diff options
author | Marc Abramowitz <marc@marc-abramowitz.com> | 2015-04-30 17:39:24 -0700 |
---|---|---|
committer | Marc Abramowitz <marc@marc-abramowitz.com> | 2015-04-30 17:39:24 -0700 |
commit | fa100c92c06d3a8a61a0dda1a2e06018437b09c6 (patch) | |
tree | a1cc50f93fbf257685c3849e03496c5e33949281 /paste/util/threadedprint.py | |
download | paste-git-test_wsgirequest_charset_use_UTF-8_instead_of_iso-8859-1.tar.gz |
test_wsgirequest_charset: Use UTF-8 instead of iso-8859-1test_wsgirequest_charset_use_UTF-8_instead_of_iso-8859-1
because it seems that the defacto standard for encoding URIs is to use UTF-8.
I've been reading about url encoding and it seems like perhaps using an
encoding other than UTF-8 is very non-standard and not well-supported (this
test is trying to use `iso-8859-1`).
From http://en.wikipedia.org/wiki/Percent-encoding
> For a non-ASCII character, it is typically converted to its byte sequence in
> UTF-8, and then each byte value is represented as above.
> The generic URI syntax mandates that new URI schemes that provide for the
> representation of character data in a URI must, in effect, represent
> characters from the unreserved set without translation, and should convert
> all other characters to bytes according to UTF-8, and then percent-encode
> those values. This requirement was introduced in January 2005 with the
> publication of RFC 3986
From http://tools.ietf.org/html/rfc3986:
> Non-ASCII characters must first be encoded according to UTF-8 [STD63], and
> then each octet of the corresponding UTF-8 sequence must be percent-encoded
> to be represented as URI characters. URI producing applications must not use
> percent-encoding in host unless it is used to represent a UTF-8 character
> sequence.
From http://tools.ietf.org/html/rfc3987:
> Conversions from URIs to IRIs MUST NOT use any character encoding other than
> UTF-8 in steps 3 and 4, even if it might be possible to guess from the
> context that another character encoding than UTF-8 was used in the URI. For
> example, the URI "http://www.example.org/r%E9sum%E9.html" might with some
> guessing be interpreted to contain two e-acute characters encoded as
> iso-8859-1. It must not be converted to an IRI containing these e-acute
> characters. Otherwise, in the future the IRI will be mapped to
> "http://www.example.org/r%C3%A9sum%C3%A9.html", which is a different URI from
> "http://www.example.org/r%E9sum%E9.html".
See issue #7, which I think this at least partially fixes.
Diffstat (limited to 'paste/util/threadedprint.py')
-rw-r--r-- | paste/util/threadedprint.py | 250 |
1 files changed, 250 insertions, 0 deletions
diff --git a/paste/util/threadedprint.py b/paste/util/threadedprint.py new file mode 100644 index 0000000..820311e --- /dev/null +++ b/paste/util/threadedprint.py @@ -0,0 +1,250 @@ +# (c) 2005 Ian Bicking and contributors; written for Paste (http://pythonpaste.org) +# Licensed under the MIT license: http://www.opensource.org/licenses/mit-license.php + +""" +threadedprint.py +================ + +:author: Ian Bicking +:date: 12 Jul 2004 + +Multi-threaded printing; allows the output produced via print to be +separated according to the thread. + +To use this, you must install the catcher, like:: + + threadedprint.install() + +The installation optionally takes one of three parameters: + +default + The default destination for print statements (e.g., ``sys.stdout``). +factory + A function that will produce the stream for a thread, given the + thread's name. +paramwriter + Instead of writing to a file-like stream, this function will be + called like ``paramwriter(thread_name, text)`` for every write. + +The thread name is the value returned by +``threading.currentThread().getName()``, a string (typically something +like Thread-N). + +You can also submit file-like objects for specific threads, which will +override any of these parameters. To do this, call ``register(stream, +[threadName])``. ``threadName`` is optional, and if not provided the +stream will be registered for the current thread. + +If no specific stream is registered for a thread, and no default has +been provided, then an error will occur when anything is written to +``sys.stdout`` (or printed). + +Note: the stream's ``write`` method will be called in the thread the +text came from, so you should consider thread safety, especially if +multiple threads share the same writer. + +Note: if you want access to the original standard out, use +``sys.__stdout__``. + +You may also uninstall this, via:: + + threadedprint.uninstall() + +TODO +---- + +* Something with ``sys.stderr``. +* Some default handlers. Maybe something that hooks into `logging`. +* Possibly cache the results of ``factory`` calls. This would be a + semantic change. + +""" + +import threading +import sys +from paste.util import filemixin + +class PrintCatcher(filemixin.FileMixin): + + def __init__(self, default=None, factory=None, paramwriter=None, + leave_stdout=False): + assert len(filter(lambda x: x is not None, + [default, factory, paramwriter])) <= 1, ( + "You can only provide one of default, factory, or paramwriter") + if leave_stdout: + assert not default, ( + "You cannot pass in both default (%r) and " + "leave_stdout=True" % default) + default = sys.stdout + if default: + self._defaultfunc = self._writedefault + elif factory: + self._defaultfunc = self._writefactory + elif paramwriter: + self._defaultfunc = self._writeparam + else: + self._defaultfunc = self._writeerror + self._default = default + self._factory = factory + self._paramwriter = paramwriter + self._catchers = {} + + def write(self, v, currentThread=threading.currentThread): + name = currentThread().getName() + catchers = self._catchers + if not catchers.has_key(name): + self._defaultfunc(name, v) + else: + catcher = catchers[name] + catcher.write(v) + + def seek(self, *args): + # Weird, but Google App Engine is seeking on stdout + name = threading.currentThread().getName() + catchers = self._catchers + if not name in catchers: + self._default.seek(*args) + else: + catchers[name].seek(*args) + + def read(self, *args): + name = threading.currentThread().getName() + catchers = self._catchers + if not name in catchers: + self._default.read(*args) + else: + catchers[name].read(*args) + + + def _writedefault(self, name, v): + self._default.write(v) + + def _writefactory(self, name, v): + self._factory(name).write(v) + + def _writeparam(self, name, v): + self._paramwriter(name, v) + + def _writeerror(self, name, v): + assert False, ( + "There is no PrintCatcher output stream for the thread %r" + % name) + + def register(self, catcher, name=None, + currentThread=threading.currentThread): + if name is None: + name = currentThread().getName() + self._catchers[name] = catcher + + def deregister(self, name=None, + currentThread=threading.currentThread): + if name is None: + name = currentThread().getName() + assert self._catchers.has_key(name), ( + "There is no PrintCatcher catcher for the thread %r" % name) + del self._catchers[name] + +_printcatcher = None +_oldstdout = None + +def install(**kw): + global _printcatcher, _oldstdout, register, deregister + if (not _printcatcher or sys.stdout is not _printcatcher): + _oldstdout = sys.stdout + _printcatcher = sys.stdout = PrintCatcher(**kw) + register = _printcatcher.register + deregister = _printcatcher.deregister + +def uninstall(): + global _printcatcher, _oldstdout, register, deregister + if _printcatcher: + sys.stdout = _oldstdout + _printcatcher = _oldstdout = None + register = not_installed_error + deregister = not_installed_error + +def not_installed_error(*args, **kw): + assert False, ( + "threadedprint has not yet been installed (call " + "threadedprint.install())") + +register = deregister = not_installed_error + +class StdinCatcher(filemixin.FileMixin): + + def __init__(self, default=None, factory=None, paramwriter=None): + assert len(filter(lambda x: x is not None, + [default, factory, paramwriter])) <= 1, ( + "You can only provide one of default, factory, or paramwriter") + if default: + self._defaultfunc = self._readdefault + elif factory: + self._defaultfunc = self._readfactory + elif paramwriter: + self._defaultfunc = self._readparam + else: + self._defaultfunc = self._readerror + self._default = default + self._factory = factory + self._paramwriter = paramwriter + self._catchers = {} + + def read(self, size=None, currentThread=threading.currentThread): + name = currentThread().getName() + catchers = self._catchers + if not catchers.has_key(name): + return self._defaultfunc(name, size) + else: + catcher = catchers[name] + return catcher.read(size) + + def _readdefault(self, name, size): + self._default.read(size) + + def _readfactory(self, name, size): + self._factory(name).read(size) + + def _readparam(self, name, size): + self._paramreader(name, size) + + def _readerror(self, name, size): + assert False, ( + "There is no StdinCatcher output stream for the thread %r" + % name) + + def register(self, catcher, name=None, + currentThread=threading.currentThread): + if name is None: + name = currentThread().getName() + self._catchers[name] = catcher + + def deregister(self, catcher, name=None, + currentThread=threading.currentThread): + if name is None: + name = currentThread().getName() + assert self._catchers.has_key(name), ( + "There is no StdinCatcher catcher for the thread %r" % name) + del self._catchers[name] + +_stdincatcher = None +_oldstdin = None + +def install_stdin(**kw): + global _stdincatcher, _oldstdin, register_stdin, deregister_stdin + if not _stdincatcher: + _oldstdin = sys.stdin + _stdincatcher = sys.stdin = StdinCatcher(**kw) + register_stdin = _stdincatcher.register + deregister_stdin = _stdincatcher.deregister + +def uninstall_stdin(): + global _stdincatcher, _oldstdin, register_stdin, deregister_stdin + if _stdincatcher: + sys.stdin = _oldstdin + _stdincatcher = _oldstdin = None + register_stdin = deregister_stdin = not_installed_error_stdin + +def not_installed_error_stdin(*args, **kw): + assert False, ( + "threadedprint has not yet been installed for stdin (call " + "threadedprint.install_stdin())") |