- Added CHANGELOG.md
- Avoid ternary operator in call to ParserCreate().
- Adding Python 3.4 to Tox test environment.
- Added full_document flag to unparse (default=True).
- Merge pull request #56 from HansWeltar/master
- Improve performance for large files
- Updated README unparse example with pretty=True.
- Fixed extra newlines in pretty print mode.
- Fixed all flake8 warnings.
- Added Tox config.
- Let expat figure out the doc encoding.
- Fixed Jython TravisCI build.
- Moved nose and coverage to tests_require.
- Dropping python 2.5 from travis.yml.
- Use system setuptools if available.
- Switch to latest setuptools.
- Include distribute_setup.py in MANIFEST.in
- Updated package classifiers (python versions, PyPy, Jython).
- Merge pull request #40 from martinblech/jython-support
- Adding Jython support.
- Fix streaming example callback (must return True)
- Merge pull request #35 from martinblech/namespace-support
- Adding support for XML namespaces.
- Merge pull request #33 from bgilb/master
- fixes whitespace style
- changes module import syntax and assertRaises
- adds unittest assertRaises
- Merge pull request #31 from martinblech/document-unparse
- Adding documentation for unparse()
- Merge pull request #30 from martinblech/prettyprint
- Adding support for pretty print in unparse()
- Merge pull request #29 from dusual/master
- ordereddict import for less 2.6 if available
- Allow using alternate versions of
expat. - Added shameless link to GitTip.
- Merge pull request #20 from kevbo/master
- Adds unparse example to README
- fix try/catch block for pypi (throws AttributeError instead of TypeError)
- prevent encoding an already encoded string
- removed unecessary try/catch for xml_input.encode(). check if file or string, EAFP style. (thanks @turicas)
- test with python 3.3 too
- avoid u'unicode' syntax (fails in python 3.2)
- handle unicode input strings properly
- add strip_whitespace option (default=True)
- Merge pull request #16 from slestak/master
- fix unittest
- working with upstream to improve #15
- remove pythonpath tweaks, change loc of #15 patch
- upstream #15
- test attribute order roundtrip only if OrderedDict is available (python >= 2.7)
- Merge branch 'master' of github.com:martinblech/xmltodict
- preserve xml attribute order (fixes #13)
- fix #12: postprocess cdata items too
- added info about official fedora package
- Merge pull request #11 from ralphbean/master
- Include REAMDE, LICENSE, and tests in the distributed tarball.
- take all characters (no need to strip and filter)
- fixed CLI (marshal only takes dict, not OrderedDict)
- ignore MANIFEST
- #8 preprocessing callback in unparse()
- implemented postprocessor callback (#6)
- update readme with install instructions
- link to travis-ci build status
- more complete info in setup.py (for uploading to PyPi)
- coverage annotations for tricky py3k workarounds
- py3k compatibility
- removed unused future print_function
- using io.StringIO on py3k
- removed unnecessary exception catching
- initial travis-ci configuration
- made _emit function private
- unparse functionality
- added tests
- updated (c) notice to acknowledge individual contributors
- added license information
- fixed README
- removed temp build directory and added a .gitignore to avoid that happening again
- Merge pull request #1 from scottscott/master
- Added setup script to make xmltodict a Python module.
- fixed bad handling of cdata in semistructured xml, changed CDATA to #text as default
- added attr_prefix, cdata_key and force_cdata parameters
- links in README
- links in README
- improved README
- initial commit