.__name__))
+
+    help_do_help = ("help", "help [topic|command]",
+                    _("print help message for the given topic/command or \
+available topics when no argument"))
+
+    def do_quit(self):
+        """quit the CLI"""
+        raise EOFError()
+
+    def help_do_quit(self):
+        return ("quit", "quit", _("quit the application"))
diff --git a/logilab/common/compat.py b/logilab/common/compat.py
new file mode 100644
index 0000000..f2eb590
--- /dev/null
+++ b/logilab/common/compat.py
@@ -0,0 +1,78 @@
+# pylint: disable=E0601,W0622,W0611
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Wrappers around some builtins introduced in python 2.3, 2.4 and
+2.5, making them available in for earlier versions of python.
+
+See another compatibility snippets from other projects:
+
+    :mod:`lib2to3.fixes`
+    :mod:`coverage.backward`
+    :mod:`unittest2.compatibility`
+"""
+
+
+__docformat__ = "restructuredtext en"
+
+import os
+import sys
+import types
+from warnings import warn
+
+# not used here, but imported to preserve API
+from six.moves import builtins
+
+if sys.version_info < (3, 0):
+    str_to_bytes = str
+    def str_encode(string, encoding):
+        if isinstance(string, unicode):
+            return string.encode(encoding)
+        return str(string)
+else:
+    def str_to_bytes(string):
+        return str.encode(string)
+    # we have to ignore the encoding in py3k to be able to write a string into a
+    # TextIOWrapper or like object (which expect an unicode string)
+    def str_encode(string, encoding):
+        return str(string)
+
+# See also http://bugs.python.org/issue11776
+if sys.version_info[0] == 3:
+    def method_type(callable, instance, klass):
+        # api change. klass is no more considered
+        return types.MethodType(callable, instance)
+else:
+    # alias types otherwise
+    method_type = types.MethodType
+
+# Pythons 2 and 3 differ on where to get StringIO
+if sys.version_info < (3, 0):
+    from cStringIO import StringIO
+    FileIO = file
+    BytesIO = StringIO
+    reload = reload
+else:
+    from io import FileIO, BytesIO, StringIO
+    from imp import reload
+
+from logilab.common.deprecation import deprecated
+
+# Other projects import these from here, keep providing them for
+# backwards compat
+any = deprecated('use builtin "any"')(any)
+all = deprecated('use builtin "all"')(all)
diff --git a/logilab/common/configuration.py b/logilab/common/configuration.py
new file mode 100644
index 0000000..b292427
--- /dev/null
+++ b/logilab/common/configuration.py
@@ -0,0 +1,1105 @@
+# copyright 2003-2012 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Classes to handle advanced configuration in simple to complex applications.
+
+Allows to load the configuration from a file or from command line
+options, to generate a sample configuration file or to display
+program's usage. Fills the gap between optik/optparse and ConfigParser
+by adding data types (which are also available as a standalone optik
+extension in the `optik_ext` module).
+
+
+Quick start: simplest usage
+---------------------------
+
+.. python ::
+
+  >>> import sys
+  >>> from logilab.common.configuration import Configuration
+  >>> options = [('dothis', {'type':'yn', 'default': True, 'metavar': '<y or n>'}),
+  ...            ('value', {'type': 'string', 'metavar': '<string>'}),
+  ...            ('multiple', {'type': 'csv', 'default': ('yop',),
+  ...                          'metavar': '<comma separated values>',
+  ...                          'help': 'you can also document the option'}),
+  ...            ('number', {'type': 'int', 'default':2, 'metavar':'<int>'}),
+  ...           ]
+  >>> config = Configuration(options=options, name='My config')
+  >>> print config['dothis']
+  True
+  >>> print config['value']
+  None
+  >>> print config['multiple']
+  ('yop',)
+  >>> print config['number']
+  2
+  >>> print config.help()
+  Usage:  [options]
+
+  Options:
+    -h, --help            show this help message and exit
+    --dothis=<y or n>
+    --value=<string>
+    --multiple=<comma separated values>
+                          you can also document the option [current: none]
+    --number=<int>
+
+  >>> f = open('myconfig.ini', 'w')
+  >>> f.write('''[MY CONFIG]
+  ... number = 3
+  ... dothis = no
+  ... multiple = 1,2,3
+  ... ''')
+  >>> f.close()
+  >>> config.load_file_configuration('myconfig.ini')
+  >>> print config['dothis']
+  False
+  >>> print config['value']
+  None
+  >>> print config['multiple']
+  ['1', '2', '3']
+  >>> print config['number']
+  3
+  >>> sys.argv = ['mon prog', '--value', 'bacon', '--multiple', '4,5,6',
+  ...             'nonoptionargument']
+  >>> print config.load_command_line_configuration()
+  ['nonoptionargument']
+  >>> print config['value']
+  bacon
+  >>> config.generate_config()
+  # class for simple configurations which don't need the
+  # manager / providers model and prefer delegation to inheritance
+  #
+  # configuration values are accessible through a dict like interface
+  #
+  [MY CONFIG]
+
+  dothis=no
+
+  value=bacon
+
+  # you can also document the option
+  multiple=4,5,6
+
+  number=3
+
+  Note : starting with Python 2.7 ConfigParser is able to take into
+  account the order of occurrences of the options into a file (by
+  using an OrderedDict). If you have two options changing some common
+  state, like a 'disable-all-stuff' and a 'enable-some-stuff-a', their
+  order of appearance will be significant : the last specified in the
+  file wins. For earlier version of python and logilab.common newer
+  than 0.61 the behaviour is unspecified.
+
+"""
+
+from __future__ import print_function
+
+__docformat__ = "restructuredtext en"
+
+__all__ = ('OptionsManagerMixIn', 'OptionsProviderMixIn',
+           'ConfigurationMixIn', 'Configuration',
+           'OptionsManager2ConfigurationAdapter')
+
+import os
+import sys
+import re
+from os.path import exists, expanduser
+from copy import copy
+from warnings import warn
+
+from six import string_types
+from six.moves import range, configparser as cp, input
+
+from logilab.common.compat import str_encode as _encode
+from logilab.common.deprecation import deprecated
+from logilab.common.textutils import normalize_text, unquote
+from logilab.common import optik_ext
+
+OptionError = optik_ext.OptionError
+
+REQUIRED = []
+
+class UnsupportedAction(Exception):
+    """raised by set_option when it doesn't know what to do for an action"""
+
+
+def _get_encoding(encoding, stream):
+    encoding = encoding or getattr(stream, 'encoding', None)
+    if not encoding:
+        import locale
+        encoding = locale.getpreferredencoding()
+    return encoding
+
+
+# validation functions ########################################################
+
+# validators will return the validated value or raise optparse.OptionValueError
+# XXX add to documentation
+
+def choice_validator(optdict, name, value):
+    """validate and return a converted value for option of type 'choice'
+    """
+    if not value in optdict['choices']:
+        msg = "option %s: invalid value: %r, should be in %s"
+        raise optik_ext.OptionValueError(msg % (name, value, optdict['choices']))
+    return value
+
+def multiple_choice_validator(optdict, name, value):
+    """validate and return a converted value for option of type 'choice'
+    """
+    choices = optdict['choices']
+    values = optik_ext.check_csv(None, name, value)
+    for value in values:
+        if not value in choices:
+            msg = "option %s: invalid value: %r, should be in %s"
+            raise optik_ext.OptionValueError(msg % (name, value, choices))
+    return values
+
+def csv_validator(optdict, name, value):
+    """validate and return a converted value for option of type 'csv'
+    """
+    return optik_ext.check_csv(None, name, value)
+
+def yn_validator(optdict, name, value):
+    """validate and return a converted value for option of type 'yn'
+    """
+    return optik_ext.check_yn(None, name, value)
+
+def named_validator(optdict, name, value):
+    """validate and return a converted value for option of type 'named'
+    """
+    return optik_ext.check_named(None, name, value)
+
+def file_validator(optdict, name, value):
+    """validate and return a filepath for option of type 'file'"""
+    return optik_ext.check_file(None, name, value)
+
+def color_validator(optdict, name, value):
+    """validate and return a valid color for option of type 'color'"""
+    return optik_ext.check_color(None, name, value)
+
+def password_validator(optdict, name, value):
+    """validate and return a string for option of type 'password'"""
+    return optik_ext.check_password(None, name, value)
+
+def date_validator(optdict, name, value):
+    """validate and return a mx DateTime object for option of type 'date'"""
+    return optik_ext.check_date(None, name, value)
+
+def time_validator(optdict, name, value):
+    """validate and return a time object for option of type 'time'"""
+    return optik_ext.check_time(None, name, value)
+
+def bytes_validator(optdict, name, value):
+    """validate and return an integer for option of type 'bytes'"""
+    return optik_ext.check_bytes(None, name, value)
+
+
+VALIDATORS = {'string': unquote,
+              'int': int,
+              'float': float,
+              'file': file_validator,
+              'font': unquote,
+              'color': color_validator,
+              'regexp': re.compile,
+              'csv': csv_validator,
+              'yn': yn_validator,
+              'bool': yn_validator,
+              'named': named_validator,
+              'password': password_validator,
+              'date': date_validator,
+              'time': time_validator,
+              'bytes': bytes_validator,
+              'choice': choice_validator,
+              'multiple_choice': multiple_choice_validator,
+              }
+
+def _call_validator(opttype, optdict, option, value):
+    if opttype not in VALIDATORS:
+        raise Exception('Unsupported type "%s"' % opttype)
+    try:
+        return VALIDATORS[opttype](optdict, option, value)
+    except TypeError:
+        try:
+            return VALIDATORS[opttype](value)
+        except optik_ext.OptionValueError:
+            raise
+        except:
+            raise optik_ext.OptionValueError('%s value (%r) should be of type %s' %
+                                   (option, value, opttype))
+
+# user input functions ########################################################
+
+# user input functions will ask the user for input on stdin then validate
+# the result and return the validated value or raise optparse.OptionValueError
+# XXX add to documentation
+
+def input_password(optdict, question='password:'):
+    from getpass import getpass
+    while True:
+        value = getpass(question)
+        value2 = getpass('confirm: ')
+        if value == value2:
+            return value
+        print('password mismatch, try again')
+
+def input_string(optdict, question):
+    value = input(question).strip()
+    return value or None
+
+def _make_input_function(opttype):
+    def input_validator(optdict, question):
+        while True:
+            value = input(question)
+            if not value.strip():
+                return None
+            try:
+                return _call_validator(opttype, optdict, None, value)
+            except optik_ext.OptionValueError as ex:
+                msg = str(ex).split(':', 1)[-1].strip()
+                print('bad value: %s' % msg)
+    return input_validator
+
+INPUT_FUNCTIONS = {
+    'string': input_string,
+    'password': input_password,
+    }
+
+for opttype in VALIDATORS.keys():
+    INPUT_FUNCTIONS.setdefault(opttype, _make_input_function(opttype))
+
+# utility functions ############################################################
+
+def expand_default(self, option):
+    """monkey patch OptionParser.expand_default since we have a particular
+    way to handle defaults to avoid overriding values in the configuration
+    file
+    """
+    if self.parser is None or not self.default_tag:
+        return option.help
+    optname = option._long_opts[0][2:]
+    try:
+        provider = self.parser.options_manager._all_options[optname]
+    except KeyError:
+        value = None
+    else:
+        optdict = provider.get_option_def(optname)
+        optname = provider.option_attrname(optname, optdict)
+        value = getattr(provider.config, optname, optdict)
+        value = format_option_value(optdict, value)
+    if value is optik_ext.NO_DEFAULT or not value:
+        value = self.NO_DEFAULT_VALUE
+    return option.help.replace(self.default_tag, str(value))
+
+
+def _validate(value, optdict, name=''):
+    """return a validated value for an option according to its type
+
+    optional argument name is only used for error message formatting
+    """
+    try:
+        _type = optdict['type']
+    except KeyError:
+        # FIXME
+        return value
+    return _call_validator(_type, optdict, name, value)
+convert = deprecated('[0.60] convert() was renamed _validate()')(_validate)
+
+# format and output functions ##################################################
+
+def comment(string):
+    """return string as a comment"""
+    lines = [line.strip() for line in string.splitlines()]
+    return '# ' + ('%s# ' % os.linesep).join(lines)
+
+def format_time(value):
+    if not value:
+        return '0'
+    if value != int(value):
+        return '%.2fs' % value
+    value = int(value)
+    nbmin, nbsec = divmod(value, 60)
+    if nbsec:
+        return '%ss' % value
+    nbhour, nbmin_ = divmod(nbmin, 60)
+    if nbmin_:
+        return '%smin' % nbmin
+    nbday, nbhour_ = divmod(nbhour, 24)
+    if nbhour_:
+        return '%sh' % nbhour
+    return '%sd' % nbday
+
+def format_bytes(value):
+    if not value:
+        return '0'
+    if value != int(value):
+        return '%.2fB' % value
+    value = int(value)
+    prevunit = 'B'
+    for unit in ('KB', 'MB', 'GB', 'TB'):
+        next, remain = divmod(value, 1024)
+        if remain:
+            return '%s%s' % (value, prevunit)
+        prevunit = unit
+        value = next
+    return '%s%s' % (value, unit)
+
+def format_option_value(optdict, value):
+    """return the user input's value from a 'compiled' value"""
+    if isinstance(value, (list, tuple)):
+        value = ','.join(value)
+    elif isinstance(value, dict):
+        value = ','.join(['%s:%s' % (k, v) for k, v in value.items()])
+    elif hasattr(value, 'match'): # optdict.get('type') == 'regexp'
+        # compiled regexp
+        value = value.pattern
+    elif optdict.get('type') == 'yn':
+        value = value and 'yes' or 'no'
+    elif isinstance(value, string_types) and value.isspace():
+        value = "'%s'" % value
+    elif optdict.get('type') == 'time' and isinstance(value, (float, int, long)):
+        value = format_time(value)
+    elif optdict.get('type') == 'bytes' and hasattr(value, '__int__'):
+        value = format_bytes(value)
+    return value
+
+def ini_format_section(stream, section, options, encoding=None, doc=None):
+    """format an options section using the INI format"""
+    encoding = _get_encoding(encoding, stream)
+    if doc:
+        print(_encode(comment(doc), encoding), file=stream)
+    print('[%s]' % section, file=stream)
+    ini_format(stream, options, encoding)
+
+def ini_format(stream, options, encoding):
+    """format options using the INI format"""
+    for optname, optdict, value in options:
+        value = format_option_value(optdict, value)
+        help = optdict.get('help')
+        if help:
+            help = normalize_text(help, line_len=79, indent='# ')
+            print(file=stream)
+            print(_encode(help, encoding), file=stream)
+        else:
+            print(file=stream)
+        if value is None:
+            print('#%s=' % optname, file=stream)
+        else:
+            value = _encode(value, encoding).strip()
+            print('%s=%s' % (optname, value), file=stream)
+
+format_section = ini_format_section
+
+def rest_format_section(stream, section, options, encoding=None, doc=None):
+    """format an options section using as ReST formatted output"""
+    encoding = _get_encoding(encoding, stream)
+    if section:
+        print('%s\n%s' % (section, "'"*len(section)), file=stream)
+    if doc:
+        print(_encode(normalize_text(doc, line_len=79, indent=''), encoding), file=stream)
+        print(file=stream)
+    for optname, optdict, value in options:
+        help = optdict.get('help')
+        print(':%s:' % optname, file=stream)
+        if help:
+            help = normalize_text(help, line_len=79, indent='  ')
+            print(_encode(help, encoding), file=stream)
+        if value:
+            value = _encode(format_option_value(optdict, value), encoding)
+            print(file=stream)
+            print('  Default: ``%s``' % value.replace("`` ", "```` ``"), file=stream)
+
+# Options Manager ##############################################################
+
+class OptionsManagerMixIn(object):
+    """MixIn to handle a configuration from both a configuration file and
+    command line options
+    """
+
+    def __init__(self, usage, config_file=None, version=None, quiet=0):
+        self.config_file = config_file
+        self.reset_parsers(usage, version=version)
+        # list of registered options providers
+        self.options_providers = []
+        # dictionary associating option name to checker
+        self._all_options = {}
+        self._short_options = {}
+        self._nocallback_options = {}
+        self._mygroups = dict()
+        # verbosity
+        self.quiet = quiet
+        self._maxlevel = 0
+
+    def reset_parsers(self, usage='', version=None):
+        # configuration file parser
+        self.cfgfile_parser = cp.ConfigParser()
+        # command line parser
+        self.cmdline_parser = optik_ext.OptionParser(usage=usage, version=version)
+        self.cmdline_parser.options_manager = self
+        self._optik_option_attrs = set(self.cmdline_parser.option_class.ATTRS)
+
+    def register_options_provider(self, provider, own_group=True):
+        """register an options provider"""
+        assert provider.priority <= 0, "provider's priority can't be >= 0"
+        for i in range(len(self.options_providers)):
+            if provider.priority > self.options_providers[i].priority:
+                self.options_providers.insert(i, provider)
+                break
+        else:
+            self.options_providers.append(provider)
+        non_group_spec_options = [option for option in provider.options
+                                  if 'group' not in option[1]]
+        groups = getattr(provider, 'option_groups', ())
+        if own_group and non_group_spec_options:
+            self.add_option_group(provider.name.upper(), provider.__doc__,
+                                  non_group_spec_options, provider)
+        else:
+            for opt, optdict in non_group_spec_options:
+                self.add_optik_option(provider, self.cmdline_parser, opt, optdict)
+        for gname, gdoc in groups:
+            gname = gname.upper()
+            goptions = [option for option in provider.options
+                        if option[1].get('group', '').upper() == gname]
+            self.add_option_group(gname, gdoc, goptions, provider)
+
+    def add_option_group(self, group_name, doc, options, provider):
+        """add an option group including the listed options
+        """
+        assert options
+        # add option group to the command line parser
+        if group_name in self._mygroups:
+            group = self._mygroups[group_name]
+        else:
+            group = optik_ext.OptionGroup(self.cmdline_parser,
+                                         title=group_name.capitalize())
+            self.cmdline_parser.add_option_group(group)
+            group.level = provider.level
+            self._mygroups[group_name] = group
+            # add section to the config file
+            if group_name != "DEFAULT":
+                self.cfgfile_parser.add_section(group_name)
+        # add provider's specific options
+        for opt, optdict in options:
+            self.add_optik_option(provider, group, opt, optdict)
+
+    def add_optik_option(self, provider, optikcontainer, opt, optdict):
+        if 'inputlevel' in optdict:
+            warn('[0.50] "inputlevel" in option dictionary for %s is deprecated,'
+                 ' use "level"' % opt, DeprecationWarning)
+            optdict['level'] = optdict.pop('inputlevel')
+        args, optdict = self.optik_option(provider, opt, optdict)
+        option = optikcontainer.add_option(*args, **optdict)
+        self._all_options[opt] = provider
+        self._maxlevel = max(self._maxlevel, option.level or 0)
+
+    def optik_option(self, provider, opt, optdict):
+        """get our personal option definition and return a suitable form for
+        use with optik/optparse
+        """
+        optdict = copy(optdict)
+        others = {}
+        if 'action' in optdict:
+            self._nocallback_options[provider] = opt
+        else:
+            optdict['action'] = 'callback'
+            optdict['callback'] = self.cb_set_provider_option
+        # default is handled here and *must not* be given to optik if you
+        # want the whole machinery to work
+        if 'default' in optdict:
+            if ('help' in optdict
+                and optdict.get('default') is not None
+                and not optdict['action'] in ('store_true', 'store_false')):
+                optdict['help'] += ' [current: %default]'
+            del optdict['default']
+        args = ['--' + str(opt)]
+        if 'short' in optdict:
+            self._short_options[optdict['short']] = opt
+            args.append('-' + optdict['short'])
+            del optdict['short']
+        # cleanup option definition dict before giving it to optik
+        for key in list(optdict.keys()):
+            if not key in self._optik_option_attrs:
+                optdict.pop(key)
+        return args, optdict
+
+    def cb_set_provider_option(self, option, opt, value, parser):
+        """optik callback for option setting"""
+        if opt.startswith('--'):
+            # remove -- on long option
+            opt = opt[2:]
+        else:
+            # short option, get its long equivalent
+            opt = self._short_options[opt[1:]]
+        # trick since we can't set action='store_true' on options
+        if value is None:
+            value = 1
+        self.global_set_option(opt, value)
+
+    def global_set_option(self, opt, value):
+        """set option on the correct option provider"""
+        self._all_options[opt].set_option(opt, value)
+
+    def generate_config(self, stream=None, skipsections=(), encoding=None):
+        """write a configuration file according to the current configuration
+        into the given stream or stdout
+        """
+        options_by_section = {}
+        sections = []
+        for provider in self.options_providers:
+            for section, options in provider.options_by_section():
+                if section is None:
+                    section = provider.name
+                if section in skipsections:
+                    continue
+                options = [(n, d, v) for (n, d, v) in options
+                           if d.get('type') is not None]
+                if not options:
+                    continue
+                if not section in sections:
+                    sections.append(section)
+                alloptions = options_by_section.setdefault(section, [])
+                alloptions += options
+        stream = stream or sys.stdout
+        encoding = _get_encoding(encoding, stream)
+        printed = False
+        for section in sections:
+            if printed:
+                print('\n', file=stream)
+            format_section(stream, section.upper(), options_by_section[section],
+                           encoding)
+            printed = True
+
+    def generate_manpage(self, pkginfo, section=1, stream=None):
+        """write a man page for the current configuration into the given
+        stream or stdout
+        """
+        self._monkeypatch_expand_default()
+        try:
+            optik_ext.generate_manpage(self.cmdline_parser, pkginfo,
+                                      section, stream=stream or sys.stdout,
+                                      level=self._maxlevel)
+        finally:
+            self._unmonkeypatch_expand_default()
+
+    # initialization methods ##################################################
+
+    def load_provider_defaults(self):
+        """initialize configuration using default values"""
+        for provider in self.options_providers:
+            provider.load_defaults()
+
+    def load_file_configuration(self, config_file=None):
+        """load the configuration from file"""
+        self.read_config_file(config_file)
+        self.load_config_file()
+
+    def read_config_file(self, config_file=None):
+        """read the configuration file but do not load it (i.e. dispatching
+        values to each options provider)
+        """
+        helplevel = 1
+        while helplevel <= self._maxlevel:
+            opt = '-'.join(['long'] * helplevel) + '-help'
+            if opt in self._all_options:
+                break # already processed
+            def helpfunc(option, opt, val, p, level=helplevel):
+                print(self.help(level))
+                sys.exit(0)
+            helpmsg = '%s verbose help.' % ' '.join(['more'] * helplevel)
+            optdict = {'action' : 'callback', 'callback' : helpfunc,
+                       'help' : helpmsg}
+            provider = self.options_providers[0]
+            self.add_optik_option(provider, self.cmdline_parser, opt, optdict)
+            provider.options += ( (opt, optdict), )
+            helplevel += 1
+        if config_file is None:
+            config_file = self.config_file
+        if config_file is not None:
+            config_file = expanduser(config_file)
+        if config_file and exists(config_file):
+            parser = self.cfgfile_parser
+            parser.read([config_file])
+            # normalize sections'title
+            for sect, values in parser._sections.items():
+                if not sect.isupper() and values:
+                    parser._sections[sect.upper()] = values
+        elif not self.quiet:
+            msg = 'No config file found, using default configuration'
+            print(msg, file=sys.stderr)
+            return
+
+    def input_config(self, onlysection=None, inputlevel=0, stream=None):
+        """interactively get configuration values by asking to the user and generate
+        a configuration file
+        """
+        if onlysection is not None:
+            onlysection = onlysection.upper()
+        for provider in self.options_providers:
+            for section, option, optdict in provider.all_options():
+                if onlysection is not None and section != onlysection:
+                    continue
+                if not 'type' in optdict:
+                    # ignore action without type (callback, store_true...)
+                    continue
+                provider.input_option(option, optdict, inputlevel)
+        # now we can generate the configuration file
+        if stream is not None:
+            self.generate_config(stream)
+
+    def load_config_file(self):
+        """dispatch values previously read from a configuration file to each
+        options provider)
+        """
+        parser = self.cfgfile_parser
+        for section in parser.sections():
+             for option, value in parser.items(section):
+                  try:
+                       self.global_set_option(option, value)
+                  except (KeyError, OptionError):
+                       # TODO handle here undeclared options appearing in the config file
+                       continue
+
+    def load_configuration(self, **kwargs):
+        """override configuration according to given parameters
+        """
+        for opt, opt_value in kwargs.items():
+            opt = opt.replace('_', '-')
+            provider = self._all_options[opt]
+            provider.set_option(opt, opt_value)
+
+    def load_command_line_configuration(self, args=None):
+        """override configuration according to command line parameters
+
+        return additional arguments
+        """
+        self._monkeypatch_expand_default()
+        try:
+            if args is None:
+                args = sys.argv[1:]
+            else:
+                args = list(args)
+            (options, args) = self.cmdline_parser.parse_args(args=args)
+            for provider in self._nocallback_options.keys():
+                config = provider.config
+                for attr in config.__dict__.keys():
+                    value = getattr(options, attr, None)
+                    if value is None:
+                        continue
+                    setattr(config, attr, value)
+            return args
+        finally:
+            self._unmonkeypatch_expand_default()
+
+
+    # help methods ############################################################
+
+    def add_help_section(self, title, description, level=0):
+        """add a dummy option section for help purpose """
+        group = optik_ext.OptionGroup(self.cmdline_parser,
+                                     title=title.capitalize(),
+                                     description=description)
+        group.level = level
+        self._maxlevel = max(self._maxlevel, level)
+        self.cmdline_parser.add_option_group(group)
+
+    def _monkeypatch_expand_default(self):
+        # monkey patch optik_ext to deal with our default values
+        try:
+            self.__expand_default_backup = optik_ext.HelpFormatter.expand_default
+            optik_ext.HelpFormatter.expand_default = expand_default
+        except AttributeError:
+            # python < 2.4: nothing to be done
+            pass
+    def _unmonkeypatch_expand_default(self):
+        # remove monkey patch
+        if hasattr(optik_ext.HelpFormatter, 'expand_default'):
+            # unpatch optik_ext to avoid side effects
+            optik_ext.HelpFormatter.expand_default = self.__expand_default_backup
+
+    def help(self, level=0):
+        """return the usage string for available options """
+        self.cmdline_parser.formatter.output_level = level
+        self._monkeypatch_expand_default()
+        try:
+            return self.cmdline_parser.format_help()
+        finally:
+            self._unmonkeypatch_expand_default()
+
+
+class Method(object):
+    """used to ease late binding of default method (so you can define options
+    on the class using default methods on the configuration instance)
+    """
+    def __init__(self, methname):
+        self.method = methname
+        self._inst = None
+
+    def bind(self, instance):
+        """bind the method to its instance"""
+        if self._inst is None:
+            self._inst = instance
+
+    def __call__(self, *args, **kwargs):
+        assert self._inst, 'unbound method'
+        return getattr(self._inst, self.method)(*args, **kwargs)
+
+# Options Provider #############################################################
+
+class OptionsProviderMixIn(object):
+    """Mixin to provide options to an OptionsManager"""
+
+    # those attributes should be overridden
+    priority = -1
+    name = 'default'
+    options = ()
+    level = 0
+
+    def __init__(self):
+        self.config = optik_ext.Values()
+        for option in self.options:
+            try:
+                option, optdict = option
+            except ValueError:
+                raise Exception('Bad option: %r' % option)
+            if isinstance(optdict.get('default'), Method):
+                optdict['default'].bind(self)
+            elif isinstance(optdict.get('callback'), Method):
+                optdict['callback'].bind(self)
+        self.load_defaults()
+
+    def load_defaults(self):
+        """initialize the provider using default values"""
+        for opt, optdict in self.options:
+            action = optdict.get('action')
+            if action != 'callback':
+                # callback action have no default
+                default = self.option_default(opt, optdict)
+                if default is REQUIRED:
+                    continue
+                self.set_option(opt, default, action, optdict)
+
+    def option_default(self, opt, optdict=None):
+        """return the default value for an option"""
+        if optdict is None:
+            optdict = self.get_option_def(opt)
+        default = optdict.get('default')
+        if callable(default):
+            default = default()
+        return default
+
+    def option_attrname(self, opt, optdict=None):
+        """get the config attribute corresponding to opt
+        """
+        if optdict is None:
+            optdict = self.get_option_def(opt)
+        return optdict.get('dest', opt.replace('-', '_'))
+    option_name = deprecated('[0.60] OptionsProviderMixIn.option_name() was renamed to option_attrname()')(option_attrname)
+
+    def option_value(self, opt):
+        """get the current value for the given option"""
+        return getattr(self.config, self.option_attrname(opt), None)
+
+    def set_option(self, opt, value, action=None, optdict=None):
+        """method called to set an option (registered in the options list)
+        """
+        if optdict is None:
+            optdict = self.get_option_def(opt)
+        if value is not None:
+            value = _validate(value, optdict, opt)
+        if action is None:
+            action = optdict.get('action', 'store')
+        if optdict.get('type') == 'named': # XXX need specific handling
+            optname = self.option_attrname(opt, optdict)
+            currentvalue = getattr(self.config, optname, None)
+            if currentvalue:
+                currentvalue.update(value)
+                value = currentvalue
+        if action == 'store':
+            setattr(self.config, self.option_attrname(opt, optdict), value)
+        elif action in ('store_true', 'count'):
+            setattr(self.config, self.option_attrname(opt, optdict), 0)
+        elif action == 'store_false':
+            setattr(self.config, self.option_attrname(opt, optdict), 1)
+        elif action == 'append':
+            opt = self.option_attrname(opt, optdict)
+            _list = getattr(self.config, opt, None)
+            if _list is None:
+                if isinstance(value, (list, tuple)):
+                    _list = value
+                elif value is not None:
+                    _list = []
+                    _list.append(value)
+                setattr(self.config, opt, _list)
+            elif isinstance(_list, tuple):
+                setattr(self.config, opt, _list + (value,))
+            else:
+                _list.append(value)
+        elif action == 'callback':
+            optdict['callback'](None, opt, value, None)
+        else:
+            raise UnsupportedAction(action)
+
+    def input_option(self, option, optdict, inputlevel=99):
+        default = self.option_default(option, optdict)
+        if default is REQUIRED:
+            defaultstr = '(required): '
+        elif optdict.get('level', 0) > inputlevel:
+            return
+        elif optdict['type'] == 'password' or default is None:
+            defaultstr = ': '
+        else:
+            defaultstr = '(default: %s): ' % format_option_value(optdict, default)
+        print(':%s:' % option)
+        print(optdict.get('help') or option)
+        inputfunc = INPUT_FUNCTIONS[optdict['type']]
+        value = inputfunc(optdict, defaultstr)
+        while default is REQUIRED and not value:
+            print('please specify a value')
+            value = inputfunc(optdict, '%s: ' % option)
+        if value is None and default is not None:
+            value = default
+        self.set_option(option, value, optdict=optdict)
+
+    def get_option_def(self, opt):
+        """return the dictionary defining an option given it's name"""
+        assert self.options
+        for option in self.options:
+            if option[0] == opt:
+                return option[1]
+        raise OptionError('no such option %s in section %r'
+                          % (opt, self.name), opt)
+
+
+    def all_options(self):
+        """return an iterator on available options for this provider
+        option are actually described by a 3-uple:
+        (section, option name, option dictionary)
+        """
+        for section, options in self.options_by_section():
+            if section is None:
+                if self.name is None:
+                    continue
+                section = self.name.upper()
+            for option, optiondict, value in options:
+                yield section, option, optiondict
+
+    def options_by_section(self):
+        """return an iterator on options grouped by section
+
+        (section, [list of (optname, optdict, optvalue)])
+        """
+        sections = {}
+        for optname, optdict in self.options:
+            sections.setdefault(optdict.get('group'), []).append(
+                (optname, optdict, self.option_value(optname)))
+        if None in sections:
+            yield None, sections.pop(None)
+        for section, options in sections.items():
+            yield section.upper(), options
+
+    def options_and_values(self, options=None):
+        if options is None:
+            options = self.options
+        for optname, optdict in options:
+            yield (optname, optdict, self.option_value(optname))
+
+# configuration ################################################################
+
+class ConfigurationMixIn(OptionsManagerMixIn, OptionsProviderMixIn):
+    """basic mixin for simple configurations which don't need the
+    manager / providers model
+    """
+    def __init__(self, *args, **kwargs):
+        if not args:
+            kwargs.setdefault('usage', '')
+        kwargs.setdefault('quiet', 1)
+        OptionsManagerMixIn.__init__(self, *args, **kwargs)
+        OptionsProviderMixIn.__init__(self)
+        if not getattr(self, 'option_groups', None):
+            self.option_groups = []
+            for option, optdict in self.options:
+                try:
+                    gdef = (optdict['group'].upper(), '')
+                except KeyError:
+                    continue
+                if not gdef in self.option_groups:
+                    self.option_groups.append(gdef)
+        self.register_options_provider(self, own_group=False)
+
+    def register_options(self, options):
+        """add some options to the configuration"""
+        options_by_group = {}
+        for optname, optdict in options:
+            options_by_group.setdefault(optdict.get('group', self.name.upper()), []).append((optname, optdict))
+        for group, options in options_by_group.items():
+            self.add_option_group(group, None, options, self)
+        self.options += tuple(options)
+
+    def load_defaults(self):
+        OptionsProviderMixIn.load_defaults(self)
+
+    def __iter__(self):
+        return iter(self.config.__dict__.iteritems())
+
+    def __getitem__(self, key):
+        try:
+            return getattr(self.config, self.option_attrname(key))
+        except (optik_ext.OptionValueError, AttributeError):
+            raise KeyError(key)
+
+    def __setitem__(self, key, value):
+        self.set_option(key, value)
+
+    def get(self, key, default=None):
+        try:
+            return getattr(self.config, self.option_attrname(key))
+        except (OptionError, AttributeError):
+            return default
+
+
+class Configuration(ConfigurationMixIn):
+    """class for simple configurations which don't need the
+    manager / providers model and prefer delegation to inheritance
+
+    configuration values are accessible through a dict like interface
+    """
+
+    def __init__(self, config_file=None, options=None, name=None,
+                 usage=None, doc=None, version=None):
+        if options is not None:
+            self.options = options
+        if name is not None:
+            self.name = name
+        if doc is not None:
+            self.__doc__ = doc
+        super(Configuration, self).__init__(config_file=config_file, usage=usage, version=version)
+
+
+class OptionsManager2ConfigurationAdapter(object):
+    """Adapt an option manager to behave like a
+    `logilab.common.configuration.Configuration` instance
+    """
+    def __init__(self, provider):
+        self.config = provider
+
+    def __getattr__(self, key):
+        return getattr(self.config, key)
+
+    def __getitem__(self, key):
+        provider = self.config._all_options[key]
+        try:
+            return getattr(provider.config, provider.option_attrname(key))
+        except AttributeError:
+            raise KeyError(key)
+
+    def __setitem__(self, key, value):
+        self.config.global_set_option(self.config.option_attrname(key), value)
+
+    def get(self, key, default=None):
+        provider = self.config._all_options[key]
+        try:
+            return getattr(provider.config, provider.option_attrname(key))
+        except AttributeError:
+            return default
+
+# other functions ##############################################################
+
+def read_old_config(newconfig, changes, configfile):
+    """initialize newconfig from a deprecated configuration file
+
+    possible changes:
+    * ('renamed', oldname, newname)
+    * ('moved', option, oldgroup, newgroup)
+    * ('typechanged', option, oldtype, newvalue)
+    """
+    # build an index of changes
+    changesindex = {}
+    for action in changes:
+        if action[0] == 'moved':
+            option, oldgroup, newgroup = action[1:]
+            changesindex.setdefault(option, []).append((action[0], oldgroup, newgroup))
+            continue
+        if action[0] == 'renamed':
+            oldname, newname = action[1:]
+            changesindex.setdefault(newname, []).append((action[0], oldname))
+            continue
+        if action[0] == 'typechanged':
+            option, oldtype, newvalue = action[1:]
+            changesindex.setdefault(option, []).append((action[0], oldtype, newvalue))
+            continue
+        if action[1] in ('added', 'removed'):
+            continue # nothing to do here
+        raise Exception('unknown change %s' % action[0])
+    # build a config object able to read the old config
+    options = []
+    for optname, optdef in newconfig.options:
+        for action in changesindex.pop(optname, ()):
+            if action[0] == 'moved':
+                oldgroup, newgroup = action[1:]
+                optdef = optdef.copy()
+                optdef['group'] = oldgroup
+            elif action[0] == 'renamed':
+                optname = action[1]
+            elif action[0] == 'typechanged':
+                oldtype = action[1]
+                optdef = optdef.copy()
+                optdef['type'] = oldtype
+        options.append((optname, optdef))
+    if changesindex:
+        raise Exception('unapplied changes: %s' % changesindex)
+    oldconfig = Configuration(options=options, name=newconfig.name)
+    # read the old config
+    oldconfig.load_file_configuration(configfile)
+    # apply values reverting changes
+    changes.reverse()
+    done = set()
+    for action in changes:
+        if action[0] == 'renamed':
+            oldname, newname = action[1:]
+            newconfig[newname] = oldconfig[oldname]
+            done.add(newname)
+        elif action[0] == 'typechanged':
+            optname, oldtype, newvalue = action[1:]
+            newconfig[optname] = newvalue
+            done.add(optname)
+    for optname, optdef in newconfig.options:
+        if optdef.get('type') and not optname in done:
+            newconfig.set_option(optname, oldconfig[optname], optdict=optdef)
+
+
+def merge_options(options, optgroup=None):
+    """preprocess a list of options and remove duplicates, returning a new list
+    (tuple actually) of options.
+
+    Options dictionaries are copied to avoid later side-effect. Also, if
+    `otpgroup` argument is specified, ensure all options are in the given group.
+    """
+    alloptions = {}
+    options = list(options)
+    for i in range(len(options)-1, -1, -1):
+        optname, optdict = options[i]
+        if optname in alloptions:
+            options.pop(i)
+            alloptions[optname].update(optdict)
+        else:
+            optdict = optdict.copy()
+            options[i] = (optname, optdict)
+            alloptions[optname] = optdict
+        if optgroup is not None:
+            alloptions[optname]['group'] = optgroup
+    return tuple(options)
diff --git a/logilab/common/contexts.py b/logilab/common/contexts.py
new file mode 100644
index 0000000..d78c327
--- /dev/null
+++ b/logilab/common/contexts.py
@@ -0,0 +1,5 @@
+from warnings import warn
+warn('logilab.common.contexts module is deprecated, use logilab.common.shellutils instead',
+     DeprecationWarning, stacklevel=1)
+
+from logilab.common.shellutils import tempfile, pushd
diff --git a/logilab/common/corbautils.py b/logilab/common/corbautils.py
new file mode 100644
index 0000000..65c301d
--- /dev/null
+++ b/logilab/common/corbautils.py
@@ -0,0 +1,117 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""A set of utility function to ease the use of OmniORBpy.
+
+
+
+
+"""
+__docformat__ = "restructuredtext en"
+
+from omniORB import CORBA, PortableServer
+import CosNaming
+
+orb = None
+
+def get_orb():
+    """
+    returns a reference to the ORB.
+    The first call to the method initialized the ORB
+    This method is mainly used internally in the module.
+    """
+
+    global orb
+    if orb is None:
+        import sys
+        orb = CORBA.ORB_init(sys.argv, CORBA.ORB_ID)
+    return orb
+
+def get_root_context():
+    """
+    returns a reference to the NameService object.
+    This method is mainly used internally in the module.
+    """
+
+    orb = get_orb()
+    nss = orb.resolve_initial_references("NameService")
+    rootContext = nss._narrow(CosNaming.NamingContext)
+    assert rootContext is not None, "Failed to narrow root naming context"
+    return rootContext
+
+def register_object_name(object, namepath):
+    """
+    Registers a object in the NamingService.
+    The name path is a list of 2-uples (id,kind) giving the path.
+
+    For instance if the path of an object is [('foo',''),('bar','')],
+    it is possible to get a reference to the object using the URL
+    'corbaname::hostname#foo/bar'.
+    [('logilab','rootmodule'),('chatbot','application'),('chatter','server')]
+    is mapped to
+    'corbaname::hostname#logilab.rootmodule/chatbot.application/chatter.server'
+
+    The get_object_reference() function can be used to resolve such a URL.
+    """
+    context = get_root_context()
+    for id, kind in namepath[:-1]:
+        name = [CosNaming.NameComponent(id, kind)]
+        try:
+            context = context.bind_new_context(name)
+        except CosNaming.NamingContext.AlreadyBound as ex:
+            context = context.resolve(name)._narrow(CosNaming.NamingContext)
+            assert context is not None, \
+                   'test context exists but is not a NamingContext'
+
+    id, kind = namepath[-1]
+    name = [CosNaming.NameComponent(id, kind)]
+    try:
+        context.bind(name, object._this())
+    except CosNaming.NamingContext.AlreadyBound as ex:
+        context.rebind(name, object._this())
+
+def activate_POA():
+    """
+    This methods activates the Portable Object Adapter.
+    You need to call it to enable the reception of messages in your code,
+    on both the client and the server.
+    """
+    orb = get_orb()
+    poa = orb.resolve_initial_references('RootPOA')
+    poaManager = poa._get_the_POAManager()
+    poaManager.activate()
+
+def run_orb():
+    """
+    Enters the ORB mainloop on the server.
+    You should not call this method on the client.
+    """
+    get_orb().run()
+
+def get_object_reference(url):
+    """
+    Resolves a corbaname URL to an object proxy.
+    See register_object_name() for examples URLs
+    """
+    return get_orb().string_to_object(url)
+
+def get_object_string(host, namepath):
+    """given an host name and a name path as described in register_object_name,
+    return a corba string identifier
+    """
+    strname = '/'.join(['.'.join(path_elt) for path_elt in namepath])
+    return 'corbaname::%s#%s' % (host, strname)
diff --git a/logilab/common/daemon.py b/logilab/common/daemon.py
new file mode 100644
index 0000000..40319a4
--- /dev/null
+++ b/logilab/common/daemon.py
@@ -0,0 +1,101 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""A daemonize function (for Unices)"""
+
+__docformat__ = "restructuredtext en"
+
+import os
+import errno
+import signal
+import sys
+import time
+import warnings
+
+from six.moves import range
+
+def setugid(user):
+    """Change process user and group ID
+
+    Argument is a numeric user id or a user name"""
+    try:
+        from pwd import getpwuid
+        passwd = getpwuid(int(user))
+    except ValueError:
+        from pwd import getpwnam
+        passwd = getpwnam(user)
+
+    if hasattr(os, 'initgroups'): # python >= 2.7
+        os.initgroups(passwd.pw_name, passwd.pw_gid)
+    else:
+        import ctypes
+        if ctypes.CDLL(None).initgroups(passwd.pw_name, passwd.pw_gid) < 0:
+            err = ctypes.c_int.in_dll(ctypes.pythonapi,"errno").value
+            raise OSError(err, os.strerror(err), 'initgroups')
+    os.setgid(passwd.pw_gid)
+    os.setuid(passwd.pw_uid)
+    os.environ['HOME'] = passwd.pw_dir
+
+
+def daemonize(pidfile=None, uid=None, umask=0o77):
+    """daemonize a Unix process. Set paranoid umask by default.
+
+    Return 1 in the original process, 2 in the first fork, and None for the
+    second fork (eg daemon process).
+    """
+    # http://www.faqs.org/faqs/unix-faq/programmer/faq/
+    #
+    # fork so the parent can exit
+    if os.fork():   # launch child and...
+        return 1
+    # disconnect from tty and create a new session
+    os.setsid()
+    # fork again so the parent, (the session group leader), can exit.
+    # as a non-session group leader, we can never regain a controlling
+    # terminal.
+    if os.fork():   # launch child again.
+        return 2
+    # move to the root to avoit mount pb
+    os.chdir('/')
+    # redirect standard descriptors
+    null = os.open('/dev/null', os.O_RDWR)
+    for i in range(3):
+        try:
+            os.dup2(null, i)
+        except OSError as e:
+            if e.errno != errno.EBADF:
+                raise
+    os.close(null)
+    # filter warnings
+    warnings.filterwarnings('ignore')
+    # write pid in a file
+    if pidfile:
+        # ensure the directory where the pid-file should be set exists (for
+        # instance /var/run/cubicweb may be deleted on computer restart)
+        piddir = os.path.dirname(pidfile)
+        if not os.path.exists(piddir):
+            os.makedirs(piddir)
+        f = file(pidfile, 'w')
+        f.write(str(os.getpid()))
+        f.close()
+    # set umask if specified
+    if umask is not None:
+        os.umask(umask)
+    # change process uid
+    if uid:
+        setugid(uid)
+    return None
diff --git a/logilab/common/date.py b/logilab/common/date.py
new file mode 100644
index 0000000..a093a8a
--- /dev/null
+++ b/logilab/common/date.py
@@ -0,0 +1,335 @@
+# copyright 2003-2012 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Date manipulation helper functions."""
+from __future__ import division
+
+__docformat__ = "restructuredtext en"
+
+import math
+import re
+import sys
+from locale import getlocale, LC_TIME
+from datetime import date, time, datetime, timedelta
+from time import strptime as time_strptime
+from calendar import monthrange, timegm
+
+from six.moves import range
+
+try:
+    from mx.DateTime import RelativeDateTime, Date, DateTimeType
+except ImportError:
+    endOfMonth = None
+    DateTimeType = datetime
+else:
+    endOfMonth = RelativeDateTime(months=1, day=-1)
+
+# NOTE: should we implement a compatibility layer between date representations
+#       as we have in lgc.db ?
+
+FRENCH_FIXED_HOLIDAYS = {
+    'jour_an': '%s-01-01',
+    'fete_travail': '%s-05-01',
+    'armistice1945': '%s-05-08',
+    'fete_nat': '%s-07-14',
+    'assomption': '%s-08-15',
+    'toussaint': '%s-11-01',
+    'armistice1918': '%s-11-11',
+    'noel': '%s-12-25',
+    }
+
+FRENCH_MOBILE_HOLIDAYS = {
+    'paques2004': '2004-04-12',
+    'ascension2004': '2004-05-20',
+    'pentecote2004': '2004-05-31',
+
+    'paques2005': '2005-03-28',
+    'ascension2005': '2005-05-05',
+    'pentecote2005': '2005-05-16',
+
+    'paques2006': '2006-04-17',
+    'ascension2006': '2006-05-25',
+    'pentecote2006': '2006-06-05',
+
+    'paques2007': '2007-04-09',
+    'ascension2007': '2007-05-17',
+    'pentecote2007': '2007-05-28',
+
+    'paques2008': '2008-03-24',
+    'ascension2008': '2008-05-01',
+    'pentecote2008': '2008-05-12',
+
+    'paques2009': '2009-04-13',
+    'ascension2009': '2009-05-21',
+    'pentecote2009': '2009-06-01',
+
+    'paques2010': '2010-04-05',
+    'ascension2010': '2010-05-13',
+    'pentecote2010': '2010-05-24',
+
+    'paques2011': '2011-04-25',
+    'ascension2011': '2011-06-02',
+    'pentecote2011': '2011-06-13',
+
+    'paques2012': '2012-04-09',
+    'ascension2012': '2012-05-17',
+    'pentecote2012': '2012-05-28',
+    }
+
+# XXX this implementation cries for multimethod dispatching
+
+def get_step(dateobj, nbdays=1):
+    # assume date is either a python datetime or a mx.DateTime object
+    if isinstance(dateobj, date):
+        return ONEDAY * nbdays
+    return nbdays # mx.DateTime is ok with integers
+
+def datefactory(year, month, day, sampledate):
+    # assume date is either a python datetime or a mx.DateTime object
+    if isinstance(sampledate, datetime):
+        return datetime(year, month, day)
+    if isinstance(sampledate, date):
+        return date(year, month, day)
+    return Date(year, month, day)
+
+def weekday(dateobj):
+    # assume date is either a python datetime or a mx.DateTime object
+    if isinstance(dateobj, date):
+        return dateobj.weekday()
+    return dateobj.day_of_week
+
+def str2date(datestr, sampledate):
+    # NOTE: datetime.strptime is not an option until we drop py2.4 compat
+    year, month, day = [int(chunk) for chunk in datestr.split('-')]
+    return datefactory(year, month, day, sampledate)
+
+def days_between(start, end):
+    if isinstance(start, date):
+        delta = end - start
+        # datetime.timedelta.days is always an integer (floored)
+        if delta.seconds:
+            return delta.days + 1
+        return delta.days
+    else:
+        return int(math.ceil((end - start).days))
+
+def get_national_holidays(begin, end):
+    """return french national days off between begin and end"""
+    begin = datefactory(begin.year, begin.month, begin.day, begin)
+    end = datefactory(end.year, end.month, end.day, end)
+    holidays = [str2date(datestr, begin)
+                for datestr in FRENCH_MOBILE_HOLIDAYS.values()]
+    for year in range(begin.year, end.year+1):
+        for datestr in FRENCH_FIXED_HOLIDAYS.values():
+            date = str2date(datestr % year, begin)
+            if date not in holidays:
+                holidays.append(date)
+    return [day for day in holidays if begin <= day < end]
+
+def add_days_worked(start, days):
+    """adds date but try to only take days worked into account"""
+    step = get_step(start)
+    weeks, plus = divmod(days, 5)
+    end = start + ((weeks * 7) + plus) * step
+    if weekday(end) >= 5: # saturday or sunday
+        end += (2 * step)
+    end += len([x for x in get_national_holidays(start, end + step)
+                if weekday(x) < 5]) * step
+    if weekday(end) >= 5: # saturday or sunday
+        end += (2 * step)
+    return end
+
+def nb_open_days(start, end):
+    assert start <= end
+    step = get_step(start)
+    days = days_between(start, end)
+    weeks, plus = divmod(days, 7)
+    if weekday(start) > weekday(end):
+        plus -= 2
+    elif weekday(end) == 6:
+        plus -= 1
+    open_days = weeks * 5 + plus
+    nb_week_holidays = len([x for x in get_national_holidays(start, end+step)
+                            if weekday(x) < 5 and x < end])
+    open_days -= nb_week_holidays
+    if open_days < 0:
+        return 0
+    return open_days
+
+def date_range(begin, end, incday=None, incmonth=None):
+    """yields each date between begin and end
+
+    :param begin: the start date
+    :param end: the end date
+    :param incr: the step to use to iterate over dates. Default is
+                 one day.
+    :param include: None (means no exclusion) or a function taking a
+                    date as parameter, and returning True if the date
+                    should be included.
+
+    When using mx datetime, you should *NOT* use incmonth argument, use instead
+    oneDay, oneHour, oneMinute, oneSecond, oneWeek or endOfMonth (to enumerate
+    months) as `incday` argument
+    """
+    assert not (incday and incmonth)
+    begin = todate(begin)
+    end = todate(end)
+    if incmonth:
+        while begin < end:
+            yield begin
+            begin = next_month(begin, incmonth)
+    else:
+        incr = get_step(begin, incday or 1)
+        while begin < end:
+           yield begin
+           begin += incr
+
+# makes py datetime usable #####################################################
+
+ONEDAY = timedelta(days=1)
+ONEWEEK = timedelta(days=7)
+
+try:
+    strptime = datetime.strptime
+except AttributeError: # py < 2.5
+    from time import strptime as time_strptime
+    def strptime(value, format):
+        return datetime(*time_strptime(value, format)[:6])
+
+def strptime_time(value, format='%H:%M'):
+    return time(*time_strptime(value, format)[3:6])
+
+def todate(somedate):
+    """return a date from a date (leaving unchanged) or a datetime"""
+    if isinstance(somedate, datetime):
+        return date(somedate.year, somedate.month, somedate.day)
+    assert isinstance(somedate, (date, DateTimeType)), repr(somedate)
+    return somedate
+
+def totime(somedate):
+    """return a time from a time (leaving unchanged), date or datetime"""
+    # XXX mx compat
+    if not isinstance(somedate, time):
+        return time(somedate.hour, somedate.minute, somedate.second)
+    assert isinstance(somedate, (time)), repr(somedate)
+    return somedate
+
+def todatetime(somedate):
+    """return a date from a date (leaving unchanged) or a datetime"""
+    # take care, datetime is a subclass of date
+    if isinstance(somedate, datetime):
+        return somedate
+    assert isinstance(somedate, (date, DateTimeType)), repr(somedate)
+    return datetime(somedate.year, somedate.month, somedate.day)
+
+def datetime2ticks(somedate):
+    return timegm(somedate.timetuple()) * 1000
+
+def ticks2datetime(ticks):
+    miliseconds, microseconds = divmod(ticks, 1000)
+    try:
+        return datetime.fromtimestamp(miliseconds)
+    except (ValueError, OverflowError):
+        epoch = datetime.fromtimestamp(0)
+        nb_days, seconds = divmod(int(miliseconds), 86400)
+        delta = timedelta(nb_days, seconds=seconds, microseconds=microseconds)
+        try:
+            return epoch + delta
+        except (ValueError, OverflowError):
+            raise
+
+def days_in_month(somedate):
+    return monthrange(somedate.year, somedate.month)[1]
+
+def days_in_year(somedate):
+    feb = date(somedate.year, 2, 1)
+    if days_in_month(feb) == 29:
+        return 366
+    else:
+        return 365
+
+def previous_month(somedate, nbmonth=1):
+    while nbmonth:
+        somedate = first_day(somedate) - ONEDAY
+        nbmonth -= 1
+    return somedate
+
+def next_month(somedate, nbmonth=1):
+    while nbmonth:
+        somedate = last_day(somedate) + ONEDAY
+        nbmonth -= 1
+    return somedate
+
+def first_day(somedate):
+    return date(somedate.year, somedate.month, 1)
+
+def last_day(somedate):
+    return date(somedate.year, somedate.month, days_in_month(somedate))
+
+def ustrftime(somedate, fmt='%Y-%m-%d'):
+    """like strftime, but returns a unicode string instead of an encoded
+    string which may be problematic with localized date.
+    """
+    if sys.version_info >= (3, 3):
+        # datetime.date.strftime() supports dates since year 1 in Python >=3.3.
+        return somedate.strftime(fmt)
+    else:
+        try:
+            if sys.version_info < (3, 0):
+                encoding = getlocale(LC_TIME)[1] or 'ascii'
+                return unicode(somedate.strftime(str(fmt)), encoding)
+            else:
+                return somedate.strftime(fmt)
+        except ValueError:
+            if somedate.year >= 1900:
+                raise
+            # datetime is not happy with dates before 1900
+            # we try to work around this, assuming a simple
+            # format string
+            fields = {'Y': somedate.year,
+                      'm': somedate.month,
+                      'd': somedate.day,
+                      }
+            if isinstance(somedate, datetime):
+                fields.update({'H': somedate.hour,
+                               'M': somedate.minute,
+                               'S': somedate.second})
+            fmt = re.sub('%([YmdHMS])', r'%(\1)02d', fmt)
+            return unicode(fmt) % fields
+
+def utcdatetime(dt):
+    if dt.tzinfo is None:
+        return dt
+    return (dt.replace(tzinfo=None) - dt.utcoffset())
+
+def utctime(dt):
+    if dt.tzinfo is None:
+        return dt
+    return (dt + dt.utcoffset() + dt.dst()).replace(tzinfo=None)
+
+def datetime_to_seconds(date):
+    """return the number of seconds since the begining of the day for that date
+    """
+    return date.second+60*date.minute + 3600*date.hour
+
+def timedelta_to_days(delta):
+    """return the time delta as a number of seconds"""
+    return delta.days + delta.seconds / (3600*24)
+
+def timedelta_to_seconds(delta):
+    """return the time delta as a fraction of days"""
+    return delta.days*(3600*24) + delta.seconds
diff --git a/logilab/common/dbf.py b/logilab/common/dbf.py
new file mode 100644
index 0000000..ab142b2
--- /dev/null
+++ b/logilab/common/dbf.py
@@ -0,0 +1,231 @@
+# -*- coding: utf-8 -*-
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""This is a DBF reader which reads Visual Fox Pro DBF format with Memo field
+
+Usage:
+
+>>> rec = readDbf('test.dbf')
+>>> for line in rec:
+>>>     print line['name']
+
+
+:date: 13/07/2007
+
+http://www.physics.ox.ac.uk/users/santoso/Software.Repository.html
+page says code is "available as is without any warranty or support".
+"""
+from __future__ import print_function
+
+import struct
+import os, os.path
+import sys
+import csv
+import tempfile
+
+from six.moves import range
+
+class Dbase:
+    def __init__(self):
+        self.fdb = None
+        self.fmemo = None
+        self.db_data = None
+        self.memo_data = None
+        self.fields = None
+        self.num_records = 0
+        self.header = None
+        self.memo_file = ''
+        self.memo_header = None
+        self.memo_block_size = 0
+        self.memo_header_len = 0
+
+    def _drop_after_NULL(self, txt):
+        for i in range(0, len(txt)):
+            if ord(struct.unpack('c', txt[i])[0])==0:
+                return txt[:i]
+        return txt
+
+    def _reverse_endian(self, num):
+        if not len(num):
+            return 0
+        val = struct.unpack('<L', num)
+        val = struct.pack('>L', val[0])
+        val = struct.unpack('>L', val)
+        return val[0]
+
+    def _assign_ids(self, lst, ids):
+        result = {}
+        idx = 0
+        for item in lst:
+            id = ids[idx]
+            result[id] = item
+            idx += 1
+        return result
+
+    def open(self, db_name):
+        filesize = os.path.getsize(db_name)
+        if filesize <= 68:
+            raise IOError('The file is not large enough to be a dbf file')
+
+        self.fdb = open(db_name, 'rb')
+
+        self.memo_file = ''
+        if os.path.isfile(db_name[0:-1] + 't'):
+            self.memo_file = db_name[0:-1] + 't'
+        elif os.path.isfile(db_name[0:-3] + 'fpt'):
+            self.memo_file = db_name[0:-3] + 'fpt'
+
+        if self.memo_file:
+            #Read memo file
+            self.fmemo = open(self.memo_file, 'rb')
+            self.memo_data = self.fmemo.read()
+            self.memo_header = self._assign_ids(struct.unpack('>6x1H', self.memo_data[:8]), ['Block size'])
+            block_size = self.memo_header['Block size']
+            if not block_size:
+                block_size = 512
+            self.memo_block_size = block_size
+            self.memo_header_len = block_size
+            memo_size = os.path.getsize(self.memo_file)
+
+        #Start reading data file
+        data = self.fdb.read(32)
+        self.header = self._assign_ids(struct.unpack('<B 3B L 2H 20x', data), ['id', 'Year', 'Month', 'Day', '# of Records', 'Header Size', 'Record Size'])
+        self.header['id'] = hex(self.header['id'])
+
+        self.num_records = self.header['# of Records']
+        data = self.fdb.read(self.header['Header Size']-34)
+        self.fields = {}
+        x = 0
+        header_pattern = '<11s c 4x B B 14x'
+        ids = ['Field Name', 'Field Type', 'Field Length', 'Field Precision']
+        pattern_len = 32
+        for offset in range(0, len(data), 32):
+            if ord(data[offset])==0x0d:
+                break
+            x += 1
+            data_subset = data[offset: offset+pattern_len]
+            if len(data_subset) < pattern_len:
+                data_subset += ' '*(pattern_len-len(data_subset))
+            self.fields[x] = self._assign_ids(struct.unpack(header_pattern, data_subset), ids)
+            self.fields[x]['Field Name'] = self._drop_after_NULL(self.fields[x]['Field Name'])
+
+        self.fdb.read(3)
+        if self.header['# of Records']:
+            data_size = (self.header['# of Records'] * self.header['Record Size']) - 1
+            self.db_data = self.fdb.read(data_size)
+        else:
+            self.db_data = ''
+        self.row_format = '<'
+        self.row_ids = []
+        self.row_len = 0
+        for key in self.fields:
+            field = self.fields[key]
+            self.row_format += '%ds ' % (field['Field Length'])
+            self.row_ids.append(field['Field Name'])
+            self.row_len += field['Field Length']
+
+    def close(self):
+        if self.fdb:
+            self.fdb.close()
+        if self.fmemo:
+            self.fmemo.close()
+
+    def get_numrecords(self):
+        return self.num_records
+
+    def get_record_with_names(self, rec_no):
+        """
+        This function accept record number from 0 to N-1
+        """
+        if rec_no < 0 or rec_no > self.num_records:
+            raise Exception('Unable to extract data outside the range')
+
+        offset = self.header['Record Size'] * rec_no
+        data = self.db_data[offset:offset+self.row_len]
+        record = self._assign_ids(struct.unpack(self.row_format, data), self.row_ids)
+
+        if self.memo_file:
+            for key in self.fields:
+                field = self.fields[key]
+                f_type = field['Field Type']
+                f_name = field['Field Name']
+                c_data = record[f_name]
+
+                if f_type=='M' or f_type=='G' or f_type=='B' or f_type=='P':
+                    c_data = self._reverse_endian(c_data)
+                    if c_data:
+                        record[f_name] = self.read_memo(c_data-1).strip()
+                else:
+                    record[f_name] = c_data.strip()
+        return record
+
+    def read_memo_record(self, num, in_length):
+        """
+        Read the record of given number. The second parameter is the length of
+        the record to read. It can be undefined, meaning read the whole record,
+        and it can be negative, meaning at most the length
+        """
+        if in_length < 0:
+            in_length = -self.memo_block_size
+
+        offset = self.memo_header_len + num * self.memo_block_size
+        self.fmemo.seek(offset)
+        if in_length<0:
+            in_length = -in_length
+        if in_length==0:
+            return ''
+        return self.fmemo.read(in_length)
+
+    def read_memo(self, num):
+        result = ''
+        buffer = self.read_memo_record(num, -1)
+        if len(buffer)<=0:
+            return ''
+        length = struct.unpack('>L', buffer[4:4+4])[0] + 8
+
+        block_size = self.memo_block_size
+        if length < block_size:
+            return buffer[8:length]
+        rest_length = length - block_size
+        rest_data = self.read_memo_record(num+1, rest_length)
+        if len(rest_data)<=0:
+            return ''
+        return buffer[8:] + rest_data
+
+def readDbf(filename):
+    """
+    Read the DBF file specified by the filename and
+    return the records as a list of dictionary.
+
+    :param: filename File name of the DBF
+    :return: List of rows
+    """
+    db = Dbase()
+    db.open(filename)
+    num = db.get_numrecords()
+    rec = []
+    for i in range(0, num):
+        record = db.get_record_with_names(i)
+        rec.append(record)
+    db.close()
+    return  rec
+
+if __name__=='__main__':
+    rec = readDbf('dbf/sptable.dbf')
+    for line in rec:
+        print('%s %s' % (line['GENUS'].strip(), line['SPECIES'].strip()))
diff --git a/logilab/common/debugger.py b/logilab/common/debugger.py
new file mode 100644
index 0000000..1f540a1
--- /dev/null
+++ b/logilab/common/debugger.py
@@ -0,0 +1,214 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Customized version of pdb's default debugger.
+
+- sets up a history file
+- uses ipython if available to colorize lines of code
+- overrides list command to search for current block instead
+  of using 5 lines of context
+
+
+
+
+"""
+
+from __future__ import print_function
+
+__docformat__ = "restructuredtext en"
+
+try:
+    import readline
+except ImportError:
+    readline = None
+import os
+import os.path as osp
+import sys
+from pdb import Pdb
+import inspect
+
+from logilab.common.compat import StringIO
+
+try:
+    from IPython import PyColorize
+except ImportError:
+    def colorize(source, *args):
+        """fallback colorize function"""
+        return source
+    def colorize_source(source, *args):
+        return source
+else:
+    def colorize(source, start_lineno, curlineno):
+        """colorize and annotate source with linenos
+        (as in pdb's list command)
+        """
+        parser = PyColorize.Parser()
+        output = StringIO()
+        parser.format(source, output)
+        annotated = []
+        for index, line in enumerate(output.getvalue().splitlines()):
+            lineno = index + start_lineno
+            if lineno == curlineno:
+                annotated.append('%4s\t->\t%s' % (lineno, line))
+            else:
+                annotated.append('%4s\t\t%s' % (lineno, line))
+        return '\n'.join(annotated)
+
+    def colorize_source(source):
+        """colorize given source"""
+        parser = PyColorize.Parser()
+        output = StringIO()
+        parser.format(source, output)
+        return output.getvalue()
+
+
+def getsource(obj):
+    """Return the text of the source code for an object.
+
+    The argument may be a module, class, method, function, traceback, frame,
+    or code object.  The source code is returned as a single string.  An
+    IOError is raised if the source code cannot be retrieved."""
+    lines, lnum = inspect.getsourcelines(obj)
+    return ''.join(lines), lnum
+
+
+################################################################
+class Debugger(Pdb):
+    """custom debugger
+
+    - sets up a history file
+    - uses ipython if available to colorize lines of code
+    - overrides list command to search for current block instead
+      of using 5 lines of context
+    """
+    def __init__(self, tcbk=None):
+        Pdb.__init__(self)
+        self.reset()
+        if tcbk:
+            while tcbk.tb_next is not None:
+                tcbk = tcbk.tb_next
+        self._tcbk = tcbk
+        self._histfile = os.path.expanduser("~/.pdbhist")
+
+    def setup_history_file(self):
+        """if readline is available, read pdb history file
+        """
+        if readline is not None:
+            try:
+                # XXX try..except shouldn't be necessary
+                # read_history_file() can accept None
+                readline.read_history_file(self._histfile)
+            except IOError:
+                pass
+
+    def start(self):
+        """starts the interactive mode"""
+        self.interaction(self._tcbk.tb_frame, self._tcbk)
+
+    def setup(self, frame, tcbk):
+        """setup hook: set up history file"""
+        self.setup_history_file()
+        Pdb.setup(self, frame, tcbk)
+
+    def set_quit(self):
+        """quit hook: save commands in the history file"""
+        if readline is not None:
+            readline.write_history_file(self._histfile)
+        Pdb.set_quit(self)
+
+    def complete_p(self, text, line, begin_idx, end_idx):
+        """provide variable names completion for the ``p`` command"""
+        namespace = dict(self.curframe.f_globals)
+        namespace.update(self.curframe.f_locals)
+        if '.' in text:
+            return self.attr_matches(text, namespace)
+        return [varname for varname in namespace if varname.startswith(text)]
+
+
+    def attr_matches(self, text, namespace):
+        """implementation coming from rlcompleter.Completer.attr_matches
+        Compute matches when text contains a dot.
+
+        Assuming the text is of the form NAME.NAME....[NAME], and is
+        evaluatable in self.namespace, it will be evaluated and its attributes
+        (as revealed by dir()) are used as possible completions.  (For class
+        instances, class members are also considered.)
+
+        WARNING: this can still invoke arbitrary C code, if an object
+        with a __getattr__ hook is evaluated.
+
+        """
+        import re
+        m = re.match(r"(\w+(\.\w+)*)\.(\w*)", text)
+        if not m:
+            return
+        expr, attr = m.group(1, 3)
+        object = eval(expr, namespace)
+        words = dir(object)
+        if hasattr(object, '__class__'):
+            words.append('__class__')
+            words = words + self.get_class_members(object.__class__)
+        matches = []
+        n = len(attr)
+        for word in words:
+            if word[:n] == attr and word != "__builtins__":
+                matches.append("%s.%s" % (expr, word))
+        return matches
+
+    def get_class_members(self, klass):
+        """implementation coming from rlcompleter.get_class_members"""
+        ret = dir(klass)
+        if hasattr(klass, '__bases__'):
+            for base in klass.__bases__:
+                ret = ret + self.get_class_members(base)
+        return ret
+
+    ## specific / overridden commands
+    def do_list(self, arg):
+        """overrides default list command to display the surrounding block
+        instead of 5 lines of context
+        """
+        self.lastcmd = 'list'
+        if not arg:
+            try:
+                source, start_lineno = getsource(self.curframe)
+                print(colorize(''.join(source), start_lineno,
+                               self.curframe.f_lineno))
+            except KeyboardInterrupt:
+                pass
+            except IOError:
+                Pdb.do_list(self, arg)
+        else:
+            Pdb.do_list(self, arg)
+    do_l = do_list
+
+    def do_open(self, arg):
+        """opens source file corresponding to the current stack level"""
+        filename = self.curframe.f_code.co_filename
+        lineno = self.curframe.f_lineno
+        cmd = 'emacsclient --no-wait +%s %s' % (lineno, filename)
+        os.system(cmd)
+
+    do_o = do_open
+
+def pm():
+    """use our custom debugger"""
+    dbg = Debugger(sys.last_traceback)
+    dbg.start()
+
+def set_trace():
+    Debugger().set_trace(sys._getframe().f_back)
diff --git a/logilab/common/decorators.py b/logilab/common/decorators.py
new file mode 100644
index 0000000..beafa20
--- /dev/null
+++ b/logilab/common/decorators.py
@@ -0,0 +1,281 @@
+# copyright 2003-2013 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+""" A few useful function/method decorators. """
+
+from __future__ import print_function
+
+__docformat__ = "restructuredtext en"
+
+import sys
+import types
+from time import clock, time
+from inspect import isgeneratorfunction, getargspec
+
+from logilab.common.compat import method_type
+
+# XXX rewrite so we can use the decorator syntax when keyarg has to be specified
+
+class cached_decorator(object):
+    def __init__(self, cacheattr=None, keyarg=None):
+        self.cacheattr = cacheattr
+        self.keyarg = keyarg
+    def __call__(self, callableobj=None):
+        assert not isgeneratorfunction(callableobj), \
+               'cannot cache generator function: %s' % callableobj
+        if len(getargspec(callableobj).args) == 1 or self.keyarg == 0:
+            cache = _SingleValueCache(callableobj, self.cacheattr)
+        elif self.keyarg:
+            cache = _MultiValuesKeyArgCache(callableobj, self.keyarg, self.cacheattr)
+        else:
+            cache = _MultiValuesCache(callableobj, self.cacheattr)
+        return cache.closure()
+
+class _SingleValueCache(object):
+    def __init__(self, callableobj, cacheattr=None):
+        self.callable = callableobj
+        if cacheattr is None:
+            self.cacheattr = '_%s_cache_' % callableobj.__name__
+        else:
+            assert cacheattr != callableobj.__name__
+            self.cacheattr = cacheattr
+
+    def __call__(__me, self, *args):
+        try:
+            return self.__dict__[__me.cacheattr]
+        except KeyError:
+            value = __me.callable(self, *args)
+            setattr(self, __me.cacheattr, value)
+            return value
+
+    def closure(self):
+        def wrapped(*args, **kwargs):
+            return self.__call__(*args, **kwargs)
+        wrapped.cache_obj = self
+        try:
+            wrapped.__doc__ = self.callable.__doc__
+            wrapped.__name__ = self.callable.__name__
+        except:
+            pass
+        return wrapped
+
+    def clear(self, holder):
+        holder.__dict__.pop(self.cacheattr, None)
+
+
+class _MultiValuesCache(_SingleValueCache):
+    def _get_cache(self, holder):
+        try:
+            _cache = holder.__dict__[self.cacheattr]
+        except KeyError:
+            _cache = {}
+            setattr(holder, self.cacheattr, _cache)
+        return _cache
+
+    def __call__(__me, self, *args, **kwargs):
+        _cache = __me._get_cache(self)
+        try:
+            return _cache[args]
+        except KeyError:
+            _cache[args] = __me.callable(self, *args)
+            return _cache[args]
+
+class _MultiValuesKeyArgCache(_MultiValuesCache):
+    def __init__(self, callableobj, keyarg, cacheattr=None):
+        super(_MultiValuesKeyArgCache, self).__init__(callableobj, cacheattr)
+        self.keyarg = keyarg
+
+    def __call__(__me, self, *args, **kwargs):
+        _cache = __me._get_cache(self)
+        key = args[__me.keyarg-1]
+        try:
+            return _cache[key]
+        except KeyError:
+            _cache[key] = __me.callable(self, *args, **kwargs)
+            return _cache[key]
+
+
+def cached(callableobj=None, keyarg=None, **kwargs):
+    """Simple decorator to cache result of method call."""
+    kwargs['keyarg'] = keyarg
+    decorator = cached_decorator(**kwargs)
+    if callableobj is None:
+        return decorator
+    else:
+        return decorator(callableobj)
+
+
+class cachedproperty(object):
+    """ Provides a cached property equivalent to the stacking of
+    @cached and @property, but more efficient.
+
+    After first usage, the <property_name> becomes part of the object's
+    __dict__. Doing:
+
+      del obj.<property_name> empties the cache.
+
+    Idea taken from the pyramid_ framework and the mercurial_ project.
+
+    .. _pyramid: http://pypi.python.org/pypi/pyramid
+    .. _mercurial: http://pypi.python.org/pypi/Mercurial
+    """
+    __slots__ = ('wrapped',)
+
+    def __init__(self, wrapped):
+        try:
+            wrapped.__name__
+        except AttributeError:
+            raise TypeError('%s must have a __name__ attribute' %
+                            wrapped)
+        self.wrapped = wrapped
+
+    @property
+    def __doc__(self):
+        doc = getattr(self.wrapped, '__doc__', None)
+        return ('<wrapped by the cachedproperty decorator>%s'
+                % ('\n%s' % doc if doc else ''))
+
+    def __get__(self, inst, objtype=None):
+        if inst is None:
+            return self
+        val = self.wrapped(inst)
+        setattr(inst, self.wrapped.__name__, val)
+        return val
+
+
+def get_cache_impl(obj, funcname):
+    cls = obj.__class__
+    member = getattr(cls, funcname)
+    if isinstance(member, property):
+        member = member.fget
+    return member.cache_obj
+
+def clear_cache(obj, funcname):
+    """Clear a cache handled by the :func:`cached` decorator. If 'x' class has
+    @cached on its method `foo`, type
+
+    >>> clear_cache(x, 'foo')
+
+    to purge this method's cache on the instance.
+    """
+    get_cache_impl(obj, funcname).clear(obj)
+
+def copy_cache(obj, funcname, cacheobj):
+    """Copy cache for <funcname> from cacheobj to obj."""
+    cacheattr = get_cache_impl(obj, funcname).cacheattr
+    try:
+        setattr(obj, cacheattr, cacheobj.__dict__[cacheattr])
+    except KeyError:
+        pass
+
+
+class wproperty(object):
+    """Simple descriptor expecting to take a modifier function as first argument
+    and looking for a _<function name> to retrieve the attribute.
+    """
+    def __init__(self, setfunc):
+        self.setfunc = setfunc
+        self.attrname = '_%s' % setfunc.__name__
+
+    def __set__(self, obj, value):
+        self.setfunc(obj, value)
+
+    def __get__(self, obj, cls):
+        assert obj is not None
+        return getattr(obj, self.attrname)
+
+
+class classproperty(object):
+    """this is a simple property-like class but for class attributes.
+    """
+    def __init__(self, get):
+        self.get = get
+    def __get__(self, inst, cls):
+        return self.get(cls)
+
+
+class iclassmethod(object):
+    '''Descriptor for method which should be available as class method if called
+    on the class or instance method if called on an instance.
+    '''
+    def __init__(self, func):
+        self.func = func
+    def __get__(self, instance, objtype):
+        if instance is None:
+            return method_type(self.func, objtype, objtype.__class__)
+        return method_type(self.func, instance, objtype)
+    def __set__(self, instance, value):
+        raise AttributeError("can't set attribute")
+
+
+def timed(f):
+    def wrap(*args, **kwargs):
+        t = time()
+        c = clock()
+        res = f(*args, **kwargs)
+        print('%s clock: %.9f / time: %.9f' % (f.__name__,
+                                               clock() - c, time() - t))
+        return res
+    return wrap
+
+
+def locked(acquire, release):
+    """Decorator taking two methods to acquire/release a lock as argument,
+    returning a decorator function which will call the inner method after
+    having called acquire(self) et will call release(self) afterwards.
+    """
+    def decorator(f):
+        def wrapper(self, *args, **kwargs):
+            acquire(self)
+            try:
+                return f(self, *args, **kwargs)
+            finally:
+                release(self)
+        return wrapper
+    return decorator
+
+
+def monkeypatch(klass, methodname=None):
+    """Decorator extending class with the decorated callable. This is basically
+    a syntactic sugar vs class assignment.
+
+    >>> class A:
+    ...     pass
+    >>> @monkeypatch(A)
+    ... def meth(self):
+    ...     return 12
+    ...
+    >>> a = A()
+    >>> a.meth()
+    12
+    >>> @monkeypatch(A, 'foo')
+    ... def meth(self):
+    ...     return 12
+    ...
+    >>> a.foo()
+    12
+    """
+    def decorator(func):
+        try:
+            name = methodname or func.__name__
+        except AttributeError:
+            raise AttributeError('%s has no __name__ attribute: '
+                                 'you should provide an explicit `methodname`'
+                                 % func)
+        setattr(klass, name, func)
+        return func
+    return decorator
diff --git a/logilab/common/deprecation.py b/logilab/common/deprecation.py
new file mode 100644
index 0000000..1c81b63
--- /dev/null
+++ b/logilab/common/deprecation.py
@@ -0,0 +1,189 @@
+# copyright 2003-2012 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Deprecation utilities."""
+
+__docformat__ = "restructuredtext en"
+
+import sys
+from warnings import warn
+
+from logilab.common.changelog import Version
+
+
+class DeprecationWrapper(object):
+    """proxy to print a warning on access to any attribute of the wrapped object
+    """
+    def __init__(self, proxied, msg=None):
+        self._proxied = proxied
+        self._msg = msg
+
+    def __getattr__(self, attr):
+        warn(self._msg, DeprecationWarning, stacklevel=2)
+        return getattr(self._proxied, attr)
+
+    def __setattr__(self, attr, value):
+        if attr in ('_proxied', '_msg'):
+            self.__dict__[attr] = value
+        else:
+            warn(self._msg, DeprecationWarning, stacklevel=2)
+            setattr(self._proxied, attr, value)
+
+
+class DeprecationManager(object):
+    """Manage the deprecation message handling. Messages are dropped for
+    versions more recent than the 'compatible' version. Example::
+
+        deprecator = deprecation.DeprecationManager("module_name")
+        deprecator.compatibility('1.3')
+
+        deprecator.warn('1.2', "message.")
+
+        @deprecator.deprecated('1.2', 'Message')
+        def any_func():
+            pass
+
+        class AnyClass(object):
+            __metaclass__ = deprecator.class_deprecated('1.2')
+    """
+    def __init__(self, module_name=None):
+        """
+        """
+        self.module_name = module_name
+        self.compatible_version = None
+
+    def compatibility(self, compatible_version):
+        """Set the compatible version.
+        """
+        self.compatible_version = Version(compatible_version)
+
+    def deprecated(self, version=None, reason=None, stacklevel=2, name=None, doc=None):
+        """Display a deprecation message only if the version is older than the
+        compatible version.
+        """
+        def decorator(func):
+            message = reason or 'The function "%s" is deprecated'
+            if '%s' in message:
+                message %= func.__name__
+            def wrapped(*args, **kwargs):
+                self.warn(version, message, stacklevel+1)
+                return func(*args, **kwargs)
+            return wrapped
+        return decorator
+
+    def class_deprecated(self, version=None):
+        class metaclass(type):
+            """metaclass to print a warning on instantiation of a deprecated class"""
+
+            def __call__(cls, *args, **kwargs):
+                msg = getattr(cls, "__deprecation_warning__",
+                              "%(cls)s is deprecated") % {'cls': cls.__name__}
+                self.warn(version, msg, stacklevel=3)
+                return type.__call__(cls, *args, **kwargs)
+        return metaclass
+
+    def moved(self, version, modpath, objname):
+        """use to tell that a callable has been moved to a new module.
+
+        It returns a callable wrapper, so that when its called a warning is printed
+        telling where the object can be found, import is done (and not before) and
+        the actual object is called.
+
+        NOTE: the usage is somewhat limited on classes since it will fail if the
+        wrapper is use in a class ancestors list, use the `class_moved` function
+        instead (which has no lazy import feature though).
+        """
+        def callnew(*args, **kwargs):
+            from logilab.common.modutils import load_module_from_name
+            message = "object %s has been moved to module %s" % (objname, modpath)
+            self.warn(version, message)
+            m = load_module_from_name(modpath)
+            return getattr(m, objname)(*args, **kwargs)
+        return callnew
+
+    def class_renamed(self, version, old_name, new_class, message=None):
+        clsdict = {}
+        if message is None:
+            message = '%s is deprecated, use %s' % (old_name, new_class.__name__)
+        clsdict['__deprecation_warning__'] = message
+        try:
+            # new-style class
+            return self.class_deprecated(version)(old_name, (new_class,), clsdict)
+        except (NameError, TypeError):
+            # old-style class
+            warn = self.warn
+            class DeprecatedClass(new_class):
+                """FIXME: There might be a better way to handle old/new-style class
+                """
+                def __init__(self, *args, **kwargs):
+                    warn(version, message, stacklevel=3)
+                    new_class.__init__(self, *args, **kwargs)
+            return DeprecatedClass
+
+    def class_moved(self, version, new_class, old_name=None, message=None):
+        """nice wrapper around class_renamed when a class has been moved into
+        another module
+        """
+        if old_name is None:
+            old_name = new_class.__name__
+        if message is None:
+            message = 'class %s is now available as %s.%s' % (
+                old_name, new_class.__module__, new_class.__name__)
+        return self.class_renamed(version, old_name, new_class, message)
+
+    def warn(self, version=None, reason="", stacklevel=2):
+        """Display a deprecation message only if the version is older than the
+        compatible version.
+        """
+        if (self.compatible_version is None
+            or version is None
+            or Version(version) < self.compatible_version):
+            if self.module_name and version:
+                reason = '[%s %s] %s' % (self.module_name, version, reason)
+            elif self.module_name:
+                reason = '[%s] %s' % (self.module_name, reason)
+            elif version:
+                reason = '[%s] %s' % (version, reason)
+            warn(reason, DeprecationWarning, stacklevel=stacklevel)
+
+_defaultdeprecator = DeprecationManager()
+
+def deprecated(reason=None, stacklevel=2, name=None, doc=None):
+    return _defaultdeprecator.deprecated(None, reason, stacklevel, name, doc)
+
+class_deprecated = _defaultdeprecator.class_deprecated()
+
+def moved(modpath, objname):
+    return _defaultdeprecator.moved(None, modpath, objname)
+moved.__doc__ = _defaultdeprecator.moved.__doc__
+
+def class_renamed(old_name, new_class, message=None):
+    """automatically creates a class which fires a DeprecationWarning
+    when instantiated.
+
+    >>> Set = class_renamed('Set', set, 'Set is now replaced by set')
+    >>> s = Set()
+    sample.py:57: DeprecationWarning: Set is now replaced by set
+    s = Set()
+    >>>
+    """
+    return _defaultdeprecator.class_renamed(None, old_name, new_class, message)
+
+def class_moved(new_class, old_name=None, message=None):
+    return _defaultdeprecator.class_moved(None, new_class, old_name, message)
+class_moved.__doc__ = _defaultdeprecator.class_moved.__doc__
+
diff --git a/logilab/common/fileutils.py b/logilab/common/fileutils.py
new file mode 100644
index 0000000..b30cf5f
--- /dev/null
+++ b/logilab/common/fileutils.py
@@ -0,0 +1,404 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""File and file-path manipulation utilities.
+
+:group path manipulation: first_level_directory, relative_path, is_binary,\
+get_by_ext, remove_dead_links
+:group file manipulation: norm_read, norm_open, lines, stream_lines, lines,\
+write_open_mode, ensure_fs_mode, export
+:sort: path manipulation, file manipulation
+"""
+
+from __future__ import print_function
+
+__docformat__ = "restructuredtext en"
+
+import sys
+import shutil
+import mimetypes
+from os.path import isabs, isdir, islink, split, exists, normpath, join
+from os.path import abspath
+from os import sep, mkdir, remove, listdir, stat, chmod, walk
+from stat import ST_MODE, S_IWRITE
+
+from logilab.common import STD_BLACKLIST as BASE_BLACKLIST, IGNORED_EXTENSIONS
+from logilab.common.shellutils import find
+from logilab.common.deprecation import deprecated
+from logilab.common.compat import FileIO
+
+def first_level_directory(path):
+    """Return the first level directory of a path.
+
+    >>> first_level_directory('home/syt/work')
+    'home'
+    >>> first_level_directory('/home/syt/work')
+    '/'
+    >>> first_level_directory('work')
+    'work'
+    >>>
+
+    :type path: str
+    :param path: the path for which we want the first level directory
+
+    :rtype: str
+    :return: the first level directory appearing in `path`
+    """
+    head, tail = split(path)
+    while head and tail:
+        head, tail = split(head)
+    if tail:
+        return tail
+    # path was absolute, head is the fs root
+    return head
+
+def abspath_listdir(path):
+    """Lists path's content using absolute paths.
+
+    >>> os.listdir('/home')
+    ['adim', 'alf', 'arthur', 'auc']
+    >>> abspath_listdir('/home')
+    ['/home/adim', '/home/alf', '/home/arthur', '/home/auc']
+    """
+    path = abspath(path)
+    return [join(path, filename) for filename in listdir(path)]
+
+
+def is_binary(filename):
+    """Return true if filename may be a binary file, according to it's
+    extension.
+
+    :type filename: str
+    :param filename: the name of the file
+
+    :rtype: bool
+    :return:
+      true if the file is a binary file (actually if it's mime type
+      isn't beginning by text/)
+    """
+    try:
+        return not mimetypes.guess_type(filename)[0].startswith('text')
+    except AttributeError:
+        return 1
+
+
+def write_open_mode(filename):
+    """Return the write mode that should used to open file.
+
+    :type filename: str
+    :param filename: the name of the file
+
+    :rtype: str
+    :return: the mode that should be use to open the file ('w' or 'wb')
+    """
+    if is_binary(filename):
+        return 'wb'
+    return 'w'
+
+
+def ensure_fs_mode(filepath, desired_mode=S_IWRITE):
+    """Check that the given file has the given mode(s) set, else try to
+    set it.
+
+    :type filepath: str
+    :param filepath: path of the file
+
+    :type desired_mode: int
+    :param desired_mode:
+      ORed flags describing the desired mode. Use constants from the
+      `stat` module for file permission's modes
+    """
+    mode = stat(filepath)[ST_MODE]
+    if not mode & desired_mode:
+        chmod(filepath, mode | desired_mode)
+
+
+# XXX (syt) unused? kill?
+class ProtectedFile(FileIO):
+    """A special file-object class that automatically does a 'chmod +w' when
+    needed.
+
+    XXX: for now, the way it is done allows 'normal file-objects' to be
+    created during the ProtectedFile object lifetime.
+    One way to circumvent this would be to chmod / unchmod on each
+    write operation.
+
+    One other way would be to :
+
+    - catch the IOError in the __init__
+
+    - if IOError, then create a StringIO object
+
+    - each write operation writes in this StringIO object
+
+    - on close()/del(), write/append the StringIO content to the file and
+      do the chmod only once
+    """
+    def __init__(self, filepath, mode):
+        self.original_mode = stat(filepath)[ST_MODE]
+        self.mode_changed = False
+        if mode in ('w', 'a', 'wb', 'ab'):
+            if not self.original_mode & S_IWRITE:
+                chmod(filepath, self.original_mode | S_IWRITE)
+                self.mode_changed = True
+        FileIO.__init__(self, filepath, mode)
+
+    def _restore_mode(self):
+        """restores the original mode if needed"""
+        if self.mode_changed:
+            chmod(self.name, self.original_mode)
+            # Don't re-chmod in case of several restore
+            self.mode_changed = False
+
+    def close(self):
+        """restore mode before closing"""
+        self._restore_mode()
+        FileIO.close(self)
+
+    def __del__(self):
+        if not self.closed:
+            self.close()
+
+
+class UnresolvableError(Exception):
+    """Exception raised by relative path when it's unable to compute relative
+    path between two paths.
+    """
+
+def relative_path(from_file, to_file):
+    """Try to get a relative path from `from_file` to `to_file`
+    (path will be absolute if to_file is an absolute file). This function
+    is useful to create link in `from_file` to `to_file`. This typical use
+    case is used in this function description.
+
+    If both files are relative, they're expected to be relative to the same
+    directory.
+
+    >>> relative_path( from_file='toto/index.html', to_file='index.html')
+    '../index.html'
+    >>> relative_path( from_file='index.html', to_file='toto/index.html')
+    'toto/index.html'
+    >>> relative_path( from_file='tutu/index.html', to_file='toto/index.html')
+    '../toto/index.html'
+    >>> relative_path( from_file='toto/index.html', to_file='/index.html')
+    '/index.html'
+    >>> relative_path( from_file='/toto/index.html', to_file='/index.html')
+    '../index.html'
+    >>> relative_path( from_file='/toto/index.html', to_file='/toto/summary.html')
+    'summary.html'
+    >>> relative_path( from_file='index.html', to_file='index.html')
+    ''
+    >>> relative_path( from_file='/index.html', to_file='toto/index.html')
+    Traceback (most recent call last):
+      File "<string>", line 1, in ?
+      File "<stdin>", line 37, in relative_path
+    UnresolvableError
+    >>> relative_path( from_file='/index.html', to_file='/index.html')
+    ''
+    >>>
+
+    :type from_file: str
+    :param from_file: source file (where links will be inserted)
+
+    :type to_file: str
+    :param to_file: target file (on which links point)
+
+    :raise UnresolvableError: if it has been unable to guess a correct path
+
+    :rtype: str
+    :return: the relative path of `to_file` from `from_file`
+    """
+    from_file = normpath(from_file)
+    to_file = normpath(to_file)
+    if from_file == to_file:
+        return ''
+    if isabs(to_file):
+        if not isabs(from_file):
+            return to_file
+    elif isabs(from_file):
+        raise UnresolvableError()
+    from_parts = from_file.split(sep)
+    to_parts = to_file.split(sep)
+    idem = 1
+    result = []
+    while len(from_parts) > 1:
+        dirname = from_parts.pop(0)
+        if idem and len(to_parts) > 1 and dirname == to_parts[0]:
+            to_parts.pop(0)
+        else:
+            idem = 0
+            result.append('..')
+    result += to_parts
+    return sep.join(result)
+
+
+def norm_read(path):
+    """Return the content of the file with normalized line feeds.
+
+    :type path: str
+    :param path: path to the file to read
+
+    :rtype: str
+    :return: the content of the file with normalized line feeds
+    """
+    return open(path, 'U').read()
+norm_read = deprecated("use \"open(path, 'U').read()\"")(norm_read)
+
+def norm_open(path):
+    """Return a stream for a file with content with normalized line feeds.
+
+    :type path: str
+    :param path: path to the file to open
+
+    :rtype: file or StringIO
+    :return: the opened file with normalized line feeds
+    """
+    return open(path, 'U')
+norm_open = deprecated("use \"open(path, 'U')\"")(norm_open)
+
+def lines(path, comments=None):
+    """Return a list of non empty lines in the file located at `path`.
+
+    :type path: str
+    :param path: path to the file
+
+    :type comments: str or None
+    :param comments:
+      optional string which can be used to comment a line in the file
+      (i.e. lines starting with this string won't be returned)
+
+    :rtype: list
+    :return:
+      a list of stripped line in the file, without empty and commented
+      lines
+
+    :warning: at some point this function will probably return an iterator
+    """
+    stream = open(path, 'U')
+    result = stream_lines(stream, comments)
+    stream.close()
+    return result
+
+
+def stream_lines(stream, comments=None):
+    """Return a list of non empty lines in the given `stream`.
+
+    :type stream: object implementing 'xreadlines' or 'readlines'
+    :param stream: file like object
+
+    :type comments: str or None
+    :param comments:
+      optional string which can be used to comment a line in the file
+      (i.e. lines starting with this string won't be returned)
+
+    :rtype: list
+    :return:
+      a list of stripped line in the file, without empty and commented
+      lines
+
+    :warning: at some point this function will probably return an iterator
+    """
+    try:
+        readlines = stream.xreadlines
+    except AttributeError:
+        readlines = stream.readlines
+    result = []
+    for line in readlines():
+        line = line.strip()
+        if line and (comments is None or not line.startswith(comments)):
+            result.append(line)
+    return result
+
+
+def export(from_dir, to_dir,
+           blacklist=BASE_BLACKLIST, ignore_ext=IGNORED_EXTENSIONS,
+           verbose=0):
+    """Make a mirror of `from_dir` in `to_dir`, omitting directories and
+    files listed in the black list or ending with one of the given
+    extensions.
+
+    :type from_dir: str
+    :param from_dir: directory to export
+
+    :type to_dir: str
+    :param to_dir: destination directory
+
+    :type blacklist: list or tuple
+    :param blacklist:
+      list of files or directories to ignore, default to the content of
+      `BASE_BLACKLIST`
+
+    :type ignore_ext: list or tuple
+    :param ignore_ext:
+      list of extensions to ignore, default to  the content of
+      `IGNORED_EXTENSIONS`
+
+    :type verbose: bool
+    :param verbose:
+      flag indicating whether information about exported files should be
+      printed to stderr, default to False
+    """
+    try:
+        mkdir(to_dir)
+    except OSError:
+        pass # FIXME we should use "exists" if the point is about existing dir
+             # else (permission problems?) shouldn't return / raise ?
+    for directory, dirnames, filenames in walk(from_dir):
+        for norecurs in blacklist:
+            try:
+                dirnames.remove(norecurs)
+            except ValueError:
+                continue
+        for dirname in dirnames:
+            src = join(directory, dirname)
+            dest = to_dir + src[len(from_dir):]
+            if isdir(src):
+                if not exists(dest):
+                    mkdir(dest)
+        for filename in filenames:
+            # don't include binary files
+            # endswith does not accept tuple in 2.4
+            if any([filename.endswith(ext) for ext in ignore_ext]):
+                continue
+            src = join(directory, filename)
+            dest = to_dir + src[len(from_dir):]
+            if verbose:
+                print(src, '->', dest, file=sys.stderr)
+            if exists(dest):
+                remove(dest)
+            shutil.copy2(src, dest)
+
+
+def remove_dead_links(directory, verbose=0):
+    """Recursively traverse directory and remove all dead links.
+
+    :type directory: str
+    :param directory: directory to cleanup
+
+    :type verbose: bool
+    :param verbose:
+      flag indicating whether information about deleted links should be
+      printed to stderr, default to False
+    """
+    for dirpath, dirname, filenames in walk(directory):
+        for filename in dirnames + filenames:
+            src = join(dirpath, filename)
+            if islink(src) and not exists(src):
+                if verbose:
+                    print('remove dead link', src)
+                remove(src)
+
diff --git a/logilab/common/graph.py b/logilab/common/graph.py
new file mode 100644
index 0000000..cef1c98
--- /dev/null
+++ b/logilab/common/graph.py
@@ -0,0 +1,282 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Graph manipulation utilities.
+
+(dot generation adapted from pypy/translator/tool/make_dot.py)
+"""
+
+__docformat__ = "restructuredtext en"
+
+__metaclass__ = type
+
+import os.path as osp
+import os
+import sys
+import tempfile
+import codecs
+import errno
+
+def escape(value):
+    """Make <value> usable in a dot file."""
+    lines = [line.replace('"', '\\"') for line in value.split('\n')]
+    data = '\\l'.join(lines)
+    return '\\n' + data
+
+def target_info_from_filename(filename):
+    """Transforms /some/path/foo.png into ('/some/path', 'foo.png', 'png')."""
+    basename = osp.basename(filename)
+    storedir = osp.dirname(osp.abspath(filename))
+    target = filename.split('.')[-1]
+    return storedir, basename, target
+
+
+class DotBackend:
+    """Dot File backend."""
+    def __init__(self, graphname, rankdir=None, size=None, ratio=None,
+            charset='utf-8', renderer='dot', additionnal_param={}):
+        self.graphname = graphname
+        self.renderer = renderer
+        self.lines = []
+        self._source = None
+        self.emit("digraph %s {" % normalize_node_id(graphname))
+        if rankdir:
+            self.emit('rankdir=%s' % rankdir)
+        if ratio:
+            self.emit('ratio=%s' % ratio)
+        if size:
+            self.emit('size="%s"' % size)
+        if charset:
+            assert charset.lower() in ('utf-8', 'iso-8859-1', 'latin1'), \
+                   'unsupported charset %s' % charset
+            self.emit('charset="%s"' % charset)
+        for param in sorted(additionnal_param.items()):
+            self.emit('='.join(param))
+
+    def get_source(self):
+        """returns self._source"""
+        if self._source is None:
+            self.emit("}\n")
+            self._source = '\n'.join(self.lines)
+            del self.lines
+        return self._source
+
+    source = property(get_source)
+
+    def generate(self, outputfile=None, dotfile=None, mapfile=None):
+        """Generates a graph file.
+
+        :param outputfile: filename and path [defaults to graphname.png]
+        :param dotfile: filename and path [defaults to graphname.dot]
+
+        :rtype: str
+        :return: a path to the generated file
+        """
+        import subprocess # introduced in py 2.4
+        name = self.graphname
+        if not dotfile:
+            # if 'outputfile' is a dot file use it as 'dotfile'
+            if outputfile and outputfile.endswith(".dot"):
+                dotfile = outputfile
+            else:
+                dotfile = '%s.dot' % name
+        if outputfile is not None:
+            storedir, basename, target = target_info_from_filename(outputfile)
+            if target != "dot":
+                pdot, dot_sourcepath = tempfile.mkstemp(".dot", name)
+                os.close(pdot)
+            else:
+                dot_sourcepath = osp.join(storedir, dotfile)
+        else:
+            target = 'png'
+            pdot, dot_sourcepath = tempfile.mkstemp(".dot", name)
+            ppng, outputfile = tempfile.mkstemp(".png", name)
+            os.close(pdot)
+            os.close(ppng)
+        pdot = codecs.open(dot_sourcepath, 'w', encoding='utf8')
+        pdot.write(self.source)
+        pdot.close()
+        if target != 'dot':
+            if sys.platform == 'win32':
+                use_shell = True
+            else:
+                use_shell = False
+            try:
+                if mapfile:
+                    subprocess.call([self.renderer,  '-Tcmapx', '-o', mapfile, '-T', target, dot_sourcepath, '-o', outputfile],
+                                    shell=use_shell)
+                else:
+                    subprocess.call([self.renderer, '-T',  target,
+                                     dot_sourcepath, '-o',  outputfile],
+                                    shell=use_shell)
+            except OSError as e:
+                if e.errno == errno.ENOENT:
+                    e.strerror = 'File not found: {0}'.format(self.renderer)
+                    raise
+            os.unlink(dot_sourcepath)
+        return outputfile
+
+    def emit(self, line):
+        """Adds <line> to final output."""
+        self.lines.append(line)
+
+    def emit_edge(self, name1, name2, **props):
+        """emit an edge from <name1> to <name2>.
+        edge properties: see http://www.graphviz.org/doc/info/attrs.html
+        """
+        attrs = ['%s="%s"' % (prop, value) for prop, value in props.items()]
+        n_from, n_to = normalize_node_id(name1), normalize_node_id(name2)
+        self.emit('%s -> %s [%s];' % (n_from, n_to, ', '.join(sorted(attrs))) )
+
+    def emit_node(self, name, **props):
+        """emit a node with given properties.
+        node properties: see http://www.graphviz.org/doc/info/attrs.html
+        """
+        attrs = ['%s="%s"' % (prop, value) for prop, value in props.items()]
+        self.emit('%s [%s];' % (normalize_node_id(name), ', '.join(sorted(attrs))))
+
+def normalize_node_id(nid):
+    """Returns a suitable DOT node id for `nid`."""
+    return '"%s"' % nid
+
+class GraphGenerator:
+    def __init__(self, backend):
+        # the backend is responsible to output the graph in a particular format
+        self.backend = backend
+
+    # XXX doesn't like space in outpufile / mapfile
+    def generate(self, visitor, propshdlr, outputfile=None, mapfile=None):
+        # the visitor
+        # the property handler is used to get node and edge properties
+        # according to the graph and to the backend
+        self.propshdlr = propshdlr
+        for nodeid, node in visitor.nodes():
+            props = propshdlr.node_properties(node)
+            self.backend.emit_node(nodeid, **props)
+        for subjnode, objnode, edge in visitor.edges():
+            props = propshdlr.edge_properties(edge, subjnode, objnode)
+            self.backend.emit_edge(subjnode, objnode, **props)
+        return self.backend.generate(outputfile=outputfile, mapfile=mapfile)
+
+
+class UnorderableGraph(Exception):
+    pass
+
+def ordered_nodes(graph):
+    """takes a dependency graph dict as arguments and return an ordered tuple of
+    nodes starting with nodes without dependencies and up to the outermost node.
+
+    If there is some cycle in the graph, :exc:`UnorderableGraph` will be raised.
+
+    Also the given graph dict will be emptied.
+    """
+    # check graph consistency
+    cycles = get_cycles(graph)
+    if cycles:
+        cycles = '\n'.join([' -> '.join(cycle) for cycle in cycles])
+        raise UnorderableGraph('cycles in graph: %s' % cycles)
+    vertices = set(graph)
+    to_vertices = set()
+    for edges in graph.values():
+        to_vertices |= set(edges)
+    missing_vertices = to_vertices - vertices
+    if missing_vertices:
+        raise UnorderableGraph('missing vertices: %s' % ', '.join(missing_vertices))
+    # order vertices
+    order = []
+    order_set = set()
+    old_len = None
+    while graph:
+        if old_len == len(graph):
+            raise UnorderableGraph('unknown problem with %s' % graph)
+        old_len = len(graph)
+        deps_ok = []
+        for node, node_deps in graph.items():
+            for dep in node_deps:
+                if dep not in order_set:
+                    break
+            else:
+                deps_ok.append(node)
+        order.append(deps_ok)
+        order_set |= set(deps_ok)
+        for node in deps_ok:
+            del graph[node]
+    result = []
+    for grp in reversed(order):
+        result.extend(sorted(grp))
+    return tuple(result)
+
+
+def get_cycles(graph_dict, vertices=None):
+    '''given a dictionary representing an ordered graph (i.e. key are vertices
+    and values is a list of destination vertices representing edges), return a
+    list of detected cycles
+    '''
+    if not graph_dict:
+        return ()
+    result = []
+    if vertices is None:
+        vertices = graph_dict.keys()
+    for vertice in vertices:
+        _get_cycles(graph_dict, [], set(), result, vertice)
+    return result
+
+def _get_cycles(graph_dict, path, visited, result, vertice):
+    """recursive function doing the real work for get_cycles"""
+    if vertice in path:
+        cycle = [vertice]
+        for node in path[::-1]:
+            if node == vertice:
+                break
+            cycle.insert(0, node)
+        # make a canonical representation
+        start_from = min(cycle)
+        index = cycle.index(start_from)
+        cycle = cycle[index:] + cycle[0:index]
+        # append it to result if not already in
+        if not cycle in result:
+            result.append(cycle)
+        return
+    path.append(vertice)
+    try:
+        for node in graph_dict[vertice]:
+            # don't check already visited nodes again
+            if node not in visited:
+                _get_cycles(graph_dict, path, visited, result, node)
+                visited.add(node)
+    except KeyError:
+        pass
+    path.pop()
+
+def has_path(graph_dict, fromnode, tonode, path=None):
+    """generic function taking a simple graph definition as a dictionary, with
+    node has key associated to a list of nodes directly reachable from it.
+
+    Return None if no path exists to go from `fromnode` to `tonode`, else the
+    first path found (as a list including the destination node at last)
+    """
+    if path is None:
+        path = []
+    elif fromnode in path:
+        return None
+    path.append(fromnode)
+    for destnode in graph_dict[fromnode]:
+        if destnode == tonode or has_path(graph_dict, destnode, tonode, path):
+            return path[1:] + [tonode]
+    path.pop()
+    return None
+
diff --git a/logilab/common/interface.py b/logilab/common/interface.py
new file mode 100644
index 0000000..3ea4ab7
--- /dev/null
+++ b/logilab/common/interface.py
@@ -0,0 +1,71 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Bases class for interfaces to provide 'light' interface handling.
+
+ TODO:
+  _ implements a check method which check that an object implements the
+    interface
+  _ Attribute objects
+
+  This module requires at least python 2.2
+"""
+__docformat__ = "restructuredtext en"
+
+
+class Interface(object):
+    """Base class for interfaces."""
+    def is_implemented_by(cls, instance):
+        return implements(instance, cls)
+    is_implemented_by = classmethod(is_implemented_by)
+
+
+def implements(obj, interface):
+    """Return true if the give object (maybe an instance or class) implements
+    the interface.
+    """
+    kimplements = getattr(obj, '__implements__', ())
+    if not isinstance(kimplements, (list, tuple)):
+        kimplements = (kimplements,)
+    for implementedinterface in kimplements:
+        if issubclass(implementedinterface, interface):
+            return True
+    return False
+
+
+def extend(klass, interface, _recurs=False):
+    """Add interface to klass'__implements__ if not already implemented in.
+
+    If klass is subclassed, ensure subclasses __implements__ it as well.
+
+    NOTE: klass should be e new class.
+    """
+    if not implements(klass, interface):
+        try:
+            kimplements = klass.__implements__
+            kimplementsklass = type(kimplements)
+            kimplements = list(kimplements)
+        except AttributeError:
+            kimplementsklass = tuple
+            kimplements = []
+        kimplements.append(interface)
+        klass.__implements__ = kimplementsklass(kimplements)
+        for subklass in klass.__subclasses__():
+            extend(subklass, interface, _recurs=True)
+    elif _recurs:
+        for subklass in klass.__subclasses__():
+            extend(subklass, interface, _recurs=True)
diff --git a/logilab/common/logging_ext.py b/logilab/common/logging_ext.py
new file mode 100644
index 0000000..3b6a580
--- /dev/null
+++ b/logilab/common/logging_ext.py
@@ -0,0 +1,195 @@
+# -*- coding: utf-8 -*-
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Extends the logging module from the standard library."""
+
+__docformat__ = "restructuredtext en"
+
+import os
+import sys
+import logging
+
+from six import string_types
+
+from logilab.common.textutils import colorize_ansi
+
+
+def set_log_methods(cls, logger):
+    """bind standard logger's methods as methods on the class"""
+    cls.__logger = logger
+    for attr in ('debug', 'info', 'warning', 'error', 'critical', 'exception'):
+        setattr(cls, attr, getattr(logger, attr))
+
+
+def xxx_cyan(record):
+    if 'XXX' in record.message:
+        return 'cyan'
+
+class ColorFormatter(logging.Formatter):
+    """
+    A color Formatter for the logging standard module.
+
+    By default, colorize CRITICAL and ERROR in red, WARNING in orange, INFO in
+    green and DEBUG in yellow.
+
+    self.colors is customizable via the 'color' constructor argument (dictionary).
+
+    self.colorfilters is a list of functions that get the LogRecord
+    and return a color name or None.
+    """
+
+    def __init__(self, fmt=None, datefmt=None, colors=None):
+        logging.Formatter.__init__(self, fmt, datefmt)
+        self.colorfilters = []
+        self.colors = {'CRITICAL': 'red',
+                       'ERROR': 'red',
+                       'WARNING': 'magenta',
+                       'INFO': 'green',
+                       'DEBUG': 'yellow',
+                       }
+        if colors is not None:
+            assert isinstance(colors, dict)
+            self.colors.update(colors)
+
+    def format(self, record):
+        msg = logging.Formatter.format(self, record)
+        if record.levelname in self.colors:
+            color = self.colors[record.levelname]
+            return colorize_ansi(msg, color)
+        else:
+            for cf in self.colorfilters:
+                color = cf(record)
+                if color:
+                    return colorize_ansi(msg, color)
+        return msg
+
+def set_color_formatter(logger=None, **kw):
+    """
+    Install a color formatter on the 'logger'. If not given, it will
+    defaults to the default logger.
+
+    Any additional keyword will be passed as-is to the ColorFormatter
+    constructor.
+    """
+    if logger is None:
+        logger = logging.getLogger()
+        if not logger.handlers:
+            logging.basicConfig()
+    format_msg = logger.handlers[0].formatter._fmt
+    fmt = ColorFormatter(format_msg, **kw)
+    fmt.colorfilters.append(xxx_cyan)
+    logger.handlers[0].setFormatter(fmt)
+
+
+LOG_FORMAT = '%(asctime)s - (%(name)s) %(levelname)s: %(message)s'
+LOG_DATE_FORMAT = '%Y-%m-%d %H:%M:%S'
+
+def get_handler(debug=False, syslog=False, logfile=None, rotation_parameters=None):
+    """get an apropriate handler according to given parameters"""
+    if os.environ.get('APYCOT_ROOT'):
+        handler = logging.StreamHandler(sys.stdout)
+    if debug:
+        handler = logging.StreamHandler()
+    elif logfile is None:
+        if syslog:
+            from logging import handlers
+            handler = handlers.SysLogHandler()
+        else:
+            handler = logging.StreamHandler()
+    else:
+        try:
+            if rotation_parameters is None:
+                if os.name == 'posix' and sys.version_info >= (2, 6):
+                    from logging.handlers import WatchedFileHandler
+                    handler = WatchedFileHandler(logfile)
+                else:
+                    handler = logging.FileHandler(logfile)
+            else:
+                from logging.handlers import TimedRotatingFileHandler
+                handler = TimedRotatingFileHandler(
+                    logfile, **rotation_parameters)
+        except IOError:
+            handler = logging.StreamHandler()
+    return handler
+
+def get_threshold(debug=False, logthreshold=None):
+    if logthreshold is None:
+        if debug:
+            logthreshold = logging.DEBUG
+        else:
+            logthreshold = logging.ERROR
+    elif isinstance(logthreshold, string_types):
+        logthreshold = getattr(logging, THRESHOLD_MAP.get(logthreshold,
+                                                          logthreshold))
+    return logthreshold
+
+def _colorable_terminal():
+    isatty = hasattr(sys.__stdout__, 'isatty') and sys.__stdout__.isatty()
+    if not isatty:
+        return False
+    if os.name == 'nt':
+        try:
+            from colorama import init as init_win32_colors
+        except ImportError:
+            return False
+        init_win32_colors()
+    return True
+
+def get_formatter(logformat=LOG_FORMAT, logdateformat=LOG_DATE_FORMAT):
+    if _colorable_terminal():
+        fmt = ColorFormatter(logformat, logdateformat)
+        def col_fact(record):
+            if 'XXX' in record.message:
+                return 'cyan'
+            if 'kick' in record.message:
+                return 'red'
+        fmt.colorfilters.append(col_fact)
+    else:
+        fmt = logging.Formatter(logformat, logdateformat)
+    return fmt
+
+def init_log(debug=False, syslog=False, logthreshold=None, logfile=None,
+             logformat=LOG_FORMAT, logdateformat=LOG_DATE_FORMAT, fmt=None,
+             rotation_parameters=None, handler=None):
+    """init the log service"""
+    logger = logging.getLogger()
+    if handler is None:
+        handler = get_handler(debug, syslog, logfile, rotation_parameters)
+    # only addHandler and removeHandler method while I would like a setHandler
+    # method, so do it this way :$
+    logger.handlers = [handler]
+    logthreshold = get_threshold(debug, logthreshold)
+    logger.setLevel(logthreshold)
+    if fmt is None:
+        if debug:
+            fmt = get_formatter(logformat=logformat, logdateformat=logdateformat)
+        else:
+            fmt = logging.Formatter(logformat, logdateformat)
+    handler.setFormatter(fmt)
+    return handler
+
+# map logilab.common.logger thresholds to logging thresholds
+THRESHOLD_MAP = {'LOG_DEBUG':   'DEBUG',
+                 'LOG_INFO':    'INFO',
+                 'LOG_NOTICE':  'INFO',
+                 'LOG_WARN':    'WARNING',
+                 'LOG_WARNING': 'WARNING',
+                 'LOG_ERR':     'ERROR',
+                 'LOG_ERROR':   'ERROR',
+                 'LOG_CRIT':    'CRITICAL',
+                 }
diff --git a/logilab/common/modutils.py b/logilab/common/modutils.py
new file mode 100644
index 0000000..a426a3a
--- /dev/null
+++ b/logilab/common/modutils.py
@@ -0,0 +1,702 @@
+# -*- coding: utf-8 -*-
+# copyright 2003-2013 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Python modules manipulation utility functions.
+
+:type PY_SOURCE_EXTS: tuple(str)
+:var PY_SOURCE_EXTS: list of possible python source file extension
+
+:type STD_LIB_DIR: str
+:var STD_LIB_DIR: directory where standard modules are located
+
+:type BUILTIN_MODULES: dict
+:var BUILTIN_MODULES: dictionary with builtin module names has key
+"""
+
+__docformat__ = "restructuredtext en"
+
+import sys
+import os
+from os.path import splitext, join, abspath, isdir, dirname, exists, basename
+from imp import find_module, load_module, C_BUILTIN, PY_COMPILED, PKG_DIRECTORY
+from distutils.sysconfig import get_config_var, get_python_lib, get_python_version
+from distutils.errors import DistutilsPlatformError
+
+from six.moves import range
+
+try:
+    import zipimport
+except ImportError:
+    zipimport = None
+
+ZIPFILE = object()
+
+from logilab.common import STD_BLACKLIST, _handle_blacklist
+
+# Notes about STD_LIB_DIR
+# Consider arch-specific installation for STD_LIB_DIR definition
+# :mod:`distutils.sysconfig` contains to much hardcoded values to rely on
+#
+# :see: `Problems with /usr/lib64 builds <http://bugs.python.org/issue1294959>`_
+# :see: `FHS <http://www.pathname.com/fhs/pub/fhs-2.3.html#LIBLTQUALGTALTERNATEFORMATESSENTIAL>`_
+if sys.platform.startswith('win'):
+    PY_SOURCE_EXTS = ('py', 'pyw')
+    PY_COMPILED_EXTS = ('dll', 'pyd')
+else:
+    PY_SOURCE_EXTS = ('py',)
+    PY_COMPILED_EXTS = ('so',)
+
+try:
+    STD_LIB_DIR = get_python_lib(standard_lib=1)
+# get_python_lib(standard_lib=1) is not available on pypy, set STD_LIB_DIR to
+# non-valid path, see https://bugs.pypy.org/issue1164
+except DistutilsPlatformError:
+    STD_LIB_DIR = '//'
+
+EXT_LIB_DIR = get_python_lib()
+
+BUILTIN_MODULES = dict(zip(sys.builtin_module_names,
+                           [1]*len(sys.builtin_module_names)))
+
+
+class NoSourceFile(Exception):
+    """exception raised when we are not able to get a python
+    source file for a precompiled file
+    """
+
+class LazyObject(object):
+    def __init__(self, module, obj):
+        self.module = module
+        self.obj = obj
+        self._imported = None
+
+    def _getobj(self):
+        if self._imported is None:
+           self._imported = getattr(load_module_from_name(self.module),
+                                    self.obj)
+        return self._imported
+
+    def __getattribute__(self, attr):
+        try:
+            return super(LazyObject, self).__getattribute__(attr)
+        except AttributeError as ex:
+            return getattr(self._getobj(), attr)
+
+    def __call__(self, *args, **kwargs):
+        return self._getobj()(*args, **kwargs)
+
+
+def load_module_from_name(dotted_name, path=None, use_sys=1):
+    """Load a Python module from its name.
+
+    :type dotted_name: str
+    :param dotted_name: python name of a module or package
+
+    :type path: list or None
+    :param path:
+      optional list of path where the module or package should be
+      searched (use sys.path if nothing or None is given)
+
+    :type use_sys: bool
+    :param use_sys:
+      boolean indicating whether the sys.modules dictionary should be
+      used or not
+
+
+    :raise ImportError: if the module or package is not found
+
+    :rtype: module
+    :return: the loaded module
+    """
+    return load_module_from_modpath(dotted_name.split('.'), path, use_sys)
+
+
+def load_module_from_modpath(parts, path=None, use_sys=1):
+    """Load a python module from its splitted name.
+
+    :type parts: list(str) or tuple(str)
+    :param parts:
+      python name of a module or package splitted on '.'
+
+    :type path: list or None
+    :param path:
+      optional list of path where the module or package should be
+      searched (use sys.path if nothing or None is given)
+
+    :type use_sys: bool
+    :param use_sys:
+      boolean indicating whether the sys.modules dictionary should be used or not
+
+    :raise ImportError: if the module or package is not found
+
+    :rtype: module
+    :return: the loaded module
+    """
+    if use_sys:
+        try:
+            return sys.modules['.'.join(parts)]
+        except KeyError:
+            pass
+    modpath = []
+    prevmodule = None
+    for part in parts:
+        modpath.append(part)
+        curname = '.'.join(modpath)
+        module = None
+        if len(modpath) != len(parts):
+            # even with use_sys=False, should try to get outer packages from sys.modules
+            module = sys.modules.get(curname)
+        elif use_sys:
+            # because it may have been indirectly loaded through a parent
+            module = sys.modules.get(curname)
+        if module is None:
+            mp_file, mp_filename, mp_desc = find_module(part, path)
+            module = load_module(curname, mp_file, mp_filename, mp_desc)
+        if prevmodule:
+            setattr(prevmodule, part, module)
+        _file = getattr(module, '__file__', '')
+        if not _file and len(modpath) != len(parts):
+            raise ImportError('no module in %s' % '.'.join(parts[len(modpath):]) )
+        path = [dirname( _file )]
+        prevmodule = module
+    return module
+
+
+def load_module_from_file(filepath, path=None, use_sys=1, extrapath=None):
+    """Load a Python module from it's path.
+
+    :type filepath: str
+    :param filepath: path to the python module or package
+
+    :type path: list or None
+    :param path:
+      optional list of path where the module or package should be
+      searched (use sys.path if nothing or None is given)
+
+    :type use_sys: bool
+    :param use_sys:
+      boolean indicating whether the sys.modules dictionary should be
+      used or not
+
+
+    :raise ImportError: if the module or package is not found
+
+    :rtype: module
+    :return: the loaded module
+    """
+    modpath = modpath_from_file(filepath, extrapath)
+    return load_module_from_modpath(modpath, path, use_sys)
+
+
+def _check_init(path, mod_path):
+    """check there are some __init__.py all along the way"""
+    for part in mod_path:
+        path = join(path, part)
+        if not _has_init(path):
+            return False
+    return True
+
+
+def modpath_from_file(filename, extrapath=None):
+    """given a file path return the corresponding splitted module's name
+    (i.e name of a module or package splitted on '.')
+
+    :type filename: str
+    :param filename: file's path for which we want the module's name
+
+    :type extrapath: dict
+    :param extrapath:
+      optional extra search path, with path as key and package name for the path
+      as value. This is usually useful to handle package splitted in multiple
+      directories using __path__ trick.
+
+
+    :raise ImportError:
+      if the corresponding module's name has not been found
+
+    :rtype: list(str)
+    :return: the corresponding splitted module's name
+    """
+    base = splitext(abspath(filename))[0]
+    if extrapath is not None:
+        for path_ in extrapath:
+            path = abspath(path_)
+            if path and base[:len(path)] == path:
+                submodpath = [pkg for pkg in base[len(path):].split(os.sep)
+                              if pkg]
+                if _check_init(path, submodpath[:-1]):
+                    return extrapath[path_].split('.') + submodpath
+    for path in sys.path:
+        path = abspath(path)
+        if path and base.startswith(path):
+            modpath = [pkg for pkg in base[len(path):].split(os.sep) if pkg]
+            if _check_init(path, modpath[:-1]):
+                return modpath
+    raise ImportError('Unable to find module for %s in %s' % (
+        filename, ', \n'.join(sys.path)))
+
+
+
+def file_from_modpath(modpath, path=None, context_file=None):
+    """given a mod path (i.e. splitted module / package name), return the
+    corresponding file, giving priority to source file over precompiled
+    file if it exists
+
+    :type modpath: list or tuple
+    :param modpath:
+      splitted module's name (i.e name of a module or package splitted
+      on '.')
+      (this means explicit relative imports that start with dots have
+      empty strings in this list!)
+
+    :type path: list or None
+    :param path:
+      optional list of path where the module or package should be
+      searched (use sys.path if nothing or None is given)
+
+    :type context_file: str or None
+    :param context_file:
+      context file to consider, necessary if the identifier has been
+      introduced using a relative import unresolvable in the actual
+      context (i.e. modutils)
+
+    :raise ImportError: if there is no such module in the directory
+
+    :rtype: str or None
+    :return:
+      the path to the module's file or None if it's an integrated
+      builtin module such as 'sys'
+    """
+    if context_file is not None:
+        context = dirname(context_file)
+    else:
+        context = context_file
+    if modpath[0] == 'xml':
+        # handle _xmlplus
+        try:
+            return _file_from_modpath(['_xmlplus'] + modpath[1:], path, context)
+        except ImportError:
+            return _file_from_modpath(modpath, path, context)
+    elif modpath == ['os', 'path']:
+        # FIXME: currently ignoring search_path...
+        return os.path.__file__
+    return _file_from_modpath(modpath, path, context)
+
+
+
+def get_module_part(dotted_name, context_file=None):
+    """given a dotted name return the module part of the name :
+
+    >>> get_module_part('logilab.common.modutils.get_module_part')
+    'logilab.common.modutils'
+
+    :type dotted_name: str
+    :param dotted_name: full name of the identifier we are interested in
+
+    :type context_file: str or None
+    :param context_file:
+      context file to consider, necessary if the identifier has been
+      introduced using a relative import unresolvable in the actual
+      context (i.e. modutils)
+
+
+    :raise ImportError: if there is no such module in the directory
+
+    :rtype: str or None
+    :return:
+      the module part of the name or None if we have not been able at
+      all to import the given name
+
+    XXX: deprecated, since it doesn't handle package precedence over module
+    (see #10066)
+    """
+    # os.path trick
+    if dotted_name.startswith('os.path'):
+        return 'os.path'
+    parts = dotted_name.split('.')
+    if context_file is not None:
+        # first check for builtin module which won't be considered latter
+        # in that case (path != None)
+        if parts[0] in BUILTIN_MODULES:
+            if len(parts) > 2:
+                raise ImportError(dotted_name)
+            return parts[0]
+        # don't use += or insert, we want a new list to be created !
+    path = None
+    starti = 0
+    if parts[0] == '':
+        assert context_file is not None, \
+                'explicit relative import, but no context_file?'
+        path = [] # prevent resolving the import non-relatively
+        starti = 1
+    while parts[starti] == '': # for all further dots: change context
+        starti += 1
+        context_file = dirname(context_file)
+    for i in range(starti, len(parts)):
+        try:
+            file_from_modpath(parts[starti:i+1],
+                    path=path, context_file=context_file)
+        except ImportError:
+            if not i >= max(1, len(parts) - 2):
+                raise
+            return '.'.join(parts[:i])
+    return dotted_name
+
+
+def get_modules(package, src_directory, blacklist=STD_BLACKLIST):
+    """given a package directory return a list of all available python
+    modules in the package and its subpackages
+
+    :type package: str
+    :param package: the python name for the package
+
+    :type src_directory: str
+    :param src_directory:
+      path of the directory corresponding to the package
+
+    :type blacklist: list or tuple
+    :param blacklist:
+      optional list of files or directory to ignore, default to
+      the value of `logilab.common.STD_BLACKLIST`
+
+    :rtype: list
+    :return:
+      the list of all available python modules in the package and its
+      subpackages
+    """
+    modules = []
+    for directory, dirnames, filenames in os.walk(src_directory):
+        _handle_blacklist(blacklist, dirnames, filenames)
+        # check for __init__.py
+        if not '__init__.py' in filenames:
+            dirnames[:] = ()
+            continue
+        if directory != src_directory:
+            dir_package = directory[len(src_directory):].replace(os.sep, '.')
+            modules.append(package + dir_package)
+        for filename in filenames:
+            if _is_python_file(filename) and filename != '__init__.py':
+                src = join(directory, filename)
+                module = package + src[len(src_directory):-3]
+                modules.append(module.replace(os.sep, '.'))
+    return modules
+
+
+
+def get_module_files(src_directory, blacklist=STD_BLACKLIST):
+    """given a package directory return a list of all available python
+    module's files in the package and its subpackages
+
+    :type src_directory: str
+    :param src_directory:
+      path of the directory corresponding to the package
+
+    :type blacklist: list or tuple
+    :param blacklist:
+      optional list of files or directory to ignore, default to the value of
+      `logilab.common.STD_BLACKLIST`
+
+    :rtype: list
+    :return:
+      the list of all available python module's files in the package and
+      its subpackages
+    """
+    files = []
+    for directory, dirnames, filenames in os.walk(src_directory):
+        _handle_blacklist(blacklist, dirnames, filenames)
+        # check for __init__.py
+        if not '__init__.py' in filenames:
+            dirnames[:] = ()
+            continue
+        for filename in filenames:
+            if _is_python_file(filename):
+                src = join(directory, filename)
+                files.append(src)
+    return files
+
+
+def get_source_file(filename, include_no_ext=False):
+    """given a python module's file name return the matching source file
+    name (the filename will be returned identically if it's a already an
+    absolute path to a python source file...)
+
+    :type filename: str
+    :param filename: python module's file name
+
+
+    :raise NoSourceFile: if no source file exists on the file system
+
+    :rtype: str
+    :return: the absolute path of the source file if it exists
+    """
+    base, orig_ext = splitext(abspath(filename))
+    for ext in PY_SOURCE_EXTS:
+        source_path = '%s.%s' % (base, ext)
+        if exists(source_path):
+            return source_path
+    if include_no_ext and not orig_ext and exists(base):
+        return base
+    raise NoSourceFile(filename)
+
+
+def cleanup_sys_modules(directories):
+    """remove submodules of `directories` from `sys.modules`"""
+    cleaned = []
+    for modname, module in list(sys.modules.items()):
+        modfile = getattr(module, '__file__', None)
+        if modfile:
+            for directory in directories:
+                if modfile.startswith(directory):
+                    cleaned.append(modname)
+                    del sys.modules[modname]
+                    break
+    return cleaned
+
+
+def is_python_source(filename):
+    """
+    rtype: bool
+    return: True if the filename is a python source file
+    """
+    return splitext(filename)[1][1:] in PY_SOURCE_EXTS
+
+
+
+def is_standard_module(modname, std_path=(STD_LIB_DIR,)):
+    """try to guess if a module is a standard python module (by default,
+    see `std_path` parameter's description)
+
+    :type modname: str
+    :param modname: name of the module we are interested in
+
+    :type std_path: list(str) or tuple(str)
+    :param std_path: list of path considered has standard
+
+
+    :rtype: bool
+    :return:
+      true if the module:
+      - is located on the path listed in one of the directory in `std_path`
+      - is a built-in module
+    """
+    modname = modname.split('.')[0]
+    try:
+        filename = file_from_modpath([modname])
+    except ImportError as ex:
+        # import failed, i'm probably not so wrong by supposing it's
+        # not standard...
+        return 0
+    # modules which are not living in a file are considered standard
+    # (sys and __builtin__ for instance)
+    if filename is None:
+        return 1
+    filename = abspath(filename)
+    if filename.startswith(EXT_LIB_DIR):
+        return 0
+    for path in std_path:
+        if filename.startswith(abspath(path)):
+            return 1
+    return False
+
+
+
+def is_relative(modname, from_file):
+    """return true if the given module name is relative to the given
+    file name
+
+    :type modname: str
+    :param modname: name of the module we are interested in
+
+    :type from_file: str
+    :param from_file:
+      path of the module from which modname has been imported
+
+    :rtype: bool
+    :return:
+      true if the module has been imported relatively to `from_file`
+    """
+    if not isdir(from_file):
+        from_file = dirname(from_file)
+    if from_file in sys.path:
+        return False
+    try:
+        find_module(modname.split('.')[0], [from_file])
+        return True
+    except ImportError:
+        return False
+
+
+# internal only functions #####################################################
+
+def _file_from_modpath(modpath, path=None, context=None):
+    """given a mod path (i.e. splitted module / package name), return the
+    corresponding file
+
+    this function is used internally, see `file_from_modpath`'s
+    documentation for more information
+    """
+    assert len(modpath) > 0
+    if context is not None:
+        try:
+            mtype, mp_filename = _module_file(modpath, [context])
+        except ImportError:
+            mtype, mp_filename = _module_file(modpath, path)
+    else:
+        mtype, mp_filename = _module_file(modpath, path)
+    if mtype == PY_COMPILED:
+        try:
+            return get_source_file(mp_filename)
+        except NoSourceFile:
+            return mp_filename
+    elif mtype == C_BUILTIN:
+        # integrated builtin module
+        return None
+    elif mtype == PKG_DIRECTORY:
+        mp_filename = _has_init(mp_filename)
+    return mp_filename
+
+def _search_zip(modpath, pic):
+    for filepath, importer in pic.items():
+        if importer is not None:
+            if importer.find_module(modpath[0]):
+                if not importer.find_module('/'.join(modpath)):
+                    raise ImportError('No module named %s in %s/%s' % (
+                        '.'.join(modpath[1:]), filepath, modpath))
+                return ZIPFILE, abspath(filepath) + '/' + '/'.join(modpath), filepath
+    raise ImportError('No module named %s' % '.'.join(modpath))
+
+try:
+    import pkg_resources
+except ImportError:
+    pkg_resources = None
+
+def _module_file(modpath, path=None):
+    """get a module type / file path
+
+    :type modpath: list or tuple
+    :param modpath:
+      splitted module's name (i.e name of a module or package splitted
+      on '.'), with leading empty strings for explicit relative import
+
+    :type path: list or None
+    :param path:
+      optional list of path where the module or package should be
+      searched (use sys.path if nothing or None is given)
+
+
+    :rtype: tuple(int, str)
+    :return: the module type flag and the file path for a module
+    """
+    # egg support compat
+    try:
+        pic = sys.path_importer_cache
+        _path = (path is None and sys.path or path)
+        for __path in _path:
+            if not __path in pic:
+                try:
+                    pic[__path] = zipimport.zipimporter(__path)
+                except zipimport.ZipImportError:
+                    pic[__path] = None
+        checkeggs = True
+    except AttributeError:
+        checkeggs = False
+    # pkg_resources support (aka setuptools namespace packages)
+    if (pkg_resources is not None
+            and modpath[0] in pkg_resources._namespace_packages
+            and modpath[0] in sys.modules
+            and len(modpath) > 1):
+        # setuptools has added into sys.modules a module object with proper
+        # __path__, get back information from there
+        module = sys.modules[modpath.pop(0)]
+        path = module.__path__
+    imported = []
+    while modpath:
+        modname = modpath[0]
+        # take care to changes in find_module implementation wrt builtin modules
+        #
+        # Python 2.6.6 (r266:84292, Sep 11 2012, 08:34:23)
+        # >>> imp.find_module('posix')
+        # (None, 'posix', ('', '', 6))
+        #
+        # Python 3.3.1 (default, Apr 26 2013, 12:08:46)
+        # >>> imp.find_module('posix')
+        # (None, None, ('', '', 6))
+        try:
+            _, mp_filename, mp_desc = find_module(modname, path)
+        except ImportError:
+            if checkeggs:
+                return _search_zip(modpath, pic)[:2]
+            raise
+        else:
+            if checkeggs and mp_filename:
+                fullabspath = [abspath(x) for x in _path]
+                try:
+                    pathindex = fullabspath.index(dirname(abspath(mp_filename)))
+                    emtype, emp_filename, zippath = _search_zip(modpath, pic)
+                    if pathindex > _path.index(zippath):
+                        # an egg takes priority
+                        return emtype, emp_filename
+                except ValueError:
+                    # XXX not in _path
+                    pass
+                except ImportError:
+                    pass
+                checkeggs = False
+        imported.append(modpath.pop(0))
+        mtype = mp_desc[2]
+        if modpath:
+            if mtype != PKG_DIRECTORY:
+                raise ImportError('No module %s in %s' % ('.'.join(modpath),
+                                                          '.'.join(imported)))
+            # XXX guess if package is using pkgutil.extend_path by looking for
+            # those keywords in the first four Kbytes
+            try:
+                with open(join(mp_filename, '__init__.py')) as stream:
+                    data = stream.read(4096)
+            except IOError:
+                path = [mp_filename]
+            else:
+                if 'pkgutil' in data and 'extend_path' in data:
+                    # extend_path is called, search sys.path for module/packages
+                    # of this name see pkgutil.extend_path documentation
+                    path = [join(p, *imported) for p in sys.path
+                            if isdir(join(p, *imported))]
+                else:
+                    path = [mp_filename]
+    return mtype, mp_filename
+
+def _is_python_file(filename):
+    """return true if the given filename should be considered as a python file
+
+    .pyc and .pyo are ignored
+    """
+    for ext in ('.py', '.so', '.pyd', '.pyw'):
+        if filename.endswith(ext):
+            return True
+    return False
+
+
+def _has_init(directory):
+    """if the given directory has a valid __init__ file, return its path,
+    else return None
+    """
+    mod_or_pack = join(directory, '__init__')
+    for ext in PY_SOURCE_EXTS + ('pyc', 'pyo'):
+        if exists(mod_or_pack + '.' + ext):
+            return mod_or_pack + '.' + ext
+    return None
diff --git a/logilab/common/optik_ext.py b/logilab/common/optik_ext.py
new file mode 100644
index 0000000..1fd2a7f
--- /dev/null
+++ b/logilab/common/optik_ext.py
@@ -0,0 +1,392 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Add an abstraction level to transparently import optik classes from optparse
+(python >= 2.3) or the optik package.
+
+It also defines three new types for optik/optparse command line parser :
+
+  * regexp
+    argument of this type will be converted using re.compile
+  * csv
+    argument of this type will be converted using split(',')
+  * yn
+    argument of this type will be true if 'y' or 'yes', false if 'n' or 'no'
+  * named
+    argument of this type are in the form <NAME>=<VALUE> or <NAME>:<VALUE>
+  * password
+    argument of this type wont be converted but this is used by other tools
+    such as interactive prompt for configuration to double check value and
+    use an invisible field
+  * multiple_choice
+    same as default "choice" type but multiple choices allowed
+  * file
+    argument of this type wont be converted but checked that the given file exists
+  * color
+    argument of this type wont be converted but checked its either a
+    named color or a color specified using hexadecimal notation (preceded by a #)
+  * time
+    argument of this type will be converted to a float value in seconds
+    according to time units (ms, s, min, h, d)
+  * bytes
+    argument of this type will be converted to a float value in bytes
+    according to byte units (b, kb, mb, gb, tb)
+"""
+from __future__ import print_function
+
+__docformat__ = "restructuredtext en"
+
+import re
+import sys
+import time
+from copy import copy
+from os.path import exists
+
+# python >= 2.3
+from optparse import OptionParser as BaseParser, Option as BaseOption, \
+     OptionGroup, OptionContainer, OptionValueError, OptionError, \
+     Values, HelpFormatter, NO_DEFAULT, SUPPRESS_HELP
+
+try:
+    from mx import DateTime
+    HAS_MX_DATETIME = True
+except ImportError:
+    HAS_MX_DATETIME = False
+
+from logilab.common.textutils import splitstrip, TIME_UNITS, BYTE_UNITS, \
+    apply_units
+
+
+def check_regexp(option, opt, value):
+    """check a regexp value by trying to compile it
+    return the compiled regexp
+    """
+    if hasattr(value, 'pattern'):
+        return value
+    try:
+        return re.compile(value)
+    except ValueError:
+        raise OptionValueError(
+            "option %s: invalid regexp value: %r" % (opt, value))
+
+def check_csv(option, opt, value):
+    """check a csv value by trying to split it
+    return the list of separated values
+    """
+    if isinstance(value, (list, tuple)):
+        return value
+    try:
+        return splitstrip(value)
+    except ValueError:
+        raise OptionValueError(
+            "option %s: invalid csv value: %r" % (opt, value))
+
+def check_yn(option, opt, value):
+    """check a yn value
+    return true for yes and false for no
+    """
+    if isinstance(value, int):
+        return bool(value)
+    if value in ('y', 'yes'):
+        return True
+    if value in ('n', 'no'):
+        return False
+    msg = "option %s: invalid yn value %r, should be in (y, yes, n, no)"
+    raise OptionValueError(msg % (opt, value))
+
+def check_named(option, opt, value):
+    """check a named value
+    return a dictionary containing (name, value) associations
+    """
+    if isinstance(value, dict):
+        return value
+    values = []
+    for value in check_csv(option, opt, value):
+        if value.find('=') != -1:
+            values.append(value.split('=', 1))
+        elif value.find(':') != -1:
+            values.append(value.split(':', 1))
+    if values:
+        return dict(values)
+    msg = "option %s: invalid named value %r, should be <NAME>=<VALUE> or \
+<NAME>:<VALUE>"
+    raise OptionValueError(msg % (opt, value))
+
+def check_password(option, opt, value):
+    """check a password value (can't be empty)
+    """
+    # no actual checking, monkey patch if you want more
+    return value
+
+def check_file(option, opt, value):
+    """check a file value
+    return the filepath
+    """
+    if exists(value):
+        return value
+    msg = "option %s: file %r does not exist"
+    raise OptionValueError(msg % (opt, value))
+
+# XXX use python datetime
+def check_date(option, opt, value):
+    """check a file value
+    return the filepath
+    """
+    try:
+        return DateTime.strptime(value, "%Y/%m/%d")
+    except DateTime.Error :
+        raise OptionValueError(
+            "expected format of %s is yyyy/mm/dd" % opt)
+
+def check_color(option, opt, value):
+    """check a color value and returns it
+    /!\ does *not* check color labels (like 'red', 'green'), only
+    checks hexadecimal forms
+    """
+    # Case (1) : color label, we trust the end-user
+    if re.match('[a-z0-9 ]+$', value, re.I):
+        return value
+    # Case (2) : only accepts hexadecimal forms
+    if re.match('#[a-f0-9]{6}', value, re.I):
+        return value
+    # Else : not a color label neither a valid hexadecimal form => error
+    msg = "option %s: invalid color : %r, should be either hexadecimal \
+    value or predefined color"
+    raise OptionValueError(msg % (opt, value))
+
+def check_time(option, opt, value):
+    if isinstance(value, (int, long, float)):
+        return value
+    return apply_units(value, TIME_UNITS)
+
+def check_bytes(option, opt, value):
+    if hasattr(value, '__int__'):
+        return value
+    return apply_units(value, BYTE_UNITS)
+
+
+class Option(BaseOption):
+    """override optik.Option to add some new option types
+    """
+    TYPES = BaseOption.TYPES + ('regexp', 'csv', 'yn', 'named', 'password',
+                                'multiple_choice', 'file', 'color',
+                                'time', 'bytes')
+    ATTRS = BaseOption.ATTRS + ['hide', 'level']
+    TYPE_CHECKER = copy(BaseOption.TYPE_CHECKER)
+    TYPE_CHECKER['regexp'] = check_regexp
+    TYPE_CHECKER['csv'] = check_csv
+    TYPE_CHECKER['yn'] = check_yn
+    TYPE_CHECKER['named'] = check_named
+    TYPE_CHECKER['multiple_choice'] = check_csv
+    TYPE_CHECKER['file'] = check_file
+    TYPE_CHECKER['color'] = check_color
+    TYPE_CHECKER['password'] = check_password
+    TYPE_CHECKER['time'] = check_time
+    TYPE_CHECKER['bytes'] = check_bytes
+    if HAS_MX_DATETIME:
+        TYPES += ('date',)
+        TYPE_CHECKER['date'] = check_date
+
+    def __init__(self, *opts, **attrs):
+        BaseOption.__init__(self, *opts, **attrs)
+        if hasattr(self, "hide") and self.hide:
+            self.help = SUPPRESS_HELP
+
+    def _check_choice(self):
+        """FIXME: need to override this due to optik misdesign"""
+        if self.type in ("choice", "multiple_choice"):
+            if self.choices is None:
+                raise OptionError(
+                    "must supply a list of choices for type 'choice'", self)
+            elif not isinstance(self.choices, (tuple, list)):
+                raise OptionError(
+                    "choices must be a list of strings ('%s' supplied)"
+                    % str(type(self.choices)).split("'")[1], self)
+        elif self.choices is not None:
+            raise OptionError(
+                "must not supply choices for type %r" % self.type, self)
+    BaseOption.CHECK_METHODS[2] = _check_choice
+
+
+    def process(self, opt, value, values, parser):
+        # First, convert the value(s) to the right type.  Howl if any
+        # value(s) are bogus.
+        value = self.convert_value(opt, value)
+        if self.type == 'named':
+            existant = getattr(values, self.dest)
+            if existant:
+                existant.update(value)
+                value = existant
+       # And then take whatever action is expected of us.
+        # This is a separate method to make life easier for
+        # subclasses to add new actions.
+        return self.take_action(
+            self.action, self.dest, opt, value, values, parser)
+
+
+class OptionParser(BaseParser):
+    """override optik.OptionParser to use our Option class
+    """
+    def __init__(self, option_class=Option, *args, **kwargs):
+        BaseParser.__init__(self, option_class=Option, *args, **kwargs)
+
+    def format_option_help(self, formatter=None):
+        if formatter is None:
+            formatter = self.formatter
+        outputlevel = getattr(formatter, 'output_level', 0)
+        formatter.store_option_strings(self)
+        result = []
+        result.append(formatter.format_heading("Options"))
+        formatter.indent()
+        if self.option_list:
+            result.append(OptionContainer.format_option_help(self, formatter))
+            result.append("\n")
+        for group in self.option_groups:
+            if group.level <= outputlevel and (
+                group.description or level_options(group, outputlevel)):
+                result.append(group.format_help(formatter))
+                result.append("\n")
+        formatter.dedent()
+        # Drop the last "\n", or the header if no options or option groups:
+        return "".join(result[:-1])
+
+
+OptionGroup.level = 0
+
+def level_options(group, outputlevel):
+    return [option for option in group.option_list
+            if (getattr(option, 'level', 0) or 0) <= outputlevel
+            and not option.help is SUPPRESS_HELP]
+
+def format_option_help(self, formatter):
+    result = []
+    outputlevel = getattr(formatter, 'output_level', 0) or 0
+    for option in level_options(self, outputlevel):
+        result.append(formatter.format_option(option))
+    return "".join(result)
+OptionContainer.format_option_help = format_option_help
+
+
+class ManHelpFormatter(HelpFormatter):
+    """Format help using man pages ROFF format"""
+
+    def __init__ (self,
+                  indent_increment=0,
+                  max_help_position=24,
+                  width=79,
+                  short_first=0):
+        HelpFormatter.__init__ (
+            self, indent_increment, max_help_position, width, short_first)
+
+    def format_heading(self, heading):
+        return '.SH %s\n' % heading.upper()
+
+    def format_description(self, description):
+        return description
+
+    def format_option(self, option):
+        try:
+            optstring = option.option_strings
+        except AttributeError:
+            optstring = self.format_option_strings(option)
+        if option.help:
+            help_text = self.expand_default(option)
+            help = ' '.join([l.strip() for l in help_text.splitlines()])
+        else:
+            help = ''
+        return '''.IP "%s"
+%s
+''' % (optstring, help)
+
+    def format_head(self, optparser, pkginfo, section=1):
+        long_desc = ""
+        try:
+            pgm = optparser._get_prog_name()
+        except AttributeError:
+            # py >= 2.4.X (dunno which X exactly, at least 2)
+            pgm = optparser.get_prog_name()
+        short_desc = self.format_short_description(pgm, pkginfo.description)
+        if hasattr(pkginfo, "long_desc"):
+            long_desc = self.format_long_description(pgm, pkginfo.long_desc)
+        return '%s\n%s\n%s\n%s' % (self.format_title(pgm, section),
+                                   short_desc, self.format_synopsis(pgm),
+                                   long_desc)
+
+    def format_title(self, pgm, section):
+        date = '-'.join([str(num) for num in time.localtime()[:3]])
+        return '.TH %s %s "%s" %s' % (pgm, section, date, pgm)
+
+    def format_short_description(self, pgm, short_desc):
+        return '''.SH NAME
+.B %s
+\- %s
+''' % (pgm, short_desc.strip())
+
+    def format_synopsis(self, pgm):
+        return '''.SH SYNOPSIS
+.B  %s
+[
+.I OPTIONS
+] [
+.I <arguments>
+]
+''' % pgm
+
+    def format_long_description(self, pgm, long_desc):
+        long_desc = '\n'.join([line.lstrip()
+                               for line in long_desc.splitlines()])
+        long_desc = long_desc.replace('\n.\n', '\n\n')
+        if long_desc.lower().startswith(pgm):
+            long_desc = long_desc[len(pgm):]
+        return '''.SH DESCRIPTION
+.B %s
+%s
+''' % (pgm, long_desc.strip())
+
+    def format_tail(self, pkginfo):
+        tail = '''.SH SEE ALSO
+/usr/share/doc/pythonX.Y-%s/
+
+.SH BUGS
+Please report bugs on the project\'s mailing list:
+%s
+
+.SH AUTHOR
+%s <%s>
+''' % (getattr(pkginfo, 'debian_name', pkginfo.modname),
+       pkginfo.mailinglist, pkginfo.author, pkginfo.author_email)
+
+        if hasattr(pkginfo, "copyright"):
+            tail += '''
+.SH COPYRIGHT
+%s
+''' % pkginfo.copyright
+
+        return tail
+
+def generate_manpage(optparser, pkginfo, section=1, stream=sys.stdout, level=0):
+    """generate a man page from an optik parser"""
+    formatter = ManHelpFormatter()
+    formatter.output_level = level
+    formatter.parser = optparser
+    print(formatter.format_head(optparser, pkginfo, section), file=stream)
+    print(optparser.format_option_help(formatter), file=stream)
+    print(formatter.format_tail(pkginfo), file=stream)
+
+
+__all__ = ('OptionParser', 'Option', 'OptionGroup', 'OptionValueError',
+           'Values')
diff --git a/logilab/common/optparser.py b/logilab/common/optparser.py
new file mode 100644
index 0000000..aa17750
--- /dev/null
+++ b/logilab/common/optparser.py
@@ -0,0 +1,92 @@
+# -*- coding: utf-8 -*-
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Extend OptionParser with commands.
+
+Example:
+
+>>> parser = OptionParser()
+>>> parser.usage = '%prog COMMAND [options] <arg> ...'
+>>> parser.add_command('build', 'mymod.build')
+>>> parser.add_command('clean', run_clean, add_opt_clean)
+>>> run, options, args = parser.parse_command(sys.argv[1:])
+>>> return run(options, args[1:])
+
+With mymod.build that defines two functions run and add_options
+"""
+from __future__ import print_function
+
+__docformat__ = "restructuredtext en"
+
+from warnings import warn
+warn('lgc.optparser module is deprecated, use lgc.clcommands instead', DeprecationWarning,
+     stacklevel=2)
+
+import sys
+import optparse
+
+class OptionParser(optparse.OptionParser):
+
+    def __init__(self, *args, **kwargs):
+        optparse.OptionParser.__init__(self, *args, **kwargs)
+        self._commands = {}
+        self.min_args, self.max_args = 0, 1
+
+    def add_command(self, name, mod_or_funcs, help=''):
+        """name of the command, name of module or tuple of functions
+        (run, add_options)
+        """
+        assert isinstance(mod_or_funcs, str) or isinstance(mod_or_funcs, tuple), \
+            "mod_or_funcs has to be a module name or a tuple of functions"
+        self._commands[name] = (mod_or_funcs, help)
+
+    def print_main_help(self):
+        optparse.OptionParser.print_help(self)
+        print('\ncommands:')
+        for cmdname, (_, help) in self._commands.items():
+            print('% 10s - %s' % (cmdname, help))
+
+    def parse_command(self, args):
+        if len(args) == 0:
+            self.print_main_help()
+            sys.exit(1)
+        cmd = args[0]
+        args = args[1:]
+        if cmd not in self._commands:
+            if cmd in ('-h', '--help'):
+                self.print_main_help()
+                sys.exit(0)
+            elif self.version is not None and cmd == "--version":
+                self.print_version()
+                sys.exit(0)
+            self.error('unknown command')
+        self.prog = '%s %s' % (self.prog, cmd)
+        mod_or_f, help = self._commands[cmd]
+        # optparse inserts self.description between usage and options help
+        self.description = help
+        if isinstance(mod_or_f, str):
+            exec('from %s import run, add_options' % mod_or_f)
+        else:
+            run, add_options = mod_or_f
+        add_options(self)
+        (options, args) = self.parse_args(args)
+        if not (self.min_args <= len(args) <= self.max_args):
+            self.error('incorrect number of arguments')
+        return run, options, args
+
+
diff --git a/logilab/common/proc.py b/logilab/common/proc.py
new file mode 100644
index 0000000..c27356c
--- /dev/null
+++ b/logilab/common/proc.py
@@ -0,0 +1,277 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""module providing:
+* process information (linux specific: rely on /proc)
+* a class for resource control (memory / time / cpu time)
+
+This module doesn't work on windows platforms (only tested on linux)
+
+:organization: Logilab
+
+
+
+"""
+__docformat__ = "restructuredtext en"
+
+import os
+import stat
+from resource import getrlimit, setrlimit, RLIMIT_CPU, RLIMIT_AS
+from signal import signal, SIGXCPU, SIGKILL, SIGUSR2, SIGUSR1
+from threading import Timer, currentThread, Thread, Event
+from time import time
+
+from logilab.common.tree import Node
+
+class NoSuchProcess(Exception): pass
+
+def proc_exists(pid):
+    """check the a pid is registered in /proc
+    raise NoSuchProcess exception if not
+    """
+    if not os.path.exists('/proc/%s' % pid):
+        raise NoSuchProcess()
+
+PPID = 3
+UTIME = 13
+STIME = 14
+CUTIME = 15
+CSTIME = 16
+VSIZE = 22
+
+class ProcInfo(Node):
+    """provide access to process information found in /proc"""
+
+    def __init__(self, pid):
+        self.pid = int(pid)
+        Node.__init__(self, self.pid)
+        proc_exists(self.pid)
+        self.file = '/proc/%s/stat' % self.pid
+        self.ppid = int(self.status()[PPID])
+
+    def memory_usage(self):
+        """return the memory usage of the process in Ko"""
+        try :
+            return int(self.status()[VSIZE])
+        except IOError:
+            return 0
+
+    def lineage_memory_usage(self):
+        return self.memory_usage() + sum([child.lineage_memory_usage()
+                                          for child in self.children])
+
+    def time(self, children=0):
+        """return the number of jiffies that this process has been scheduled
+        in user and kernel mode"""
+        status = self.status()
+        time = int(status[UTIME]) + int(status[STIME])
+        if children:
+            time += int(status[CUTIME]) + int(status[CSTIME])
+        return time
+
+    def status(self):
+        """return the list of fields found in /proc/<pid>/stat"""
+        return open(self.file).read().split()
+
+    def name(self):
+        """return the process name found in /proc/<pid>/stat
+        """
+        return self.status()[1].strip('()')
+
+    def age(self):
+        """return the age of the process
+        """
+        return os.stat(self.file)[stat.ST_MTIME]
+
+class ProcInfoLoader:
+    """manage process information"""
+
+    def __init__(self):
+        self._loaded = {}
+
+    def list_pids(self):
+        """return a list of existent process ids"""
+        for subdir in os.listdir('/proc'):
+            if subdir.isdigit():
+                yield int(subdir)
+
+    def load(self, pid):
+        """get a ProcInfo object for a given pid"""
+        pid = int(pid)
+        try:
+            return self._loaded[pid]
+        except KeyError:
+            procinfo = ProcInfo(pid)
+            procinfo.manager = self
+            self._loaded[pid] = procinfo
+            return procinfo
+
+
+    def load_all(self):
+        """load all processes information"""
+        for pid in self.list_pids():
+            try:
+                procinfo = self.load(pid)
+                if procinfo.parent is None and procinfo.ppid:
+                    pprocinfo = self.load(procinfo.ppid)
+                    pprocinfo.append(procinfo)
+            except NoSuchProcess:
+                pass
+
+
+try:
+    class ResourceError(BaseException):
+        """Error raise when resource limit is reached"""
+        limit = "Unknown Resource Limit"
+except NameError:
+    class ResourceError(Exception):
+        """Error raise when resource limit is reached"""
+        limit = "Unknown Resource Limit"
+
+
+class XCPUError(ResourceError):
+    """Error raised when CPU Time limit is reached"""
+    limit = "CPU Time"
+
+class LineageMemoryError(ResourceError):
+    """Error raised when the total amount of memory used by a process and
+    it's child is reached"""
+    limit = "Lineage total Memory"
+
+class TimeoutError(ResourceError):
+    """Error raised when the process is running for to much time"""
+    limit = "Real Time"
+
+# Can't use subclass because the StandardError MemoryError raised
+RESOURCE_LIMIT_EXCEPTION = (ResourceError, MemoryError)
+
+
+class MemorySentinel(Thread):
+    """A class checking a process don't use too much memory in a separated
+    daemonic thread
+    """
+    def __init__(self, interval, memory_limit, gpid=os.getpid()):
+        Thread.__init__(self, target=self._run, name="Test.Sentinel")
+        self.memory_limit = memory_limit
+        self._stop = Event()
+        self.interval = interval
+        self.setDaemon(True)
+        self.gpid = gpid
+
+    def stop(self):
+        """stop ap"""
+        self._stop.set()
+
+    def _run(self):
+        pil = ProcInfoLoader()
+        while not self._stop.isSet():
+            if self.memory_limit <= pil.load(self.gpid).lineage_memory_usage():
+                os.killpg(self.gpid, SIGUSR1)
+            self._stop.wait(self.interval)
+
+
+class ResourceController:
+
+    def __init__(self, max_cpu_time=None, max_time=None, max_memory=None,
+                 max_reprieve=60):
+        if SIGXCPU == -1:
+            raise RuntimeError("Unsupported platform")
+        self.max_time = max_time
+        self.max_memory = max_memory
+        self.max_cpu_time = max_cpu_time
+        self._reprieve = max_reprieve
+        self._timer = None
+        self._msentinel = None
+        self._old_max_memory = None
+        self._old_usr1_hdlr = None
+        self._old_max_cpu_time = None
+        self._old_usr2_hdlr = None
+        self._old_sigxcpu_hdlr = None
+        self._limit_set = 0
+        self._abort_try = 0
+        self._start_time = None
+        self._elapse_time = 0
+
+    def _hangle_sig_timeout(self, sig, frame):
+        raise TimeoutError()
+
+    def _hangle_sig_memory(self, sig, frame):
+        if self._abort_try < self._reprieve:
+            self._abort_try += 1
+            raise LineageMemoryError("Memory limit reached")
+        else:
+            os.killpg(os.getpid(), SIGKILL)
+
+    def _handle_sigxcpu(self, sig, frame):
+        if self._abort_try < self._reprieve:
+            self._abort_try += 1
+            raise XCPUError("Soft CPU time limit reached")
+        else:
+            os.killpg(os.getpid(), SIGKILL)
+
+    def _time_out(self):
+        if self._abort_try < self._reprieve:
+            self._abort_try += 1
+            os.killpg(os.getpid(), SIGUSR2)
+            if self._limit_set > 0:
+                self._timer = Timer(1, self._time_out)
+                self._timer.start()
+        else:
+            os.killpg(os.getpid(), SIGKILL)
+
+    def setup_limit(self):
+        """set up the process limit"""
+        assert currentThread().getName() == 'MainThread'
+        os.setpgrp()
+        if self._limit_set <= 0:
+            if self.max_time is not None:
+                self._old_usr2_hdlr = signal(SIGUSR2, self._hangle_sig_timeout)
+                self._timer = Timer(max(1, int(self.max_time) - self._elapse_time),
+                                    self._time_out)
+                self._start_time = int(time())
+                self._timer.start()
+            if self.max_cpu_time is not None:
+                self._old_max_cpu_time = getrlimit(RLIMIT_CPU)
+                cpu_limit = (int(self.max_cpu_time), self._old_max_cpu_time[1])
+                self._old_sigxcpu_hdlr = signal(SIGXCPU, self._handle_sigxcpu)
+                setrlimit(RLIMIT_CPU, cpu_limit)
+            if self.max_memory is not None:
+                self._msentinel = MemorySentinel(1, int(self.max_memory) )
+                self._old_max_memory = getrlimit(RLIMIT_AS)
+                self._old_usr1_hdlr = signal(SIGUSR1, self._hangle_sig_memory)
+                as_limit = (int(self.max_memory), self._old_max_memory[1])
+                setrlimit(RLIMIT_AS, as_limit)
+                self._msentinel.start()
+        self._limit_set += 1
+
+    def clean_limit(self):
+        """reinstall the old process limit"""
+        if self._limit_set > 0:
+            if self.max_time is not None:
+                self._timer.cancel()
+                self._elapse_time += int(time())-self._start_time
+                self._timer = None
+                signal(SIGUSR2, self._old_usr2_hdlr)
+            if self.max_cpu_time is not None:
+                setrlimit(RLIMIT_CPU, self._old_max_cpu_time)
+                signal(SIGXCPU, self._old_sigxcpu_hdlr)
+            if self.max_memory is not None:
+                self._msentinel.stop()
+                self._msentinel = None
+                setrlimit(RLIMIT_AS, self._old_max_memory)
+                signal(SIGUSR1, self._old_usr1_hdlr)
+        self._limit_set -= 1
diff --git a/logilab/common/pyro_ext.py b/logilab/common/pyro_ext.py
new file mode 100644
index 0000000..5204b1b
--- /dev/null
+++ b/logilab/common/pyro_ext.py
@@ -0,0 +1,180 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Python Remote Object utilities
+
+Main functions available:
+
+* `register_object` to expose arbitrary object through pyro using delegation
+  approach and register it in the nameserver.
+* `ns_unregister` unregister an object identifier from the nameserver.
+* `ns_get_proxy` get a pyro proxy from a nameserver object identifier.
+"""
+
+__docformat__ = "restructuredtext en"
+
+import logging
+import tempfile
+
+from Pyro import core, naming, errors, util, config
+
+_LOGGER = logging.getLogger('pyro')
+_MARKER = object()
+
+config.PYRO_STORAGE = tempfile.gettempdir()
+
+def ns_group_and_id(idstr, defaultnsgroup=_MARKER):
+    try:
+        nsgroup, nsid = idstr.rsplit('.', 1)
+    except ValueError:
+        if defaultnsgroup is _MARKER:
+            nsgroup = config.PYRO_NS_DEFAULTGROUP
+        else:
+            nsgroup = defaultnsgroup
+        nsid = idstr
+    if nsgroup is not None and not nsgroup.startswith(':'):
+        nsgroup = ':' + nsgroup
+    return nsgroup, nsid
+
+def host_and_port(hoststr):
+    if not hoststr:
+        return None, None
+    try:
+        hoststr, port = hoststr.split(':')
+    except ValueError:
+        port = None
+    else:
+        port = int(port)
+    return hoststr, port
+
+_DAEMONS = {}
+_PYRO_OBJS = {}
+def _get_daemon(daemonhost, start=True):
+    if not daemonhost in _DAEMONS:
+        if not start:
+            raise Exception('no daemon for %s' % daemonhost)
+        if not _DAEMONS:
+            core.initServer(banner=0)
+        host, port = host_and_port(daemonhost)
+        daemon = core.Daemon(host=host, port=port)
+        _DAEMONS[daemonhost] = daemon
+    return _DAEMONS[daemonhost]
+
+
+def locate_ns(nshost):
+    """locate and return the pyro name server to the daemon"""
+    core.initClient(banner=False)
+    return naming.NameServerLocator().getNS(*host_and_port(nshost))
+
+
+def register_object(object, nsid, defaultnsgroup=_MARKER,
+                    daemonhost=None, nshost=None, use_pyrons=True):
+    """expose the object as a pyro object and register it in the name-server
+
+    if use_pyrons is False, then the object is exposed, but no
+    attempt to register it to a pyro nameserver is made.
+
+    return the pyro daemon object
+    """
+    nsgroup, nsid = ns_group_and_id(nsid, defaultnsgroup)
+    daemon = _get_daemon(daemonhost)
+    if use_pyrons:
+        nsd = locate_ns(nshost)
+        # make sure our namespace group exists
+        try:
+            nsd.createGroup(nsgroup)
+        except errors.NamingError:
+            pass
+        daemon.useNameServer(nsd)
+    # use Delegation approach
+    impl = core.ObjBase()
+    impl.delegateTo(object)
+    qnsid = '%s.%s' % (nsgroup, nsid)
+    uri = daemon.connect(impl, qnsid)
+    _PYRO_OBJS[qnsid] = str(uri)
+    _LOGGER.info('registered %s a pyro object using group %s and id %s',
+                 object, nsgroup, nsid)
+    return daemon
+
+def get_object_uri(qnsid):
+    return _PYRO_OBJS[qnsid]
+
+def ns_unregister(nsid, defaultnsgroup=_MARKER, nshost=None):
+    """unregister the object with the given nsid from the pyro name server"""
+    nsgroup, nsid = ns_group_and_id(nsid, defaultnsgroup)
+    try:
+        nsd = locate_ns(nshost)
+    except errors.PyroError as ex:
+        # name server not responding
+        _LOGGER.error('can\'t locate pyro name server: %s', ex)
+    else:
+        try:
+            nsd.unregister('%s.%s' % (nsgroup, nsid))
+            _LOGGER.info('%s unregistered from pyro name server', nsid)
+        except errors.NamingError:
+            _LOGGER.warning('%s not registered in pyro name server', nsid)
+
+
+def ns_reregister(nsid, defaultnsgroup=_MARKER, nshost=None):
+    """reregister a pyro object into the name server. You only have to specify
+    the name-server id of the object (though you MUST have gone through
+    `register_object` for the given object previously).
+
+    This is especially useful for long running server while the name server may
+    have been restarted, and its records lost.
+    """
+    nsgroup, nsid = ns_group_and_id(nsid, defaultnsgroup)
+    qnsid = '%s.%s' % (nsgroup, nsid)
+    nsd = locate_ns(nshost)
+    try:
+        nsd.unregister(qnsid)
+    except errors.NamingError:
+        # make sure our namespace group exists
+        try:
+            nsd.createGroup(nsgroup)
+        except errors.NamingError:
+            pass
+    nsd.register(qnsid, _PYRO_OBJS[qnsid])
+
+def ns_get_proxy(nsid, defaultnsgroup=_MARKER, nshost=None):
+    """
+    if nshost is None, the nameserver is found by a broadcast.
+    """
+    # resolve the Pyro object
+    nsgroup, nsid = ns_group_and_id(nsid, defaultnsgroup)
+    try:
+        nsd = locate_ns(nshost)
+        pyrouri = nsd.resolve('%s.%s' % (nsgroup, nsid))
+    except errors.ProtocolError as ex:
+        raise errors.PyroError(
+            'Could not connect to the Pyro name server (host: %s)' % nshost)
+    except errors.NamingError:
+        raise errors.PyroError(
+            'Could not get proxy for %s (not registered in Pyro), '
+            'you may have to restart your server-side application' % nsid)
+    return core.getProxyForURI(pyrouri)
+
+def get_proxy(pyro_uri):
+    """get a proxy for the passed pyro uri without using a nameserver
+    """
+    return core.getProxyForURI(pyro_uri)
+
+def set_pyro_log_threshold(level):
+    pyrologger = logging.getLogger('Pyro.%s' % str(id(util.Log)))
+    # remove handlers so only the root handler is used
+    pyrologger.handlers = []
+    pyrologger.setLevel(level)
diff --git a/logilab/common/pytest.py b/logilab/common/pytest.py
new file mode 100644
index 0000000..58515a9
--- /dev/null
+++ b/logilab/common/pytest.py
@@ -0,0 +1,1199 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""pytest is a tool that eases test running and debugging.
+
+To be able to use pytest, you should either write tests using
+the logilab.common.testlib's framework or the unittest module of the
+Python's standard library.
+
+You can customize pytest's behaviour by defining a ``pytestconf.py`` file
+somewhere in your test directory. In this file, you can add options or
+change the way tests are run.
+
+To add command line options, you must define a ``update_parser`` function in
+your ``pytestconf.py`` file. The function must accept a single parameter
+that will be the OptionParser's instance to customize.
+
+If you wish to customize the tester, you'll have to define a class named
+``CustomPyTester``. This class should extend the default `PyTester` class
+defined in the pytest module. Take a look at the `PyTester` and `DjangoTester`
+classes for more information about what can be done.
+
+For instance, if you wish to add a custom -l option to specify a loglevel, you
+could define the following ``pytestconf.py`` file ::
+
+    import logging
+    from logilab.common.pytest import PyTester
+
+    def update_parser(parser):
+        parser.add_option('-l', '--loglevel', dest='loglevel', action='store',
+                          choices=('debug', 'info', 'warning', 'error', 'critical'),
+                          default='critical', help="the default log level possible choices are "
+                          "('debug', 'info', 'warning', 'error', 'critical')")
+        return parser
+
+
+    class CustomPyTester(PyTester):
+        def __init__(self, cvg, options):
+            super(CustomPyTester, self).__init__(cvg, options)
+            loglevel = options.loglevel.upper()
+            logger = logging.getLogger('erudi')
+            logger.setLevel(logging.getLevelName(loglevel))
+
+
+In your TestCase class you can then get the value of a specific option with
+the ``optval`` method::
+
+    class MyTestCase(TestCase):
+        def test_foo(self):
+            loglevel = self.optval('loglevel')
+            # ...
+
+
+You can also tag your tag your test for fine filtering
+
+With those tag::
+
+    from logilab.common.testlib import tag, TestCase
+
+    class Exemple(TestCase):
+
+        @tag('rouge', 'carre')
+        def toto(self):
+            pass
+
+        @tag('carre', 'vert')
+        def tata(self):
+            pass
+
+        @tag('rouge')
+        def titi(test):
+            pass
+
+you can filter the function with a simple python expression
+
+ * ``toto`` and ``titi`` match ``rouge``
+ * ``toto``, ``tata`` and ``titi``, match ``rouge or carre``
+ * ``tata`` and ``titi`` match``rouge ^ carre``
+ * ``titi`` match ``rouge and not carre``
+"""
+
+from __future__ import print_function
+
+__docformat__ = "restructuredtext en"
+
+PYTEST_DOC = """%prog [OPTIONS] [testfile [testpattern]]
+
+examples:
+
+pytest path/to/mytests.py
+pytest path/to/mytests.py TheseTests
+pytest path/to/mytests.py TheseTests.test_thisone
+pytest path/to/mytests.py -m '(not long and database) or regr'
+
+pytest one (will run both test_thisone and test_thatone)
+pytest path/to/mytests.py -s not (will skip test_notthisone)
+"""
+
+ENABLE_DBC = False
+FILE_RESTART = ".pytest.restart"
+
+import os, sys, re
+import os.path as osp
+from time import time, clock
+import warnings
+import types
+from inspect import isgeneratorfunction, isclass
+from contextlib import contextmanager
+
+from logilab.common.fileutils import abspath_listdir
+from logilab.common import textutils
+from logilab.common import testlib, STD_BLACKLIST
+# use the same unittest module as testlib
+from logilab.common.testlib import unittest, start_interactive_mode
+from logilab.common.deprecation import deprecated
+import doctest
+
+import unittest as unittest_legacy
+if not getattr(unittest_legacy, "__package__", None):
+    try:
+        import unittest2.suite as unittest_suite
+    except ImportError:
+        sys.exit("You have to install python-unittest2 to use this module")
+else:
+    import unittest.suite as unittest_suite
+
+try:
+    import django
+    from logilab.common.modutils import modpath_from_file, load_module_from_modpath
+    DJANGO_FOUND = True
+except ImportError:
+    DJANGO_FOUND = False
+
+CONF_FILE = 'pytestconf.py'
+
+## coverage pausing tools
+
+@contextmanager
+def replace_trace(trace=None):
+    """A context manager that temporary replaces the trace function"""
+    oldtrace = sys.gettrace()
+    sys.settrace(trace)
+    try:
+        yield
+    finally:
+        # specific hack to work around a bug in pycoverage, see
+        # https://bitbucket.org/ned/coveragepy/issue/123
+        if (oldtrace is not None and not callable(oldtrace) and
+            hasattr(oldtrace, 'pytrace')):
+            oldtrace = oldtrace.pytrace
+        sys.settrace(oldtrace)
+
+
+def pause_trace():
+    """A context manager that temporary pauses any tracing"""
+    return replace_trace()
+
+class TraceController(object):
+    ctx_stack = []
+
+    @classmethod
+    @deprecated('[lgc 0.63.1] Use the pause_trace() context manager')
+    def pause_tracing(cls):
+        cls.ctx_stack.append(pause_trace())
+        cls.ctx_stack[-1].__enter__()
+
+    @classmethod
+    @deprecated('[lgc 0.63.1] Use the pause_trace() context manager')
+    def resume_tracing(cls):
+        cls.ctx_stack.pop().__exit__(None, None, None)
+
+
+pause_tracing = TraceController.pause_tracing
+resume_tracing = TraceController.resume_tracing
+
+
+def nocoverage(func):
+    """Function decorator that pauses tracing functions"""
+    if hasattr(func, 'uncovered'):
+        return func
+    func.uncovered = True
+
+    def not_covered(*args, **kwargs):
+        with pause_trace():
+            return func(*args, **kwargs)
+    not_covered.uncovered = True
+    return not_covered
+
+## end of coverage pausing tools
+
+
+TESTFILE_RE = re.compile("^((unit)?test.*|smoketest)\.py$")
+def this_is_a_testfile(filename):
+    """returns True if `filename` seems to be a test file"""
+    return TESTFILE_RE.match(osp.basename(filename))
+
+TESTDIR_RE = re.compile("^(unit)?tests?$")
+def this_is_a_testdir(dirpath):
+    """returns True if `filename` seems to be a test directory"""
+    return TESTDIR_RE.match(osp.basename(dirpath))
+
+
+def load_pytest_conf(path, parser):
+    """loads a ``pytestconf.py`` file and update default parser
+    and / or tester.
+    """
+    namespace = {}
+    exec(open(path, 'rb').read(), namespace)
+    if 'update_parser' in namespace:
+        namespace['update_parser'](parser)
+    return namespace.get('CustomPyTester', PyTester)
+
+
+def project_root(parser, projdir=os.getcwd()):
+    """try to find project's root and add it to sys.path"""
+    previousdir = curdir = osp.abspath(projdir)
+    testercls = PyTester
+    conf_file_path = osp.join(curdir, CONF_FILE)
+    if osp.isfile(conf_file_path):
+        testercls = load_pytest_conf(conf_file_path, parser)
+    while this_is_a_testdir(curdir) or \
+              osp.isfile(osp.join(curdir, '__init__.py')):
+        newdir = osp.normpath(osp.join(curdir, os.pardir))
+        if newdir == curdir:
+            break
+        previousdir = curdir
+        curdir = newdir
+        conf_file_path = osp.join(curdir, CONF_FILE)
+        if osp.isfile(conf_file_path):
+            testercls = load_pytest_conf(conf_file_path, parser)
+    return previousdir, testercls
+
+
+class GlobalTestReport(object):
+    """this class holds global test statistics"""
+    def __init__(self):
+        self.ran = 0
+        self.skipped = 0
+        self.failures = 0
+        self.errors = 0
+        self.ttime = 0
+        self.ctime = 0
+        self.modulescount = 0
+        self.errmodules = []
+
+    def feed(self, filename, testresult, ttime, ctime):
+        """integrates new test information into internal statistics"""
+        ran = testresult.testsRun
+        self.ran += ran
+        self.skipped += len(getattr(testresult, 'skipped', ()))
+        self.failures += len(testresult.failures)
+        self.errors += len(testresult.errors)
+        self.ttime += ttime
+        self.ctime += ctime
+        self.modulescount += 1
+        if not testresult.wasSuccessful():
+            problems = len(testresult.failures) + len(testresult.errors)
+            self.errmodules.append((filename[:-3], problems, ran))
+
+    def failed_to_test_module(self, filename):
+        """called when the test module could not be imported by unittest
+        """
+        self.errors += 1
+        self.modulescount += 1
+        self.ran += 1
+        self.errmodules.append((filename[:-3], 1, 1))
+
+    def skip_module(self, filename):
+        self.modulescount += 1
+        self.ran += 1
+        self.errmodules.append((filename[:-3], 0, 0))
+
+    def __str__(self):
+        """this is just presentation stuff"""
+        line1 = ['Ran %s test cases in %.2fs (%.2fs CPU)'
+                 % (self.ran, self.ttime, self.ctime)]
+        if self.errors:
+            line1.append('%s errors' % self.errors)
+        if self.failures:
+            line1.append('%s failures' % self.failures)
+        if self.skipped:
+            line1.append('%s skipped' % self.skipped)
+        modulesok = self.modulescount - len(self.errmodules)
+        if self.errors or self.failures:
+            line2 = '%s modules OK (%s failed)' % (modulesok,
+                                                   len(self.errmodules))
+            descr = ', '.join(['%s [%s/%s]' % info for info in self.errmodules])
+            line3 = '\nfailures: %s' % descr
+        elif modulesok:
+            line2 = 'All %s modules OK' % modulesok
+            line3 = ''
+        else:
+            return ''
+        return '%s\n%s%s' % (', '.join(line1), line2, line3)
+
+
+
+def remove_local_modules_from_sys(testdir):
+    """remove all modules from cache that come from `testdir`
+
+    This is used to avoid strange side-effects when using the
+    testall() mode of pytest.
+    For instance, if we run pytest on this tree::
+
+      A/test/test_utils.py
+      B/test/test_utils.py
+
+    we **have** to clean sys.modules to make sure the correct test_utils
+    module is ran in B
+    """
+    for modname, mod in list(sys.modules.items()):
+        if mod is None:
+            continue
+        if not hasattr(mod, '__file__'):
+            # this is the case of some built-in modules like sys, imp, marshal
+            continue
+        modfile = mod.__file__
+        # if modfile is not an absolute path, it was probably loaded locally
+        # during the tests
+        if not osp.isabs(modfile) or modfile.startswith(testdir):
+            del sys.modules[modname]
+
+
+
+class PyTester(object):
+    """encapsulates testrun logic"""
+
+    def __init__(self, cvg, options):
+        self.report = GlobalTestReport()
+        self.cvg = cvg
+        self.options = options
+        self.firstwrite = True
+        self._errcode = None
+
+    def show_report(self):
+        """prints the report and returns appropriate exitcode"""
+        # everything has been ran, print report
+        print("*" * 79)
+        print(self.report)
+
+    def get_errcode(self):
+        # errcode set explicitly
+        if self._errcode is not None:
+            return self._errcode
+        return self.report.failures + self.report.errors
+
+    def set_errcode(self, errcode):
+        self._errcode = errcode
+    errcode = property(get_errcode, set_errcode)
+
+    def testall(self, exitfirst=False):
+        """walks through current working directory, finds something
+        which can be considered as a testdir and runs every test there
+        """
+        here = os.getcwd()
+        for dirname, dirs, _ in os.walk(here):
+            for skipped in STD_BLACKLIST:
+                if skipped in dirs:
+                    dirs.remove(skipped)
+            basename = osp.basename(dirname)
+            if this_is_a_testdir(basename):
+                print("going into", dirname)
+                # we found a testdir, let's explore it !
+                if not self.testonedir(dirname, exitfirst):
+                    break
+                dirs[:] = []
+        if self.report.ran == 0:
+            print("no test dir found testing here:", here)
+            # if no test was found during the visit, consider
+            # the local directory as a test directory even if
+            # it doesn't have a traditional test directory name
+            self.testonedir(here)
+
+    def testonedir(self, testdir, exitfirst=False):
+        """finds each testfile in the `testdir` and runs it
+
+        return true when all tests has been executed, false if exitfirst and
+        some test has failed.
+        """
+        for filename in abspath_listdir(testdir):
+            if this_is_a_testfile(filename):
+                if self.options.exitfirst and not self.options.restart:
+                    # overwrite restart file
+                    try:
+                        restartfile = open(FILE_RESTART, "w")
+                        restartfile.close()
+                    except Exception:
+                        print("Error while overwriting succeeded test file :",
+                              osp.join(os.getcwd(), FILE_RESTART),
+                              file=sys.__stderr__)
+                        raise
+                # run test and collect information
+                prog = self.testfile(filename, batchmode=True)
+                if exitfirst and (prog is None or not prog.result.wasSuccessful()):
+                    return False
+                self.firstwrite = True
+        # clean local modules
+        remove_local_modules_from_sys(testdir)
+        return True
+
+    def testfile(self, filename, batchmode=False):
+        """runs every test in `filename`
+
+        :param filename: an absolute path pointing to a unittest file
+        """
+        here = os.getcwd()
+        dirname = osp.dirname(filename)
+        if dirname:
+            os.chdir(dirname)
+        # overwrite restart file if it has not been done already
+        if self.options.exitfirst and not self.options.restart and self.firstwrite:
+            try:
+                restartfile = open(FILE_RESTART, "w")
+                restartfile.close()
+            except Exception:
+                print("Error while overwriting succeeded test file :",
+                      osp.join(os.getcwd(), FILE_RESTART), file=sys.__stderr__)
+                raise
+        modname = osp.basename(filename)[:-3]
+        print(('  %s  ' % osp.basename(filename)).center(70, '='),
+              file=sys.__stderr__)
+        try:
+            tstart, cstart = time(), clock()
+            try:
+                testprog = SkipAwareTestProgram(modname, batchmode=batchmode, cvg=self.cvg,
+                                                 options=self.options, outstream=sys.stderr)
+            except KeyboardInterrupt:
+                raise
+            except SystemExit as exc:
+                self.errcode = exc.code
+                raise
+            except testlib.SkipTest:
+                print("Module skipped:", filename)
+                self.report.skip_module(filename)
+                return None
+            except Exception:
+                self.report.failed_to_test_module(filename)
+                print('unhandled exception occurred while testing', modname,
+                      file=sys.stderr)
+                import traceback
+                traceback.print_exc(file=sys.stderr)
+                return None
+
+            tend, cend = time(), clock()
+            ttime, ctime = (tend - tstart), (cend - cstart)
+            self.report.feed(filename, testprog.result, ttime, ctime)
+            return testprog
+        finally:
+            if dirname:
+                os.chdir(here)
+
+
+
+class DjangoTester(PyTester):
+
+    def load_django_settings(self, dirname):
+        """try to find project's setting and load it"""
+        curdir = osp.abspath(dirname)
+        previousdir = curdir
+        while not osp.isfile(osp.join(curdir, 'settings.py')) and \
+                  osp.isfile(osp.join(curdir, '__init__.py')):
+            newdir = osp.normpath(osp.join(curdir, os.pardir))
+            if newdir == curdir:
+                raise AssertionError('could not find settings.py')
+            previousdir = curdir
+            curdir = newdir
+        # late django initialization
+        settings = load_module_from_modpath(modpath_from_file(osp.join(curdir, 'settings.py')))
+        from django.core.management import setup_environ
+        setup_environ(settings)
+        settings.DEBUG = False
+        self.settings = settings
+        # add settings dir to pythonpath since it's the project's root
+        if curdir not in sys.path:
+            sys.path.insert(1, curdir)
+
+    def before_testfile(self):
+        # Those imports must be done **after** setup_environ was called
+        from django.test.utils import setup_test_environment
+        from django.test.utils import create_test_db
+        setup_test_environment()
+        create_test_db(verbosity=0)
+        self.dbname = self.settings.TEST_DATABASE_NAME
+
+    def after_testfile(self):
+        # Those imports must be done **after** setup_environ was called
+        from django.test.utils import teardown_test_environment
+        from django.test.utils import destroy_test_db
+        teardown_test_environment()
+        print('destroying', self.dbname)
+        destroy_test_db(self.dbname, verbosity=0)
+
+    def testall(self, exitfirst=False):
+        """walks through current working directory, finds something
+        which can be considered as a testdir and runs every test there
+        """
+        for dirname, dirs, files in os.walk(os.getcwd()):
+            for skipped in ('CVS', '.svn', '.hg'):
+                if skipped in dirs:
+                    dirs.remove(skipped)
+            if 'tests.py' in files:
+                if not self.testonedir(dirname, exitfirst):
+                    break
+                dirs[:] = []
+            else:
+                basename = osp.basename(dirname)
+                if basename in ('test', 'tests'):
+                    print("going into", dirname)
+                    # we found a testdir, let's explore it !
+                    if not self.testonedir(dirname, exitfirst):
+                        break
+                    dirs[:] = []
+
+    def testonedir(self, testdir, exitfirst=False):
+        """finds each testfile in the `testdir` and runs it
+
+        return true when all tests has been executed, false if exitfirst and
+        some test has failed.
+        """
+        # special django behaviour : if tests are splitted in several files,
+        # remove the main tests.py file and tests each test file separately
+        testfiles = [fpath for fpath in abspath_listdir(testdir)
+                     if this_is_a_testfile(fpath)]
+        if len(testfiles) > 1:
+            try:
+                testfiles.remove(osp.join(testdir, 'tests.py'))
+            except ValueError:
+                pass
+        for filename in testfiles:
+            # run test and collect information
+            prog = self.testfile(filename, batchmode=True)
+            if exitfirst and (prog is None or not prog.result.wasSuccessful()):
+                return False
+        # clean local modules
+        remove_local_modules_from_sys(testdir)
+        return True
+
+    def testfile(self, filename, batchmode=False):
+        """runs every test in `filename`
+
+        :param filename: an absolute path pointing to a unittest file
+        """
+        here = os.getcwd()
+        dirname = osp.dirname(filename)
+        if dirname:
+            os.chdir(dirname)
+        self.load_django_settings(dirname)
+        modname = osp.basename(filename)[:-3]
+        print(('  %s  ' % osp.basename(filename)).center(70, '='),
+              file=sys.stderr)
+        try:
+            try:
+                tstart, cstart = time(), clock()
+                self.before_testfile()
+                testprog = SkipAwareTestProgram(modname, batchmode=batchmode, cvg=self.cvg)
+                tend, cend = time(), clock()
+                ttime, ctime = (tend - tstart), (cend - cstart)
+                self.report.feed(filename, testprog.result, ttime, ctime)
+                return testprog
+            except SystemExit:
+                raise
+            except Exception as exc:
+                import traceback
+                traceback.print_exc()
+                self.report.failed_to_test_module(filename)
+                print('unhandled exception occurred while testing', modname)
+                print('error: %s' % exc)
+                return None
+        finally:
+            self.after_testfile()
+            if dirname:
+                os.chdir(here)
+
+
+def make_parser():
+    """creates the OptionParser instance
+    """
+    from optparse import OptionParser
+    parser = OptionParser(usage=PYTEST_DOC)
+
+    parser.newargs = []
+    def rebuild_cmdline(option, opt, value, parser):
+        """carry the option to unittest_main"""
+        parser.newargs.append(opt)
+
+    def rebuild_and_store(option, opt, value, parser):
+        """carry the option to unittest_main and store
+        the value on current parser
+        """
+        parser.newargs.append(opt)
+        setattr(parser.values, option.dest, True)
+
+    def capture_and_rebuild(option, opt, value, parser):
+        warnings.simplefilter('ignore', DeprecationWarning)
+        rebuild_cmdline(option, opt, value, parser)
+
+    # pytest options
+    parser.add_option('-t', dest='testdir', default=None,
+                      help="directory where the tests will be found")
+    parser.add_option('-d', dest='dbc', default=False,
+                      action="store_true", help="enable design-by-contract")
+    # unittest_main options provided and passed through pytest
+    parser.add_option('-v', '--verbose', callback=rebuild_cmdline,
+                      action="callback", help="Verbose output")
+    parser.add_option('-i', '--pdb', callback=rebuild_and_store,
+                      dest="pdb", action="callback",
+                      help="Enable test failure inspection")
+    parser.add_option('-x', '--exitfirst', callback=rebuild_and_store,
+                      dest="exitfirst", default=False,
+                      action="callback", help="Exit on first failure "
+                      "(only make sense when pytest run one test file)")
+    parser.add_option('-R', '--restart', callback=rebuild_and_store,
+                      dest="restart", default=False,
+                      action="callback",
+                      help="Restart tests from where it failed (implies exitfirst) "
+                        "(only make sense if tests previously ran with exitfirst only)")
+    parser.add_option('--color', callback=rebuild_cmdline,
+                      action="callback",
+                      help="colorize tracebacks")
+    parser.add_option('-s', '--skip',
+                      # XXX: I wish I could use the callback action but it
+                      #      doesn't seem to be able to get the value
+                      #      associated to the option
+                      action="store", dest="skipped", default=None,
+                      help="test names matching this name will be skipped "
+                      "to skip several patterns, use commas")
+    parser.add_option('-q', '--quiet', callback=rebuild_cmdline,
+                      action="callback", help="Minimal output")
+    parser.add_option('-P', '--profile', default=None, dest='profile',
+                      help="Profile execution and store data in the given file")
+    parser.add_option('-m', '--match', default=None, dest='tags_pattern',
+                      help="only execute test whose tag match the current pattern")
+
+    if DJANGO_FOUND:
+        parser.add_option('-J', '--django', dest='django', default=False,
+                          action="store_true",
+                          help='use pytest for django test cases')
+    return parser
+
+
+def parseargs(parser):
+    """Parse the command line and return (options processed), (options to pass to
+    unittest_main()), (explicitfile or None).
+    """
+    # parse the command line
+    options, args = parser.parse_args()
+    filenames = [arg for arg in args if arg.endswith('.py')]
+    if filenames:
+        if len(filenames) > 1:
+            parser.error("only one filename is acceptable")
+        explicitfile = filenames[0]
+        args.remove(explicitfile)
+    else:
+        explicitfile = None
+    # someone wants DBC
+    testlib.ENABLE_DBC = options.dbc
+    newargs = parser.newargs
+    if options.skipped:
+        newargs.extend(['--skip', options.skipped])
+    # restart implies exitfirst
+    if options.restart:
+        options.exitfirst = True
+    # append additional args to the new sys.argv and let unittest_main
+    # do the rest
+    newargs += args
+    return options, explicitfile
+
+
+
+def run():
+    parser = make_parser()
+    rootdir, testercls = project_root(parser)
+    options, explicitfile = parseargs(parser)
+    # mock a new command line
+    sys.argv[1:] = parser.newargs
+    cvg = None
+    if not '' in sys.path:
+        sys.path.insert(0, '')
+    if DJANGO_FOUND and options.django:
+        tester = DjangoTester(cvg, options)
+    else:
+        tester = testercls(cvg, options)
+    if explicitfile:
+        cmd, args = tester.testfile, (explicitfile,)
+    elif options.testdir:
+        cmd, args = tester.testonedir, (options.testdir, options.exitfirst)
+    else:
+        cmd, args = tester.testall, (options.exitfirst,)
+    try:
+        try:
+            if options.profile:
+                import hotshot
+                prof = hotshot.Profile(options.profile)
+                prof.runcall(cmd, *args)
+                prof.close()
+                print('profile data saved in', options.profile)
+            else:
+                cmd(*args)
+        except SystemExit:
+            raise
+        except:
+            import traceback
+            traceback.print_exc()
+    finally:
+        tester.show_report()
+        sys.exit(tester.errcode)
+
+class SkipAwareTestProgram(unittest.TestProgram):
+    # XXX: don't try to stay close to unittest.py, use optparse
+    USAGE = """\
+Usage: %(progName)s [options] [test] [...]
+
+Options:
+  -h, --help       Show this message
+  -v, --verbose    Verbose output
+  -i, --pdb        Enable test failure inspection
+  -x, --exitfirst  Exit on first failure
+  -s, --skip       skip test matching this pattern (no regexp for now)
+  -q, --quiet      Minimal output
+  --color          colorize tracebacks
+
+  -m, --match      Run only test whose tag match this pattern
+
+  -P, --profile    FILE: Run the tests using cProfile and saving results
+                   in FILE
+
+Examples:
+  %(progName)s                               - run default set of tests
+  %(progName)s MyTestSuite                   - run suite 'MyTestSuite'
+  %(progName)s MyTestCase.testSomething      - run MyTestCase.testSomething
+  %(progName)s MyTestCase                    - run all 'test*' test methods
+                                               in MyTestCase
+"""
+    def __init__(self, module='__main__', defaultTest=None, batchmode=False,
+                 cvg=None, options=None, outstream=sys.stderr):
+        self.batchmode = batchmode
+        self.cvg = cvg
+        self.options = options
+        self.outstream = outstream
+        super(SkipAwareTestProgram, self).__init__(
+            module=module, defaultTest=defaultTest,
+            testLoader=NonStrictTestLoader())
+
+    def parseArgs(self, argv):
+        self.pdbmode = False
+        self.exitfirst = False
+        self.skipped_patterns = []
+        self.test_pattern = None
+        self.tags_pattern = None
+        self.colorize = False
+        self.profile_name = None
+        import getopt
+        try:
+            options, args = getopt.getopt(argv[1:], 'hHvixrqcp:s:m:P:',
+                                          ['help', 'verbose', 'quiet', 'pdb',
+                                           'exitfirst', 'restart',
+                                           'skip=', 'color', 'match=', 'profile='])
+            for opt, value in options:
+                if opt in ('-h', '-H', '--help'):
+                    self.usageExit()
+                if opt in ('-i', '--pdb'):
+                    self.pdbmode = True
+                if opt in ('-x', '--exitfirst'):
+                    self.exitfirst = True
+                if opt in ('-r', '--restart'):
+                    self.restart = True
+                    self.exitfirst = True
+                if opt in ('-q', '--quiet'):
+                    self.verbosity = 0
+                if opt in ('-v', '--verbose'):
+                    self.verbosity = 2
+                if opt in ('-s', '--skip'):
+                    self.skipped_patterns = [pat.strip() for pat in
+                                             value.split(', ')]
+                if opt == '--color':
+                    self.colorize = True
+                if opt in ('-m', '--match'):
+                    #self.tags_pattern = value
+                    self.options["tag_pattern"] = value
+                if opt in ('-P', '--profile'):
+                    self.profile_name = value
+            self.testLoader.skipped_patterns = self.skipped_patterns
+            if len(args) == 0 and self.defaultTest is None:
+                suitefunc = getattr(self.module, 'suite', None)
+                if isinstance(suitefunc, (types.FunctionType,
+                        types.MethodType)):
+                    self.test = self.module.suite()
+                else:
+                    self.test = self.testLoader.loadTestsFromModule(self.module)
+                return
+            if len(args) > 0:
+                self.test_pattern = args[0]
+                self.testNames = args
+            else:
+                self.testNames = (self.defaultTest, )
+            self.createTests()
+        except getopt.error as msg:
+            self.usageExit(msg)
+
+    def runTests(self):
+        if self.profile_name:
+            import cProfile
+            cProfile.runctx('self._runTests()', globals(), locals(), self.profile_name )
+        else:
+            return self._runTests()
+
+    def _runTests(self):
+        self.testRunner = SkipAwareTextTestRunner(verbosity=self.verbosity,
+                                                  stream=self.outstream,
+                                                  exitfirst=self.exitfirst,
+                                                  pdbmode=self.pdbmode,
+                                                  cvg=self.cvg,
+                                                  test_pattern=self.test_pattern,
+                                                  skipped_patterns=self.skipped_patterns,
+                                                  colorize=self.colorize,
+                                                  batchmode=self.batchmode,
+                                                  options=self.options)
+
+        def removeSucceededTests(obj, succTests):
+            """ Recursive function that removes succTests from
+            a TestSuite or TestCase
+            """
+            if isinstance(obj, unittest.TestSuite):
+                removeSucceededTests(obj._tests, succTests)
+            if isinstance(obj, list):
+                for el in obj[:]:
+                    if isinstance(el, unittest.TestSuite):
+                        removeSucceededTests(el, succTests)
+                    elif isinstance(el, unittest.TestCase):
+                        descr = '.'.join((el.__class__.__module__,
+                                el.__class__.__name__,
+                                el._testMethodName))
+                        if descr in succTests:
+                            obj.remove(el)
+        # take care, self.options may be None
+        if getattr(self.options, 'restart', False):
+            # retrieve succeeded tests from FILE_RESTART
+            try:
+                restartfile = open(FILE_RESTART, 'r')
+                try:
+                    succeededtests = list(elem.rstrip('\n\r') for elem in
+                                          restartfile.readlines())
+                    removeSucceededTests(self.test, succeededtests)
+                finally:
+                    restartfile.close()
+            except Exception as ex:
+                raise Exception("Error while reading succeeded tests into %s: %s"
+                                % (osp.join(os.getcwd(), FILE_RESTART), ex))
+
+        result = self.testRunner.run(self.test)
+        # help garbage collection: we want TestSuite, which hold refs to every
+        # executed TestCase, to be gc'ed
+        del self.test
+        if getattr(result, "debuggers", None) and \
+           getattr(self, "pdbmode", None):
+            start_interactive_mode(result)
+        if not getattr(self, "batchmode", None):
+            sys.exit(not result.wasSuccessful())
+        self.result = result
+
+
+class SkipAwareTextTestRunner(unittest.TextTestRunner):
+
+    def __init__(self, stream=sys.stderr, verbosity=1,
+                 exitfirst=False, pdbmode=False, cvg=None, test_pattern=None,
+                 skipped_patterns=(), colorize=False, batchmode=False,
+                 options=None):
+        super(SkipAwareTextTestRunner, self).__init__(stream=stream,
+                                                      verbosity=verbosity)
+        self.exitfirst = exitfirst
+        self.pdbmode = pdbmode
+        self.cvg = cvg
+        self.test_pattern = test_pattern
+        self.skipped_patterns = skipped_patterns
+        self.colorize = colorize
+        self.batchmode = batchmode
+        self.options = options
+
+    def _this_is_skipped(self, testedname):
+        return any([(pat in testedname) for pat in self.skipped_patterns])
+
+    def _runcondition(self, test, skipgenerator=True):
+        if isinstance(test, testlib.InnerTest):
+            testname = test.name
+        else:
+            if isinstance(test, testlib.TestCase):
+                meth = test._get_test_method()
+                testname = '%s.%s' % (test.__name__, meth.__name__)
+            elif isinstance(test, types.FunctionType):
+                func = test
+                testname = func.__name__
+            elif isinstance(test, types.MethodType):
+                cls = test.__self__.__class__
+                testname = '%s.%s' % (cls.__name__, test.__name__)
+            else:
+                return True # Not sure when this happens
+            if isgeneratorfunction(test) and skipgenerator:
+                return self.does_match_tags(test) # Let inner tests decide at run time
+        if self._this_is_skipped(testname):
+            return False # this was explicitly skipped
+        if self.test_pattern is not None:
+            try:
+                classpattern, testpattern = self.test_pattern.split('.')
+                klass, name = testname.split('.')
+                if classpattern not in klass or testpattern not in name:
+                    return False
+            except ValueError:
+                if self.test_pattern not in testname:
+                    return False
+
+        return self.does_match_tags(test)
+
+    def does_match_tags(self, test):
+        if self.options is not None:
+            tags_pattern = getattr(self.options, 'tags_pattern', None)
+            if tags_pattern is not None:
+                tags = getattr(test, 'tags', testlib.Tags())
+                if tags.inherit and isinstance(test, types.MethodType):
+                    tags = tags | getattr(test.im_class, 'tags', testlib.Tags())
+                return tags.match(tags_pattern)
+        return True # no pattern
+
+    def _makeResult(self):
+        return testlib.SkipAwareTestResult(self.stream, self.descriptions,
+                                   self.verbosity, self.exitfirst,
+                                   self.pdbmode, self.cvg, self.colorize)
+
+    def run(self, test):
+        "Run the given test case or test suite."
+        result = self._makeResult()
+        startTime = time()
+        test(result, runcondition=self._runcondition, options=self.options)
+        stopTime = time()
+        timeTaken = stopTime - startTime
+        result.printErrors()
+        if not self.batchmode:
+            self.stream.writeln(result.separator2)
+            run = result.testsRun
+            self.stream.writeln("Ran %d test%s in %.3fs" %
+                                (run, run != 1 and "s" or "", timeTaken))
+            self.stream.writeln()
+            if not result.wasSuccessful():
+                if self.colorize:
+                    self.stream.write(textutils.colorize_ansi("FAILED", color='red'))
+                else:
+                    self.stream.write("FAILED")
+            else:
+                if self.colorize:
+                    self.stream.write(textutils.colorize_ansi("OK", color='green'))
+                else:
+                    self.stream.write("OK")
+            failed, errored, skipped = map(len, (result.failures,
+                                                 result.errors,
+                                                 result.skipped))
+
+            det_results = []
+            for name, value in (("failures", result.failures),
+                                ("errors",result.errors),
+                                ("skipped", result.skipped)):
+                if value:
+                    det_results.append("%s=%i" % (name, len(value)))
+            if det_results:
+                self.stream.write(" (")
+                self.stream.write(', '.join(det_results))
+                self.stream.write(")")
+            self.stream.writeln("")
+        return result
+
+class NonStrictTestLoader(unittest.TestLoader):
+    """
+    Overrides default testloader to be able to omit classname when
+    specifying tests to run on command line.
+
+    For example, if the file test_foo.py contains ::
+
+        class FooTC(TestCase):
+            def test_foo1(self): # ...
+            def test_foo2(self): # ...
+            def test_bar1(self): # ...
+
+        class BarTC(TestCase):
+            def test_bar2(self): # ...
+
+    'python test_foo.py' will run the 3 tests in FooTC
+    'python test_foo.py FooTC' will run the 3 tests in FooTC
+    'python test_foo.py test_foo' will run test_foo1 and test_foo2
+    'python test_foo.py test_foo1' will run test_foo1
+    'python test_foo.py test_bar' will run FooTC.test_bar1 and BarTC.test_bar2
+    """
+
+    def __init__(self):
+        self.skipped_patterns = ()
+
+    # some magic here to accept empty list by extending
+    # and to provide callable capability
+    def loadTestsFromNames(self, names, module=None):
+        suites = []
+        for name in names:
+            suites.extend(self.loadTestsFromName(name, module))
+        return self.suiteClass(suites)
+
+    def _collect_tests(self, module):
+        tests = {}
+        for obj in vars(module).values():
+            if isclass(obj) and issubclass(obj, unittest.TestCase):
+                classname = obj.__name__
+                if classname[0] == '_' or self._this_is_skipped(classname):
+                    continue
+                methodnames = []
+                # obj is a TestCase class
+                for attrname in dir(obj):
+                    if attrname.startswith(self.testMethodPrefix):
+                        attr = getattr(obj, attrname)
+                        if callable(attr):
+                            methodnames.append(attrname)
+                # keep track of class (obj) for convenience
+                tests[classname] = (obj, methodnames)
+        return tests
+
+    def loadTestsFromSuite(self, module, suitename):
+        try:
+            suite = getattr(module, suitename)()
+        except AttributeError:
+            return []
+        assert hasattr(suite, '_tests'), \
+               "%s.%s is not a valid TestSuite" % (module.__name__, suitename)
+        # python2.3 does not implement __iter__ on suites, we need to return
+        # _tests explicitly
+        return suite._tests
+
+    def loadTestsFromName(self, name, module=None):
+        parts = name.split('.')
+        if module is None or len(parts) > 2:
+            # let the base class do its job here
+            return [super(NonStrictTestLoader, self).loadTestsFromName(name)]
+        tests = self._collect_tests(module)
+        collected = []
+        if len(parts) == 1:
+            pattern = parts[0]
+            if callable(getattr(module, pattern, None)
+                    )  and pattern not in tests:
+                # consider it as a suite
+                return self.loadTestsFromSuite(module, pattern)
+            if pattern in tests:
+                # case python unittest_foo.py MyTestTC
+                klass, methodnames = tests[pattern]
+                for methodname in methodnames:
+                    collected = [klass(methodname)
+                        for methodname in methodnames]
+            else:
+                # case python unittest_foo.py something
+                for klass, methodnames in tests.values():
+                    # skip methodname if matched by skipped_patterns
+                    for skip_pattern in self.skipped_patterns:
+                        methodnames = [methodname
+                                      for methodname in methodnames
+                                      if skip_pattern not in methodname]
+                    collected += [klass(methodname)
+                                  for methodname in methodnames
+                                  if pattern in methodname]
+        elif len(parts) == 2:
+            # case "MyClass.test_1"
+            classname, pattern = parts
+            klass, methodnames = tests.get(classname, (None, []))
+            for methodname in methodnames:
+                collected = [klass(methodname) for methodname in methodnames
+                            if pattern in methodname]
+        return collected
+
+    def _this_is_skipped(self, testedname):
+        return any([(pat in testedname) for pat in self.skipped_patterns])
+
+    def getTestCaseNames(self, testCaseClass):
+        """Return a sorted sequence of method names found within testCaseClass
+        """
+        is_skipped = self._this_is_skipped
+        classname = testCaseClass.__name__
+        if classname[0] == '_' or is_skipped(classname):
+            return []
+        testnames = super(NonStrictTestLoader, self).getTestCaseNames(
+                testCaseClass)
+        return [testname for testname in testnames if not is_skipped(testname)]
+
+
+# The 2 functions below are modified versions of the TestSuite.run method
+# that is provided with unittest2 for python 2.6, in unittest2/suite.py
+# It is used to monkeypatch the original implementation to support
+# extra runcondition and options arguments (see in testlib.py)
+
+def _ts_run(self, result, runcondition=None, options=None):
+    self._wrapped_run(result, runcondition=runcondition, options=options)
+    self._tearDownPreviousClass(None, result)
+    self._handleModuleTearDown(result)
+    return result
+
+def _ts_wrapped_run(self, result, debug=False, runcondition=None, options=None):
+    for test in self:
+        if result.shouldStop:
+            break
+        if unittest_suite._isnotsuite(test):
+            self._tearDownPreviousClass(test, result)
+            self._handleModuleFixture(test, result)
+            self._handleClassSetUp(test, result)
+            result._previousTestClass = test.__class__
+            if (getattr(test.__class__, '_classSetupFailed', False) or
+                getattr(result, '_moduleSetUpFailed', False)):
+                continue
+
+        # --- modifications to deal with _wrapped_run ---
+        # original code is:
+        #
+        # if not debug:
+        #     test(result)
+        # else:
+        #     test.debug()
+        if hasattr(test, '_wrapped_run'):
+            try:
+                test._wrapped_run(result, debug, runcondition=runcondition, options=options)
+            except TypeError:
+                test._wrapped_run(result, debug)
+        elif not debug:
+            try:
+                test(result, runcondition, options)
+            except TypeError:
+                test(result)
+        else:
+            test.debug()
+        # --- end of modifications to deal with _wrapped_run ---
+    return result
+
+if sys.version_info >= (2, 7):
+    # The function below implements a modified version of the
+    # TestSuite.run method that is provided with python 2.7, in
+    # unittest/suite.py
+    def _ts_run(self, result, debug=False, runcondition=None, options=None):
+        topLevel = False
+        if getattr(result, '_testRunEntered', False) is False:
+            result._testRunEntered = topLevel = True
+
+        self._wrapped_run(result, debug, runcondition, options)
+
+        if topLevel:
+            self._tearDownPreviousClass(None, result)
+            self._handleModuleTearDown(result)
+            result._testRunEntered = False
+        return result
+
+
+def enable_dbc(*args):
+    """
+    Without arguments, return True if contracts can be enabled and should be
+    enabled (see option -d), return False otherwise.
+
+    With arguments, return False if contracts can't or shouldn't be enabled,
+    otherwise weave ContractAspect with items passed as arguments.
+    """
+    if not ENABLE_DBC:
+        return False
+    try:
+        from logilab.aspects.weaver import weaver
+        from logilab.aspects.lib.contracts import ContractAspect
+    except ImportError:
+        sys.stderr.write(
+            'Warning: logilab.aspects is not available. Contracts disabled.')
+        return False
+    for arg in args:
+        weaver.weave_module(arg, ContractAspect)
+    return True
+
+
+# monkeypatch unittest and doctest (ouch !)
+unittest._TextTestResult = testlib.SkipAwareTestResult
+unittest.TextTestRunner = SkipAwareTextTestRunner
+unittest.TestLoader = NonStrictTestLoader
+unittest.TestProgram = SkipAwareTestProgram
+
+if sys.version_info >= (2, 4):
+    doctest.DocTestCase.__bases__ = (testlib.TestCase,)
+    # XXX check python2.6 compatibility
+    #doctest.DocTestCase._cleanups = []
+    #doctest.DocTestCase._out = []
+else:
+    unittest.FunctionTestCase.__bases__ = (testlib.TestCase,)
+unittest.TestSuite.run = _ts_run
+unittest.TestSuite._wrapped_run = _ts_wrapped_run
diff --git a/logilab/common/registry.py b/logilab/common/registry.py
new file mode 100644
index 0000000..a52b2eb
--- /dev/null
+++ b/logilab/common/registry.py
@@ -0,0 +1,1119 @@
+# copyright 2003-2013 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of Logilab-common.
+#
+# Logilab-common is free software: you can redistribute it and/or modify it
+# under the terms of the GNU Lesser General Public License as published by the
+# Free Software Foundation, either version 2.1 of the License, or (at your
+# option) any later version.
+#
+# Logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with Logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""This module provides bases for predicates dispatching (the pattern in use
+here is similar to what's refered as multi-dispatch or predicate-dispatch in the
+literature, though a bit different since the idea is to select across different
+implementation 'e.g. classes), not to dispatch a message to a function or
+method. It contains the following classes:
+
+* :class:`RegistryStore`, the top level object which loads implementation
+  objects and stores them into registries. You'll usually use it to access
+  registries and their contained objects;
+
+* :class:`Registry`, the base class which contains objects semantically grouped
+  (for instance, sharing a same API, hence the 'implementation' name). You'll
+  use it to select the proper implementation according to a context. Notice you
+  may use registries on their own without using the store.
+
+.. Note::
+
+  implementation objects are usually designed to be accessed through the
+  registry and not by direct instantiation, besides to use it as base classe.
+
+The selection procedure is delegated to a selector, which is responsible for
+scoring the object according to some context. At the end of the selection, if an
+implementation has been found, an instance of this class is returned. A selector
+is built from one or more predicates combined together using AND, OR, NOT
+operators (actually `&`, `|` and `~`). You'll thus find some base classes to
+build predicates:
+
+* :class:`Predicate`, the abstract base predicate class
+
+* :class:`AndPredicate`, :class:`OrPredicate`, :class:`NotPredicate`, which you
+  shouldn't have to use directly. You'll use `&`, `|` and '~' operators between
+  predicates directly
+
+* :func:`objectify_predicate`
+
+You'll eventually find one concrete predicate: :class:`yes`
+
+.. autoclass:: RegistryStore
+.. autoclass:: Registry
+
+Predicates
+----------
+.. autoclass:: Predicate
+.. autofunc:: objectify_predicate
+.. autoclass:: yes
+
+Debugging
+---------
+.. autoclass:: traced_selection
+
+Exceptions
+----------
+.. autoclass:: RegistryException
+.. autoclass:: RegistryNotFound
+.. autoclass:: ObjectNotFound
+.. autoclass:: NoSelectableObject
+"""
+
+from __future__ import print_function
+
+__docformat__ = "restructuredtext en"
+
+import sys
+import types
+import weakref
+import traceback as tb
+from os import listdir, stat
+from os.path import join, isdir, exists
+from logging import getLogger
+from warnings import warn
+
+from six import string_types, add_metaclass
+
+from logilab.common.modutils import modpath_from_file
+from logilab.common.logging_ext import set_log_methods
+from logilab.common.decorators import classproperty
+
+
+class RegistryException(Exception):
+    """Base class for registry exception."""
+
+class RegistryNotFound(RegistryException):
+    """Raised when an unknown registry is requested.
+
+    This is usually a programming/typo error.
+    """
+
+class ObjectNotFound(RegistryException):
+    """Raised when an unregistered object is requested.
+
+    This may be a programming/typo or a misconfiguration error.
+    """
+
+class NoSelectableObject(RegistryException):
+    """Raised when no object is selectable for a given context."""
+    def __init__(self, args, kwargs, objects):
+        self.args = args
+        self.kwargs = kwargs
+        self.objects = objects
+
+    def __str__(self):
+        return ('args: %s, kwargs: %s\ncandidates: %s'
+                % (self.args, self.kwargs.keys(), self.objects))
+
+
+def _modname_from_path(path, extrapath=None):
+    modpath = modpath_from_file(path, extrapath)
+    # omit '__init__' from package's name to avoid loading that module
+    # once for each name when it is imported by some other object
+    # module. This supposes import in modules are done as::
+    #
+    #   from package import something
+    #
+    # not::
+    #
+    #   from package.__init__ import something
+    #
+    # which seems quite correct.
+    if modpath[-1] == '__init__':
+        modpath.pop()
+    return '.'.join(modpath)
+
+
+def _toload_info(path, extrapath, _toload=None):
+    """Return a dictionary of <modname>: <modpath> and an ordered list of
+    (file, module name) to load
+    """
+    if _toload is None:
+        assert isinstance(path, list)
+        _toload = {}, []
+    for fileordir in path:
+        if isdir(fileordir) and exists(join(fileordir, '__init__.py')):
+            subfiles = [join(fileordir, fname) for fname in listdir(fileordir)]
+            _toload_info(subfiles, extrapath, _toload)
+        elif fileordir[-3:] == '.py':
+            modname = _modname_from_path(fileordir, extrapath)
+            _toload[0][modname] = fileordir
+            _toload[1].append((fileordir, modname))
+    return _toload
+
+
+class RegistrableObject(object):
+    """This is the base class for registrable objects which are selected
+    according to a context.
+
+    :attr:`__registry__`
+      name of the registry for this object (string like 'views',
+      'templates'...). You may want to define `__registries__` directly if your
+      object should be registered in several registries.
+
+    :attr:`__regid__`
+      object's identifier in the registry (string like 'main',
+      'primary', 'folder_box')
+
+    :attr:`__select__`
+      class'selector
+
+    Moreover, the `__abstract__` attribute may be set to True to indicate that a
+    class is abstract and should not be registered.
+
+    You don't have to inherit from this class to put it in a registry (having
+    `__regid__` and `__select__` is enough), though this is needed for classes
+    that should be automatically registered.
+    """
+
+    __registry__ = None
+    __regid__ = None
+    __select__ = None
+    __abstract__ = True # see doc snipppets below (in Registry class)
+
+    @classproperty
+    def __registries__(cls):
+        if cls.__registry__ is None:
+            return ()
+        return (cls.__registry__,)
+
+
+class RegistrableInstance(RegistrableObject):
+    """Inherit this class if you want instances of the classes to be
+    automatically registered.
+    """
+
+    def __new__(cls, *args, **kwargs):
+        """Add a __module__ attribute telling the module where the instance was
+        created, for automatic registration.
+        """
+        obj = super(RegistrableInstance, cls).__new__(cls)
+        # XXX subclass must no override __new__
+        filepath = tb.extract_stack(limit=2)[0][0]
+        obj.__module__ = _modname_from_path(filepath)
+        return obj
+
+
+class Registry(dict):
+    """The registry store a set of implementations associated to identifier:
+
+    * to each identifier are associated a list of implementations
+
+    * to select an implementation of a given identifier, you should use one of the
+      :meth:`select` or :meth:`select_or_none` method
+
+    * to select a list of implementations for a context, you should use the
+      :meth:`possible_objects` method
+
+    * dictionary like access to an identifier will return the bare list of
+      implementations for this identifier.
+
+    To be usable in a registry, the only requirement is to have a `__select__`
+    attribute.
+
+    At the end of the registration process, the :meth:`__registered__`
+    method is called on each registered object which have them, given the
+    registry in which it's registered as argument.
+
+    Registration methods:
+
+    .. automethod: register
+    .. automethod: unregister
+
+    Selection methods:
+
+    .. automethod: select
+    .. automethod: select_or_none
+    .. automethod: possible_objects
+    .. automethod: object_by_id
+    """
+    def __init__(self, debugmode):
+        super(Registry, self).__init__()
+        self.debugmode = debugmode
+
+    def __getitem__(self, name):
+        """return the registry (list of implementation objects) associated to
+        this name
+        """
+        try:
+            return super(Registry, self).__getitem__(name)
+        except KeyError:
+            exc = ObjectNotFound(name)
+            exc.__traceback__ = sys.exc_info()[-1]
+            raise exc
+
+    @classmethod
+    def objid(cls, obj):
+        """returns a unique identifier for an object stored in the registry"""
+        return '%s.%s' % (obj.__module__, cls.objname(obj))
+
+    @classmethod
+    def objname(cls, obj):
+        """returns a readable name for an object stored in the registry"""
+        return getattr(obj, '__name__', id(obj))
+
+    def initialization_completed(self):
+        """call method __registered__() on registered objects when the callback
+        is defined"""
+        for objects in self.values():
+            for objectcls in objects:
+                registered = getattr(objectcls, '__registered__', None)
+                if registered:
+                    registered(self)
+        if self.debugmode:
+            wrap_predicates(_lltrace)
+
+    def register(self, obj, oid=None, clear=False):
+        """base method to add an object in the registry"""
+        assert not '__abstract__' in obj.__dict__, obj
+        assert obj.__select__, obj
+        oid = oid or obj.__regid__
+        assert oid, ('no explicit name supplied to register object %s, '
+                     'which has no __regid__ set' % obj)
+        if clear:
+            objects = self[oid] =  []
+        else:
+            objects = self.setdefault(oid, [])
+        assert not obj in objects, 'object %s is already registered' % obj
+        objects.append(obj)
+
+    def register_and_replace(self, obj, replaced):
+        """remove <replaced> and register <obj>"""
+        # XXXFIXME this is a duplication of unregister()
+        # remove register_and_replace in favor of unregister + register
+        # or simplify by calling unregister then register here
+        if not isinstance(replaced, string_types):
+            replaced = self.objid(replaced)
+        # prevent from misspelling
+        assert obj is not replaced, 'replacing an object by itself: %s' % obj
+        registered_objs = self.get(obj.__regid__, ())
+        for index, registered in enumerate(registered_objs):
+            if self.objid(registered) == replaced:
+                del registered_objs[index]
+                break
+        else:
+            self.warning('trying to replace %s that is not registered with %s',
+                         replaced, obj)
+        self.register(obj)
+
+    def unregister(self, obj):
+        """remove object <obj> from this registry"""
+        objid = self.objid(obj)
+        oid = obj.__regid__
+        for registered in self.get(oid, ()):
+            # use self.objid() to compare objects because vreg will probably
+            # have its own version of the object, loaded through execfile
+            if self.objid(registered) == objid:
+                self[oid].remove(registered)
+                break
+        else:
+            self.warning('can\'t remove %s, no id %s in the registry',
+                         objid, oid)
+
+    def all_objects(self):
+        """return a list containing all objects in this registry.
+        """
+        result = []
+        for objs in self.values():
+            result += objs
+        return result
+
+    # dynamic selection methods ################################################
+
+    def object_by_id(self, oid, *args, **kwargs):
+        """return object with the `oid` identifier. Only one object is expected
+        to be found.
+
+        raise :exc:`ObjectNotFound` if there are no object with id `oid` in this
+        registry
+
+        raise :exc:`AssertionError` if there is more than one object there
+        """
+        objects = self[oid]
+        assert len(objects) == 1, objects
+        return objects[0](*args, **kwargs)
+
+    def select(self, __oid, *args, **kwargs):
+        """return the most specific object among those with the given oid
+        according to the given context.
+
+        raise :exc:`ObjectNotFound` if there are no object with id `oid` in this
+        registry
+
+        raise :exc:`NoSelectableObject` if no object can be selected
+        """
+        obj =  self._select_best(self[__oid], *args, **kwargs)
+        if obj is None:
+            raise NoSelectableObject(args, kwargs, self[__oid] )
+        return obj
+
+    def select_or_none(self, __oid, *args, **kwargs):
+        """return the most specific object among those with the given oid
+        according to the given context, or None if no object applies.
+        """
+        try:
+            return self._select_best(self[__oid], *args, **kwargs)
+        except ObjectNotFound:
+            return None
+
+    def possible_objects(self, *args, **kwargs):
+        """return an iterator on possible objects in this registry for the given
+        context
+        """
+        for objects in self.values():
+            obj = self._select_best(objects,  *args, **kwargs)
+            if obj is None:
+                continue
+            yield obj
+
+    def _select_best(self, objects, *args, **kwargs):
+        """return an instance of the most specific object according
+        to parameters
+
+        return None if not object apply (don't raise `NoSelectableObject` since
+        it's costly when searching objects using `possible_objects`
+        (e.g. searching for hooks).
+        """
+        score, winners = 0, None
+        for obj in objects:
+            objectscore = obj.__select__(obj, *args, **kwargs)
+            if objectscore > score:
+                score, winners = objectscore, [obj]
+            elif objectscore > 0 and objectscore == score:
+                winners.append(obj)
+        if winners is None:
+            return None
+        if len(winners) > 1:
+            # log in production environement / test, error while debugging
+            msg = 'select ambiguity: %s\n(args: %s, kwargs: %s)'
+            if self.debugmode:
+                # raise bare exception in debug mode
+                raise Exception(msg % (winners, args, kwargs.keys()))
+            self.error(msg, winners, args, kwargs.keys())
+        # return the result of calling the object
+        return self.selected(winners[0], args, kwargs)
+
+    def selected(self, winner, args, kwargs):
+        """override here if for instance you don't want "instanciation"
+        """
+        return winner(*args, **kwargs)
+
+    # these are overridden by set_log_methods below
+    # only defining here to prevent pylint from complaining
+    info = warning = error = critical = exception = debug = lambda msg, *a, **kw: None
+
+
+def obj_registries(cls, registryname=None):
+    """return a tuple of registry names (see __registries__)"""
+    if registryname:
+        return (registryname,)
+    return cls.__registries__
+
+
+class RegistryStore(dict):
+    """This class is responsible for loading objects and storing them
+    in their registry which is created on the fly as needed.
+
+    It handles dynamic registration of objects and provides a
+    convenient api to access them. To be recognized as an object that
+    should be stored into one of the store's registry
+    (:class:`Registry`), an object must provide the following
+    attributes, used control how they interact with the registry:
+
+    :attr:`__registries__`
+      list of registry names (string like 'views', 'templates'...) into which
+      the object should be registered
+
+    :attr:`__regid__`
+      object identifier in the registry (string like 'main',
+      'primary', 'folder_box')
+
+    :attr:`__select__`
+      the object predicate selectors
+
+    Moreover, the :attr:`__abstract__` attribute may be set to `True`
+    to indicate that an object is abstract and should not be registered
+    (such inherited attributes not considered).
+
+    .. Note::
+
+      When using the store to load objects dynamically, you *always* have
+      to use **super()** to get the methods and attributes of the
+      superclasses, and not use the class identifier. If not, you'll get into
+      trouble at reload time.
+
+      For example, instead of writing::
+
+          class Thing(Parent):
+              __regid__ = 'athing'
+              __select__ = yes()
+
+              def f(self, arg1):
+                  Parent.f(self, arg1)
+
+      You must write::
+
+          class Thing(Parent):
+              __regid__ = 'athing'
+              __select__ = yes()
+
+              def f(self, arg1):
+                  super(Thing, self).f(arg1)
+
+    Controlling object registration
+    -------------------------------
+
+    Dynamic loading is triggered by calling the
+    :meth:`register_objects` method, given a list of directories to
+    inspect for python modules.
+
+    .. automethod: register_objects
+
+    For each module, by default, all compatible objects are registered
+    automatically. However if some objects come as replacement of
+    other objects, or have to be included only if some condition is
+    met, you'll have to define a `registration_callback(vreg)`
+    function in the module and explicitly register **all objects** in
+    this module, using the api defined below.
+
+
+    .. automethod:: RegistryStore.register_all
+    .. automethod:: RegistryStore.register_and_replace
+    .. automethod:: RegistryStore.register
+    .. automethod:: RegistryStore.unregister
+
+    .. Note::
+        Once the function `registration_callback(vreg)` is implemented in a
+        module, all the objects from this module have to be explicitly
+        registered as it disables the automatic object registration.
+
+
+    Examples:
+
+    .. sourcecode:: python
+
+       def registration_callback(store):
+          # register everything in the module except BabarClass
+          store.register_all(globals().values(), __name__, (BabarClass,))
+
+          # conditionally register BabarClass
+          if 'babar_relation' in store.schema:
+              store.register(BabarClass)
+
+    In this example, we register all application object classes defined in the module
+    except `BabarClass`. This class is then registered only if the 'babar_relation'
+    relation type is defined in the instance schema.
+
+    .. sourcecode:: python
+
+       def registration_callback(store):
+          store.register(Elephant)
+          # replace Babar by Celeste
+          store.register_and_replace(Celeste, Babar)
+
+    In this example, we explicitly register classes one by one:
+
+    * the `Elephant` class
+    * the `Celeste` to replace `Babar`
+
+    If at some point we register a new appobject class in this module, it won't be
+    registered at all without modification to the `registration_callback`
+    implementation. The first example will register it though, thanks to the call
+    to the `register_all` method.
+
+    Controlling registry instantiation
+    ----------------------------------
+
+    The `REGISTRY_FACTORY` class dictionary allows to specify which class should
+    be instantiated for a given registry name. The class associated to `None`
+    key will be the class used when there is no specific class for a name.
+    """
+
+    def __init__(self, debugmode=False):
+        super(RegistryStore, self).__init__()
+        self.debugmode = debugmode
+
+    def reset(self):
+        """clear all registries managed by this store"""
+        # don't use self.clear, we want to keep existing subdictionaries
+        for subdict in self.values():
+            subdict.clear()
+        self._lastmodifs = {}
+
+    def __getitem__(self, name):
+        """return the registry (dictionary of class objects) associated to
+        this name
+        """
+        try:
+            return super(RegistryStore, self).__getitem__(name)
+        except KeyError:
+            exc = RegistryNotFound(name)
+            exc.__traceback__ = sys.exc_info()[-1]
+            raise exc
+
+    # methods for explicit (un)registration ###################################
+
+    # default class, when no specific class set
+    REGISTRY_FACTORY = {None: Registry}
+
+    def registry_class(self, regid):
+        """return existing registry named regid or use factory to create one and
+        return it"""
+        try:
+            return self.REGISTRY_FACTORY[regid]
+        except KeyError:
+            return self.REGISTRY_FACTORY[None]
+
+    def setdefault(self, regid):
+        try:
+            return self[regid]
+        except RegistryNotFound:
+            self[regid] = self.registry_class(regid)(self.debugmode)
+            return self[regid]
+
+    def register_all(self, objects, modname, butclasses=()):
+        """register registrable objects into `objects`.
+
+        Registrable objects are properly configured subclasses of
+        :class:`RegistrableObject`.  Objects which are not defined in the module
+        `modname` or which are in `butclasses` won't be registered.
+
+        Typical usage is:
+
+        .. sourcecode:: python
+
+            store.register_all(globals().values(), __name__, (ClassIWantToRegisterExplicitly,))
+
+        So you get partially automatic registration, keeping manual registration
+        for some object (to use
+        :meth:`~logilab.common.registry.RegistryStore.register_and_replace` for
+        instance).
+        """
+        assert isinstance(modname, string_types), \
+            'modname expected to be a module name (ie string), got %r' % modname
+        for obj in objects:
+            if self.is_registrable(obj) and obj.__module__ == modname and not obj in butclasses:
+                if isinstance(obj, type):
+                    self._load_ancestors_then_object(modname, obj, butclasses)
+                else:
+                    self.register(obj)
+
+    def register(self, obj, registryname=None, oid=None, clear=False):
+        """register `obj` implementation into `registryname` or
+        `obj.__registries__` if not specified, with identifier `oid` or
+        `obj.__regid__` if not specified.
+
+        If `clear` is true, all objects with the same identifier will be
+        previously unregistered.
+        """
+        assert not obj.__dict__.get('__abstract__'), obj
+        for registryname in obj_registries(obj, registryname):
+            registry = self.setdefault(registryname)
+            registry.register(obj, oid=oid, clear=clear)
+            self.debug("register %s in %s['%s']",
+                       registry.objname(obj), registryname, oid or obj.__regid__)
+            self._loadedmods.setdefault(obj.__module__, {})[registry.objid(obj)] = obj
+
+    def unregister(self, obj, registryname=None):
+        """unregister `obj` object from the registry `registryname` or
+        `obj.__registries__` if not specified.
+        """
+        for registryname in obj_registries(obj, registryname):
+            registry = self[registryname]
+            registry.unregister(obj)
+            self.debug("unregister %s from %s['%s']",
+                       registry.objname(obj), registryname, obj.__regid__)
+
+    def register_and_replace(self, obj, replaced, registryname=None):
+        """register `obj` object into `registryname` or
+        `obj.__registries__` if not specified. If found, the `replaced` object
+        will be unregistered first (else a warning will be issued as it is
+        generally unexpected).
+        """
+        for registryname in obj_registries(obj, registryname):
+            registry = self[registryname]
+            registry.register_and_replace(obj, replaced)
+            self.debug("register %s in %s['%s'] instead of %s",
+                       registry.objname(obj), registryname, obj.__regid__,
+                       registry.objname(replaced))
+
+    # initialization methods ###################################################
+
+    def init_registration(self, path, extrapath=None):
+        """reset registry and walk down path to return list of (path, name)
+        file modules to be loaded"""
+        # XXX make this private by renaming it to _init_registration ?
+        self.reset()
+        # compute list of all modules that have to be loaded
+        self._toloadmods, filemods = _toload_info(path, extrapath)
+        # XXX is _loadedmods still necessary ? It seems like it's useful
+        #     to avoid loading same module twice, especially with the
+        #     _load_ancestors_then_object logic but this needs to be checked
+        self._loadedmods = {}
+        return filemods
+
+    def register_objects(self, path, extrapath=None):
+        """register all objects found walking down <path>"""
+        # load views from each directory in the instance's path
+        # XXX inline init_registration ?
+        filemods = self.init_registration(path, extrapath)
+        for filepath, modname in filemods:
+            self.load_file(filepath, modname)
+        self.initialization_completed()
+
+    def initialization_completed(self):
+        """call initialization_completed() on all known registries"""
+        for reg in self.values():
+            reg.initialization_completed()
+
+    def _mdate(self, filepath):
+        """ return the modification date of a file path """
+        try:
+            return stat(filepath)[-2]
+        except OSError:
+            # this typically happens on emacs backup files (.#foo.py)
+            self.warning('Unable to load %s. It is likely to be a backup file',
+                         filepath)
+            return None
+
+    def is_reload_needed(self, path):
+        """return True if something module changed and the registry should be
+        reloaded
+        """
+        lastmodifs = self._lastmodifs
+        for fileordir in path:
+            if isdir(fileordir) and exists(join(fileordir, '__init__.py')):
+                if self.is_reload_needed([join(fileordir, fname)
+                                          for fname in listdir(fileordir)]):
+                    return True
+            elif fileordir[-3:] == '.py':
+                mdate = self._mdate(fileordir)
+                if mdate is None:
+                    continue # backup file, see _mdate implementation
+                elif "flymake" in fileordir:
+                    # flymake + pylint in use, don't consider these they will corrupt the registry
+                    continue
+                if fileordir not in lastmodifs or lastmodifs[fileordir] < mdate:
+                    self.info('File %s changed since last visit', fileordir)
+                    return True
+        return False
+
+    def load_file(self, filepath, modname):
+        """ load registrable objects (if any) from a python file """
+        from logilab.common.modutils import load_module_from_name
+        if modname in self._loadedmods:
+            return
+        self._loadedmods[modname] = {}
+        mdate = self._mdate(filepath)
+        if mdate is None:
+            return # backup file, see _mdate implementation
+        elif "flymake" in filepath:
+            # flymake + pylint in use, don't consider these they will corrupt the registry
+            return
+        # set update time before module loading, else we get some reloading
+        # weirdness in case of syntax error or other error while importing the
+        # module
+        self._lastmodifs[filepath] = mdate
+        # load the module
+        module = load_module_from_name(modname)
+        self.load_module(module)
+
+    def load_module(self, module):
+        """Automatically handle module objects registration.
+
+        Instances are registered as soon as they are hashable and have the
+        following attributes:
+
+        * __regid__ (a string)
+        * __select__ (a callable)
+        * __registries__ (a tuple/list of string)
+
+        For classes this is a bit more complicated :
+
+        - first ensure parent classes are already registered
+
+        - class with __abstract__ == True in their local dictionary are skipped
+
+        - object class needs to have registries and identifier properly set to a
+          non empty string to be registered.
+        """
+        self.info('loading %s from %s', module.__name__, module.__file__)
+        if hasattr(module, 'registration_callback'):
+            module.registration_callback(self)
+        else:
+            self.register_all(vars(module).values(), module.__name__)
+
+    def _load_ancestors_then_object(self, modname, objectcls, butclasses=()):
+        """handle class registration according to rules defined in
+        :meth:`load_module`
+        """
+        # backward compat, we used to allow whatever else than classes
+        if not isinstance(objectcls, type):
+            if self.is_registrable(objectcls) and objectcls.__module__ == modname:
+                self.register(objectcls)
+            return
+        # imported classes
+        objmodname = objectcls.__module__
+        if objmodname != modname:
+            # The module of the object is not the same as the currently
+            # worked on module, or this is actually an instance, which
+            # has no module at all
+            if objmodname in self._toloadmods:
+                # if this is still scheduled for loading, let's proceed immediately,
+                # but using the object module
+                self.load_file(self._toloadmods[objmodname], objmodname)
+            return
+        # ensure object hasn't been already processed
+        clsid = '%s.%s' % (modname, objectcls.__name__)
+        if clsid in self._loadedmods[modname]:
+            return
+        self._loadedmods[modname][clsid] = objectcls
+        # ensure ancestors are registered
+        for parent in objectcls.__bases__:
+            self._load_ancestors_then_object(modname, parent, butclasses)
+        # ensure object is registrable
+        if objectcls in butclasses or not self.is_registrable(objectcls):
+            return
+        # backward compat
+        reg = self.setdefault(obj_registries(objectcls)[0])
+        if reg.objname(objectcls)[0] == '_':
+            warn("[lgc 0.59] object whose name start with '_' won't be "
+                 "skipped anymore at some point, use __abstract__ = True "
+                 "instead (%s)" % objectcls, DeprecationWarning)
+            return
+        # register, finally
+        self.register(objectcls)
+
+    @classmethod
+    def is_registrable(cls, obj):
+        """ensure `obj` should be registered
+
+        as arbitrary stuff may be registered, do a lot of check and warn about
+        weird cases (think to dumb proxy objects)
+        """
+        if isinstance(obj, type):
+            if not issubclass(obj, RegistrableObject):
+                # ducktyping backward compat
+                if not (getattr(obj, '__registries__', None)
+                        and getattr(obj, '__regid__', None)
+                        and getattr(obj, '__select__', None)):
+                    return False
+            elif issubclass(obj, RegistrableInstance):
+                return False
+        elif not isinstance(obj, RegistrableInstance):
+            return False
+        if not obj.__regid__:
+            return False # no regid
+        registries = obj.__registries__
+        if not registries:
+            return False # no registries
+        selector = obj.__select__
+        if not selector:
+            return False # no selector
+        if obj.__dict__.get('__abstract__', False):
+            return False
+        # then detect potential problems that should be warned
+        if not isinstance(registries, (tuple, list)):
+            cls.warning('%s has __registries__ which is not a list or tuple', obj)
+            return False
+        if not callable(selector):
+            cls.warning('%s has not callable __select__', obj)
+            return False
+        return True
+
+    # these are overridden by set_log_methods below
+    # only defining here to prevent pylint from complaining
+    info = warning = error = critical = exception = debug = lambda msg, *a, **kw: None
+
+
+# init logging
+set_log_methods(RegistryStore, getLogger('registry.store'))
+set_log_methods(Registry, getLogger('registry'))
+
+
+# helpers for debugging selectors
+TRACED_OIDS = None
+
+def _trace_selector(cls, selector, args, ret):
+    vobj = args[0]
+    if TRACED_OIDS == 'all' or vobj.__regid__ in TRACED_OIDS:
+        print('%s -> %s for %s(%s)' % (cls, ret, vobj, vobj.__regid__))
+
+def _lltrace(selector):
+    """use this decorator on your predicates so they become traceable with
+    :class:`traced_selection`
+    """
+    def traced(cls, *args, **kwargs):
+        ret = selector(cls, *args, **kwargs)
+        if TRACED_OIDS is not None:
+            _trace_selector(cls, selector, args, ret)
+        return ret
+    traced.__name__ = selector.__name__
+    traced.__doc__ = selector.__doc__
+    return traced
+
+class traced_selection(object): # pylint: disable=C0103
+    """
+    Typical usage is :
+
+    .. sourcecode:: python
+
+        >>> from logilab.common.registry import traced_selection
+        >>> with traced_selection():
+        ...     # some code in which you want to debug selectors
+        ...     # for all objects
+
+    This will yield lines like this in the logs::
+
+        selector one_line_rset returned 0 for <class 'elephant.Babar'>
+
+    You can also give to :class:`traced_selection` the identifiers of objects on
+    which you want to debug selection ('oid1' and 'oid2' in the example above).
+
+    .. sourcecode:: python
+
+        >>> with traced_selection( ('regid1', 'regid2') ):
+        ...     # some code in which you want to debug selectors
+        ...     # for objects with __regid__ 'regid1' and 'regid2'
+
+    A potentially useful point to set up such a tracing function is
+    the `logilab.common.registry.Registry.select` method body.
+    """
+
+    def __init__(self, traced='all'):
+        self.traced = traced
+
+    def __enter__(self):
+        global TRACED_OIDS
+        TRACED_OIDS = self.traced
+
+    def __exit__(self, exctype, exc, traceback):
+        global TRACED_OIDS
+        TRACED_OIDS = None
+        return traceback is None
+
+# selector base classes and operations ########################################
+
+def objectify_predicate(selector_func):
+    """Most of the time, a simple score function is enough to build a selector.
+    The :func:`objectify_predicate` decorator turn it into a proper selector
+    class::
+
+        @objectify_predicate
+        def one(cls, req, rset=None, **kwargs):
+            return 1
+
+        class MyView(View):
+            __select__ = View.__select__ & one()
+
+    """
+    return type(selector_func.__name__, (Predicate,),
+                {'__doc__': selector_func.__doc__,
+                 '__call__': lambda self, *a, **kw: selector_func(*a, **kw)})
+
+
+_PREDICATES = {}
+
+def wrap_predicates(decorator):
+    for predicate in _PREDICATES.values():
+        if not '_decorators' in predicate.__dict__:
+            predicate._decorators = set()
+        if decorator in predicate._decorators:
+            continue
+        predicate._decorators.add(decorator)
+        predicate.__call__ = decorator(predicate.__call__)
+
+class PredicateMetaClass(type):
+    def __new__(mcs, *args, **kwargs):
+        # use __new__ so subclasses doesn't have to call Predicate.__init__
+        inst = type.__new__(mcs, *args, **kwargs)
+        proxy = weakref.proxy(inst, lambda p: _PREDICATES.pop(id(p)))
+        _PREDICATES[id(proxy)] = proxy
+        return inst
+
+
+@add_metaclass(PredicateMetaClass)
+class Predicate(object):
+    """base class for selector classes providing implementation
+    for operators ``&``, ``|`` and  ``~``
+
+    This class is only here to give access to binary operators, the selector
+    logic itself should be implemented in the :meth:`__call__` method. Notice it
+    should usually accept any arbitrary arguments (the context), though that may
+    vary depending on your usage of the registry.
+
+    a selector is called to help choosing the correct object for a
+    particular context by returning a score (`int`) telling how well
+    the implementation given as first argument fit to the given context.
+
+    0 score means that the class doesn't apply.
+    """
+
+    @property
+    def func_name(self):
+        # backward compatibility
+        return self.__class__.__name__
+
+    def search_selector(self, selector):
+        """search for the given selector, selector instance or tuple of
+        selectors in the selectors tree. Return None if not found.
+        """
+        if self is selector:
+            return self
+        if (isinstance(selector, type) or isinstance(selector, tuple)) and \
+               isinstance(self, selector):
+            return self
+        return None
+
+    def __str__(self):
+        return self.__class__.__name__
+
+    def __and__(self, other):
+        return AndPredicate(self, other)
+    def __rand__(self, other):
+        return AndPredicate(other, self)
+    def __iand__(self, other):
+        return AndPredicate(self, other)
+    def __or__(self, other):
+        return OrPredicate(self, other)
+    def __ror__(self, other):
+        return OrPredicate(other, self)
+    def __ior__(self, other):
+        return OrPredicate(self, other)
+
+    def __invert__(self):
+        return NotPredicate(self)
+
+    # XXX (function | function) or (function & function) not managed yet
+
+    def __call__(self, cls, *args, **kwargs):
+        return NotImplementedError("selector %s must implement its logic "
+                                   "in its __call__ method" % self.__class__)
+
+    def __repr__(self):
+        return u'<Predicate %s at %x>' % (self.__class__.__name__, id(self))
+
+
+class MultiPredicate(Predicate):
+    """base class for compound selector classes"""
+
+    def __init__(self, *selectors):
+        self.selectors = self.merge_selectors(selectors)
+
+    def __str__(self):
+        return '%s(%s)' % (self.__class__.__name__,
+                           ','.join(str(s) for s in self.selectors))
+
+    @classmethod
+    def merge_selectors(cls, selectors):
+        """deal with selector instanciation when necessary and merge
+        multi-selectors if possible:
+
+        AndPredicate(AndPredicate(sel1, sel2), AndPredicate(sel3, sel4))
+        ==> AndPredicate(sel1, sel2, sel3, sel4)
+        """
+        merged_selectors = []
+        for selector in selectors:
+            # XXX do we really want magic-transformations below?
+            # if so, wanna warn about them?
+            if isinstance(selector, types.FunctionType):
+                selector = objectify_predicate(selector)()
+            if isinstance(selector, type) and issubclass(selector, Predicate):
+                selector = selector()
+            assert isinstance(selector, Predicate), selector
+            if isinstance(selector, cls):
+                merged_selectors += selector.selectors
+            else:
+                merged_selectors.append(selector)
+        return merged_selectors
+
+    def search_selector(self, selector):
+        """search for the given selector or selector instance (or tuple of
+        selectors) in the selectors tree. Return None if not found
+        """
+        for childselector in self.selectors:
+            if childselector is selector:
+                return childselector
+            found = childselector.search_selector(selector)
+            if found is not None:
+                return found
+        # if not found in children, maybe we are looking for self?
+        return super(MultiPredicate, self).search_selector(selector)
+
+
+class AndPredicate(MultiPredicate):
+    """and-chained selectors"""
+    def __call__(self, cls, *args, **kwargs):
+        score = 0
+        for selector in self.selectors:
+            partscore = selector(cls, *args, **kwargs)
+            if not partscore:
+                return 0
+            score += partscore
+        return score
+
+
+class OrPredicate(MultiPredicate):
+    """or-chained selectors"""
+    def __call__(self, cls, *args, **kwargs):
+        for selector in self.selectors:
+            partscore = selector(cls, *args, **kwargs)
+            if partscore:
+                return partscore
+        return 0
+
+class NotPredicate(Predicate):
+    """negation selector"""
+    def __init__(self, selector):
+        self.selector = selector
+
+    def __call__(self, cls, *args, **kwargs):
+        score = self.selector(cls, *args, **kwargs)
+        return int(not score)
+
+    def __str__(self):
+        return 'NOT(%s)' % self.selector
+
+
+class yes(Predicate): # pylint: disable=C0103
+    """Return the score given as parameter, with a default score of 0.5 so any
+    other selector take precedence.
+
+    Usually used for objects which can be selected whatever the context, or
+    also sometimes to add arbitrary points to a score.
+
+    Take care, `yes(0)` could be named 'no'...
+    """
+    def __init__(self, score=0.5):
+        self.score = score
+
+    def __call__(self, *args, **kwargs):
+        return self.score
+
+
+# deprecated stuff #############################################################
+
+from logilab.common.deprecation import deprecated
+
+@deprecated('[lgc 0.59] use Registry.objid class method instead')
+def classid(cls):
+    return '%s.%s' % (cls.__module__, cls.__name__)
+
+@deprecated('[lgc 0.59] use obj_registries function instead')
+def class_registries(cls, registryname):
+    return obj_registries(cls, registryname)
+
diff --git a/logilab/common/shellutils.py b/logilab/common/shellutils.py
new file mode 100644
index 0000000..4e68956
--- /dev/null
+++ b/logilab/common/shellutils.py
@@ -0,0 +1,462 @@
+# copyright 2003-2014 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""shell/term utilities, useful to write some python scripts instead of shell
+scripts.
+"""
+
+from __future__ import print_function
+
+__docformat__ = "restructuredtext en"
+
+import os
+import glob
+import shutil
+import stat
+import sys
+import tempfile
+import time
+import fnmatch
+import errno
+import string
+import random
+import subprocess
+from os.path import exists, isdir, islink, basename, join
+
+from six import string_types
+from six.moves import range, input as raw_input
+
+from logilab.common import STD_BLACKLIST, _handle_blacklist
+from logilab.common.compat import str_to_bytes
+from logilab.common.deprecation import deprecated
+
+try:
+    from logilab.common.proc import ProcInfo, NoSuchProcess
+except ImportError:
+    # windows platform
+    class NoSuchProcess(Exception): pass
+
+    def ProcInfo(pid):
+        raise NoSuchProcess()
+
+
+class tempdir(object):
+
+    def __enter__(self):
+        self.path = tempfile.mkdtemp()
+        return self.path
+
+    def __exit__(self, exctype, value, traceback):
+        # rmtree in all cases
+        shutil.rmtree(self.path)
+        return traceback is None
+
+
+class pushd(object):
+    def __init__(self, directory):
+        self.directory = directory
+
+    def __enter__(self):
+        self.cwd = os.getcwd()
+        os.chdir(self.directory)
+        return self.directory
+
+    def __exit__(self, exctype, value, traceback):
+        os.chdir(self.cwd)
+
+
+def chown(path, login=None, group=None):
+    """Same as `os.chown` function but accepting user login or group name as
+    argument. If login or group is omitted, it's left unchanged.
+
+    Note: you must own the file to chown it (or be root). Otherwise OSError is raised.
+    """
+    if login is None:
+        uid = -1
+    else:
+        try:
+            uid = int(login)
+        except ValueError:
+            import pwd # Platforms: Unix
+            uid = pwd.getpwnam(login).pw_uid
+    if group is None:
+        gid = -1
+    else:
+        try:
+            gid = int(group)
+        except ValueError:
+            import grp
+            gid = grp.getgrnam(group).gr_gid
+    os.chown(path, uid, gid)
+
+def mv(source, destination, _action=shutil.move):
+    """A shell-like mv, supporting wildcards.
+    """
+    sources = glob.glob(source)
+    if len(sources) > 1:
+        assert isdir(destination)
+        for filename in sources:
+            _action(filename, join(destination, basename(filename)))
+    else:
+        try:
+            source = sources[0]
+        except IndexError:
+            raise OSError('No file matching %s' % source)
+        if isdir(destination) and exists(destination):
+            destination = join(destination, basename(source))
+        try:
+            _action(source, destination)
+        except OSError as ex:
+            raise OSError('Unable to move %r to %r (%s)' % (
+                source, destination, ex))
+
+def rm(*files):
+    """A shell-like rm, supporting wildcards.
+    """
+    for wfile in files:
+        for filename in glob.glob(wfile):
+            if islink(filename):
+                os.remove(filename)
+            elif isdir(filename):
+                shutil.rmtree(filename)
+            else:
+                os.remove(filename)
+
+def cp(source, destination):
+    """A shell-like cp, supporting wildcards.
+    """
+    mv(source, destination, _action=shutil.copy)
+
+def find(directory, exts, exclude=False, blacklist=STD_BLACKLIST):
+    """Recursively find files ending with the given extensions from the directory.
+
+    :type directory: str
+    :param directory:
+      directory where the search should start
+
+    :type exts: basestring or list or tuple
+    :param exts:
+      extensions or lists or extensions to search
+
+    :type exclude: boolean
+    :param exts:
+      if this argument is True, returning files NOT ending with the given
+      extensions
+
+    :type blacklist: list or tuple
+    :param blacklist:
+      optional list of files or directory to ignore, default to the value of
+      `logilab.common.STD_BLACKLIST`
+
+    :rtype: list
+    :return:
+      the list of all matching files
+    """
+    if isinstance(exts, string_types):
+        exts = (exts,)
+    if exclude:
+        def match(filename, exts):
+            for ext in exts:
+                if filename.endswith(ext):
+                    return False
+            return True
+    else:
+        def match(filename, exts):
+            for ext in exts:
+                if filename.endswith(ext):
+                    return True
+            return False
+    files = []
+    for dirpath, dirnames, filenames in os.walk(directory):
+        _handle_blacklist(blacklist, dirnames, filenames)
+        # don't append files if the directory is blacklisted
+        dirname = basename(dirpath)
+        if dirname in blacklist:
+            continue
+        files.extend([join(dirpath, f) for f in filenames if match(f, exts)])
+    return files
+
+
+def globfind(directory, pattern, blacklist=STD_BLACKLIST):
+    """Recursively finds files matching glob `pattern` under `directory`.
+
+    This is an alternative to `logilab.common.shellutils.find`.
+
+    :type directory: str
+    :param directory:
+      directory where the search should start
+
+    :type pattern: basestring
+    :param pattern:
+      the glob pattern (e.g *.py, foo*.py, etc.)
+
+    :type blacklist: list or tuple
+    :param blacklist:
+      optional list of files or directory to ignore, default to the value of
+      `logilab.common.STD_BLACKLIST`
+
+    :rtype: iterator
+    :return:
+      iterator over the list of all matching files
+    """
+    for curdir, dirnames, filenames in os.walk(directory):
+        _handle_blacklist(blacklist, dirnames, filenames)
+        for fname in fnmatch.filter(filenames, pattern):
+            yield join(curdir, fname)
+
+def unzip(archive, destdir):
+    import zipfile
+    if not exists(destdir):
+        os.mkdir(destdir)
+    zfobj = zipfile.ZipFile(archive)
+    for name in zfobj.namelist():
+        if name.endswith('/'):
+            os.mkdir(join(destdir, name))
+        else:
+            outfile = open(join(destdir, name), 'wb')
+            outfile.write(zfobj.read(name))
+            outfile.close()
+
+
+class Execute:
+    """This is a deadlock safe version of popen2 (no stdin), that returns
+    an object with errorlevel, out and err.
+    """
+
+    def __init__(self, command):
+        cmd = subprocess.Popen(command, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+        self.out, self.err = cmd.communicate()
+        self.status = os.WEXITSTATUS(cmd.returncode)
+
+Execute = deprecated('Use subprocess.Popen instead')(Execute)
+
+
+def acquire_lock(lock_file, max_try=10, delay=10, max_delay=3600):
+    """Acquire a lock represented by a file on the file system
+
+    If the process written in lock file doesn't exist anymore, we remove the
+    lock file immediately
+    If age of the lock_file is greater than max_delay, then we raise a UserWarning
+    """
+    count = abs(max_try)
+    while count:
+        try:
+            fd = os.open(lock_file, os.O_EXCL | os.O_RDWR | os.O_CREAT)
+            os.write(fd, str_to_bytes(str(os.getpid())) )
+            os.close(fd)
+            return True
+        except OSError as e:
+            if e.errno == errno.EEXIST:
+                try:
+                    fd = open(lock_file, "r")
+                    pid = int(fd.readline())
+                    pi = ProcInfo(pid)
+                    age = (time.time() - os.stat(lock_file)[stat.ST_MTIME])
+                    if age / max_delay > 1 :
+                        raise UserWarning("Command '%s' (pid %s) has locked the "
+                                          "file '%s' for %s minutes"
+                                          % (pi.name(), pid, lock_file, age/60))
+                except UserWarning:
+                    raise
+                except NoSuchProcess:
+                    os.remove(lock_file)
+                except Exception:
+                    # The try block is not essential. can be skipped.
+                    # Note: ProcInfo object is only available for linux
+                    # process information are not accessible...
+                    # or lock_file is no more present...
+                    pass
+            else:
+                raise
+            count -= 1
+            time.sleep(delay)
+    else:
+        raise Exception('Unable to acquire %s' % lock_file)
+
+def release_lock(lock_file):
+    """Release a lock represented by a file on the file system."""
+    os.remove(lock_file)
+
+
+class ProgressBar(object):
+    """A simple text progression bar."""
+
+    def __init__(self, nbops, size=20, stream=sys.stdout, title=''):
+        if title:
+            self._fstr = '\r%s [%%-%ss]' % (title, int(size))
+        else:
+            self._fstr = '\r[%%-%ss]' % int(size)
+        self._stream = stream
+        self._total = nbops
+        self._size = size
+        self._current = 0
+        self._progress = 0
+        self._current_text = None
+        self._last_text_write_size = 0
+
+    def _get_text(self):
+        return self._current_text
+
+    def _set_text(self, text=None):
+        if text != self._current_text:
+            self._current_text = text
+            self.refresh()
+
+    def _del_text(self):
+        self.text = None
+
+    text = property(_get_text, _set_text, _del_text)
+
+    def update(self, offset=1, exact=False):
+        """Move FORWARD to new cursor position (cursor will never go backward).
+
+        :offset: fraction of ``size``
+
+        :exact:
+
+          - False: offset relative to current cursor position if True
+          - True: offset as an asbsolute position
+
+        """
+        if exact:
+            self._current = offset
+        else:
+            self._current += offset
+
+        progress = int((float(self._current)/float(self._total))*self._size)
+        if progress > self._progress:
+            self._progress = progress
+            self.refresh()
+
+    def refresh(self):
+        """Refresh the progression bar display."""
+        self._stream.write(self._fstr % ('=' * min(self._progress, self._size)) )
+        if self._last_text_write_size or self._current_text:
+            template = ' %%-%is' % (self._last_text_write_size)
+            text = self._current_text
+            if text is None:
+                text = ''
+            self._stream.write(template % text)
+            self._last_text_write_size = len(text.rstrip())
+        self._stream.flush()
+
+    def finish(self):
+        self._stream.write('\n')
+        self._stream.flush()
+
+
+class DummyProgressBar(object):
+    __slot__ = ('text',)
+
+    def refresh(self):
+        pass
+    def update(self):
+        pass
+    def finish(self):
+        pass
+
+
+_MARKER = object()
+class progress(object):
+
+    def __init__(self, nbops=_MARKER, size=_MARKER, stream=_MARKER, title=_MARKER, enabled=True):
+        self.nbops = nbops
+        self.size = size
+        self.stream = stream
+        self.title = title
+        self.enabled = enabled
+
+    def __enter__(self):
+        if self.enabled:
+            kwargs = {}
+            for attr in ('nbops', 'size', 'stream', 'title'):
+                value = getattr(self, attr)
+                if value is not _MARKER:
+                    kwargs[attr] = value
+            self.pb = ProgressBar(**kwargs)
+        else:
+            self.pb =  DummyProgressBar()
+        return self.pb
+
+    def __exit__(self, exc_type, exc_val, exc_tb):
+        self.pb.finish()
+
+class RawInput(object):
+
+    def __init__(self, input=None, printer=None):
+        self._input = input or raw_input
+        self._print = printer
+
+    def ask(self, question, options, default):
+        assert default in options
+        choices = []
+        for option in options:
+            if option == default:
+                label = option[0].upper()
+            else:
+                label = option[0].lower()
+            if len(option) > 1:
+                label += '(%s)' % option[1:].lower()
+            choices.append((option, label))
+        prompt = "%s [%s]: " % (question,
+                                '/'.join([opt[1] for opt in choices]))
+        tries = 3
+        while tries > 0:
+            answer = self._input(prompt).strip().lower()
+            if not answer:
+                return default
+            possible = [option for option, label in choices
+                        if option.lower().startswith(answer)]
+            if len(possible) == 1:
+                return possible[0]
+            elif len(possible) == 0:
+                msg = '%s is not an option.' % answer
+            else:
+                msg = ('%s is an ambiguous answer, do you mean %s ?' % (
+                        answer, ' or '.join(possible)))
+            if self._print:
+                self._print(msg)
+            else:
+                print(msg)
+            tries -= 1
+        raise Exception('unable to get a sensible answer')
+
+    def confirm(self, question, default_is_yes=True):
+        default = default_is_yes and 'y' or 'n'
+        answer = self.ask(question, ('y', 'n'), default)
+        return answer == 'y'
+
+ASK = RawInput()
+
+
+def getlogin():
+    """avoid using os.getlogin() because of strange tty / stdin problems
+    (man 3 getlogin)
+    Another solution would be to use $LOGNAME, $USER or $USERNAME
+    """
+    if sys.platform != 'win32':
+        import pwd # Platforms: Unix
+        return pwd.getpwuid(os.getuid())[0]
+    else:
+        return os.environ['USERNAME']
+
+def generate_password(length=8, vocab=string.ascii_letters + string.digits):
+    """dumb password generation function"""
+    pwd = ''
+    for i in range(length):
+        pwd += random.choice(vocab)
+    return pwd
diff --git a/logilab/common/sphinx_ext.py b/logilab/common/sphinx_ext.py
new file mode 100644
index 0000000..a24608c
--- /dev/null
+++ b/logilab/common/sphinx_ext.py
@@ -0,0 +1,87 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+from logilab.common.decorators import monkeypatch
+
+from sphinx.ext import autodoc
+
+class DocstringOnlyModuleDocumenter(autodoc.ModuleDocumenter):
+    objtype = 'docstring'
+    def format_signature(self):
+        pass
+    def add_directive_header(self, sig):
+        pass
+    def document_members(self, all_members=False):
+        pass
+
+    def resolve_name(self, modname, parents, path, base):
+        if modname is not None:
+            return modname, parents + [base]
+        return (path or '') + base, []
+
+
+#autodoc.add_documenter(DocstringOnlyModuleDocumenter)
+
+def setup(app):
+    app.add_autodocumenter(DocstringOnlyModuleDocumenter)
+
+
+
+from sphinx.ext.autodoc import (ViewList, Options, AutodocReporter, nodes,
+                                assemble_option_dict, nested_parse_with_titles)
+
+@monkeypatch(autodoc.AutoDirective)
+def run(self):
+    self.filename_set = set()  # a set of dependent filenames
+    self.reporter = self.state.document.reporter
+    self.env = self.state.document.settings.env
+    self.warnings = []
+    self.result = ViewList()
+
+    # find out what documenter to call
+    objtype = self.name[4:]
+    doc_class = self._registry[objtype]
+    # process the options with the selected documenter's option_spec
+    self.genopt = Options(assemble_option_dict(
+        self.options.items(), doc_class.option_spec))
+    # generate the output
+    documenter = doc_class(self, self.arguments[0])
+    documenter.generate(more_content=self.content)
+    if not self.result:
+        return self.warnings
+
+    # record all filenames as dependencies -- this will at least
+    # partially make automatic invalidation possible
+    for fn in self.filename_set:
+        self.env.note_dependency(fn)
+
+    # use a custom reporter that correctly assigns lines to source
+    # filename/description and lineno
+    old_reporter = self.state.memo.reporter
+    self.state.memo.reporter = AutodocReporter(self.result,
+                                               self.state.memo.reporter)
+    if self.name in ('automodule', 'autodocstring'):
+        node = nodes.section()
+        # necessary so that the child nodes get the right source/line set
+        node.document = self.state.document
+        nested_parse_with_titles(self.state, self.result, node)
+    else:
+        node = nodes.paragraph()
+        node.document = self.state.document
+        self.state.nested_parse(self.result, 0, node)
+    self.state.memo.reporter = old_reporter
+    return self.warnings + node.children
diff --git a/logilab/common/sphinxutils.py b/logilab/common/sphinxutils.py
new file mode 100644
index 0000000..ab6e8a1
--- /dev/null
+++ b/logilab/common/sphinxutils.py
@@ -0,0 +1,122 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Sphinx utils
+
+ModuleGenerator: Generate a file that lists all the modules of a list of
+packages in order to pull all the docstring.
+This should not be used in a makefile to systematically generate sphinx
+documentation!
+
+Typical usage:
+
+>>> from logilab.common.sphinxutils import ModuleGenerator
+>>> mgen = ModuleGenerator('logilab common', '/home/adim/src/logilab/common')
+>>> mgen.generate('api_logilab_common.rst', exclude_dirs=('test',))
+"""
+
+import os, sys
+import os.path as osp
+import inspect
+
+from logilab.common import STD_BLACKLIST
+from logilab.common.shellutils import globfind
+from logilab.common.modutils import load_module_from_file, modpath_from_file
+
+def module_members(module):
+    members = []
+    for name, value in inspect.getmembers(module):
+        if getattr(value, '__module__', None) == module.__name__:
+            members.append( (name, value) )
+    return sorted(members)
+
+
+def class_members(klass):
+    return sorted([name for name in vars(klass)
+                   if name not in ('__doc__', '__module__',
+                                   '__dict__', '__weakref__')])
+
+class ModuleGenerator:
+    file_header = """.. -*- coding: utf-8 -*-\n\n%s\n"""
+    module_def = """
+:mod:`%s`
+=======%s
+
+.. automodule:: %s
+   :members: %s
+"""
+    class_def = """
+
+.. autoclass:: %s
+   :members: %s
+
+"""
+
+    def __init__(self, project_title, code_dir):
+        self.title = project_title
+        self.code_dir = osp.abspath(code_dir)
+
+    def generate(self, dest_file, exclude_dirs=STD_BLACKLIST):
+        """make the module file"""
+        self.fn = open(dest_file, 'w')
+        num = len(self.title) + 6
+        title = "=" * num + "\n %s API\n" % self.title + "=" * num
+        self.fn.write(self.file_header % title)
+        self.gen_modules(exclude_dirs=exclude_dirs)
+        self.fn.close()
+
+    def gen_modules(self, exclude_dirs):
+        """generate all modules"""
+        for module in self.find_modules(exclude_dirs):
+            modname = module.__name__
+            classes = []
+            modmembers = []
+            for objname, obj in module_members(module):
+                if inspect.isclass(obj):
+                    classmembers = class_members(obj)
+                    classes.append( (objname, classmembers) )
+                else:
+                    modmembers.append(objname)
+            self.fn.write(self.module_def % (modname, '=' * len(modname),
+                                             modname,
+                                             ', '.join(modmembers)))
+            for klass, members in classes:
+                self.fn.write(self.class_def % (klass, ', '.join(members)))
+
+    def find_modules(self, exclude_dirs):
+        basepath = osp.dirname(self.code_dir)
+        basedir = osp.basename(basepath) + osp.sep
+        if basedir not in sys.path:
+            sys.path.insert(1, basedir)
+        for filepath in globfind(self.code_dir, '*.py', exclude_dirs):
+            if osp.basename(filepath) in ('setup.py', '__pkginfo__.py'):
+                continue
+            try:
+                module = load_module_from_file(filepath)
+            except: # module might be broken or magic
+                dotted_path = modpath_from_file(filepath)
+                module = type('.'.join(dotted_path), (), {}) # mock it
+            yield module
+
+
+if __name__ == '__main__':
+    # example :
+    title, code_dir, outfile = sys.argv[1:]
+    generator = ModuleGenerator(title, code_dir)
+    # XXX modnames = ['logilab']
+    generator.generate(outfile, ('test', 'tests', 'examples',
+                             'data', 'doc', '.hg', 'migration'))
diff --git a/logilab/common/table.py b/logilab/common/table.py
new file mode 100644
index 0000000..2f3df69
--- /dev/null
+++ b/logilab/common/table.py
@@ -0,0 +1,929 @@
+# copyright 2003-2012 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Table management module."""
+
+from __future__ import print_function
+
+__docformat__ = "restructuredtext en"
+
+from six.moves import range
+
+class Table(object):
+    """Table defines a data table with column and row names.
+    inv:
+        len(self.data) <= len(self.row_names)
+        forall(self.data, lambda x: len(x) <= len(self.col_names))
+    """
+
+    def __init__(self, default_value=0, col_names=None, row_names=None):
+        self.col_names = []
+        self.row_names = []
+        self.data = []
+        self.default_value = default_value
+        if col_names:
+            self.create_columns(col_names)
+        if row_names:
+            self.create_rows(row_names)
+
+    def _next_row_name(self):
+        return 'row%s' % (len(self.row_names)+1)
+
+    def __iter__(self):
+        return iter(self.data)
+
+    def __eq__(self, other):
+        if other is None:
+            return False
+        else:
+            return list(self) == list(other)
+
+    __hash__ = object.__hash__
+
+    def __ne__(self, other):
+        return not self == other
+
+    def __len__(self):
+        return len(self.row_names)
+
+    ## Rows / Columns creation #################################################
+    def create_rows(self, row_names):
+        """Appends row_names to the list of existing rows
+        """
+        self.row_names.extend(row_names)
+        for row_name in row_names:
+            self.data.append([self.default_value]*len(self.col_names))
+
+    def create_columns(self, col_names):
+        """Appends col_names to the list of existing columns
+        """
+        for col_name in col_names:
+            self.create_column(col_name)
+
+    def create_row(self, row_name=None):
+        """Creates a rowname to the row_names list
+        """
+        row_name = row_name or self._next_row_name()
+        self.row_names.append(row_name)
+        self.data.append([self.default_value]*len(self.col_names))
+
+
+    def create_column(self, col_name):
+        """Creates a colname to the col_names list
+        """
+        self.col_names.append(col_name)
+        for row in self.data:
+            row.append(self.default_value)
+
+    ## Sort by column ##########################################################
+    def sort_by_column_id(self, col_id, method = 'asc'):
+        """Sorts the table (in-place) according to data stored in col_id
+        """
+        try:
+            col_index = self.col_names.index(col_id)
+            self.sort_by_column_index(col_index, method)
+        except ValueError:
+            raise KeyError("Col (%s) not found in table" % (col_id))
+
+
+    def sort_by_column_index(self, col_index, method = 'asc'):
+        """Sorts the table 'in-place' according to data stored in col_index
+
+        method should be in ('asc', 'desc')
+        """
+        sort_list = sorted([(row[col_index], row, row_name)
+                     for row, row_name in zip(self.data, self.row_names)])
+        # Sorting sort_list will sort according to col_index
+        # If we want reverse sort, then reverse list
+        if method.lower() == 'desc':
+            sort_list.reverse()
+
+        # Rebuild data / row names
+        self.data = []
+        self.row_names = []
+        for val, row, row_name in sort_list:
+            self.data.append(row)
+            self.row_names.append(row_name)
+
+    def groupby(self, colname, *others):
+        """builds indexes of data
+        :returns: nested dictionaries pointing to actual rows
+        """
+        groups = {}
+        colnames = (colname,) + others
+        col_indexes = [self.col_names.index(col_id) for col_id in colnames]
+        for row in self.data:
+            ptr = groups
+            for col_index in col_indexes[:-1]:
+                ptr = ptr.setdefault(row[col_index], {})
+            ptr = ptr.setdefault(row[col_indexes[-1]],
+                                 Table(default_value=self.default_value,
+                                       col_names=self.col_names))
+            ptr.append_row(tuple(row))
+        return groups
+
+    def select(self, colname, value):
+        grouped = self.groupby(colname)
+        try:
+            return grouped[value]
+        except KeyError:
+            return []
+
+    def remove(self, colname, value):
+        col_index = self.col_names.index(colname)
+        for row in self.data[:]:
+            if row[col_index] == value:
+                self.data.remove(row)
+
+
+    ## The 'setter' part #######################################################
+    def set_cell(self, row_index, col_index, data):
+        """sets value of cell 'row_indew', 'col_index' to data
+        """
+        self.data[row_index][col_index] = data
+
+
+    def set_cell_by_ids(self, row_id, col_id, data):
+        """sets value of cell mapped by row_id and col_id to data
+        Raises a KeyError if row_id or col_id are not found in the table
+        """
+        try:
+            row_index = self.row_names.index(row_id)
+        except ValueError:
+            raise KeyError("Row (%s) not found in table" % (row_id))
+        else:
+            try:
+                col_index = self.col_names.index(col_id)
+                self.data[row_index][col_index] = data
+            except ValueError:
+                raise KeyError("Column (%s) not found in table" % (col_id))
+
+
+    def set_row(self, row_index, row_data):
+        """sets the 'row_index' row
+        pre:
+            type(row_data) == types.ListType
+            len(row_data) == len(self.col_names)
+        """
+        self.data[row_index] = row_data
+
+
+    def set_row_by_id(self, row_id, row_data):
+        """sets the 'row_id' column
+        pre:
+            type(row_data) == types.ListType
+            len(row_data) == len(self.row_names)
+        Raises a KeyError if row_id is not found
+        """
+        try:
+            row_index = self.row_names.index(row_id)
+            self.set_row(row_index, row_data)
+        except ValueError:
+            raise KeyError('Row (%s) not found in table' % (row_id))
+
+
+    def append_row(self, row_data, row_name=None):
+        """Appends a row to the table
+        pre:
+            type(row_data) == types.ListType
+            len(row_data) == len(self.col_names)
+        """
+        row_name = row_name or self._next_row_name()
+        self.row_names.append(row_name)
+        self.data.append(row_data)
+        return len(self.data) - 1
+
+    def insert_row(self, index, row_data, row_name=None):
+        """Appends row_data before 'index' in the table. To make 'insert'
+        behave like 'list.insert', inserting in an out of range index will
+        insert row_data to the end of the list
+        pre:
+            type(row_data) == types.ListType
+            len(row_data) == len(self.col_names)
+        """
+        row_name = row_name or self._next_row_name()
+        self.row_names.insert(index, row_name)
+        self.data.insert(index, row_data)
+
+
+    def delete_row(self, index):
+        """Deletes the 'index' row in the table, and returns it.
+        Raises an IndexError if index is out of range
+        """
+        self.row_names.pop(index)
+        return self.data.pop(index)
+
+
+    def delete_row_by_id(self, row_id):
+        """Deletes the 'row_id' row in the table.
+        Raises a KeyError if row_id was not found.
+        """
+        try:
+            row_index = self.row_names.index(row_id)
+            self.delete_row(row_index)
+        except ValueError:
+            raise KeyError('Row (%s) not found in table' % (row_id))
+
+
+    def set_column(self, col_index, col_data):
+        """sets the 'col_index' column
+        pre:
+            type(col_data) == types.ListType
+            len(col_data) == len(self.row_names)
+        """
+
+        for row_index, cell_data in enumerate(col_data):
+            self.data[row_index][col_index] = cell_data
+
+
+    def set_column_by_id(self, col_id, col_data):
+        """sets the 'col_id' column
+        pre:
+            type(col_data) == types.ListType
+            len(col_data) == len(self.col_names)
+        Raises a KeyError if col_id is not found
+        """
+        try:
+            col_index = self.col_names.index(col_id)
+            self.set_column(col_index, col_data)
+        except ValueError:
+            raise KeyError('Column (%s) not found in table' % (col_id))
+
+
+    def append_column(self, col_data, col_name):
+        """Appends the 'col_index' column
+        pre:
+            type(col_data) == types.ListType
+            len(col_data) == len(self.row_names)
+        """
+        self.col_names.append(col_name)
+        for row_index, cell_data in enumerate(col_data):
+            self.data[row_index].append(cell_data)
+
+
+    def insert_column(self, index, col_data, col_name):
+        """Appends col_data before 'index' in the table. To make 'insert'
+        behave like 'list.insert', inserting in an out of range index will
+        insert col_data to the end of the list
+        pre:
+            type(col_data) == types.ListType
+            len(col_data) == len(self.row_names)
+        """
+        self.col_names.insert(index, col_name)
+        for row_index, cell_data in enumerate(col_data):
+            self.data[row_index].insert(index, cell_data)
+
+
+    def delete_column(self, index):
+        """Deletes the 'index' column in the table, and returns it.
+        Raises an IndexError if index is out of range
+        """
+        self.col_names.pop(index)
+        return [row.pop(index) for row in self.data]
+
+
+    def delete_column_by_id(self, col_id):
+        """Deletes the 'col_id' col in the table.
+        Raises a KeyError if col_id was not found.
+        """
+        try:
+            col_index = self.col_names.index(col_id)
+            self.delete_column(col_index)
+        except ValueError:
+            raise KeyError('Column (%s) not found in table' % (col_id))
+
+
+    ## The 'getter' part #######################################################
+
+    def get_shape(self):
+        """Returns a tuple which represents the table's shape
+        """
+        return len(self.row_names), len(self.col_names)
+    shape = property(get_shape)
+
+    def __getitem__(self, indices):
+        """provided for convenience"""
+        rows, multirows = None, False
+        cols, multicols = None, False
+        if isinstance(indices, tuple):
+            rows = indices[0]
+            if len(indices) > 1:
+                cols = indices[1]
+        else:
+            rows = indices
+        # define row slice
+        if isinstance(rows, str):
+            try:
+                rows = self.row_names.index(rows)
+            except ValueError:
+                raise KeyError("Row (%s) not found in table" % (rows))
+        if isinstance(rows, int):
+            rows = slice(rows, rows+1)
+            multirows = False
+        else:
+            rows = slice(None)
+            multirows = True
+        # define col slice
+        if isinstance(cols, str):
+            try:
+                cols = self.col_names.index(cols)
+            except ValueError:
+                raise KeyError("Column (%s) not found in table" % (cols))
+        if isinstance(cols, int):
+            cols = slice(cols, cols+1)
+            multicols = False
+        else:
+            cols = slice(None)
+            multicols = True
+        # get sub-table
+        tab = Table()
+        tab.default_value = self.default_value
+        tab.create_rows(self.row_names[rows])
+        tab.create_columns(self.col_names[cols])
+        for idx, row in enumerate(self.data[rows]):
+            tab.set_row(idx, row[cols])
+        if multirows :
+            if multicols:
+                return tab
+            else:
+                return [item[0] for item in tab.data]
+        else:
+            if multicols:
+                return tab.data[0]
+            else:
+                return tab.data[0][0]
+
+    def get_cell_by_ids(self, row_id, col_id):
+        """Returns the element at [row_id][col_id]
+        """
+        try:
+            row_index = self.row_names.index(row_id)
+        except ValueError:
+            raise KeyError("Row (%s) not found in table" % (row_id))
+        else:
+            try:
+                col_index = self.col_names.index(col_id)
+            except ValueError:
+                raise KeyError("Column (%s) not found in table" % (col_id))
+        return self.data[row_index][col_index]
+
+    def get_row_by_id(self, row_id):
+        """Returns the 'row_id' row
+        """
+        try:
+            row_index = self.row_names.index(row_id)
+        except ValueError:
+            raise KeyError("Row (%s) not found in table" % (row_id))
+        return self.data[row_index]
+
+    def get_column_by_id(self, col_id, distinct=False):
+        """Returns the 'col_id' col
+        """
+        try:
+            col_index = self.col_names.index(col_id)
+        except ValueError:
+            raise KeyError("Column (%s) not found in table" % (col_id))
+        return self.get_column(col_index, distinct)
+
+    def get_columns(self):
+        """Returns all the columns in the table
+        """
+        return [self[:, index] for index in range(len(self.col_names))]
+
+    def get_column(self, col_index, distinct=False):
+        """get a column by index"""
+        col = [row[col_index] for row in self.data]
+        if distinct:
+            col = list(set(col))
+        return col
+
+    def apply_stylesheet(self, stylesheet):
+        """Applies the stylesheet to this table
+        """
+        for instruction in stylesheet.instructions:
+            eval(instruction)
+
+
+    def transpose(self):
+        """Keeps the self object intact, and returns the transposed (rotated)
+        table.
+        """
+        transposed = Table()
+        transposed.create_rows(self.col_names)
+        transposed.create_columns(self.row_names)
+        for col_index, column in enumerate(self.get_columns()):
+            transposed.set_row(col_index, column)
+        return transposed
+
+
+    def pprint(self):
+        """returns a string representing the table in a pretty
+        printed 'text' format.
+        """
+        # The maximum row name (to know the start_index of the first col)
+        max_row_name = 0
+        for row_name in self.row_names:
+            if len(row_name) > max_row_name:
+                max_row_name = len(row_name)
+        col_start = max_row_name + 5
+
+        lines = []
+        # Build the 'first' line <=> the col_names one
+        # The first cell <=> an empty one
+        col_names_line = [' '*col_start]
+        for col_name in self.col_names:
+            col_names_line.append(col_name + ' '*5)
+        lines.append('|' + '|'.join(col_names_line) + '|')
+        max_line_length = len(lines[0])
+
+        # Build the table
+        for row_index, row in enumerate(self.data):
+            line = []
+            # First, build the row_name's cell
+            row_name = self.row_names[row_index]
+            line.append(row_name + ' '*(col_start-len(row_name)))
+
+            # Then, build all the table's cell for this line.
+            for col_index, cell in enumerate(row):
+                col_name_length = len(self.col_names[col_index]) + 5
+                data = str(cell)
+                line.append(data + ' '*(col_name_length - len(data)))
+            lines.append('|' + '|'.join(line) + '|')
+            if len(lines[-1]) > max_line_length:
+                max_line_length = len(lines[-1])
+
+        # Wrap the table with '-' to make a frame
+        lines.insert(0, '-'*max_line_length)
+        lines.append('-'*max_line_length)
+        return '\n'.join(lines)
+
+
+    def __repr__(self):
+        return repr(self.data)
+
+    def as_text(self):
+        data = []
+        # We must convert cells into strings before joining them
+        for row in self.data:
+            data.append([str(cell) for cell in row])
+        lines = ['\t'.join(row) for row in data]
+        return '\n'.join(lines)
+
+
+
+class TableStyle:
+    """Defines a table's style
+    """
+
+    def __init__(self, table):
+
+        self._table = table
+        self.size = dict([(col_name, '1*') for col_name in table.col_names])
+        # __row_column__ is a special key to define the first column which
+        # actually has no name (<=> left most column <=> row names column)
+        self.size['__row_column__'] = '1*'
+        self.alignment = dict([(col_name, 'right')
+                               for col_name in table.col_names])
+        self.alignment['__row_column__'] = 'right'
+
+        # We shouldn't have to create an entry for
+        # the 1st col (the row_column one)
+        self.units = dict([(col_name, '') for col_name in table.col_names])
+        self.units['__row_column__'] = ''
+
+    # XXX FIXME : params order should be reversed for all set() methods
+    def set_size(self, value, col_id):
+        """sets the size of the specified col_id to value
+        """
+        self.size[col_id] = value
+
+    def set_size_by_index(self, value, col_index):
+        """Allows to set the size according to the column index rather than
+        using the column's id.
+        BE CAREFUL : the '0' column is the '__row_column__' one !
+        """
+        if col_index == 0:
+            col_id = '__row_column__'
+        else:
+            col_id = self._table.col_names[col_index-1]
+
+        self.size[col_id] = value
+
+
+    def set_alignment(self, value, col_id):
+        """sets the alignment of the specified col_id to value
+        """
+        self.alignment[col_id] = value
+
+
+    def set_alignment_by_index(self, value, col_index):
+        """Allows to set the alignment according to the column index rather than
+        using the column's id.
+        BE CAREFUL : the '0' column is the '__row_column__' one !
+        """
+        if col_index == 0:
+            col_id = '__row_column__'
+        else:
+            col_id = self._table.col_names[col_index-1]
+
+        self.alignment[col_id] = value
+
+
+    def set_unit(self, value, col_id):
+        """sets the unit of the specified col_id to value
+        """
+        self.units[col_id] = value
+
+
+    def set_unit_by_index(self, value, col_index):
+        """Allows to set the unit according to the column index rather than
+        using the column's id.
+        BE CAREFUL :  the '0' column is the '__row_column__' one !
+        (Note that in the 'unit' case, you shouldn't have to set a unit
+        for the 1st column (the __row__column__ one))
+        """
+        if col_index == 0:
+            col_id = '__row_column__'
+        else:
+            col_id = self._table.col_names[col_index-1]
+
+        self.units[col_id] = value
+
+
+    def get_size(self, col_id):
+        """Returns the size of the specified col_id
+        """
+        return self.size[col_id]
+
+
+    def get_size_by_index(self, col_index):
+        """Allows to get the size  according to the column index rather than
+        using the column's id.
+        BE CAREFUL : the '0' column is the '__row_column__' one !
+        """
+        if col_index == 0:
+            col_id = '__row_column__'
+        else:
+            col_id = self._table.col_names[col_index-1]
+
+        return self.size[col_id]
+
+
+    def get_alignment(self, col_id):
+        """Returns the alignment of the specified col_id
+        """
+        return self.alignment[col_id]
+
+
+    def get_alignment_by_index(self, col_index):
+        """Allors to get the alignment according to the column index rather than
+        using the column's id.
+        BE CAREFUL : the '0' column is the '__row_column__' one !
+        """
+        if col_index == 0:
+            col_id = '__row_column__'
+        else:
+            col_id = self._table.col_names[col_index-1]
+
+        return self.alignment[col_id]
+
+
+    def get_unit(self, col_id):
+        """Returns the unit of the specified col_id
+        """
+        return self.units[col_id]
+
+
+    def get_unit_by_index(self, col_index):
+        """Allors to get the unit according to the column index rather than
+        using the column's id.
+        BE CAREFUL : the '0' column is the '__row_column__' one !
+        """
+        if col_index == 0:
+            col_id = '__row_column__'
+        else:
+            col_id = self._table.col_names[col_index-1]
+
+        return self.units[col_id]
+
+
+import re
+CELL_PROG = re.compile("([0-9]+)_([0-9]+)")
+
+class TableStyleSheet:
+    """A simple Table stylesheet
+    Rules are expressions where cells are defined by the row_index
+    and col_index separated by an underscore ('_').
+    For example, suppose you want to say that the (2,5) cell must be
+    the sum of its two preceding cells in the row, you would create
+    the following rule :
+        2_5 = 2_3 + 2_4
+    You can also use all the math.* operations you want. For example:
+        2_5 = sqrt(2_3**2 + 2_4**2)
+    """
+
+    def __init__(self, rules = None):
+        rules = rules or []
+        self.rules = []
+        self.instructions = []
+        for rule in rules:
+            self.add_rule(rule)
+
+
+    def add_rule(self, rule):
+        """Adds a rule to the stylesheet rules
+        """
+        try:
+            source_code = ['from math import *']
+            source_code.append(CELL_PROG.sub(r'self.data[\1][\2]', rule))
+            self.instructions.append(compile('\n'.join(source_code),
+                'table.py', 'exec'))
+            self.rules.append(rule)
+        except SyntaxError:
+            print("Bad Stylesheet Rule : %s [skipped]" % rule)
+
+
+    def add_rowsum_rule(self, dest_cell, row_index, start_col, end_col):
+        """Creates and adds a rule to sum over the row at row_index from
+        start_col to end_col.
+        dest_cell is a tuple of two elements (x,y) of the destination cell
+        No check is done for indexes ranges.
+        pre:
+            start_col >= 0
+            end_col > start_col
+        """
+        cell_list = ['%d_%d'%(row_index, index) for index in range(start_col,
+                                                                   end_col + 1)]
+        rule = '%d_%d=' % dest_cell + '+'.join(cell_list)
+        self.add_rule(rule)
+
+
+    def add_rowavg_rule(self, dest_cell, row_index, start_col, end_col):
+        """Creates and adds a rule to make the row average (from start_col
+        to end_col)
+        dest_cell is a tuple of two elements (x,y) of the destination cell
+        No check is done for indexes ranges.
+        pre:
+            start_col >= 0
+            end_col > start_col
+        """
+        cell_list = ['%d_%d'%(row_index, index) for index in range(start_col,
+                                                                   end_col + 1)]
+        num = (end_col - start_col + 1)
+        rule = '%d_%d=' % dest_cell + '('+'+'.join(cell_list)+')/%f'%num
+        self.add_rule(rule)
+
+
+    def add_colsum_rule(self, dest_cell, col_index, start_row, end_row):
+        """Creates and adds a rule to sum over the col at col_index from
+        start_row to end_row.
+        dest_cell is a tuple of two elements (x,y) of the destination cell
+        No check is done for indexes ranges.
+        pre:
+            start_row >= 0
+            end_row > start_row
+        """
+        cell_list = ['%d_%d'%(index, col_index) for index in range(start_row,
+                                                                   end_row + 1)]
+        rule = '%d_%d=' % dest_cell + '+'.join(cell_list)
+        self.add_rule(rule)
+
+
+    def add_colavg_rule(self, dest_cell, col_index, start_row, end_row):
+        """Creates and adds a rule to make the col average (from start_row
+        to end_row)
+        dest_cell is a tuple of two elements (x,y) of the destination cell
+        No check is done for indexes ranges.
+        pre:
+            start_row >= 0
+            end_row > start_row
+        """
+        cell_list = ['%d_%d'%(index, col_index) for index in range(start_row,
+                                                                   end_row + 1)]
+        num = (end_row - start_row + 1)
+        rule = '%d_%d=' % dest_cell + '('+'+'.join(cell_list)+')/%f'%num
+        self.add_rule(rule)
+
+
+
+class TableCellRenderer:
+    """Defines a simple text renderer
+    """
+
+    def __init__(self, **properties):
+        """keywords should be properties with an associated boolean as value.
+        For example :
+            renderer = TableCellRenderer(units = True, alignment = False)
+        An unspecified property will have a 'False' value by default.
+        Possible properties are :
+            alignment, unit
+        """
+        self.properties = properties
+
+
+    def render_cell(self, cell_coord, table, table_style):
+        """Renders the cell at 'cell_coord' in the table, using table_style
+        """
+        row_index, col_index = cell_coord
+        cell_value = table.data[row_index][col_index]
+        final_content = self._make_cell_content(cell_value,
+                                                table_style, col_index  +1)
+        return self._render_cell_content(final_content,
+                                         table_style, col_index + 1)
+
+
+    def render_row_cell(self, row_name, table, table_style):
+        """Renders the cell for 'row_id' row
+        """
+        cell_value = row_name
+        return self._render_cell_content(cell_value, table_style, 0)
+
+
+    def render_col_cell(self, col_name, table, table_style):
+        """Renders the cell for 'col_id' row
+        """
+        cell_value = col_name
+        col_index = table.col_names.index(col_name)
+        return self._render_cell_content(cell_value, table_style, col_index +1)
+
+
+
+    def _render_cell_content(self, content, table_style, col_index):
+        """Makes the appropriate rendering for this cell content.
+        Rendering properties will be searched using the
+        *table_style.get_xxx_by_index(col_index)' methods
+
+        **This method should be overridden in the derived renderer classes.**
+        """
+        return content
+
+
+    def _make_cell_content(self, cell_content, table_style, col_index):
+        """Makes the cell content (adds decoration data, like units for
+        example)
+        """
+        final_content = cell_content
+        if 'skip_zero' in self.properties:
+            replacement_char = self.properties['skip_zero']
+        else:
+            replacement_char = 0
+        if replacement_char and final_content == 0:
+            return replacement_char
+
+        try:
+            units_on = self.properties['units']
+            if units_on:
+                final_content = self._add_unit(
+                    cell_content, table_style, col_index)
+        except KeyError:
+            pass
+
+        return final_content
+
+
+    def _add_unit(self, cell_content, table_style, col_index):
+        """Adds unit to the cell_content if needed
+        """
+        unit = table_style.get_unit_by_index(col_index)
+        return str(cell_content) + " " + unit
+
+
+
+class DocbookRenderer(TableCellRenderer):
+    """Defines how to render a cell for a docboook table
+    """
+
+    def define_col_header(self, col_index, table_style):
+        """Computes the colspec element according to the style
+        """
+        size = table_style.get_size_by_index(col_index)
+        return '<colspec colname="c%d" colwidth="%s"/>\n' % \
+               (col_index, size)
+
+
+    def _render_cell_content(self, cell_content, table_style, col_index):
+        """Makes the appropriate rendering for this cell content.
+        Rendering properties will be searched using the
+        table_style.get_xxx_by_index(col_index)' methods.
+        """
+        try:
+            align_on = self.properties['alignment']
+            alignment = table_style.get_alignment_by_index(col_index)
+            if align_on:
+                return "<entry align='%s'>%s</entry>\n" % \
+                       (alignment, cell_content)
+        except KeyError:
+            # KeyError <=> Default alignment
+            return "<entry>%s</entry>\n" % cell_content
+
+
+class TableWriter:
+    """A class to write tables
+    """
+
+    def __init__(self, stream, table, style, **properties):
+        self._stream = stream
+        self.style = style or TableStyle(table)
+        self._table = table
+        self.properties = properties
+        self.renderer = None
+
+
+    def set_style(self, style):
+        """sets the table's associated style
+        """
+        self.style = style
+
+
+    def set_renderer(self, renderer):
+        """sets the way to render cell
+        """
+        self.renderer = renderer
+
+
+    def update_properties(self, **properties):
+        """Updates writer's properties (for cell rendering)
+        """
+        self.properties.update(properties)
+
+
+    def write_table(self, title = ""):
+        """Writes the table
+        """
+        raise NotImplementedError("write_table must be implemented !")
+
+
+
+class DocbookTableWriter(TableWriter):
+    """Defines an implementation of TableWriter to write a table in Docbook
+    """
+
+    def _write_headers(self):
+        """Writes col headers
+        """
+        # Define col_headers (colstpec elements)
+        for col_index in range(len(self._table.col_names)+1):
+            self._stream.write(self.renderer.define_col_header(col_index,
+                                                              self.style))
+
+        self._stream.write("<thead>\n<row>\n")
+        # XXX FIXME : write an empty entry <=> the first (__row_column) column
+        self._stream.write('<entry></entry>\n')
+        for col_name in self._table.col_names:
+            self._stream.write(self.renderer.render_col_cell(
+                col_name, self._table,
+                self.style))
+
+        self._stream.write("</row>\n</thead>\n")
+
+
+    def _write_body(self):
+        """Writes the table body
+        """
+        self._stream.write('<tbody>\n')
+
+        for row_index, row in enumerate(self._table.data):
+            self._stream.write('<row>\n')
+            row_name = self._table.row_names[row_index]
+            # Write the first entry (row_name)
+            self._stream.write(self.renderer.render_row_cell(row_name,
+                                                            self._table,
+                                                            self.style))
+
+            for col_index, cell in enumerate(row):
+                self._stream.write(self.renderer.render_cell(
+                    (row_index, col_index),
+                    self._table, self.style))
+
+            self._stream.write('</row>\n')
+
+        self._stream.write('</tbody>\n')
+
+
+    def write_table(self, title = ""):
+        """Writes the table
+        """
+        self._stream.write('<table>\n<title>%s></title>\n'%(title))
+        self._stream.write(
+            '<tgroup cols="%d" align="left" colsep="1" rowsep="1">\n'%
+            (len(self._table.col_names)+1))
+        self._write_headers()
+        self._write_body()
+
+        self._stream.write('</tgroup>\n</table>\n')
+
+
diff --git a/logilab/common/tasksqueue.py b/logilab/common/tasksqueue.py
new file mode 100644
index 0000000..ed74cf5
--- /dev/null
+++ b/logilab/common/tasksqueue.py
@@ -0,0 +1,101 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Prioritized tasks queue"""
+
+__docformat__ = "restructuredtext en"
+
+from bisect import insort_left
+
+from six.moves import queue
+
+LOW = 0
+MEDIUM = 10
+HIGH = 100
+
+PRIORITY = {
+    'LOW': LOW,
+    'MEDIUM': MEDIUM,
+    'HIGH': HIGH,
+    }
+REVERSE_PRIORITY = dict((values, key) for key, values in PRIORITY.items())
+
+
+
+class PrioritizedTasksQueue(queue.Queue):
+
+    def _init(self, maxsize):
+        """Initialize the queue representation"""
+        self.maxsize = maxsize
+        # ordered list of task, from the lowest to the highest priority
+        self.queue = []
+
+    def _put(self, item):
+        """Put a new item in the queue"""
+        for i, task in enumerate(self.queue):
+            # equivalent task
+            if task == item:
+                # if new task has a higher priority, remove the one already
+                # queued so the new priority will be considered
+                if task < item:
+                    item.merge(task)
+                    del self.queue[i]
+                    break
+                # else keep it so current order is kept
+                task.merge(item)
+                return
+        insort_left(self.queue, item)
+
+    def _get(self):
+        """Get an item from the queue"""
+        return self.queue.pop()
+
+    def __iter__(self):
+        return iter(self.queue)
+
+    def remove(self, tid):
+        """remove a specific task from the queue"""
+        # XXX acquire lock
+        for i, task in enumerate(self):
+            if task.id == tid:
+                self.queue.pop(i)
+                return
+        raise ValueError('not task of id %s in queue' % tid)
+
+class Task(object):
+    def __init__(self, tid, priority=LOW):
+        # task id
+        self.id = tid
+        # task priority
+        self.priority = priority
+
+    def __repr__(self):
+        return '<Task %s @%#x>' % (self.id, id(self))
+
+    def __cmp__(self, other):
+        return cmp(self.priority, other.priority)
+
+    def __lt__(self, other):
+        return self.priority < other.priority
+
+    def __eq__(self, other):
+        return self.id == other.id
+
+    __hash__ = object.__hash__
+
+    def merge(self, other):
+        pass
diff --git a/logilab/common/testlib.py b/logilab/common/testlib.py
new file mode 100644
index 0000000..31efe56
--- /dev/null
+++ b/logilab/common/testlib.py
@@ -0,0 +1,1392 @@
+# -*- coding: utf-8 -*-
+# copyright 2003-2012 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Run tests.
+
+This will find all modules whose name match a given prefix in the test
+directory, and run them. Various command line options provide
+additional facilities.
+
+Command line options:
+
+ -v  verbose -- run tests in verbose mode with output to stdout
+ -q  quiet   -- don't print anything except if a test fails
+ -t  testdir -- directory where the tests will be found
+ -x  exclude -- add a test to exclude
+ -p  profile -- profiled execution
+ -d  dbc     -- enable design-by-contract
+ -m  match   -- only run test matching the tag pattern which follow
+
+If no non-option arguments are present, prefixes used are 'test',
+'regrtest', 'smoketest' and 'unittest'.
+
+"""
+
+from __future__ import print_function
+
+__docformat__ = "restructuredtext en"
+# modified copy of some functions from test/regrtest.py from PyXml
+# disable camel case warning
+# pylint: disable=C0103
+
+import sys
+import os, os.path as osp
+import re
+import traceback
+import inspect
+import difflib
+import tempfile
+import math
+import warnings
+from shutil import rmtree
+from operator import itemgetter
+from itertools import dropwhile
+from inspect import isgeneratorfunction
+
+from six import string_types
+from six.moves import builtins, range, configparser, input
+
+from logilab.common.deprecation import deprecated
+
+import unittest as unittest_legacy
+if not getattr(unittest_legacy, "__package__", None):
+    try:
+        import unittest2 as unittest
+        from unittest2 import SkipTest
+    except ImportError:
+        raise ImportError("You have to install python-unittest2 to use %s" % __name__)
+else:
+    import unittest
+    from unittest import SkipTest
+
+from functools import wraps
+
+from logilab.common.debugger import Debugger, colorize_source
+from logilab.common.decorators import cached, classproperty
+from logilab.common import textutils
+
+
+__all__ = ['main', 'unittest_main', 'find_tests', 'run_test', 'spawn']
+
+DEFAULT_PREFIXES = ('test', 'regrtest', 'smoketest', 'unittest',
+                    'func', 'validation')
+
+is_generator = deprecated('[lgc 0.63] use inspect.isgeneratorfunction')(isgeneratorfunction)
+
+# used by unittest to count the number of relevant levels in the traceback
+__unittest = 1
+
+
+def with_tempdir(callable):
+    """A decorator ensuring no temporary file left when the function return
+    Work only for temporary file create with the tempfile module"""
+    if isgeneratorfunction(callable):
+        def proxy(*args, **kwargs):
+            old_tmpdir = tempfile.gettempdir()
+            new_tmpdir = tempfile.mkdtemp(prefix="temp-lgc-")
+            tempfile.tempdir = new_tmpdir
+            try:
+                for x in callable(*args, **kwargs):
+                    yield x
+            finally:
+                try:
+                    rmtree(new_tmpdir, ignore_errors=True)
+                finally:
+                    tempfile.tempdir = old_tmpdir
+        return proxy
+
+    @wraps(callable)
+    def proxy(*args, **kargs):
+
+        old_tmpdir = tempfile.gettempdir()
+        new_tmpdir = tempfile.mkdtemp(prefix="temp-lgc-")
+        tempfile.tempdir = new_tmpdir
+        try:
+            return callable(*args, **kargs)
+        finally:
+            try:
+                rmtree(new_tmpdir, ignore_errors=True)
+            finally:
+                tempfile.tempdir = old_tmpdir
+    return proxy
+
+def in_tempdir(callable):
+    """A decorator moving the enclosed function inside the tempfile.tempfdir
+    """
+    @wraps(callable)
+    def proxy(*args, **kargs):
+
+        old_cwd = os.getcwd()
+        os.chdir(tempfile.tempdir)
+        try:
+            return callable(*args, **kargs)
+        finally:
+            os.chdir(old_cwd)
+    return proxy
+
+def within_tempdir(callable):
+    """A decorator run the enclosed function inside a tmpdir removed after execution
+    """
+    proxy = with_tempdir(in_tempdir(callable))
+    proxy.__name__ = callable.__name__
+    return proxy
+
+def find_tests(testdir,
+               prefixes=DEFAULT_PREFIXES, suffix=".py",
+               excludes=(),
+               remove_suffix=True):
+    """
+    Return a list of all applicable test modules.
+    """
+    tests = []
+    for name in os.listdir(testdir):
+        if not suffix or name.endswith(suffix):
+            for prefix in prefixes:
+                if name.startswith(prefix):
+                    if remove_suffix and name.endswith(suffix):
+                        name = name[:-len(suffix)]
+                    if name not in excludes:
+                        tests.append(name)
+    tests.sort()
+    return tests
+
+
+## PostMortem Debug facilities #####
+def start_interactive_mode(result):
+    """starts an interactive shell so that the user can inspect errors
+    """
+    debuggers = result.debuggers
+    descrs = result.error_descrs + result.fail_descrs
+    if len(debuggers) == 1:
+        # don't ask for test name if there's only one failure
+        debuggers[0].start()
+    else:
+        while True:
+            testindex = 0
+            print("Choose a test to debug:")
+            # order debuggers in the same way than errors were printed
+            print("\n".join(['\t%s : %s' % (i, descr) for i, (_, descr)
+                  in enumerate(descrs)]))
+            print("Type 'exit' (or ^D) to quit")
+            print()
+            try:
+                todebug = input('Enter a test name: ')
+                if todebug.strip().lower() == 'exit':
+                    print()
+                    break
+                else:
+                    try:
+                        testindex = int(todebug)
+                        debugger = debuggers[descrs[testindex][0]]
+                    except (ValueError, IndexError):
+                        print("ERROR: invalid test number %r" % (todebug, ))
+                    else:
+                        debugger.start()
+            except (EOFError, KeyboardInterrupt):
+                print()
+                break
+
+
+# test utils ##################################################################
+
+class SkipAwareTestResult(unittest._TextTestResult):
+
+    def __init__(self, stream, descriptions, verbosity,
+                 exitfirst=False, pdbmode=False, cvg=None, colorize=False):
+        super(SkipAwareTestResult, self).__init__(stream,
+                                                  descriptions, verbosity)
+        self.skipped = []
+        self.debuggers = []
+        self.fail_descrs = []
+        self.error_descrs = []
+        self.exitfirst = exitfirst
+        self.pdbmode = pdbmode
+        self.cvg = cvg
+        self.colorize = colorize
+        self.pdbclass = Debugger
+        self.verbose = verbosity > 1
+
+    def descrs_for(self, flavour):
+        return getattr(self, '%s_descrs' % flavour.lower())
+
+    def _create_pdb(self, test_descr, flavour):
+        self.descrs_for(flavour).append( (len(self.debuggers), test_descr) )
+        if self.pdbmode:
+            self.debuggers.append(self.pdbclass(sys.exc_info()[2]))
+
+    def _iter_valid_frames(self, frames):
+        """only consider non-testlib frames when formatting  traceback"""
+        lgc_testlib = osp.abspath(__file__)
+        std_testlib = osp.abspath(unittest.__file__)
+        invalid = lambda fi: osp.abspath(fi[1]) in (lgc_testlib, std_testlib)
+        for frameinfo in dropwhile(invalid, frames):
+            yield frameinfo
+
+    def _exc_info_to_string(self, err, test):
+        """Converts a sys.exc_info()-style tuple of values into a string.
+
+        This method is overridden here because we want to colorize
+        lines if --color is passed, and display local variables if
+        --verbose is passed
+        """
+        exctype, exc, tb = err
+        output = ['Traceback (most recent call last)']
+        frames = inspect.getinnerframes(tb)
+        colorize = self.colorize
+        frames = enumerate(self._iter_valid_frames(frames))
+        for index, (frame, filename, lineno, funcname, ctx, ctxindex) in frames:
+            filename = osp.abspath(filename)
+            if ctx is None: # pyc files or C extensions for instance
+                source = '<no source available>'
+            else:
+                source = ''.join(ctx)
+            if colorize:
+                filename = textutils.colorize_ansi(filename, 'magenta')
+                source = colorize_source(source)
+            output.append('  File "%s", line %s, in %s' % (filename, lineno, funcname))
+            output.append('    %s' % source.strip())
+            if self.verbose:
+                output.append('%r == %r' % (dir(frame), test.__module__))
+                output.append('')
+                output.append('    ' + ' local variables '.center(66, '-'))
+                for varname, value in sorted(frame.f_locals.items()):
+                    output.append('    %s: %r' % (varname, value))
+                    if varname == 'self': # special handy processing for self
+                        for varname, value in sorted(vars(value).items()):
+                            output.append('      self.%s: %r' % (varname, value))
+                output.append('    ' + '-' * 66)
+                output.append('')
+        output.append(''.join(traceback.format_exception_only(exctype, exc)))
+        return '\n'.join(output)
+
+    def addError(self, test, err):
+        """err ->  (exc_type, exc, tcbk)"""
+        exc_type, exc, _ = err
+        if isinstance(exc, SkipTest):
+            assert exc_type == SkipTest
+            self.addSkip(test, exc)
+        else:
+            if self.exitfirst:
+                self.shouldStop = True
+            descr = self.getDescription(test)
+            super(SkipAwareTestResult, self).addError(test, err)
+            self._create_pdb(descr, 'error')
+
+    def addFailure(self, test, err):
+        if self.exitfirst:
+            self.shouldStop = True
+        descr = self.getDescription(test)
+        super(SkipAwareTestResult, self).addFailure(test, err)
+        self._create_pdb(descr, 'fail')
+
+    def addSkip(self, test, reason):
+        self.skipped.append((test, reason))
+        if self.showAll:
+            self.stream.writeln("SKIPPED")
+        elif self.dots:
+            self.stream.write('S')
+
+    def printErrors(self):
+        super(SkipAwareTestResult, self).printErrors()
+        self.printSkippedList()
+
+    def printSkippedList(self):
+        # format (test, err) compatible with unittest2
+        for test, err in self.skipped:
+            descr = self.getDescription(test)
+            self.stream.writeln(self.separator1)
+            self.stream.writeln("%s: %s" % ('SKIPPED', descr))
+            self.stream.writeln("\t%s" % err)
+
+    def printErrorList(self, flavour, errors):
+        for (_, descr), (test, err) in zip(self.descrs_for(flavour), errors):
+            self.stream.writeln(self.separator1)
+            self.stream.writeln("%s: %s" % (flavour, descr))
+            self.stream.writeln(self.separator2)
+            self.stream.writeln(err)
+            self.stream.writeln('no stdout'.center(len(self.separator2)))
+            self.stream.writeln('no stderr'.center(len(self.separator2)))
+
+# Add deprecation warnings about new api used by module level fixtures in unittest2
+# http://www.voidspace.org.uk/python/articles/unittest2.shtml#setupmodule-and-teardownmodule
+class _DebugResult(object): # simplify import statement among unittest flavors..
+    "Used by the TestSuite to hold previous class when running in debug."
+    _previousTestClass = None
+    _moduleSetUpFailed = False
+    shouldStop = False
+
+from logilab.common.decorators import monkeypatch
+@monkeypatch(unittest.TestSuite)
+def _handleModuleTearDown(self, result):
+    previousModule = self._get_previous_module(result)
+    if previousModule is None:
+        return
+    if result._moduleSetUpFailed:
+        return
+    try:
+        module = sys.modules[previousModule]
+    except KeyError:
+        return
+    # add testlib specific deprecation warning and switch to new api
+    if hasattr(module, 'teardown_module'):
+        warnings.warn('Please rename teardown_module() to tearDownModule() instead.',
+                      DeprecationWarning)
+        setattr(module, 'tearDownModule', module.teardown_module)
+    # end of monkey-patching
+    tearDownModule = getattr(module, 'tearDownModule', None)
+    if tearDownModule is not None:
+        try:
+            tearDownModule()
+        except Exception as e:
+            if isinstance(result, _DebugResult):
+                raise
+            errorName = 'tearDownModule (%s)' % previousModule
+            self._addClassOrModuleLevelException(result, e, errorName)
+
+@monkeypatch(unittest.TestSuite)
+def _handleModuleFixture(self, test, result):
+    previousModule = self._get_previous_module(result)
+    currentModule = test.__class__.__module__
+    if currentModule == previousModule:
+        return
+    self._handleModuleTearDown(result)
+    result._moduleSetUpFailed = False
+    try:
+        module = sys.modules[currentModule]
+    except KeyError:
+        return
+    # add testlib specific deprecation warning and switch to new api
+    if hasattr(module, 'setup_module'):
+        warnings.warn('Please rename setup_module() to setUpModule() instead.',
+                      DeprecationWarning)
+        setattr(module, 'setUpModule', module.setup_module)
+    # end of monkey-patching
+    setUpModule = getattr(module, 'setUpModule', None)
+    if setUpModule is not None:
+        try:
+            setUpModule()
+        except Exception as e:
+            if isinstance(result, _DebugResult):
+                raise
+            result._moduleSetUpFailed = True
+            errorName = 'setUpModule (%s)' % currentModule
+            self._addClassOrModuleLevelException(result, e, errorName)
+
+# backward compatibility: TestSuite might be imported from lgc.testlib
+TestSuite = unittest.TestSuite
+
+class keywords(dict):
+    """Keyword args (**kwargs) support for generative tests."""
+
+class starargs(tuple):
+    """Variable arguments (*args) for generative tests."""
+    def __new__(cls, *args):
+        return tuple.__new__(cls, args)
+
+unittest_main = unittest.main
+
+
+class InnerTestSkipped(SkipTest):
+    """raised when a test is skipped"""
+    pass
+
+def parse_generative_args(params):
+    args = []
+    varargs = ()
+    kwargs = {}
+    flags = 0 # 2 <=> starargs, 4 <=> kwargs
+    for param in params:
+        if isinstance(param, starargs):
+            varargs = param
+            if flags:
+                raise TypeError('found starargs after keywords !')
+            flags |= 2
+            args += list(varargs)
+        elif isinstance(param, keywords):
+            kwargs = param
+            if flags & 4:
+                raise TypeError('got multiple keywords parameters')
+            flags |= 4
+        elif flags & 2 or flags & 4:
+            raise TypeError('found parameters after kwargs or args')
+        else:
+            args.append(param)
+
+    return args, kwargs
+
+
+class InnerTest(tuple):
+    def __new__(cls, name, *data):
+        instance = tuple.__new__(cls, data)
+        instance.name = name
+        return instance
+
+class Tags(set):
+    """A set of tag able validate an expression"""
+
+    def __init__(self, *tags, **kwargs):
+        self.inherit = kwargs.pop('inherit', True)
+        if kwargs:
+           raise TypeError("%s are an invalid keyword argument for this function" % kwargs.keys())
+
+        if len(tags) == 1 and not isinstance(tags[0], string_types):
+            tags = tags[0]
+        super(Tags, self).__init__(tags, **kwargs)
+
+    def __getitem__(self, key):
+        return key in self
+
+    def match(self, exp):
+        return eval(exp, {}, self)
+
+
+# duplicate definition from unittest2 of the _deprecate decorator
+def _deprecate(original_func):
+    def deprecated_func(*args, **kwargs):
+        warnings.warn(
+            ('Please use %s instead.' % original_func.__name__),
+            DeprecationWarning, 2)
+        return original_func(*args, **kwargs)
+    return deprecated_func
+
+class TestCase(unittest.TestCase):
+    """A unittest.TestCase extension with some additional methods."""
+    maxDiff = None
+    pdbclass = Debugger
+    tags = Tags()
+
+    def __init__(self, methodName='runTest'):
+        super(TestCase, self).__init__(methodName)
+        self.__exc_info = sys.exc_info
+        self.__testMethodName = self._testMethodName
+        self._current_test_descr = None
+        self._options_ = None
+
+    @classproperty
+    @cached
+    def datadir(cls): # pylint: disable=E0213
+        """helper attribute holding the standard test's data directory
+
+        NOTE: this is a logilab's standard
+        """
+        mod = __import__(cls.__module__)
+        return osp.join(osp.dirname(osp.abspath(mod.__file__)), 'data')
+    # cache it (use a class method to cache on class since TestCase is
+    # instantiated for each test run)
+
+    @classmethod
+    def datapath(cls, *fname):
+        """joins the object's datadir and `fname`"""
+        return osp.join(cls.datadir, *fname)
+
+    def set_description(self, descr):
+        """sets the current test's description.
+        This can be useful for generative tests because it allows to specify
+        a description per yield
+        """
+        self._current_test_descr = descr
+
+    # override default's unittest.py feature
+    def shortDescription(self):
+        """override default unittest shortDescription to handle correctly
+        generative tests
+        """
+        if self._current_test_descr is not None:
+            return self._current_test_descr
+        return super(TestCase, self).shortDescription()
+
+    def quiet_run(self, result, func, *args, **kwargs):
+        try:
+            func(*args, **kwargs)
+        except (KeyboardInterrupt, SystemExit):
+            raise
+        except unittest.SkipTest as e:
+            if hasattr(result, 'addSkip'):
+                result.addSkip(self, str(e))
+            else:
+                warnings.warn("TestResult has no addSkip method, skips not reported",
+                              RuntimeWarning, 2)
+                result.addSuccess(self)
+            return False
+        except:
+            result.addError(self, self.__exc_info())
+            return False
+        return True
+
+    def _get_test_method(self):
+        """return the test method"""
+        return getattr(self, self._testMethodName)
+
+    def optval(self, option, default=None):
+        """return the option value or default if the option is not define"""
+        return getattr(self._options_, option, default)
+
+    def __call__(self, result=None, runcondition=None, options=None):
+        """rewrite TestCase.__call__ to support generative tests
+        This is mostly a copy/paste from unittest.py (i.e same
+        variable names, same logic, except for the generative tests part)
+        """
+        from logilab.common.pytest import FILE_RESTART
+        if result is None:
+            result = self.defaultTestResult()
+        result.pdbclass = self.pdbclass
+        self._options_ = options
+        # if result.cvg:
+        #     result.cvg.start()
+        testMethod = self._get_test_method()
+        if (getattr(self.__class__, "__unittest_skip__", False) or
+            getattr(testMethod, "__unittest_skip__", False)):
+            # If the class or method was skipped.
+            try:
+                skip_why = (getattr(self.__class__, '__unittest_skip_why__', '')
+                            or getattr(testMethod, '__unittest_skip_why__', ''))
+                self._addSkip(result, skip_why)
+            finally:
+                result.stopTest(self)
+            return
+        if runcondition and not runcondition(testMethod):
+            return # test is skipped
+        result.startTest(self)
+        try:
+            if not self.quiet_run(result, self.setUp):
+                return
+            generative = isgeneratorfunction(testMethod)
+            # generative tests
+            if generative:
+                self._proceed_generative(result, testMethod,
+                                         runcondition)
+            else:
+                status = self._proceed(result, testMethod)
+                success = (status == 0)
+            if not self.quiet_run(result, self.tearDown):
+                return
+            if not generative and success:
+                if hasattr(options, "exitfirst") and options.exitfirst:
+                    # add this test to restart file
+                    try:
+                        restartfile = open(FILE_RESTART, 'a')
+                        try:
+                            descr = '.'.join((self.__class__.__module__,
+                                              self.__class__.__name__,
+                                              self._testMethodName))
+                            restartfile.write(descr+os.linesep)
+                        finally:
+                            restartfile.close()
+                    except Exception:
+                        print("Error while saving succeeded test into",
+                              osp.join(os.getcwd(), FILE_RESTART),
+                              file=sys.__stderr__)
+                        raise
+                result.addSuccess(self)
+        finally:
+            # if result.cvg:
+            #     result.cvg.stop()
+            result.stopTest(self)
+
+    def _proceed_generative(self, result, testfunc, runcondition=None):
+        # cancel startTest()'s increment
+        result.testsRun -= 1
+        success = True
+        try:
+            for params in testfunc():
+                if runcondition and not runcondition(testfunc,
+                        skipgenerator=False):
+                    if not (isinstance(params, InnerTest)
+                            and runcondition(params)):
+                        continue
+                if not isinstance(params, (tuple, list)):
+                    params = (params, )
+                func = params[0]
+                args, kwargs = parse_generative_args(params[1:])
+                # increment test counter manually
+                result.testsRun += 1
+                status = self._proceed(result, func, args, kwargs)
+                if status == 0:
+                    result.addSuccess(self)
+                    success = True
+                else:
+                    success = False
+                    # XXX Don't stop anymore if an error occured
+                    #if status == 2:
+                    #    result.shouldStop = True
+                if result.shouldStop: # either on error or on exitfirst + error
+                    break
+        except:
+            # if an error occurs between two yield
+            result.addError(self, self.__exc_info())
+            success = False
+        return success
+
+    def _proceed(self, result, testfunc, args=(), kwargs=None):
+        """proceed the actual test
+        returns 0 on success, 1 on failure, 2 on error
+
+        Note: addSuccess can't be called here because we have to wait
+        for tearDown to be successfully executed to declare the test as
+        successful
+        """
+        kwargs = kwargs or {}
+        try:
+            testfunc(*args, **kwargs)
+        except self.failureException:
+            result.addFailure(self, self.__exc_info())
+            return 1
+        except KeyboardInterrupt:
+            raise
+        except InnerTestSkipped as e:
+            result.addSkip(self, e)
+            return 1
+        except SkipTest as e:
+            result.addSkip(self, e)
+            return 0
+        except:
+            result.addError(self, self.__exc_info())
+            return 2
+        return 0
+
+    def defaultTestResult(self):
+        """return a new instance of the defaultTestResult"""
+        return SkipAwareTestResult()
+
+    skip = _deprecate(unittest.TestCase.skipTest)
+    assertEquals = _deprecate(unittest.TestCase.assertEqual)
+    assertNotEquals = _deprecate(unittest.TestCase.assertNotEqual)
+    assertAlmostEquals = _deprecate(unittest.TestCase.assertAlmostEqual)
+    assertNotAlmostEquals = _deprecate(unittest.TestCase.assertNotAlmostEqual)
+
+    def innerSkip(self, msg=None):
+        """mark a generative test as skipped for the <msg> reason"""
+        msg = msg or 'test was skipped'
+        raise InnerTestSkipped(msg)
+
+    @deprecated('Please use assertDictEqual instead.')
+    def assertDictEquals(self, dict1, dict2, msg=None, context=None):
+        """compares two dicts
+
+        If the two dict differ, the first difference is shown in the error
+        message
+        :param dict1: a Python Dictionary
+        :param dict2: a Python Dictionary
+        :param msg: custom message (String) in case of failure
+        """
+        dict1 = dict(dict1)
+        msgs = []
+        for key, value in dict2.items():
+            try:
+                if dict1[key] != value:
+                    msgs.append('%r != %r for key %r' % (dict1[key], value,
+                        key))
+                del dict1[key]
+            except KeyError:
+                msgs.append('missing %r key' % key)
+        if dict1:
+            msgs.append('dict2 is lacking %r' % dict1)
+        if msg:
+            self.failureException(msg)
+        elif msgs:
+            if context is not None:
+                base = '%s\n' % context
+            else:
+                base = ''
+            self.fail(base + '\n'.join(msgs))
+
+    @deprecated('Please use assertCountEqual instead.')
+    def assertUnorderedIterableEquals(self, got, expected, msg=None):
+        """compares two iterable and shows difference between both
+
+        :param got: the unordered Iterable that we found
+        :param expected: the expected unordered Iterable
+        :param msg: custom message (String) in case of failure
+        """
+        got, expected = list(got), list(expected)
+        self.assertSetEqual(set(got), set(expected), msg)
+        if len(got) != len(expected):
+            if msg is None:
+                msg = ['Iterable have the same elements but not the same number',
+                       '\t<element>\t<expected>i\t<got>']
+                got_count = {}
+                expected_count = {}
+                for element in got:
+                    got_count[element] = got_count.get(element, 0) + 1
+                for element in expected:
+                    expected_count[element] = expected_count.get(element, 0) + 1
+                # we know that got_count.key() == expected_count.key()
+                # because of assertSetEqual
+                for element, count in got_count.iteritems():
+                    other_count = expected_count[element]
+                    if other_count != count:
+                        msg.append('\t%s\t%s\t%s' % (element, other_count, count))
+
+            self.fail(msg)
+
+    assertUnorderedIterableEqual = assertUnorderedIterableEquals
+    assertUnordIterEquals = assertUnordIterEqual = assertUnorderedIterableEqual
+
+    @deprecated('Please use assertSetEqual instead.')
+    def assertSetEquals(self,got,expected, msg=None):
+        """compares two sets and shows difference between both
+
+        Don't use it for iterables other than sets.
+
+        :param got: the Set that we found
+        :param expected: the second Set to be compared to the first one
+        :param msg: custom message (String) in case of failure
+        """
+
+        if not(isinstance(got, set) and isinstance(expected, set)):
+            warnings.warn("the assertSetEquals function if now intended for set only."\
+                          "use assertUnorderedIterableEquals instead.",
+                DeprecationWarning, 2)
+            return self.assertUnorderedIterableEquals(got, expected, msg)
+
+        items={}
+        items['missing'] = expected - got
+        items['unexpected'] = got - expected
+        if any(items.itervalues()):
+            if msg is None:
+                msg = '\n'.join('%s:\n\t%s' % (key, "\n\t".join(str(value) for value in values))
+                    for key, values in items.iteritems() if values)
+            self.fail(msg)
+
+    @deprecated('Please use assertListEqual instead.')
+    def assertListEquals(self, list_1, list_2, msg=None):
+        """compares two lists
+
+        If the two list differ, the first difference is shown in the error
+        message
+
+        :param list_1: a Python List
+        :param list_2: a second Python List
+        :param msg: custom message (String) in case of failure
+        """
+        _l1 = list_1[:]
+        for i, value in enumerate(list_2):
+            try:
+                if _l1[0] != value:
+                    from pprint import pprint
+                    pprint(list_1)
+                    pprint(list_2)
+                    self.fail('%r != %r for index %d' % (_l1[0], value, i))
+                del _l1[0]
+            except IndexError:
+                if msg is None:
+                    msg = 'list_1 has only %d elements, not %s '\
+                        '(at least %r missing)'% (i, len(list_2), value)
+                self.fail(msg)
+        if _l1:
+            if msg is None:
+                msg = 'list_2 is lacking %r' % _l1
+            self.fail(msg)
+
+    @deprecated('Non-standard. Please use assertMultiLineEqual instead.')
+    def assertLinesEquals(self, string1, string2, msg=None, striplines=False):
+        """compare two strings and assert that the text lines of the strings
+        are equal.
+
+        :param string1: a String
+        :param string2: a String
+        :param msg: custom message (String) in case of failure
+        :param striplines: Boolean to trigger line stripping before comparing
+        """
+        lines1 = string1.splitlines()
+        lines2 = string2.splitlines()
+        if striplines:
+            lines1 = [l.strip() for l in lines1]
+            lines2 = [l.strip() for l in lines2]
+        self.assertListEqual(lines1, lines2, msg)
+    assertLineEqual = assertLinesEquals
+
+    @deprecated('Non-standard: please copy test method to your TestCase class')
+    def assertXMLWellFormed(self, stream, msg=None, context=2):
+        """asserts the XML stream is well-formed (no DTD conformance check)
+
+        :param context: number of context lines in standard message
+                        (show all data if negative).
+                        Only available with element tree
+        """
+        try:
+            from xml.etree.ElementTree import parse
+            self._assertETXMLWellFormed(stream, parse, msg)
+        except ImportError:
+            from xml.sax import make_parser, SAXParseException
+            parser = make_parser()
+            try:
+                parser.parse(stream)
+            except SAXParseException as ex:
+                if msg is None:
+                    stream.seek(0)
+                    for _ in range(ex.getLineNumber()):
+                        line = stream.readline()
+                    pointer = ('' * (ex.getLineNumber() - 1)) + '^'
+                    msg = 'XML stream not well formed: %s\n%s%s' % (ex, line, pointer)
+                self.fail(msg)
+
+    @deprecated('Non-standard: please copy test method to your TestCase class')
+    def assertXMLStringWellFormed(self, xml_string, msg=None, context=2):
+        """asserts the XML string is well-formed (no DTD conformance check)
+
+        :param context: number of context lines in standard message
+                        (show all data if negative).
+                        Only available with element tree
+        """
+        try:
+            from xml.etree.ElementTree import fromstring
+        except ImportError:
+            from elementtree.ElementTree import fromstring
+        self._assertETXMLWellFormed(xml_string, fromstring, msg)
+
+    def _assertETXMLWellFormed(self, data, parse, msg=None, context=2):
+        """internal function used by /assertXML(String)?WellFormed/ functions
+
+        :param data: xml_data
+        :param parse: appropriate parser function for this data
+        :param msg: error message
+        :param context: number of context lines in standard message
+                        (show all data if negative).
+                        Only available with element tree
+        """
+        from xml.parsers.expat import ExpatError
+        try:
+            from xml.etree.ElementTree import ParseError
+        except ImportError:
+            # compatibility for <python2.7
+            ParseError = ExpatError
+        try:
+            parse(data)
+        except (ExpatError, ParseError) as ex:
+            if msg is None:
+                if hasattr(data, 'readlines'): #file like object
+                    data.seek(0)
+                    lines = data.readlines()
+                else:
+                    lines = data.splitlines(True)
+                nb_lines = len(lines)
+                context_lines = []
+
+                # catch when ParseError doesn't set valid lineno
+                if ex.lineno is not None:
+                    if context < 0:
+                        start = 1
+                        end   = nb_lines
+                    else:
+                        start = max(ex.lineno-context, 1)
+                        end   = min(ex.lineno+context, nb_lines)
+                    line_number_length = len('%i' % end)
+                    line_pattern = " %%%ii: %%s" % line_number_length
+
+                    for line_no in range(start, ex.lineno):
+                        context_lines.append(line_pattern % (line_no, lines[line_no-1]))
+                    context_lines.append(line_pattern % (ex.lineno, lines[ex.lineno-1]))
+                    context_lines.append('%s^\n' % (' ' * (1 + line_number_length + 2 +ex.offset)))
+                    for line_no in range(ex.lineno+1, end+1):
+                        context_lines.append(line_pattern % (line_no, lines[line_no-1]))
+
+                rich_context = ''.join(context_lines)
+                msg = 'XML stream not well formed: %s\n%s' % (ex, rich_context)
+            self.fail(msg)
+
+    @deprecated('Non-standard: please copy test method to your TestCase class')
+    def assertXMLEqualsTuple(self, element, tup):
+        """compare an ElementTree Element to a tuple formatted as follow:
+        (tagname, [attrib[, children[, text[, tail]]]])"""
+        # check tag
+        self.assertTextEquals(element.tag, tup[0])
+        # check attrib
+        if len(element.attrib) or len(tup)>1:
+            if len(tup)<=1:
+                self.fail( "tuple %s has no attributes (%s expected)"%(tup,
+                    dict(element.attrib)))
+            self.assertDictEqual(element.attrib, tup[1])
+        # check children
+        if len(element) or len(tup)>2:
+            if len(tup)<=2:
+                self.fail( "tuple %s has no children (%i expected)"%(tup,
+                    len(element)))
+            if len(element) != len(tup[2]):
+                self.fail( "tuple %s has %i children%s (%i expected)"%(tup,
+                    len(tup[2]),
+                        ('', 's')[len(tup[2])>1], len(element)))
+            for index in range(len(tup[2])):
+                self.assertXMLEqualsTuple(element[index], tup[2][index])
+        #check text
+        if element.text or len(tup)>3:
+            if len(tup)<=3:
+                self.fail( "tuple %s has no text value (%r expected)"%(tup,
+                    element.text))
+            self.assertTextEquals(element.text, tup[3])
+        #check tail
+        if element.tail or len(tup)>4:
+            if len(tup)<=4:
+                self.fail( "tuple %s has no tail value (%r expected)"%(tup,
+                    element.tail))
+            self.assertTextEquals(element.tail, tup[4])
+
+    def _difftext(self, lines1, lines2, junk=None, msg_prefix='Texts differ'):
+        junk = junk or (' ', '\t')
+        # result is a generator
+        result = difflib.ndiff(lines1, lines2, charjunk=lambda x: x in junk)
+        read = []
+        for line in result:
+            read.append(line)
+            # lines that don't start with a ' ' are diff ones
+            if not line.startswith(' '):
+                self.fail('\n'.join(['%s\n'%msg_prefix]+read + list(result)))
+
+    @deprecated('Non-standard. Please use assertMultiLineEqual instead.')
+    def assertTextEquals(self, text1, text2, junk=None,
+            msg_prefix='Text differ', striplines=False):
+        """compare two multiline strings (using difflib and splitlines())
+
+        :param text1: a Python BaseString
+        :param text2: a second Python Basestring
+        :param junk: List of Caracters
+        :param msg_prefix: String (message prefix)
+        :param striplines: Boolean to trigger line stripping before comparing
+        """
+        msg = []
+        if not isinstance(text1, string_types):
+            msg.append('text1 is not a string (%s)'%(type(text1)))
+        if not isinstance(text2, string_types):
+            msg.append('text2 is not a string (%s)'%(type(text2)))
+        if msg:
+            self.fail('\n'.join(msg))
+        lines1 = text1.strip().splitlines(True)
+        lines2 = text2.strip().splitlines(True)
+        if striplines:
+            lines1 = [line.strip() for line in lines1]
+            lines2 = [line.strip() for line in lines2]
+        self._difftext(lines1, lines2, junk,  msg_prefix)
+    assertTextEqual = assertTextEquals
+
+    @deprecated('Non-standard: please copy test method to your TestCase class')
+    def assertStreamEquals(self, stream1, stream2, junk=None,
+            msg_prefix='Stream differ'):
+        """compare two streams (using difflib and readlines())"""
+        # if stream2 is stream2, readlines() on stream1 will also read lines
+        # in stream2, so they'll appear different, although they're not
+        if stream1 is stream2:
+            return
+        # make sure we compare from the beginning of the stream
+        stream1.seek(0)
+        stream2.seek(0)
+        # compare
+        self._difftext(stream1.readlines(), stream2.readlines(), junk,
+             msg_prefix)
+
+    assertStreamEqual = assertStreamEquals
+
+    @deprecated('Non-standard: please copy test method to your TestCase class')
+    def assertFileEquals(self, fname1, fname2, junk=(' ', '\t')):
+        """compares two files using difflib"""
+        self.assertStreamEqual(open(fname1), open(fname2), junk,
+            msg_prefix='Files differs\n-:%s\n+:%s\n'%(fname1, fname2))
+
+    assertFileEqual = assertFileEquals
+
+    @deprecated('Non-standard: please copy test method to your TestCase class')
+    def assertDirEquals(self, path_a, path_b):
+        """compares two files using difflib"""
+        assert osp.exists(path_a), "%s doesn't exists" % path_a
+        assert osp.exists(path_b), "%s doesn't exists" % path_b
+
+        all_a = [ (ipath[len(path_a):].lstrip('/'), idirs, ifiles)
+                    for ipath, idirs, ifiles in os.walk(path_a)]
+        all_a.sort(key=itemgetter(0))
+
+        all_b = [ (ipath[len(path_b):].lstrip('/'), idirs, ifiles)
+                    for ipath, idirs, ifiles in os.walk(path_b)]
+        all_b.sort(key=itemgetter(0))
+
+        iter_a, iter_b = iter(all_a), iter(all_b)
+        partial_iter = True
+        ipath_a, idirs_a, ifiles_a = data_a = None, None, None
+        while True:
+            try:
+                ipath_a, idirs_a, ifiles_a = datas_a = next(iter_a)
+                partial_iter = False
+                ipath_b, idirs_b, ifiles_b = datas_b = next(iter_b)
+                partial_iter = True
+
+
+                self.assertTrue(ipath_a == ipath_b,
+                    "unexpected %s in %s while looking %s from %s" %
+                    (ipath_a, path_a, ipath_b, path_b))
+
+
+                errors = {}
+                sdirs_a = set(idirs_a)
+                sdirs_b = set(idirs_b)
+                errors["unexpected directories"] = sdirs_a - sdirs_b
+                errors["missing directories"] = sdirs_b - sdirs_a
+
+                sfiles_a = set(ifiles_a)
+                sfiles_b = set(ifiles_b)
+                errors["unexpected files"] = sfiles_a - sfiles_b
+                errors["missing files"] = sfiles_b - sfiles_a
+
+
+                msgs = [ "%s: %s"% (name, items)
+                    for name, items in errors.items() if items]
+
+                if msgs:
+                    msgs.insert(0, "%s and %s differ :" % (
+                        osp.join(path_a, ipath_a),
+                        osp.join(path_b, ipath_b),
+                        ))
+                    self.fail("\n".join(msgs))
+
+                for files in (ifiles_a, ifiles_b):
+                    files.sort()
+
+                for index, path in enumerate(ifiles_a):
+                    self.assertFileEquals(osp.join(path_a, ipath_a, path),
+                        osp.join(path_b, ipath_b, ifiles_b[index]))
+
+            except StopIteration:
+                break
+
+    assertDirEqual = assertDirEquals
+
+    def assertIsInstance(self, obj, klass, msg=None, strict=False):
+        """check if an object is an instance of a class
+
+        :param obj: the Python Object to be checked
+        :param klass: the target class
+        :param msg: a String for a custom message
+        :param strict: if True, check that the class of <obj> is <klass>;
+                       else check with 'isinstance'
+        """
+        if strict:
+            warnings.warn('[API] Non-standard. Strict parameter has vanished',
+                          DeprecationWarning, stacklevel=2)
+        if msg is None:
+            if strict:
+                msg = '%r is not of class %s but of %s'
+            else:
+                msg = '%r is not an instance of %s but of %s'
+            msg = msg % (obj, klass, type(obj))
+        if strict:
+            self.assertTrue(obj.__class__ is klass, msg)
+        else:
+            self.assertTrue(isinstance(obj, klass), msg)
+
+    @deprecated('Please use assertIsNone instead.')
+    def assertNone(self, obj, msg=None):
+        """assert obj is None
+
+        :param obj: Python Object to be tested
+        """
+        if msg is None:
+            msg = "reference to %r when None expected"%(obj,)
+        self.assertTrue( obj is None, msg )
+
+    @deprecated('Please use assertIsNotNone instead.')
+    def assertNotNone(self, obj, msg=None):
+        """assert obj is not None"""
+        if msg is None:
+            msg = "unexpected reference to None"
+        self.assertTrue( obj is not None, msg )
+
+    @deprecated('Non-standard. Please use assertAlmostEqual instead.')
+    def assertFloatAlmostEquals(self, obj, other, prec=1e-5,
+                                relative=False, msg=None):
+        """compares if two floats have a distance smaller than expected
+        precision.
+
+        :param obj: a Float
+        :param other: another Float to be comparted to <obj>
+        :param prec: a Float describing the precision
+        :param relative: boolean switching to relative/absolute precision
+        :param msg: a String for a custom message
+        """
+        if msg is None:
+            msg = "%r != %r" % (obj, other)
+        if relative:
+            prec = prec*math.fabs(obj)
+        self.assertTrue(math.fabs(obj - other) < prec, msg)
+
+    def failUnlessRaises(self, excClass, callableObj=None, *args, **kwargs):
+        """override default failUnlessRaises method to return the raised
+        exception instance.
+
+        Fail unless an exception of class excClass is thrown
+        by callableObj when invoked with arguments args and keyword
+        arguments kwargs. If a different type of exception is
+        thrown, it will not be caught, and the test case will be
+        deemed to have suffered an error, exactly as for an
+        unexpected exception.
+
+        CAUTION! There are subtle differences between Logilab and unittest2
+        - exc is not returned in standard version
+        - context capabilities in standard version
+        - try/except/else construction (minor)
+
+        :param excClass: the Exception to be raised
+        :param callableObj: a callable Object which should raise <excClass>
+        :param args: a List of arguments for <callableObj>
+        :param kwargs: a List of keyword arguments  for <callableObj>
+        """
+        # XXX cube vcslib : test_branches_from_app
+        if callableObj is None:
+            _assert = super(TestCase, self).assertRaises
+            return _assert(excClass, callableObj, *args, **kwargs)
+        try:
+            callableObj(*args, **kwargs)
+        except excClass as exc:
+            class ProxyException:
+                def __init__(self, obj):
+                    self._obj = obj
+                def __getattr__(self, attr):
+                    warn_msg = ("This exception was retrieved with the old testlib way "
+                                "`exc = self.assertRaises(Exc, callable)`, please use "
+                                "the context manager instead'")
+                    warnings.warn(warn_msg, DeprecationWarning, 2)
+                    return self._obj.__getattribute__(attr)
+            return ProxyException(exc)
+        else:
+            if hasattr(excClass, '__name__'):
+                excName = excClass.__name__
+            else:
+                excName = str(excClass)
+            raise self.failureException("%s not raised" % excName)
+
+    assertRaises = failUnlessRaises
+
+    if sys.version_info >= (3,2):
+        assertItemsEqual = unittest.TestCase.assertCountEqual
+    else:
+        assertCountEqual = unittest.TestCase.assertItemsEqual
+        if sys.version_info < (2,7):
+            def assertIsNotNone(self, value, *args, **kwargs):
+                self.assertNotEqual(None, value, *args, **kwargs)
+
+TestCase.assertItemsEqual = deprecated('assertItemsEqual is deprecated, use assertCountEqual')(
+    TestCase.assertItemsEqual)
+
+import doctest
+
+class SkippedSuite(unittest.TestSuite):
+    def test(self):
+        """just there to trigger test execution"""
+        self.skipped_test('doctest module has no DocTestSuite class')
+
+
+class DocTestFinder(doctest.DocTestFinder):
+
+    def __init__(self, *args, **kwargs):
+        self.skipped = kwargs.pop('skipped', ())
+        doctest.DocTestFinder.__init__(self, *args, **kwargs)
+
+    def _get_test(self, obj, name, module, globs, source_lines):
+        """override default _get_test method to be able to skip tests
+        according to skipped attribute's value
+        """
+        if getattr(obj, '__name__', '') in self.skipped:
+            return None
+        return doctest.DocTestFinder._get_test(self, obj, name, module,
+                                               globs, source_lines)
+
+
+class DocTest(TestCase):
+    """trigger module doctest
+    I don't know how to make unittest.main consider the DocTestSuite instance
+    without this hack
+    """
+    skipped = ()
+    def __call__(self, result=None, runcondition=None, options=None):\
+        # pylint: disable=W0613
+        try:
+            finder = DocTestFinder(skipped=self.skipped)
+            suite = doctest.DocTestSuite(self.module, test_finder=finder)
+            # XXX iirk
+            doctest.DocTestCase._TestCase__exc_info = sys.exc_info
+        except AttributeError:
+            suite = SkippedSuite()
+        # doctest may gork the builtins dictionnary
+        # This happen to the "_" entry used by gettext
+        old_builtins = builtins.__dict__.copy()
+        try:
+            return suite.run(result)
+        finally:
+            builtins.__dict__.clear()
+            builtins.__dict__.update(old_builtins)
+    run = __call__
+
+    def test(self):
+        """just there to trigger test execution"""
+
+MAILBOX = None
+
+class MockSMTP:
+    """fake smtplib.SMTP"""
+
+    def __init__(self, host, port):
+        self.host = host
+        self.port = port
+        global MAILBOX
+        self.reveived = MAILBOX = []
+
+    def set_debuglevel(self, debuglevel):
+        """ignore debug level"""
+
+    def sendmail(self, fromaddr, toaddres, body):
+        """push sent mail in the mailbox"""
+        self.reveived.append((fromaddr, toaddres, body))
+
+    def quit(self):
+        """ignore quit"""
+
+
+class MockConfigParser(configparser.ConfigParser):
+    """fake ConfigParser.ConfigParser"""
+
+    def __init__(self, options):
+        configparser.ConfigParser.__init__(self)
+        for section, pairs in options.iteritems():
+            self.add_section(section)
+            for key, value in pairs.iteritems():
+                self.set(section, key, value)
+    def write(self, _):
+        raise NotImplementedError()
+
+
+class MockConnection:
+    """fake DB-API 2.0 connexion AND cursor (i.e. cursor() return self)"""
+
+    def __init__(self, results):
+        self.received = []
+        self.states = []
+        self.results = results
+
+    def cursor(self):
+        """Mock cursor method"""
+        return self
+    def execute(self, query, args=None):
+        """Mock execute method"""
+        self.received.append( (query, args) )
+    def fetchone(self):
+        """Mock fetchone method"""
+        return self.results[0]
+    def fetchall(self):
+        """Mock fetchall method"""
+        return self.results
+    def commit(self):
+        """Mock commiy method"""
+        self.states.append( ('commit', len(self.received)) )
+    def rollback(self):
+        """Mock rollback method"""
+        self.states.append( ('rollback', len(self.received)) )
+    def close(self):
+        """Mock close method"""
+        pass
+
+
+def mock_object(**params):
+    """creates an object using params to set attributes
+    >>> option = mock_object(verbose=False, index=range(5))
+    >>> option.verbose
+    False
+    >>> option.index
+    [0, 1, 2, 3, 4]
+    """
+    return type('Mock', (), params)()
+
+
+def create_files(paths, chroot):
+    """Creates directories and files found in <path>.
+
+    :param paths: list of relative paths to files or directories
+    :param chroot: the root directory in which paths will be created
+
+    >>> from os.path import isdir, isfile
+    >>> isdir('/tmp/a')
+    False
+    >>> create_files(['a/b/foo.py', 'a/b/c/', 'a/b/c/d/e.py'], '/tmp')
+    >>> isdir('/tmp/a')
+    True
+    >>> isdir('/tmp/a/b/c')
+    True
+    >>> isfile('/tmp/a/b/c/d/e.py')
+    True
+    >>> isfile('/tmp/a/b/foo.py')
+    True
+    """
+    dirs, files = set(), set()
+    for path in paths:
+        path = osp.join(chroot, path)
+        filename = osp.basename(path)
+        # path is a directory path
+        if filename == '':
+            dirs.add(path)
+        # path is a filename path
+        else:
+            dirs.add(osp.dirname(path))
+            files.add(path)
+    for dirpath in dirs:
+        if not osp.isdir(dirpath):
+            os.makedirs(dirpath)
+    for filepath in files:
+        open(filepath, 'w').close()
+
+
+class AttrObject: # XXX cf mock_object
+    def __init__(self, **kwargs):
+        self.__dict__.update(kwargs)
+
+def tag(*args, **kwargs):
+    """descriptor adding tag to a function"""
+    def desc(func):
+        assert not hasattr(func, 'tags')
+        func.tags = Tags(*args, **kwargs)
+        return func
+    return desc
+
+def require_version(version):
+    """ Compare version of python interpreter to the given one. Skip the test
+    if older.
+    """
+    def check_require_version(f):
+        version_elements = version.split('.')
+        try:
+            compare = tuple([int(v) for v in version_elements])
+        except ValueError:
+            raise ValueError('%s is not a correct version : should be X.Y[.Z].' % version)
+        current = sys.version_info[:3]
+        if current < compare:
+            def new_f(self, *args, **kwargs):
+                self.skipTest('Need at least %s version of python. Current version is %s.' % (version, '.'.join([str(element) for element in current])))
+            new_f.__name__ = f.__name__
+            return new_f
+        else:
+            return f
+    return check_require_version
+
+def require_module(module):
+    """ Check if the given module is loaded. Skip the test if not.
+    """
+    def check_require_module(f):
+        try:
+            __import__(module)
+            return f
+        except ImportError:
+            def new_f(self, *args, **kwargs):
+                self.skipTest('%s can not be imported.' % module)
+            new_f.__name__ = f.__name__
+            return new_f
+    return check_require_module
+
diff --git a/logilab/common/textutils.py b/logilab/common/textutils.py
new file mode 100644
index 0000000..9046f97
--- /dev/null
+++ b/logilab/common/textutils.py
@@ -0,0 +1,537 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Some text manipulation utility functions.
+
+
+:group text formatting: normalize_text, normalize_paragraph, pretty_match,\
+unquote, colorize_ansi
+:group text manipulation: searchall, splitstrip
+:sort: text formatting, text manipulation
+
+:type ANSI_STYLES: dict(str)
+:var ANSI_STYLES: dictionary mapping style identifier to ANSI terminal code
+
+:type ANSI_COLORS: dict(str)
+:var ANSI_COLORS: dictionary mapping color identifier to ANSI terminal code
+
+:type ANSI_PREFIX: str
+:var ANSI_PREFIX:
+  ANSI terminal code notifying the start of an ANSI escape sequence
+
+:type ANSI_END: str
+:var ANSI_END:
+  ANSI terminal code notifying the end of an ANSI escape sequence
+
+:type ANSI_RESET: str
+:var ANSI_RESET:
+  ANSI terminal code resetting format defined by a previous ANSI escape sequence
+"""
+__docformat__ = "restructuredtext en"
+
+import sys
+import re
+import os.path as osp
+from warnings import warn
+from unicodedata import normalize as _uninormalize
+try:
+    from os import linesep
+except ImportError:
+    linesep = '\n' # gae
+
+from logilab.common.deprecation import deprecated
+
+MANUAL_UNICODE_MAP = {
+    u'\xa1': u'!',    # INVERTED EXCLAMATION MARK
+    u'\u0142': u'l',  # LATIN SMALL LETTER L WITH STROKE
+    u'\u2044': u'/',  # FRACTION SLASH
+    u'\xc6': u'AE',   # LATIN CAPITAL LETTER AE
+    u'\xa9': u'(c)',  # COPYRIGHT SIGN
+    u'\xab': u'"',    # LEFT-POINTING DOUBLE ANGLE QUOTATION MARK
+    u'\xe6': u'ae',   # LATIN SMALL LETTER AE
+    u'\xae': u'(r)',  # REGISTERED SIGN
+    u'\u0153': u'oe', # LATIN SMALL LIGATURE OE
+    u'\u0152': u'OE', # LATIN CAPITAL LIGATURE OE
+    u'\xd8': u'O',    # LATIN CAPITAL LETTER O WITH STROKE
+    u'\xf8': u'o',    # LATIN SMALL LETTER O WITH STROKE
+    u'\xbb': u'"',    # RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK
+    u'\xdf': u'ss',   # LATIN SMALL LETTER SHARP S
+    }
+
+def unormalize(ustring, ignorenonascii=None, substitute=None):
+    """replace diacritical characters with their corresponding ascii characters
+
+    Convert the unicode string to its long normalized form (unicode character
+    will be transform into several characters) and keep the first one only.
+    The normal form KD (NFKD) will apply the compatibility decomposition, i.e.
+    replace all compatibility characters with their equivalents.
+
+    :type substitute: str
+    :param substitute: replacement character to use if decomposition fails
+
+    :see: Another project about ASCII transliterations of Unicode text
+          http://pypi.python.org/pypi/Unidecode
+    """
+    # backward compatibility, ignorenonascii was a boolean
+    if ignorenonascii is not None:
+        warn("ignorenonascii is deprecated, use substitute named parameter instead",
+             DeprecationWarning, stacklevel=2)
+        if ignorenonascii:
+            substitute = ''
+    res = []
+    for letter in ustring[:]:
+        try:
+            replacement = MANUAL_UNICODE_MAP[letter]
+        except KeyError:
+            replacement = _uninormalize('NFKD', letter)[0]
+            if ord(replacement) >= 2 ** 7:
+                if substitute is None:
+                    raise ValueError("can't deal with non-ascii based characters")
+                replacement = substitute
+        res.append(replacement)
+    return u''.join(res)
+
+def unquote(string):
+    """remove optional quotes (simple or double) from the string
+
+    :type string: str or unicode
+    :param string: an optionally quoted string
+
+    :rtype: str or unicode
+    :return: the unquoted string (or the input string if it wasn't quoted)
+    """
+    if not string:
+        return string
+    if string[0] in '"\'':
+        string = string[1:]
+    if string[-1] in '"\'':
+        string = string[:-1]
+    return string
+
+
+_BLANKLINES_RGX = re.compile('\r?\n\r?\n')
+_NORM_SPACES_RGX = re.compile('\s+')
+
+def normalize_text(text, line_len=80, indent='', rest=False):
+    """normalize a text to display it with a maximum line size and
+    optionally arbitrary indentation. Line jumps are normalized but blank
+    lines are kept. The indentation string may be used to insert a
+    comment (#) or a quoting (>) mark  for instance.
+
+    :type text: str or unicode
+    :param text: the input text to normalize
+
+    :type line_len: int
+    :param line_len: expected maximum line's length, default to 80
+
+    :type indent: str or unicode
+    :param indent: optional string to use as indentation
+
+    :rtype: str or unicode
+    :return:
+      the input text normalized to fit on lines with a maximized size
+      inferior to `line_len`, and optionally prefixed by an
+      indentation string
+    """
+    if rest:
+        normp = normalize_rest_paragraph
+    else:
+        normp = normalize_paragraph
+    result = []
+    for text in _BLANKLINES_RGX.split(text):
+        result.append(normp(text, line_len, indent))
+    return ('%s%s%s' % (linesep, indent, linesep)).join(result)
+
+
+def normalize_paragraph(text, line_len=80, indent=''):
+    """normalize a text to display it with a maximum line size and
+    optionally arbitrary indentation. Line jumps are normalized. The
+    indentation string may be used top insert a comment mark for
+    instance.
+
+    :type text: str or unicode
+    :param text: the input text to normalize
+
+    :type line_len: int
+    :param line_len: expected maximum line's length, default to 80
+
+    :type indent: str or unicode
+    :param indent: optional string to use as indentation
+
+    :rtype: str or unicode
+    :return:
+      the input text normalized to fit on lines with a maximized size
+      inferior to `line_len`, and optionally prefixed by an
+      indentation string
+    """
+    text = _NORM_SPACES_RGX.sub(' ', text)
+    line_len = line_len - len(indent)
+    lines = []
+    while text:
+        aline, text = splittext(text.strip(), line_len)
+        lines.append(indent + aline)
+    return linesep.join(lines)
+
+def normalize_rest_paragraph(text, line_len=80, indent=''):
+    """normalize a ReST text to display it with a maximum line size and
+    optionally arbitrary indentation. Line jumps are normalized. The
+    indentation string may be used top insert a comment mark for
+    instance.
+
+    :type text: str or unicode
+    :param text: the input text to normalize
+
+    :type line_len: int
+    :param line_len: expected maximum line's length, default to 80
+
+    :type indent: str or unicode
+    :param indent: optional string to use as indentation
+
+    :rtype: str or unicode
+    :return:
+      the input text normalized to fit on lines with a maximized size
+      inferior to `line_len`, and optionally prefixed by an
+      indentation string
+    """
+    toreport = ''
+    lines = []
+    line_len = line_len - len(indent)
+    for line in text.splitlines():
+        line = toreport + _NORM_SPACES_RGX.sub(' ', line.strip())
+        toreport = ''
+        while len(line) > line_len:
+            # too long line, need split
+            line, toreport = splittext(line, line_len)
+            lines.append(indent + line)
+            if toreport:
+                line = toreport + ' '
+                toreport = ''
+            else:
+                line = ''
+        if line:
+            lines.append(indent + line.strip())
+    return linesep.join(lines)
+
+
+def splittext(text, line_len):
+    """split the given text on space according to the given max line size
+
+    return a 2-uple:
+    * a line <= line_len if possible
+    * the rest of the text which has to be reported on another line
+    """
+    if len(text) <= line_len:
+        return text, ''
+    pos = min(len(text)-1, line_len)
+    while pos > 0 and text[pos] != ' ':
+        pos -= 1
+    if pos == 0:
+        pos = min(len(text), line_len)
+        while len(text) > pos and text[pos] != ' ':
+            pos += 1
+    return text[:pos], text[pos+1:].strip()
+
+
+def splitstrip(string, sep=','):
+    """return a list of stripped string by splitting the string given as
+    argument on `sep` (',' by default). Empty string are discarded.
+
+    >>> splitstrip('a, b, c   ,  4,,')
+    ['a', 'b', 'c', '4']
+    >>> splitstrip('a')
+    ['a']
+    >>>
+
+    :type string: str or unicode
+    :param string: a csv line
+
+    :type sep: str or unicode
+    :param sep: field separator, default to the comma (',')
+
+    :rtype: str or unicode
+    :return: the unquoted string (or the input string if it wasn't quoted)
+    """
+    return [word.strip() for word in string.split(sep) if word.strip()]
+
+get_csv = deprecated('get_csv is deprecated, use splitstrip')(splitstrip)
+
+
+def split_url_or_path(url_or_path):
+    """return the latest component of a string containing either an url of the
+    form <scheme>://<path> or a local file system path
+    """
+    if '://' in url_or_path:
+        return url_or_path.rstrip('/').rsplit('/', 1)
+    return osp.split(url_or_path.rstrip(osp.sep))
+
+
+def text_to_dict(text):
+    """parse multilines text containing simple 'key=value' lines and return a
+    dict of {'key': 'value'}. When the same key is encountered multiple time,
+    value is turned into a list containing all values.
+
+    >>> d = text_to_dict('''multiple=1
+    ... multiple= 2
+    ... single =3
+    ... ''')
+    >>> d['single']
+    '3'
+    >>> d['multiple']
+    ['1', '2']
+
+    """
+    res = {}
+    if not text:
+        return res
+    for line in text.splitlines():
+        line = line.strip()
+        if line and not line.startswith('#'):
+            key, value = [w.strip() for w in line.split('=', 1)]
+            if key in res:
+                try:
+                    res[key].append(value)
+                except AttributeError:
+                    res[key] = [res[key], value]
+            else:
+                res[key] = value
+    return res
+
+
+_BLANK_URE = r'(\s|,)+'
+_BLANK_RE = re.compile(_BLANK_URE)
+__VALUE_URE = r'-?(([0-9]+\.[0-9]*)|((0x?)?[0-9]+))'
+__UNITS_URE = r'[a-zA-Z]+'
+_VALUE_RE = re.compile(r'(?P<value>%s)(?P<unit>%s)?'%(__VALUE_URE, __UNITS_URE))
+_VALIDATION_RE = re.compile(r'^((%s)(%s))*(%s)?$' % (__VALUE_URE, __UNITS_URE,
+                                                    __VALUE_URE))
+
+BYTE_UNITS = {
+    "b": 1,
+    "kb": 1024,
+    "mb": 1024 ** 2,
+    "gb": 1024 ** 3,
+    "tb": 1024 ** 4,
+}
+
+TIME_UNITS = {
+    "ms": 0.0001,
+    "s": 1,
+    "min": 60,
+    "h": 60 * 60,
+    "d": 60 * 60 *24,
+}
+
+def apply_units(string, units, inter=None, final=float, blank_reg=_BLANK_RE,
+                value_reg=_VALUE_RE):
+    """Parse the string applying the units defined in units
+    (e.g.: "1.5m",{'m',60} -> 80).
+
+    :type string: str or unicode
+    :param string: the string to parse
+
+    :type units: dict (or any object with __getitem__ using basestring key)
+    :param units: a dict mapping a unit string repr to its value
+
+    :type inter: type
+    :param inter: used to parse every intermediate value (need __sum__)
+
+    :type blank_reg: regexp
+    :param blank_reg: should match every blank char to ignore.
+
+    :type value_reg: regexp with "value" and optional "unit" group
+    :param value_reg: match a value and it's unit into the
+    """
+    if inter is None:
+        inter = final
+    fstring = _BLANK_RE.sub('', string)
+    if not (fstring and _VALIDATION_RE.match(fstring)):
+        raise ValueError("Invalid unit string: %r." % string)
+    values = []
+    for match in value_reg.finditer(fstring):
+        dic = match.groupdict()
+        lit, unit = dic["value"], dic.get("unit")
+        value = inter(lit)
+        if unit is not None:
+            try:
+                value *= units[unit.lower()]
+            except KeyError:
+                raise KeyError('invalid unit %s. valid units are %s' %
+                               (unit, units.keys()))
+        values.append(value)
+    return final(sum(values))
+
+
+_LINE_RGX = re.compile('\r\n|\r+|\n')
+
+def pretty_match(match, string, underline_char='^'):
+    """return a string with the match location underlined:
+
+    >>> import re
+    >>> print(pretty_match(re.search('mange', 'il mange du bacon'), 'il mange du bacon'))
+    il mange du bacon
+       ^^^^^
+    >>>
+
+    :type match: _sre.SRE_match
+    :param match: object returned by re.match, re.search or re.finditer
+
+    :type string: str or unicode
+    :param string:
+      the string on which the regular expression has been applied to
+      obtain the `match` object
+
+    :type underline_char: str or unicode
+    :param underline_char:
+      character to use to underline the matched section, default to the
+      carret '^'
+
+    :rtype: str or unicode
+    :return:
+      the original string with an inserted line to underline the match
+      location
+    """
+    start = match.start()
+    end = match.end()
+    string = _LINE_RGX.sub(linesep, string)
+    start_line_pos = string.rfind(linesep, 0, start)
+    if start_line_pos == -1:
+        start_line_pos = 0
+        result = []
+    else:
+        result = [string[:start_line_pos]]
+        start_line_pos += len(linesep)
+    offset = start - start_line_pos
+    underline = ' ' * offset + underline_char * (end - start)
+    end_line_pos = string.find(linesep, end)
+    if end_line_pos == -1:
+        string = string[start_line_pos:]
+        result.append(string)
+        result.append(underline)
+    else:
+        end = string[end_line_pos + len(linesep):]
+        string = string[start_line_pos:end_line_pos]
+        result.append(string)
+        result.append(underline)
+        result.append(end)
+    return linesep.join(result).rstrip()
+
+
+# Ansi colorization ###########################################################
+
+ANSI_PREFIX = '\033['
+ANSI_END = 'm'
+ANSI_RESET = '\033[0m'
+ANSI_STYLES = {
+    'reset': "0",
+    'bold': "1",
+    'italic': "3",
+    'underline': "4",
+    'blink': "5",
+    'inverse': "7",
+    'strike': "9",
+}
+ANSI_COLORS = {
+    'reset': "0",
+    'black': "30",
+    'red': "31",
+    'green': "32",
+    'yellow': "33",
+    'blue': "34",
+    'magenta': "35",
+    'cyan': "36",
+    'white': "37",
+}
+
+def _get_ansi_code(color=None, style=None):
+    """return ansi escape code corresponding to color and style
+
+    :type color: str or None
+    :param color:
+      the color name (see `ANSI_COLORS` for available values)
+      or the color number when 256 colors are available
+
+    :type style: str or None
+    :param style:
+      style string (see `ANSI_COLORS` for available values). To get
+      several style effects at the same time, use a coma as separator.
+
+    :raise KeyError: if an unexistent color or style identifier is given
+
+    :rtype: str
+    :return: the built escape code
+    """
+    ansi_code = []
+    if style:
+        style_attrs = splitstrip(style)
+        for effect in style_attrs:
+            ansi_code.append(ANSI_STYLES[effect])
+    if color:
+        if color.isdigit():
+            ansi_code.extend(['38', '5'])
+            ansi_code.append(color)
+        else:
+            ansi_code.append(ANSI_COLORS[color])
+    if ansi_code:
+        return ANSI_PREFIX + ';'.join(ansi_code) + ANSI_END
+    return ''
+
+def colorize_ansi(msg, color=None, style=None):
+    """colorize message by wrapping it with ansi escape codes
+
+    :type msg: str or unicode
+    :param msg: the message string to colorize
+
+    :type color: str or None
+    :param color:
+      the color identifier (see `ANSI_COLORS` for available values)
+
+    :type style: str or None
+    :param style:
+      style string (see `ANSI_COLORS` for available values). To get
+      several style effects at the same time, use a coma as separator.
+
+    :raise KeyError: if an unexistent color or style identifier is given
+
+    :rtype: str or unicode
+    :return: the ansi escaped string
+    """
+    # If both color and style are not defined, then leave the text as is
+    if color is None and style is None:
+        return msg
+    escape_code = _get_ansi_code(color, style)
+    # If invalid (or unknown) color, don't wrap msg with ansi codes
+    if escape_code:
+        return '%s%s%s' % (escape_code, msg, ANSI_RESET)
+    return msg
+
+DIFF_STYLE = {'separator': 'cyan', 'remove': 'red', 'add': 'green'}
+
+def diff_colorize_ansi(lines, out=sys.stdout, style=DIFF_STYLE):
+    for line in lines:
+        if line[:4] in ('--- ', '+++ '):
+            out.write(colorize_ansi(line, style['separator']))
+        elif line[0] == '-':
+            out.write(colorize_ansi(line, style['remove']))
+        elif line[0] == '+':
+            out.write(colorize_ansi(line, style['add']))
+        elif line[:4] == '--- ':
+            out.write(colorize_ansi(line, style['separator']))
+        elif line[:4] == '+++ ':
+            out.write(colorize_ansi(line, style['separator']))
+        else:
+            out.write(line)
+
diff --git a/logilab/common/tree.py b/logilab/common/tree.py
new file mode 100644
index 0000000..885eb0f
--- /dev/null
+++ b/logilab/common/tree.py
@@ -0,0 +1,369 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Base class to represent a tree structure.
+
+
+
+
+"""
+__docformat__ = "restructuredtext en"
+
+import sys
+
+from logilab.common import flatten
+from logilab.common.visitor import VisitedMixIn, FilteredIterator, no_filter
+
+## Exceptions #################################################################
+
+class NodeNotFound(Exception):
+    """raised when a node has not been found"""
+
+EX_SIBLING_NOT_FOUND = "No such sibling as '%s'"
+EX_CHILD_NOT_FOUND = "No such child as '%s'"
+EX_NODE_NOT_FOUND = "No such node as '%s'"
+
+
+# Base node ###################################################################
+
+class Node(object):
+    """a basic tree node, characterized by an id"""
+
+    def __init__(self, nid=None) :
+        self.id = nid
+        # navigation
+        self.parent = None
+        self.children = []
+
+    def __iter__(self):
+        return iter(self.children)
+
+    def __str__(self, indent=0):
+        s = ['%s%s %s' % (' '*indent, self.__class__.__name__, self.id)]
+        indent += 2
+        for child in self.children:
+            try:
+                s.append(child.__str__(indent))
+            except TypeError:
+                s.append(child.__str__())
+        return '\n'.join(s)
+
+    def is_leaf(self):
+        return not self.children
+
+    def append(self, child):
+        """add a node to children"""
+        self.children.append(child)
+        child.parent = self
+
+    def remove(self, child):
+        """remove a child node"""
+        self.children.remove(child)
+        child.parent = None
+
+    def insert(self, index, child):
+        """insert a child node"""
+        self.children.insert(index, child)
+        child.parent = self
+
+    def replace(self, old_child, new_child):
+        """replace a child node with another"""
+        i = self.children.index(old_child)
+        self.children.pop(i)
+        self.children.insert(i, new_child)
+        new_child.parent = self
+
+    def get_sibling(self, nid):
+        """return the sibling node that has given id"""
+        try:
+            return self.parent.get_child_by_id(nid)
+        except NodeNotFound :
+            raise NodeNotFound(EX_SIBLING_NOT_FOUND % nid)
+
+    def next_sibling(self):
+        """
+        return the next sibling for this node if any
+        """
+        parent = self.parent
+        if parent is None:
+            # root node has no sibling
+            return None
+        index = parent.children.index(self)
+        try:
+            return parent.children[index+1]
+        except IndexError:
+            return None
+
+    def previous_sibling(self):
+        """
+        return the previous sibling for this node if any
+        """
+        parent = self.parent
+        if parent is None:
+            # root node has no sibling
+            return None
+        index = parent.children.index(self)
+        if index > 0:
+            return parent.children[index-1]
+        return None
+
+    def get_node_by_id(self, nid):
+        """
+        return node in whole hierarchy that has given id
+        """
+        root = self.root()
+        try:
+            return root.get_child_by_id(nid, 1)
+        except NodeNotFound :
+            raise NodeNotFound(EX_NODE_NOT_FOUND % nid)
+
+    def get_child_by_id(self, nid, recurse=None):
+        """
+        return child of given id
+        """
+        if self.id == nid:
+            return self
+        for c in self.children :
+            if recurse:
+                try:
+                    return c.get_child_by_id(nid, 1)
+                except NodeNotFound :
+                    continue
+            if c.id == nid :
+                return c
+        raise NodeNotFound(EX_CHILD_NOT_FOUND % nid)
+
+    def get_child_by_path(self, path):
+        """
+        return child of given path (path is a list of ids)
+        """
+        if len(path) > 0 and path[0] == self.id:
+            if len(path) == 1 :
+                return self
+            else :
+                for c in self.children :
+                    try:
+                        return c.get_child_by_path(path[1:])
+                    except NodeNotFound :
+                        pass
+        raise NodeNotFound(EX_CHILD_NOT_FOUND % path)
+
+    def depth(self):
+        """
+        return depth of this node in the tree
+        """
+        if self.parent is not None:
+            return 1 + self.parent.depth()
+        else :
+            return 0
+
+    def depth_down(self):
+        """
+        return depth of the tree from this node
+        """
+        if self.children:
+            return 1 + max([c.depth_down() for c in self.children])
+        return 1
+
+    def width(self):
+        """
+        return the width of the tree from this node
+        """
+        return len(self.leaves())
+
+    def root(self):
+        """
+        return the root node of the tree
+        """
+        if self.parent is not None:
+            return self.parent.root()
+        return self
+
+    def leaves(self):
+        """
+        return a list with all the leaves nodes descendant from this node
+        """
+        leaves = []
+        if self.children:
+            for child in self.children:
+                leaves += child.leaves()
+            return leaves
+        else:
+            return [self]
+
+    def flatten(self, _list=None):
+        """
+        return a list with all the nodes descendant from this node
+        """
+        if _list is None:
+            _list = []
+        _list.append(self)
+        for c in self.children:
+            c.flatten(_list)
+        return _list
+
+    def lineage(self):
+        """
+        return list of parents up to root node
+        """
+        lst = [self]
+        if self.parent is not None:
+            lst.extend(self.parent.lineage())
+        return lst
+
+class VNode(Node, VisitedMixIn):
+    """a visitable node
+    """
+    pass
+
+
+class BinaryNode(VNode):
+    """a binary node (i.e. only two children
+    """
+    def __init__(self, lhs=None, rhs=None) :
+        VNode.__init__(self)
+        if lhs is not None or rhs is not None:
+            assert lhs and rhs
+            self.append(lhs)
+            self.append(rhs)
+
+    def remove(self, child):
+        """remove the child and replace this node with the other child
+        """
+        self.children.remove(child)
+        self.parent.replace(self, self.children[0])
+
+    def get_parts(self):
+        """
+        return the left hand side and the right hand side of this node
+        """
+        return self.children[0], self.children[1]
+
+
+
+if sys.version_info[0:2] >= (2, 2):
+    list_class = list
+else:
+    from UserList import UserList
+    list_class = UserList
+
+class ListNode(VNode, list_class):
+    """Used to manipulate Nodes as Lists
+    """
+    def __init__(self):
+        list_class.__init__(self)
+        VNode.__init__(self)
+        self.children = self
+
+    def __str__(self, indent=0):
+        return '%s%s %s' % (indent*' ', self.__class__.__name__,
+                            ', '.join([str(v) for v in self]))
+
+    def append(self, child):
+        """add a node to children"""
+        list_class.append(self, child)
+        child.parent = self
+
+    def insert(self, index, child):
+        """add a node to children"""
+        list_class.insert(self, index, child)
+        child.parent = self
+
+    def remove(self, child):
+        """add a node to children"""
+        list_class.remove(self, child)
+        child.parent = None
+
+    def pop(self, index):
+        """add a node to children"""
+        child = list_class.pop(self, index)
+        child.parent = None
+
+    def __iter__(self):
+        return list_class.__iter__(self)
+
+# construct list from tree ####################################################
+
+def post_order_list(node, filter_func=no_filter):
+    """
+    create a list with tree nodes for which the <filter> function returned true
+    in a post order fashion
+    """
+    l, stack = [], []
+    poped, index = 0, 0
+    while node:
+        if filter_func(node):
+            if node.children and not poped:
+                stack.append((node, index))
+                index = 0
+                node = node.children[0]
+            else:
+                l.append(node)
+                index += 1
+                try:
+                    node = stack[-1][0].children[index]
+                except IndexError:
+                    node = None
+        else:
+            node = None
+        poped = 0
+        if node is None and stack:
+            node, index = stack.pop()
+            poped = 1
+    return l
+
+def pre_order_list(node, filter_func=no_filter):
+    """
+    create a list with tree nodes for which the <filter> function returned true
+    in a pre order fashion
+    """
+    l, stack = [], []
+    poped, index = 0, 0
+    while node:
+        if filter_func(node):
+            if not poped:
+                l.append(node)
+            if node.children and not poped:
+                stack.append((node, index))
+                index = 0
+                node = node.children[0]
+            else:
+                index += 1
+                try:
+                    node = stack[-1][0].children[index]
+                except IndexError:
+                    node = None
+        else:
+            node = None
+        poped = 0
+        if node is None and len(stack) > 1:
+            node, index = stack.pop()
+            poped = 1
+    return l
+
+class PostfixedDepthFirstIterator(FilteredIterator):
+    """a postfixed depth first iterator, designed to be used with visitors
+    """
+    def __init__(self, node, filter_func=None):
+        FilteredIterator.__init__(self, node, post_order_list, filter_func)
+
+class PrefixedDepthFirstIterator(FilteredIterator):
+    """a prefixed depth first iterator, designed to be used with visitors
+    """
+    def __init__(self, node, filter_func=None):
+        FilteredIterator.__init__(self, node, pre_order_list, filter_func)
+
diff --git a/logilab/common/umessage.py b/logilab/common/umessage.py
new file mode 100644
index 0000000..a5e4799
--- /dev/null
+++ b/logilab/common/umessage.py
@@ -0,0 +1,194 @@
+# copyright 2003-2012 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Unicode email support (extends email from stdlib)"""
+
+__docformat__ = "restructuredtext en"
+
+import email
+from encodings import search_function
+import sys
+if sys.version_info >= (2, 5):
+    from email.utils import parseaddr, parsedate
+    from email.header import decode_header
+else:
+    from email.Utils import parseaddr, parsedate
+    from email.Header import decode_header
+
+from datetime import datetime
+
+try:
+    from mx.DateTime import DateTime
+except ImportError:
+    DateTime = datetime
+
+import logilab.common as lgc
+
+
+def decode_QP(string):
+    parts = []
+    for decoded, charset in decode_header(string):
+        if not charset :
+            charset = 'iso-8859-15'
+        parts.append(decoded.decode(charset, 'replace'))
+
+    if sys.version_info < (3, 3):
+        # decoding was non-RFC compliant wrt to whitespace handling
+        # see http://bugs.python.org/issue1079
+        return u' '.join(parts)
+    return u''.join(parts)
+
+def message_from_file(fd):
+    try:
+        return UMessage(email.message_from_file(fd))
+    except email.Errors.MessageParseError:
+        return ''
+
+def message_from_string(string):
+    try:
+        return UMessage(email.message_from_string(string))
+    except email.Errors.MessageParseError:
+        return ''
+
+class UMessage:
+    """Encapsulates an email.Message instance and returns only unicode objects.
+    """
+
+    def __init__(self, message):
+        self.message = message
+
+    # email.Message interface #################################################
+
+    def get(self, header, default=None):
+        value = self.message.get(header, default)
+        if value:
+            return decode_QP(value)
+        return value
+
+    def __getitem__(self, header):
+        return self.get(header)
+
+    def get_all(self, header, default=()):
+        return [decode_QP(val) for val in self.message.get_all(header, default)
+                if val is not None]
+
+    def is_multipart(self):
+        return self.message.is_multipart()
+
+    def get_boundary(self):
+        return self.message.get_boundary()
+
+    def walk(self):
+        for part in self.message.walk():
+            yield UMessage(part)
+
+    if sys.version_info < (3, 0):
+
+        def get_payload(self, index=None, decode=False):
+            message = self.message
+            if index is None:
+                payload = message.get_payload(index, decode)
+                if isinstance(payload, list):
+                    return [UMessage(msg) for msg in payload]
+                if message.get_content_maintype() != 'text':
+                    return payload
+
+                charset = message.get_content_charset() or 'iso-8859-1'
+                if search_function(charset) is None:
+                    charset = 'iso-8859-1'
+                return unicode(payload or '', charset, "replace")
+            else:
+                payload = UMessage(message.get_payload(index, decode))
+            return payload
+
+        def get_content_maintype(self):
+            return unicode(self.message.get_content_maintype())
+
+        def get_content_type(self):
+            return unicode(self.message.get_content_type())
+
+        def get_filename(self, failobj=None):
+            value = self.message.get_filename(failobj)
+            if value is failobj:
+                return value
+            try:
+                return unicode(value)
+            except UnicodeDecodeError:
+                return u'error decoding filename'
+
+    else:
+
+        def get_payload(self, index=None, decode=False):
+            message = self.message
+            if index is None:
+                payload = message.get_payload(index, decode)
+                if isinstance(payload, list):
+                    return [UMessage(msg) for msg in payload]
+                return payload
+            else:
+                payload = UMessage(message.get_payload(index, decode))
+            return payload
+
+        def get_content_maintype(self):
+            return self.message.get_content_maintype()
+
+        def get_content_type(self):
+            return self.message.get_content_type()
+
+        def get_filename(self, failobj=None):
+            return self.message.get_filename(failobj)
+
+    # other convenience methods ###############################################
+
+    def headers(self):
+        """return an unicode string containing all the message's headers"""
+        values = []
+        for header in self.message.keys():
+            values.append(u'%s: %s' % (header, self.get(header)))
+        return '\n'.join(values)
+
+    def multi_addrs(self, header):
+        """return a list of 2-uple (name, address) for the given address (which
+        is expected to be an header containing address such as from, to, cc...)
+        """
+        persons = []
+        for person in self.get_all(header, ()):
+            name, mail = parseaddr(person)
+            persons.append((name, mail))
+        return persons
+
+    def date(self, alternative_source=False, return_str=False):
+        """return a datetime object for the email's date or None if no date is
+        set or if it can't be parsed
+        """
+        value = self.get('date')
+        if value is None and alternative_source:
+            unix_from = self.message.get_unixfrom()
+            if unix_from is not None:
+                try:
+                    value = unix_from.split(" ", 2)[2]
+                except IndexError:
+                    pass
+        if value is not None:
+            datetuple = parsedate(value)
+            if datetuple:
+                if lgc.USE_MX_DATETIME:
+                    return DateTime(*datetuple[:6])
+                return datetime(*datetuple[:6])
+            elif not return_str:
+                return None
+        return value
diff --git a/logilab/common/ureports/__init__.py b/logilab/common/ureports/__init__.py
new file mode 100644
index 0000000..d76ebe5
--- /dev/null
+++ b/logilab/common/ureports/__init__.py
@@ -0,0 +1,172 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Universal report objects and some formatting drivers.
+
+A way to create simple reports using python objects, primarily designed to be
+formatted as text and html.
+"""
+__docformat__ = "restructuredtext en"
+
+import sys
+
+from logilab.common.compat import StringIO
+from logilab.common.textutils import linesep
+
+
+def get_nodes(node, klass):
+    """return an iterator on all children node of the given klass"""
+    for child in node.children:
+        if isinstance(child, klass):
+            yield child
+        # recurse (FIXME: recursion controled by an option)
+        for grandchild in get_nodes(child, klass):
+            yield grandchild
+
+def layout_title(layout):
+    """try to return the layout's title as string, return None if not found
+    """
+    for child in layout.children:
+        if isinstance(child, Title):
+            return u' '.join([node.data for node in get_nodes(child, Text)])
+
+def build_summary(layout, level=1):
+    """make a summary for the report, including X level"""
+    assert level > 0
+    level -= 1
+    summary = List(klass=u'summary')
+    for child in layout.children:
+        if not isinstance(child, Section):
+            continue
+        label = layout_title(child)
+        if not label and not child.id:
+            continue
+        if not child.id:
+            child.id = label.replace(' ', '-')
+        node = Link(u'#'+child.id, label=label or child.id)
+        # FIXME: Three following lines produce not very compliant
+        # docbook: there are some useless <para><para>. They might be
+        # replaced by the three commented lines but this then produces
+        # a bug in html display...
+        if level and [n for n in child.children if isinstance(n, Section)]:
+            node = Paragraph([node, build_summary(child, level)])
+        summary.append(node)
+#         summary.append(node)
+#         if level and [n for n in child.children if isinstance(n, Section)]:
+#             summary.append(build_summary(child, level))
+    return summary
+
+
+class BaseWriter(object):
+    """base class for ureport writers"""
+
+    def format(self, layout, stream=None, encoding=None):
+        """format and write the given layout into the stream object
+
+        unicode policy: unicode strings may be found in the layout;
+        try to call stream.write with it, but give it back encoded using
+        the given encoding if it fails
+        """
+        if stream is None:
+            stream = sys.stdout
+        if not encoding:
+            encoding = getattr(stream, 'encoding', 'UTF-8')
+        self.encoding = encoding or 'UTF-8'
+        self.__compute_funcs = []
+        self.out = stream
+        self.begin_format(layout)
+        layout.accept(self)
+        self.end_format(layout)
+
+    def format_children(self, layout):
+        """recurse on the layout children and call their accept method
+        (see the Visitor pattern)
+        """
+        for child in getattr(layout, 'children', ()):
+            child.accept(self)
+
+    def writeln(self, string=u''):
+        """write a line in the output buffer"""
+        self.write(string + linesep)
+
+    def write(self, string):
+        """write a string in the output buffer"""
+        try:
+            self.out.write(string)
+        except UnicodeEncodeError:
+            self.out.write(string.encode(self.encoding))
+
+    def begin_format(self, layout):
+        """begin to format a layout"""
+        self.section = 0
+
+    def end_format(self, layout):
+        """finished to format a layout"""
+
+    def get_table_content(self, table):
+        """trick to get table content without actually writing it
+
+        return an aligned list of lists containing table cells values as string
+        """
+        result = [[]]
+        cols = table.cols
+        for cell in self.compute_content(table):
+            if cols == 0:
+                result.append([])
+                cols = table.cols
+            cols -= 1
+            result[-1].append(cell)
+        # fill missing cells
+        while len(result[-1]) < cols:
+            result[-1].append(u'')
+        return result
+
+    def compute_content(self, layout):
+        """trick to compute the formatting of children layout before actually
+        writing it
+
+        return an iterator on strings (one for each child element)
+        """
+        # use cells !
+        def write(data):
+            try:
+                stream.write(data)
+            except UnicodeEncodeError:
+                stream.write(data.encode(self.encoding))
+        def writeln(data=u''):
+            try:
+                stream.write(data+linesep)
+            except UnicodeEncodeError:
+                stream.write(data.encode(self.encoding)+linesep)
+        self.write = write
+        self.writeln = writeln
+        self.__compute_funcs.append((write, writeln))
+        for child in layout.children:
+            stream = StringIO()
+            child.accept(self)
+            yield stream.getvalue()
+        self.__compute_funcs.pop()
+        try:
+            self.write, self.writeln = self.__compute_funcs[-1]
+        except IndexError:
+            del self.write
+            del self.writeln
+
+
+from logilab.common.ureports.nodes import *
+from logilab.common.ureports.text_writer import TextWriter
+from logilab.common.ureports.html_writer import HTMLWriter
diff --git a/logilab/common/ureports/docbook_writer.py b/logilab/common/ureports/docbook_writer.py
new file mode 100644
index 0000000..857068c
--- /dev/null
+++ b/logilab/common/ureports/docbook_writer.py
@@ -0,0 +1,140 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""HTML formatting drivers for ureports"""
+__docformat__ = "restructuredtext en"
+
+from six.moves import range
+
+from logilab.common.ureports import HTMLWriter
+
+class DocbookWriter(HTMLWriter):
+    """format layouts as HTML"""
+
+    def begin_format(self, layout):
+        """begin to format a layout"""
+        super(HTMLWriter, self).begin_format(layout)
+        if self.snippet is None:
+            self.writeln('<?xml version="1.0" encoding="ISO-8859-1"?>')
+            self.writeln("""
+<book xmlns:xi='http://www.w3.org/2001/XInclude'
+      lang='fr'>
+""")
+
+    def end_format(self, layout):
+        """finished to format a layout"""
+        if self.snippet is None:
+            self.writeln('</book>')
+
+    def visit_section(self, layout):
+        """display a section (using <chapter> (level 0) or <section>)"""
+        if self.section == 0:
+            tag = "chapter"
+        else:
+            tag = "section"
+        self.section += 1
+        self.writeln(self._indent('<%s%s>' % (tag, self.handle_attrs(layout))))
+        self.format_children(layout)
+        self.writeln(self._indent('</%s>'% tag))
+        self.section -= 1
+
+    def visit_title(self, layout):
+        """display a title using <title>"""
+        self.write(self._indent('  <title%s>' % self.handle_attrs(layout)))
+        self.format_children(layout)
+        self.writeln('</title>')
+
+    def visit_table(self, layout):
+        """display a table as html"""
+        self.writeln(self._indent('  <table%s><title>%s</title>' \
+                     % (self.handle_attrs(layout), layout.title)))
+        self.writeln(self._indent('    <tgroup cols="%s">'% layout.cols))
+        for i in range(layout.cols):
+            self.writeln(self._indent('      <colspec colname="c%s" colwidth="1*"/>' % i))
+
+        table_content = self.get_table_content(layout)
+        # write headers
+        if layout.cheaders:
+            self.writeln(self._indent('      <thead>'))
+            self._write_row(table_content[0])
+            self.writeln(self._indent('      </thead>'))
+            table_content = table_content[1:]
+        elif layout.rcheaders:
+            self.writeln(self._indent('      <thead>'))
+            self._write_row(table_content[-1])
+            self.writeln(self._indent('      </thead>'))
+            table_content = table_content[:-1]
+        # write body
+        self.writeln(self._indent('      <tbody>'))
+        for i in range(len(table_content)):
+            row = table_content[i]
+            self.writeln(self._indent('        <row>'))
+            for j in range(len(row)):
+                cell = row[j] or '&#160;'
+                self.writeln(self._indent('          <entry>%s</entry>' % cell))
+            self.writeln(self._indent('        </row>'))
+        self.writeln(self._indent('      </tbody>'))
+        self.writeln(self._indent('    </tgroup>'))
+        self.writeln(self._indent('  </table>'))
+
+    def _write_row(self, row):
+        """write content of row (using <row> <entry>)"""
+        self.writeln('        <row>')
+        for j in range(len(row)):
+            cell = row[j] or '&#160;'
+            self.writeln('          <entry>%s</entry>' % cell)
+        self.writeln(self._indent('        </row>'))
+
+    def visit_list(self, layout):
+        """display a list (using <itemizedlist>)"""
+        self.writeln(self._indent('  <itemizedlist%s>' % self.handle_attrs(layout)))
+        for row in list(self.compute_content(layout)):
+            self.writeln('    <listitem><para>%s</para></listitem>' % row)
+        self.writeln(self._indent('  </itemizedlist>'))
+
+    def visit_paragraph(self, layout):
+        """display links (using <para>)"""
+        self.write(self._indent('  <para>'))
+        self.format_children(layout)
+        self.writeln('</para>')
+
+    def visit_span(self, layout):
+        """display links (using <p>)"""
+        #TODO: translate in docbook
+        self.write('<literal %s>' % self.handle_attrs(layout))
+        self.format_children(layout)
+        self.write('</literal>')
+
+    def visit_link(self, layout):
+        """display links (using <ulink>)"""
+        self.write('<ulink url="%s"%s>%s</ulink>' % (layout.url,
+                                               self.handle_attrs(layout),
+                                               layout.label))
+
+    def visit_verbatimtext(self, layout):
+        """display verbatim text (using <programlisting>)"""
+        self.writeln(self._indent('  <programlisting>'))
+        self.write(layout.data.replace('&', '&amp;').replace('<', '&lt;'))
+        self.writeln(self._indent('  </programlisting>'))
+
+    def visit_text(self, layout):
+        """add some text"""
+        self.write(layout.data.replace('&', '&amp;').replace('<', '&lt;'))
+
+    def _indent(self, string):
+        """correctly indent string according to section"""
+        return ' ' * 2*(self.section) + string
diff --git a/logilab/common/ureports/html_writer.py b/logilab/common/ureports/html_writer.py
new file mode 100644
index 0000000..eba34ea
--- /dev/null
+++ b/logilab/common/ureports/html_writer.py
@@ -0,0 +1,133 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""HTML formatting drivers for ureports"""
+__docformat__ = "restructuredtext en"
+
+from cgi import escape
+
+from six.moves import range
+
+from logilab.common.ureports import BaseWriter
+
+
+class HTMLWriter(BaseWriter):
+    """format layouts as HTML"""
+
+    def __init__(self, snippet=None):
+        super(HTMLWriter, self).__init__()
+        self.snippet = snippet
+
+    def handle_attrs(self, layout):
+        """get an attribute string from layout member attributes"""
+        attrs = u''
+        klass = getattr(layout, 'klass', None)
+        if klass:
+            attrs += u' class="%s"' % klass
+        nid = getattr(layout, 'id', None)
+        if nid:
+            attrs += u' id="%s"' % nid
+        return attrs
+
+    def begin_format(self, layout):
+        """begin to format a layout"""
+        super(HTMLWriter, self).begin_format(layout)
+        if self.snippet is None:
+            self.writeln(u'<html>')
+            self.writeln(u'<body>')
+
+    def end_format(self, layout):
+        """finished to format a layout"""
+        if self.snippet is None:
+            self.writeln(u'</body>')
+            self.writeln(u'</html>')
+
+
+    def visit_section(self, layout):
+        """display a section as html, using div + h[section level]"""
+        self.section += 1
+        self.writeln(u'<div%s>' % self.handle_attrs(layout))
+        self.format_children(layout)
+        self.writeln(u'</div>')
+        self.section -= 1
+
+    def visit_title(self, layout):
+        """display a title using <hX>"""
+        self.write(u'<h%s%s>' % (self.section, self.handle_attrs(layout)))
+        self.format_children(layout)
+        self.writeln(u'</h%s>' % self.section)
+
+    def visit_table(self, layout):
+        """display a table as html"""
+        self.writeln(u'<table%s>' % self.handle_attrs(layout))
+        table_content = self.get_table_content(layout)
+        for i in range(len(table_content)):
+            row = table_content[i]
+            if i == 0 and layout.rheaders:
+                self.writeln(u'<tr class="header">')
+            elif i+1 == len(table_content) and layout.rrheaders:
+                self.writeln(u'<tr class="header">')
+            else:
+                self.writeln(u'<tr class="%s">' % (i%2 and 'even' or 'odd'))
+            for j in range(len(row)):
+                cell = row[j] or u'&#160;'
+                if (layout.rheaders and i == 0) or \
+                   (layout.cheaders and j == 0) or \
+                   (layout.rrheaders and i+1 == len(table_content)) or \
+                   (layout.rcheaders and j+1 == len(row)):
+                    self.writeln(u'<th>%s</th>' % cell)
+                else:
+                    self.writeln(u'<td>%s</td>' % cell)
+            self.writeln(u'</tr>')
+        self.writeln(u'</table>')
+
+    def visit_list(self, layout):
+        """display a list as html"""
+        self.writeln(u'<ul%s>' % self.handle_attrs(layout))
+        for row in list(self.compute_content(layout)):
+            self.writeln(u'<li>%s</li>' % row)
+        self.writeln(u'</ul>')
+
+    def visit_paragraph(self, layout):
+        """display links (using <p>)"""
+        self.write(u'<p>')
+        self.format_children(layout)
+        self.write(u'</p>')
+
+    def visit_span(self, layout):
+        """display links (using <p>)"""
+        self.write(u'<span%s>' % self.handle_attrs(layout))
+        self.format_children(layout)
+        self.write(u'</span>')
+
+    def visit_link(self, layout):
+        """display links (using <a>)"""
+        self.write(u' <a href="%s"%s>%s</a>' % (layout.url,
+                                                self.handle_attrs(layout),
+                                                layout.label))
+    def visit_verbatimtext(self, layout):
+        """display verbatim text (using <pre>)"""
+        self.write(u'<pre>')
+        self.write(layout.data.replace(u'&', u'&amp;').replace(u'<', u'&lt;'))
+        self.write(u'</pre>')
+
+    def visit_text(self, layout):
+        """add some text"""
+        data = layout.data
+        if layout.escaped:
+            data = data.replace(u'&', u'&amp;').replace(u'<', u'&lt;')
+        self.write(data)
diff --git a/logilab/common/ureports/nodes.py b/logilab/common/ureports/nodes.py
new file mode 100644
index 0000000..a9585b3
--- /dev/null
+++ b/logilab/common/ureports/nodes.py
@@ -0,0 +1,203 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Micro reports objects.
+
+A micro report is a tree of layout and content objects.
+"""
+__docformat__ = "restructuredtext en"
+
+from logilab.common.tree import VNode
+
+from six import string_types
+
+class BaseComponent(VNode):
+    """base report component
+
+    attributes
+    * id : the component's optional id
+    * klass : the component's optional klass
+    """
+    def __init__(self, id=None, klass=None):
+        VNode.__init__(self, id)
+        self.klass = klass
+
+class BaseLayout(BaseComponent):
+    """base container node
+
+    attributes
+    * BaseComponent attributes
+    * children : components in this table (i.e. the table's cells)
+    """
+    def __init__(self, children=(), **kwargs):
+        super(BaseLayout, self).__init__(**kwargs)
+        for child in children:
+            if isinstance(child, BaseComponent):
+                self.append(child)
+            else:
+                self.add_text(child)
+
+    def append(self, child):
+        """overridden to detect problems easily"""
+        assert child not in self.parents()
+        VNode.append(self, child)
+
+    def parents(self):
+        """return the ancestor nodes"""
+        assert self.parent is not self
+        if self.parent is None:
+            return []
+        return [self.parent] + self.parent.parents()
+
+    def add_text(self, text):
+        """shortcut to add text data"""
+        self.children.append(Text(text))
+
+
+# non container nodes #########################################################
+
+class Text(BaseComponent):
+    """a text portion
+
+    attributes :
+    * BaseComponent attributes
+    * data : the text value as an encoded or unicode string
+    """
+    def __init__(self, data, escaped=True, **kwargs):
+        super(Text, self).__init__(**kwargs)
+        #if isinstance(data, unicode):
+        #    data = data.encode('ascii')
+        assert isinstance(data, string_types), data.__class__
+        self.escaped = escaped
+        self.data = data
+
+class VerbatimText(Text):
+    """a verbatim text, display the raw data
+
+    attributes :
+    * BaseComponent attributes
+    * data : the text value as an encoded or unicode string
+    """
+
+class Link(BaseComponent):
+    """a labelled link
+
+    attributes :
+    * BaseComponent attributes
+    * url : the link's target (REQUIRED)
+    * label : the link's label as a string (use the url by default)
+    """
+    def __init__(self, url, label=None, **kwargs):
+        super(Link, self).__init__(**kwargs)
+        assert url
+        self.url = url
+        self.label = label or url
+
+
+class Image(BaseComponent):
+    """an embedded or a single image
+
+    attributes :
+    * BaseComponent attributes
+    * filename : the image's filename (REQUIRED)
+    * stream : the stream object containing the image data (REQUIRED)
+    * title : the image's optional title
+    """
+    def __init__(self, filename, stream, title=None, **kwargs):
+        super(Image, self).__init__(**kwargs)
+        assert filename
+        assert stream
+        self.filename = filename
+        self.stream = stream
+        self.title = title
+
+
+# container nodes #############################################################
+
+class Section(BaseLayout):
+    """a section
+
+    attributes :
+    * BaseLayout attributes
+
+    a title may also be given to the constructor, it'll be added
+    as a first element
+    a description may also be given to the constructor, it'll be added
+    as a first paragraph
+    """
+    def __init__(self, title=None, description=None, **kwargs):
+        super(Section, self).__init__(**kwargs)
+        if description:
+            self.insert(0, Paragraph([Text(description)]))
+        if title:
+            self.insert(0, Title(children=(title,)))
+
+class Title(BaseLayout):
+    """a title
+
+    attributes :
+    * BaseLayout attributes
+
+    A title must not contains a section nor a paragraph!
+    """
+
+class Span(BaseLayout):
+    """a title
+
+    attributes :
+    * BaseLayout attributes
+
+    A span should only contains Text and Link nodes (in-line elements)
+    """
+
+class Paragraph(BaseLayout):
+    """a simple text paragraph
+
+    attributes :
+    * BaseLayout attributes
+
+    A paragraph must not contains a section !
+    """
+
+class Table(BaseLayout):
+    """some tabular data
+
+    attributes :
+    * BaseLayout attributes
+    * cols : the number of columns of the table (REQUIRED)
+    * rheaders : the first row's elements are table's header
+    * cheaders : the first col's elements are table's header
+    * title : the table's optional title
+    """
+    def __init__(self, cols, title=None,
+                 rheaders=0, cheaders=0, rrheaders=0, rcheaders=0,
+                 **kwargs):
+        super(Table, self).__init__(**kwargs)
+        assert isinstance(cols, int)
+        self.cols = cols
+        self.title = title
+        self.rheaders = rheaders
+        self.cheaders = cheaders
+        self.rrheaders = rrheaders
+        self.rcheaders = rcheaders
+
+class List(BaseLayout):
+    """some list data
+
+    attributes :
+    * BaseLayout attributes
+    """
diff --git a/logilab/common/ureports/text_writer.py b/logilab/common/ureports/text_writer.py
new file mode 100644
index 0000000..c87613c
--- /dev/null
+++ b/logilab/common/ureports/text_writer.py
@@ -0,0 +1,145 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Text formatting drivers for ureports"""
+
+from __future__ import print_function
+
+__docformat__ = "restructuredtext en"
+
+from six.moves import range
+
+from logilab.common.textutils import linesep
+from logilab.common.ureports import BaseWriter
+
+
+TITLE_UNDERLINES = [u'', u'=', u'-', u'`', u'.', u'~', u'^']
+BULLETS = [u'*', u'-']
+
+class TextWriter(BaseWriter):
+    """format layouts as text
+    (ReStructured inspiration but not totally handled yet)
+    """
+    def begin_format(self, layout):
+        super(TextWriter, self).begin_format(layout)
+        self.list_level = 0
+        self.pending_urls = []
+
+    def visit_section(self, layout):
+        """display a section as text
+        """
+        self.section += 1
+        self.writeln()
+        self.format_children(layout)
+        if self.pending_urls:
+            self.writeln()
+            for label, url in self.pending_urls:
+                self.writeln(u'.. _`%s`: %s' % (label, url))
+            self.pending_urls = []
+        self.section -= 1
+        self.writeln()
+
+    def visit_title(self, layout):
+        title = u''.join(list(self.compute_content(layout)))
+        self.writeln(title)
+        try:
+            self.writeln(TITLE_UNDERLINES[self.section] * len(title))
+        except IndexError:
+            print("FIXME TITLE TOO DEEP. TURNING TITLE INTO TEXT")
+
+    def visit_paragraph(self, layout):
+        """enter a paragraph"""
+        self.format_children(layout)
+        self.writeln()
+
+    def visit_span(self, layout):
+        """enter a span"""
+        self.format_children(layout)
+
+    def visit_table(self, layout):
+        """display a table as text"""
+        table_content = self.get_table_content(layout)
+        # get columns width
+        cols_width = [0]*len(table_content[0])
+        for row in table_content:
+            for index in range(len(row)):
+                col = row[index]
+                cols_width[index] = max(cols_width[index], len(col))
+        if layout.klass == 'field':
+            self.field_table(layout, table_content, cols_width)
+        else:
+            self.default_table(layout, table_content, cols_width)
+        self.writeln()
+
+    def default_table(self, layout, table_content, cols_width):
+        """format a table"""
+        cols_width = [size+1 for size in cols_width]
+        format_strings = u' '.join([u'%%-%ss'] * len(cols_width))
+        format_strings = format_strings % tuple(cols_width)
+        format_strings = format_strings.split(' ')
+        table_linesep = u'\n+' + u'+'.join([u'-'*w for w in cols_width]) + u'+\n'
+        headsep = u'\n+' + u'+'.join([u'='*w for w in cols_width]) + u'+\n'
+        # FIXME: layout.cheaders
+        self.write(table_linesep)
+        for i in range(len(table_content)):
+            self.write(u'|')
+            line = table_content[i]
+            for j in range(len(line)):
+                self.write(format_strings[j] % line[j])
+                self.write(u'|')
+            if i == 0 and layout.rheaders:
+                self.write(headsep)
+            else:
+                self.write(table_linesep)
+
+    def field_table(self, layout, table_content, cols_width):
+        """special case for field table"""
+        assert layout.cols == 2
+        format_string = u'%s%%-%ss: %%s' % (linesep, cols_width[0])
+        for field, value in table_content:
+            self.write(format_string % (field, value))
+
+
+    def visit_list(self, layout):
+        """display a list layout as text"""
+        bullet = BULLETS[self.list_level % len(BULLETS)]
+        indent = '  ' * self.list_level
+        self.list_level += 1
+        for child in layout.children:
+            self.write(u'%s%s%s ' % (linesep, indent, bullet))
+            child.accept(self)
+        self.list_level -= 1
+
+    def visit_link(self, layout):
+        """add a hyperlink"""
+        if layout.label != layout.url:
+            self.write(u'`%s`_' % layout.label)
+            self.pending_urls.append( (layout.label, layout.url) )
+        else:
+            self.write(layout.url)
+
+    def visit_verbatimtext(self, layout):
+        """display a verbatim layout as text (so difficult ;)
+        """
+        self.writeln(u'::\n')
+        for line in layout.data.splitlines():
+            self.writeln(u'    ' + line)
+        self.writeln()
+
+    def visit_text(self, layout):
+        """add some text"""
+        self.write(u'%s' % layout.data)
diff --git a/logilab/common/urllib2ext.py b/logilab/common/urllib2ext.py
new file mode 100644
index 0000000..339aec0
--- /dev/null
+++ b/logilab/common/urllib2ext.py
@@ -0,0 +1,89 @@
+from __future__ import print_function
+
+import logging
+import urllib2
+
+import kerberos as krb
+
+class GssapiAuthError(Exception):
+    """raised on error during authentication process"""
+
+import re
+RGX = re.compile('(?:.*,)*\s*Negotiate\s*([^,]*),?', re.I)
+
+def get_negociate_value(headers):
+    for authreq in headers.getheaders('www-authenticate'):
+        match = RGX.search(authreq)
+        if match:
+            return match.group(1)
+
+class HTTPGssapiAuthHandler(urllib2.BaseHandler):
+    """Negotiate HTTP authentication using context from GSSAPI"""
+
+    handler_order = 400 # before Digest Auth
+
+    def __init__(self):
+        self._reset()
+
+    def _reset(self):
+        self._retried = 0
+        self._context = None
+
+    def clean_context(self):
+        if self._context is not None:
+            krb.authGSSClientClean(self._context)
+
+    def http_error_401(self, req, fp, code, msg, headers):
+        try:
+            if self._retried > 5:
+                raise urllib2.HTTPError(req.get_full_url(), 401,
+                                        "negotiate auth failed", headers, None)
+            self._retried += 1
+            logging.debug('gssapi handler, try %s' % self._retried)
+            negotiate = get_negociate_value(headers)
+            if negotiate is None:
+                logging.debug('no negociate found in a www-authenticate header')
+                return None
+            logging.debug('HTTPGssapiAuthHandler: negotiate 1 is %r' % negotiate)
+            result, self._context = krb.authGSSClientInit("HTTP@%s" % req.get_host())
+            if result < 1:
+                raise GssapiAuthError("HTTPGssapiAuthHandler: init failed with %d" % result)
+            result = krb.authGSSClientStep(self._context, negotiate)
+            if result < 0:
+                raise GssapiAuthError("HTTPGssapiAuthHandler: step 1 failed with %d" % result)
+            client_response = krb.authGSSClientResponse(self._context)
+            logging.debug('HTTPGssapiAuthHandler: client response is %s...' % client_response[:10])
+            req.add_unredirected_header('Authorization', "Negotiate %s" % client_response)
+            server_response = self.parent.open(req)
+            negotiate = get_negociate_value(server_response.info())
+            if negotiate is None:
+                logging.warning('HTTPGssapiAuthHandler: failed to authenticate server')
+            else:
+                logging.debug('HTTPGssapiAuthHandler negotiate 2: %s' % negotiate)
+                result = krb.authGSSClientStep(self._context, negotiate)
+                if result < 1:
+                    raise GssapiAuthError("HTTPGssapiAuthHandler: step 2 failed with %d" % result)
+            return server_response
+        except GssapiAuthError as exc:
+            logging.error(repr(exc))
+        finally:
+            self.clean_context()
+            self._reset()
+
+if __name__ == '__main__':
+    import sys
+    # debug
+    import httplib
+    httplib.HTTPConnection.debuglevel = 1
+    httplib.HTTPSConnection.debuglevel = 1
+    # debug
+    import logging
+    logging.basicConfig(level=logging.DEBUG)
+    # handle cookies
+    import cookielib
+    cj = cookielib.CookieJar()
+    ch = urllib2.HTTPCookieProcessor(cj)
+    # test with url sys.argv[1]
+    h = HTTPGssapiAuthHandler()
+    response = urllib2.build_opener(h, ch).open(sys.argv[1])
+    print('\nresponse: %s\n--------------\n' % response.code, response.info())
diff --git a/logilab/common/vcgutils.py b/logilab/common/vcgutils.py
new file mode 100644
index 0000000..9cd2acd
--- /dev/null
+++ b/logilab/common/vcgutils.py
@@ -0,0 +1,216 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""Functions to generate files readable with Georg Sander's vcg
+(Visualization of Compiler Graphs).
+
+You can download vcg at http://rw4.cs.uni-sb.de/~sander/html/gshome.html
+Note that vcg exists as a debian package.
+
+See vcg's documentation for explanation about the different values that
+maybe used for the functions parameters.
+
+
+
+
+"""
+__docformat__ = "restructuredtext en"
+
+import string
+
+ATTRS_VAL = {
+    'algos':       ('dfs', 'tree', 'minbackward',
+                    'left_to_right', 'right_to_left',
+                    'top_to_bottom', 'bottom_to_top',
+                    'maxdepth', 'maxdepthslow', 'mindepth', 'mindepthslow',
+                    'mindegree', 'minindegree', 'minoutdegree',
+                    'maxdegree', 'maxindegree', 'maxoutdegree'),
+    'booleans':    ('yes', 'no'),
+    'colors':      ('black', 'white', 'blue', 'red', 'green', 'yellow',
+                    'magenta', 'lightgrey',
+                    'cyan', 'darkgrey', 'darkblue', 'darkred', 'darkgreen',
+                    'darkyellow', 'darkmagenta', 'darkcyan', 'gold',
+                    'lightblue', 'lightred', 'lightgreen', 'lightyellow',
+                    'lightmagenta', 'lightcyan', 'lilac', 'turquoise',
+                    'aquamarine', 'khaki', 'purple', 'yellowgreen', 'pink',
+                    'orange', 'orchid'),
+    'shapes':      ('box', 'ellipse', 'rhomb', 'triangle'),
+    'textmodes':   ('center', 'left_justify', 'right_justify'),
+    'arrowstyles': ('solid', 'line', 'none'),
+    'linestyles':  ('continuous', 'dashed', 'dotted', 'invisible'),
+    }
+
+# meaning of possible values:
+#   O    -> string
+#   1    -> int
+#   list -> value in list
+GRAPH_ATTRS = {
+    'title':              0,
+    'label':              0,
+    'color':               ATTRS_VAL['colors'],
+    'textcolor':           ATTRS_VAL['colors'],
+    'bordercolor':         ATTRS_VAL['colors'],
+    'width':               1,
+    'height':              1,
+    'borderwidth':         1,
+    'textmode':            ATTRS_VAL['textmodes'],
+    'shape':               ATTRS_VAL['shapes'],
+    'shrink':              1,
+    'stretch':             1,
+    'orientation':         ATTRS_VAL['algos'],
+    'vertical_order':      1,
+    'horizontal_order':    1,
+    'xspace':              1,
+    'yspace':              1,
+    'layoutalgorithm':    ATTRS_VAL['algos'],
+    'late_edge_labels':   ATTRS_VAL['booleans'],
+    'display_edge_labels': ATTRS_VAL['booleans'],
+    'dirty_edge_labels':  ATTRS_VAL['booleans'],
+    'finetuning':          ATTRS_VAL['booleans'],
+    'manhattan_edges':     ATTRS_VAL['booleans'],
+    'smanhattan_edges':    ATTRS_VAL['booleans'],
+    'port_sharing':        ATTRS_VAL['booleans'],
+    'edges':               ATTRS_VAL['booleans'],
+    'nodes':               ATTRS_VAL['booleans'],
+    'splines':             ATTRS_VAL['booleans'],
+    }
+NODE_ATTRS = {
+    'title':              0,
+    'label':              0,
+    'color':               ATTRS_VAL['colors'],
+    'textcolor':           ATTRS_VAL['colors'],
+    'bordercolor':         ATTRS_VAL['colors'],
+    'width':               1,
+    'height':              1,
+    'borderwidth':         1,
+    'textmode':            ATTRS_VAL['textmodes'],
+    'shape':               ATTRS_VAL['shapes'],
+    'shrink':              1,
+    'stretch':             1,
+    'vertical_order':      1,
+    'horizontal_order':    1,
+    }
+EDGE_ATTRS = {
+    'sourcename':         0,
+    'targetname':         0,
+    'label':              0,
+    'linestyle':          ATTRS_VAL['linestyles'],
+    'class':              1,
+    'thickness':          0,
+    'color':               ATTRS_VAL['colors'],
+    'textcolor':           ATTRS_VAL['colors'],
+    'arrowcolor':          ATTRS_VAL['colors'],
+    'backarrowcolor':      ATTRS_VAL['colors'],
+    'arrowsize':           1,
+    'backarrowsize':       1,
+    'arrowstyle':          ATTRS_VAL['arrowstyles'],
+    'backarrowstyle':      ATTRS_VAL['arrowstyles'],
+    'textmode':            ATTRS_VAL['textmodes'],
+    'priority':            1,
+    'anchor':              1,
+    'horizontal_order':    1,
+    }
+
+
+# Misc utilities ###############################################################
+
+def latin_to_vcg(st):
+    """Convert latin characters using vcg escape sequence.
+    """
+    for char in st:
+        if char not in string.ascii_letters:
+            try:
+                num = ord(char)
+                if num >= 192:
+                    st = st.replace(char, r'\fi%d'%ord(char))
+            except:
+                pass
+    return st
+
+
+class VCGPrinter:
+    """A vcg graph writer.
+    """
+
+    def __init__(self, output_stream):
+        self._stream = output_stream
+        self._indent = ''
+
+    def open_graph(self, **args):
+        """open a vcg graph
+        """
+        self._stream.write('%sgraph:{\n'%self._indent)
+        self._inc_indent()
+        self._write_attributes(GRAPH_ATTRS, **args)
+
+    def close_graph(self):
+        """close a vcg graph
+        """
+        self._dec_indent()
+        self._stream.write('%s}\n'%self._indent)
+
+
+    def node(self, title, **args):
+        """draw a node
+        """
+        self._stream.write('%snode: {title:"%s"' % (self._indent, title))
+        self._write_attributes(NODE_ATTRS, **args)
+        self._stream.write('}\n')
+
+
+    def edge(self, from_node, to_node, edge_type='', **args):
+        """draw an edge from a node to another.
+        """
+        self._stream.write(
+            '%s%sedge: {sourcename:"%s" targetname:"%s"' % (
+            self._indent, edge_type, from_node, to_node))
+        self._write_attributes(EDGE_ATTRS, **args)
+        self._stream.write('}\n')
+
+
+    # private ##################################################################
+
+    def _write_attributes(self, attributes_dict, **args):
+        """write graph, node or edge attributes
+        """
+        for key, value in args.items():
+            try:
+                _type =  attributes_dict[key]
+            except KeyError:
+                raise Exception('''no such attribute %s
+possible attributes are %s''' % (key, attributes_dict.keys()))
+
+            if not _type:
+                self._stream.write('%s%s:"%s"\n' % (self._indent, key, value))
+            elif _type == 1:
+                self._stream.write('%s%s:%s\n' % (self._indent, key,
+                                                  int(value)))
+            elif value in _type:
+                self._stream.write('%s%s:%s\n' % (self._indent, key, value))
+            else:
+                raise Exception('''value %s isn\'t correct for attribute %s
+correct values are %s''' % (value, key, _type))
+
+    def _inc_indent(self):
+        """increment indentation
+        """
+        self._indent = '  %s' % self._indent
+
+    def _dec_indent(self):
+        """decrement indentation
+        """
+        self._indent = self._indent[:-2]
diff --git a/logilab/common/visitor.py b/logilab/common/visitor.py
new file mode 100644
index 0000000..ed2b70f
--- /dev/null
+++ b/logilab/common/visitor.py
@@ -0,0 +1,109 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""A generic visitor abstract implementation.
+
+
+
+
+"""
+__docformat__ = "restructuredtext en"
+
+def no_filter(_):
+    return 1
+
+# Iterators ###################################################################
+class FilteredIterator(object):
+
+    def __init__(self, node, list_func, filter_func=None):
+        self._next = [(node, 0)]
+        if filter_func is None:
+            filter_func = no_filter
+        self._list = list_func(node, filter_func)
+
+    def __next__(self):
+        try:
+            return self._list.pop(0)
+        except :
+            return None
+
+    next = __next__
+
+# Base Visitor ################################################################
+class Visitor(object):
+
+    def __init__(self, iterator_class, filter_func=None):
+        self._iter_class = iterator_class
+        self.filter = filter_func
+
+    def visit(self, node, *args, **kargs):
+        """
+        launch the visit on a given node
+
+        call 'open_visit' before the beginning of the visit, with extra args
+        given
+        when all nodes have been visited, call the 'close_visit' method
+        """
+        self.open_visit(node, *args, **kargs)
+        return self.close_visit(self._visit(node))
+
+    def _visit(self, node):
+        iterator = self._get_iterator(node)
+        n = next(iterator)
+        while n:
+            result = n.accept(self)
+            n = next(iterator)
+        return result
+
+    def _get_iterator(self, node):
+        return self._iter_class(node, self.filter)
+
+    def open_visit(self, *args, **kargs):
+        """
+        method called at the beginning of the visit
+        """
+        pass
+
+    def close_visit(self, result):
+        """
+        method called at the end of the visit
+        """
+        return result
+
+# standard visited mixin ######################################################
+class VisitedMixIn(object):
+    """
+    Visited interface allow node visitors to use the node
+    """
+    def get_visit_name(self):
+        """
+        return the visit name for the mixed class. When calling 'accept', the
+        method <'visit_' + name returned by this method> will be called on the
+        visitor
+        """
+        try:
+            return self.TYPE.replace('-', '_')
+        except:
+            return self.__class__.__name__.lower()
+
+    def accept(self, visitor, *args, **kwargs):
+        func = getattr(visitor, 'visit_%s' % self.get_visit_name())
+        return func(self, *args, **kwargs)
+
+    def leave(self, visitor, *args, **kwargs):
+        func = getattr(visitor, 'leave_%s' % self.get_visit_name())
+        return func(self, *args, **kwargs)
diff --git a/logilab/common/xmlrpcutils.py b/logilab/common/xmlrpcutils.py
new file mode 100644
index 0000000..1d30d82
--- /dev/null
+++ b/logilab/common/xmlrpcutils.py
@@ -0,0 +1,131 @@
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""XML-RPC utilities."""
+__docformat__ = "restructuredtext en"
+
+import xmlrpclib
+from base64 import encodestring
+#from cStringIO import StringIO
+
+ProtocolError = xmlrpclib.ProtocolError
+
+## class BasicAuthTransport(xmlrpclib.Transport):
+##     def __init__(self, username=None, password=None):
+##         self.username = username
+##         self.password = password
+##         self.verbose = None
+##         self.has_ssl = httplib.__dict__.has_key("HTTPConnection")
+
+##     def request(self, host, handler, request_body, verbose=None):
+##         # issue XML-RPC request
+##         if self.has_ssl:
+##             if host.startswith("https:"): h = httplib.HTTPSConnection(host)
+##             else: h = httplib.HTTPConnection(host)
+##         else: h = httplib.HTTP(host)
+
+##         h.putrequest("POST", handler)
+
+##         # required by HTTP/1.1
+##         if not self.has_ssl: # HTTPConnection already does 1.1
+##             h.putheader("Host", host)
+##         h.putheader("Connection", "close")
+
+##         if request_body: h.send(request_body)
+##         if self.has_ssl:
+##             response = h.getresponse()
+##             if response.status != 200:
+##                 raise xmlrpclib.ProtocolError(host + handler,
+##                                               response.status,
+##                                               response.reason,
+##                                               response.msg)
+##             file = response.fp
+##         else:
+##             errcode, errmsg, headers = h.getreply()
+##             if errcode != 200:
+##                 raise xmlrpclib.ProtocolError(host + handler, errcode,
+##                                               errmsg, headers)
+
+##             file = h.getfile()
+
+##         return self.parse_response(file)
+
+
+
+class AuthMixin:
+    """basic http authentication mixin for xmlrpc transports"""
+
+    def __init__(self, username, password, encoding):
+        self.verbose = 0
+        self.username = username
+        self.password = password
+        self.encoding = encoding
+
+    def request(self, host, handler, request_body, verbose=0):
+        """issue XML-RPC request"""
+        h = self.make_connection(host)
+        h.putrequest("POST", handler)
+        # required by XML-RPC
+        h.putheader("User-Agent", self.user_agent)
+        h.putheader("Content-Type", "text/xml")
+        h.putheader("Content-Length", str(len(request_body)))
+        h.putheader("Host", host)
+        h.putheader("Connection", "close")
+        # basic auth
+        if self.username is not None and self.password is not None:
+            h.putheader("AUTHORIZATION", "Basic %s" % encodestring(
+                "%s:%s" % (self.username, self.password)).replace("\012", ""))
+        h.endheaders()
+        # send body
+        if request_body:
+            h.send(request_body)
+        # get and check reply
+        errcode, errmsg, headers = h.getreply()
+        if errcode != 200:
+            raise ProtocolError(host + handler, errcode, errmsg, headers)
+        file = h.getfile()
+##         # FIXME: encoding ??? iirc, this fix a bug in xmlrpclib but...
+##         data = h.getfile().read()
+##         if self.encoding != 'UTF-8':
+##             data = data.replace("version='1.0'",
+##                                 "version='1.0' encoding='%s'" % self.encoding)
+##         result = StringIO()
+##         result.write(data)
+##         result.seek(0)
+##         return self.parse_response(result)
+        return self.parse_response(file)
+
+class BasicAuthTransport(AuthMixin, xmlrpclib.Transport):
+    """basic http authentication transport"""
+
+class BasicAuthSafeTransport(AuthMixin, xmlrpclib.SafeTransport):
+    """basic https authentication transport"""
+
+
+def connect(url, user=None, passwd=None, encoding='ISO-8859-1'):
+    """return an xml rpc server on <url>, using user / password if specified
+    """
+    if user or passwd:
+        assert user and passwd is not None
+        if url.startswith('https://'):
+            transport = BasicAuthSafeTransport(user, passwd, encoding)
+        else:
+            transport = BasicAuthTransport(user, passwd, encoding)
+    else:
+        transport = None
+    server = xmlrpclib.ServerProxy(url, transport, encoding=encoding)
+    return server
diff --git a/logilab/common/xmlutils.py b/logilab/common/xmlutils.py
new file mode 100644
index 0000000..d383b9d
--- /dev/null
+++ b/logilab/common/xmlutils.py
@@ -0,0 +1,61 @@
+# -*- coding: utf-8 -*-
+# copyright 2003-2011 LOGILAB S.A. (Paris, FRANCE), all rights reserved.
+# contact http://www.logilab.fr/ -- mailto:contact@logilab.fr
+#
+# This file is part of logilab-common.
+#
+# logilab-common is free software: you can redistribute it and/or modify it under
+# the terms of the GNU Lesser General Public License as published by the Free
+# Software Foundation, either version 2.1 of the License, or (at your option) any
+# later version.
+#
+# logilab-common is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+# FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public License for more
+# details.
+#
+# You should have received a copy of the GNU Lesser General Public License along
+# with logilab-common.  If not, see <http://www.gnu.org/licenses/>.
+"""XML utilities.
+
+This module contains useful functions for parsing and using XML data. For the
+moment, there is only one function that can parse the data inside a processing
+instruction and return a Python dictionary.
+
+
+
+
+"""
+__docformat__ = "restructuredtext en"
+
+import re
+
+RE_DOUBLE_QUOTE = re.compile('([\w\-\.]+)="([^"]+)"')
+RE_SIMPLE_QUOTE = re.compile("([\w\-\.]+)='([^']+)'")
+
+def parse_pi_data(pi_data):
+    """
+    Utility function that parses the data contained in an XML
+    processing instruction and returns a dictionary of keywords and their
+    associated values (most of the time, the processing instructions contain
+    data like ``keyword="value"``, if a keyword is not associated to a value,
+    for example ``keyword``, it will be associated to ``None``).
+
+    :param pi_data: data contained in an XML processing instruction.
+    :type pi_data: unicode
+
+    :returns: Dictionary of the keywords (Unicode strings) associated to
+              their values (Unicode strings) as they were defined in the
+              data.
+    :rtype: dict
+    """
+    results = {}
+    for elt in pi_data.split():
+        if RE_DOUBLE_QUOTE.match(elt):
+            kwd, val = RE_DOUBLE_QUOTE.match(elt).groups()
+        elif RE_SIMPLE_QUOTE.match(elt):
+            kwd, val = RE_SIMPLE_QUOTE.match(elt).groups()
+        else:
+            kwd, val = elt, None
+        results[kwd] = val
+    return results