summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* [debian] Account for doc directory renamedebian/1.4.3-11.4.3Denis Laxalde2019-07-262-2/+2
|
* [debian] Drop documentation buildDenis Laxalde2019-07-263-16/+2
| | | | | The system has changed and we lack resources to maintain it in Debian.
* [pkg] Version 1.4.3Denis Laxalde2019-07-262-1/+7
|
* [py3] Wrap dict.keys() in list()Denis Laxalde2019-07-261-1/+1
| | | | | | Otherwise, the error message looks like: invalid unit minutes. valid units are dict_keys(['d', 'h', 'ms', 's', 'min'])
* Raise a ValueError in case of invalid unit in textutils.apply_units()Denis Laxalde2019-07-262-2/+5
| | | | | | | | | The original KeyError comes from 57e242dffe89 which no justification. I believe it was meant to be a ValueError as in other errors in this function. Also, _ensure_correctly_typed() in cubicweb/web/formfields.py will catch ValueError. Adding a test for this.
* [tox] Add a "docs" environmentDenis Laxalde2019-07-091-0/+7
| | | | | We use python2 since building the documentation with python3 does not work at the moment.
* [doc] fix sphinx warningsLaurent Peuch2019-07-033-19/+36
|
* [mod] mock kerberos in sphinx to avoid installing itLaurent Peuch2019-06-202-1/+2
|
* [doc] rename doc/ into docs/Laurent Peuch2019-06-2010-0/+0
| | | | readthedocs likes documentation to be in a docS folder.
* [doc] add instructions on how to build the docLaurent Peuch2019-06-191-0/+23
|
* [doc] add instruction on how to install using pypiLaurent Peuch2019-06-191-1/+5
|
* [doc] include README into doc index.rst to avoid duplicationLaurent Peuch2019-06-192-187/+41
|
* [doc] point to the cubicweb-devel mailing list since python-projects seems deadLaurent Peuch2019-06-191-3/+3
|
* [doc] migrate documentation to sphinx using sphinx-apidoc and including READMELaurent Peuch2019-05-2311-8/+834
| | | | | | | | | | | | The rationnal to do this migration are: * be able to host the documentation on readthedocs.org, right now, according to the README, the user needs to built it locally to be able to read it * epydoc is not maintained anymore * sphinx generally offer a better user experience and allow to write real documentation instead of just generating an API * it was quick to do with sphinx-apidoc and seems worth it to be able to put it online
* Bump debian packagingPhilippe Pepiot2018-10-181-0/+6
|
* Added tag 1.4.2 for changeset 11c1e111da21Denis Laxalde2018-07-060-0/+0
|
* [pkg] Version 1.4.21.4.2Denis Laxalde2018-07-061-1/+1
|
* Use getfullargspec instead of getargspec in python 3Frank Bessou2018-07-061-2/+9
| | | | | | | In python 3 "getargspec" is deprecated and "getfullargspec" is added. The only difference between these two functions is that the "keywords" tuple's field is renamed into "varkw" (which does not impact us in this case).
* [tox] Use py3 environment instead of py34Sylvain Thénault2017-07-061-1/+1
| | | | The former would select any Python 3 interpreter.
* Added tag 1.4.1, debian/1.4.1-1, centos/1.4.1-1 for changeset 096358768ef9Denis Laxalde2017-07-110-0/+0
|
* [pkg] Version 1.4.1debian/1.4.1-1centos/1.4.1-11.4.1Denis Laxalde2017-07-113-2/+8
|
* [configuration] fix registration of multiple groups at onceAdrien Di Mascio2017-05-112-2/+18
| | | | | | | | The "options" iteration variable hid the original function parameters, therefore only the options of the last group were registered. Rename the iteration variable and everything's fine.
* Use a context manager for io.open()Denis Laxalde2017-07-042-5/+4
|
* Use io.open() to avoid deprecation warning about 'U' modeDenis Laxalde2017-07-042-2/+4
| | | | | | Use io.open() which has a consistent interface between Python 2 and Python 3. In particular, `newline` parameter is meant to control "universal new lines" and it is None by default so we don't need to specify it.
* configuration: fix action query in read_old_configDenis Laxalde2017-07-051-1/+1
| | | | | | From the context, using action[0] seems correct. The problem comes from changeset 0a22ff6e5207 and shows up when using "option_added()" CubicWeb migration command.
* Added tag debian/1.4.0-1, centos/1.4.0-1 for changeset 5d5770c6d3c8Denis Laxalde2017-03-200-0/+0
|
* Added tag 1.4.0 for changeset 5d5770c6d3c8Denis Laxalde2017-03-200-0/+0
|
* [pkg] Version 1.4.0debian/1.4.0-1centos/1.4.0-11.4.0Denis Laxalde2017-03-173-2/+8
|
* [textutils] Add two more manual mappings to convert unicode into ASCIIYann Voté2017-02-282-0/+4
|
* [registry] fix autoreload with register_modnames()Philippe Pepiot2017-01-251-0/+3
| | | | | Don't register bytecode files in '_lastmodifs'. With python 2 loader get_filename() seems to return either pyc or py file.
* [registry] RegistrableInstance should be instantiated with __module__=__name__Philippe Pepiot2017-01-203-6/+21
| | | | | | | | | To detect in whih module the instance was created we previously detect the filename in python stack and then use modpath_from_file(). Since the later is now deprecated, we should force passing the module at instantiation with __module__=__name__. Deprecate old usage
* [registry] deprecate register_objects()Philippe Pepiot2017-01-202-13/+11
| | | | | This method is to be replaced by register_modnames() to avoid using deprecated modpath_from_file()
* [modutils] deprecate modpath_from_filePhilippe Pepiot2017-01-202-3/+11
| | | | | | The method will likely fail if there somes symlinks in your python environment and will certainly fail on a file that isn't in sys.path (like custom importers using sys.meta_path).
* [registry] add register_modnames()Philippe Pepiot2016-12-232-0/+29
| | | | | | Alternative to register_objects() to inspect modules names instead of directories. Avoiding using bogus logilab.common.modutils.modpath_from_file()
* [modutils] add clean_sys_modules()Philippe Pepiot2016-12-231-0/+12
| | | | | A variant of cleanup_sys_modules using a list of modnames to remove from sys.modules.
* Added tag 1.3.0, debian/1.3.0-1, centos/1.3.0-1 for changeset bbc78dd79f6dDenis Laxalde2016-11-180-0/+0
|
* [debian] Use logilab-pytest executable directlydebian/1.3.0-1centos/1.3.0-11.3.0Denis Laxalde2016-11-181-2/+2
| | | | To please damned Apycot, hopefully.
* [test] Do not use "modutils" in test_site_packagesDenis Laxalde2016-11-181-2/+3
| | | | | | Because "modutils" module may also be found in source directory, this test may fail under some circumstances (like Debian package build, for instance).
* [pkg] Version 1.3.0Denis Laxalde2016-11-183-2/+8
|
* Fix test breakage in modutils with python 3Sylvain Thénault2016-11-161-1/+3
|
* [configuration] Fix multi-line strings handlingChristophe de Vienne2016-01-112-0/+47
| | | | Closes #3691182
* [testlib] Deprecate doctest things in favour of stdlib moduleDenis Laxalde2016-01-195-29/+23
| | | | | | And adjust all tests accordingly. Drop docstring example from fileutils.abspath_listdir since I could not have it skipped.
* Deprecate logilab-pytestDenis Laxalde2016-10-033-3/+4
|
* Rename pytest executable into logilab-pytestDenis Laxalde2016-10-0312-42/+48
| | | | Closes #8396846.
* [testlib] Deprecate with_tempdir (Closes #8267966)Denis Laxalde2016-10-241-1/+3
| | | | | | | Recommend tempfile.TemporaryDirectory, available in tempfile from Python 3.4 or from `backports.tempfile`_ otherwise. .. _`backports.tempfile`: https://pypi.python.org/pypi/backports.tempfile
* [modutils] Copy back modpath_from_file from astroidDenis Laxalde2016-11-141-9/+27
| | | | | | | Astroid has copy of modutils, originally from logilab-common, which has been modified and improved since. Copy back the `modpath_from_file` function in logilab-common, so that it is more robust to file path "subtelities" (case, symlinks, user, in particular).
* [shellutils] Typo slot -> slotsDenis Laxalde2016-06-301-1/+1
| | | | Closes #6820061.
* Added tag 1.2.2, debian/1.2.2-1, centos/1.2.2-1 for changeset da1d07661e4eDenis Laxalde2016-06-210-0/+0
|
* [pkg] Version 1.2.2debian/1.2.2-1centos/1.2.2-11.2.2Denis Laxalde2016-06-213-2/+8
|
* [doc] fixesDavid Douard2015-07-031-8/+11
|