Skip to content

Scipy 2013 sprinting notes

Michael Droettboom edited this page · 6 revisions

The matplotlib sprint during Scipy 2013 was a major success.

We had 15 participants, ranging in experience working on matplotlib from "never submitted a pull request" up to long-time core contributors.

We helped new contributors get their development environments set up to start making pull requests to the project. Given the obvious talent of everyone involved, this will certainly lead to many valuable contributions in the future. Unfortunately, this is still harder than it needs to be -- we should keep working on making setting up a matplotlib development environment easier.

22 issues and pull requests were created or worked on during the sprint, summarized below:

  • @dmcdougall worked on functional plotting. He also created a warning when horizontal alignment is used with anchored text #2166. He was also very helpful getting new Mac users set up with their development environments.

  • @NelleV changed thousands of lines of code by refactoring the axes.py module (~8000 lines) into a package. #1931, #2182

  • @oscareli98 discovered a few bugs and inconsistencies and helped track them down and make tests. The "short" (single letter) color names are not consistent with the long ones #2164. When making ticks really thick, in 1.2.0, they had an obvious "slant".

  • @kthyng fixed a bug so that contour colors are scaled so the colorbar matches #2176

  • @gabraganca, @leouieda and @treethinker helped reorganize some of our examples #2181, #2160

  • @mrterry pointed out some shortcomings in the docs #2171, fixed a bug where table kwargs were not getting forwarded

  • @WeatherGod worked on a number of mplot3d-related bugs. #2169, #2172, #2162

  • @alsaihn reorganized our list of external resources in the docs

  • @tbekolay added text to the gallery images in the docs. This makes it easier to search in the gallery, for example. He also worked on increasing test coverage in the toolkits, particularly mplot3d #2177, #2163

  • @katyhuff cleaned out some cruft #2168, and worked on cleaning up the rcParam template #2193.

  • David (github id?) worked on multi-colored lines, with an aim of creating a convenient interface for them.

  • Cory (github id?) was interested in plotting many rectangles very quickly.

  • @deeplycloudy worked on syncing of mutlidimensional data across arbitrary axes sets, with evented data processing pipeline. See https://github.com/deeplycloudy/stormdrain/, moderately well-commented example in ./examples.

  • @pelson was around helping others and having productive discussions that should grow into good MEPs.

  • As for myself, @mdboom, I managed to open a few issues related to discussions had with various participants:

    • It would be nice to be able to test without installing first #2175

    • Updating coding guidelines wrt. documentation #2173

    • Scatter plots are too slow #2156

There were also offline discussions of the following:

  • How is retina display support going? (Seems to work on OSX backend, not sure about others)

  • GUI events don't seem to fire with Anaconda -- perhaps because it's not a framework build? #2120

Something went wrong with that request. Please try again.