A guide for developers who are doing a matplotlib release
- Edit :file:`__init__.py` and bump the version number
When doing a release
- Make sure :file:`examples/tests/backend_driver.py` runs without errors and check the output of the PNG, PDF, PS and SVG backends
- Run :file:`unit/memleak_hawaii3.py` and make sure there are no memory leaks
- Run :file:`unit/nose_tests.py` and make sure all the unit tests are passing
- try some GUI examples, eg :file:`simple_plot.py` with GTKAgg, TkAgg, etc...
- remove font cache and tex cache from :file:`.matplotlib` and test with and without cache on some example script
Once all the tests are passing and you are ready to do a release, you need to create a release branch:
git checkout -b v1.1.x git push [email protected]:matplotlib/matplotlib.git v1.1.x
On the branch, do any additional testing you want to do, and then build binaries and source distributions for testing as release candidates.
- Make sure the :file:`MANIFEST.in` us up to date and remove :file:`MANIFEST` so it will be rebuilt by MANIFEST.in
- run git clean in the mpl git directory before building the sdist
- unpack the sdist and make sure you can build from that directory
- Use :file:`setup.cfg` to set the default backends. For windows and OSX, the default backend should be TkAgg. You should also turn on or off any platform specific build options you need. Importantly, you also need to make sure that you delete the :file:`build` dir after any changes to file:setup.cfg before rebuilding since cruft in the :file:`build` dir can get carried along.
- on windows, unix2dos the rc file
- We have a Makefile for the OS X builds in the mpl source dir :file:`release/osx`, so use this to prepare the OS X releases.
- We have a Makefile for the win32 mingw builds in the mpl source dir :file:`release/win32` which you can use this to prepare the windows releases, but this is currently broken for python2.6 as described at http://www.nabble.com/binary-installers-for-python2.6--libpng-segfault%2C-MSVCR90.DLL-and-%09mingw-td23971661.html
Post the release candidates to http://matplotlib.sf.net/release-candidates and post a message to matplotlib-users and devel requesting testing. To post to the server, you can do:
> scp somefile.tgz jdh2358,[email protected]:/home/groups/m/ma/matplotlib/htdocs/release-candidates/
replacing 'jdh2358' with your sourceforge login.
Any changes to fix bugs in the release candidate should be fixed in the release branch and merged into the trunk.
Post the win32 and OS-X binaries for testing and make a request on matplotlib-devel for testing. Pester us if we don't respond
ftp the source and binaries to the anonymous FTP site:
mpl> git clean mpl> python setup.py sdist mpl> cd dist/ dist> sftp [email protected] Connecting to frs.sourceforge.net... sftp> cd uploads sftp> ls sftp> lls matplotlib-0.98.2.tar.gz sftp> put matplotlib-0.98.2.tar.gz Uploading matplotlib-0.98.2.tar.gz to /incoming/j/jd/jdh2358/uploads/matplotlib-0.98.2.tar.gz
go https://sourceforge.net/project/admin/explorer.php?group_id=80706 and do a file release. Click on the "Admin" tab to log in as an admin, and then the "File Releases" tab. Go to the bottom and click "add release" and enter the package name but not the version number in the "Package Name" box. You will then be prompted for the "New release name" at which point you can add the version number, eg somepackage-0.1 and click "Create this release".
You will then be taken to a fairly self explanatory page where you can enter the Change notes, the release notes, and select which packages from the incoming ftp archive you want to include in this release. For each binary, you will need to select the platform and file type, and when you are done you click on the "notify users who are monitoring this package link"
Announce the release on matplotlib-announce, matplotlib-users and matplotlib-devel. Include a summary of highlights from the CHANGELOG and/or post the whole CHANGELOG since the last release.