2018-06-12 conda-forge meeting¶
2018-05-29 conda-forge meeting Pinned Items
Zoom instructions: +How to connect to zoom
Actionable things to check in at the June 12 meeting
add something to the bot to add new PRs that manage the cb2 → cb3
e.g., if a compiler is detected, add the right compilers for the right files (Justin, nominally — @Christopher W @Mike S and @Anthony S will probably be reviewing those PRs) See issue: https://github.com/regro/cf-scripts/issues/162
@Mike S is working on a secondary channel to push the rebuilt cb3 packages
Finding a good solution to sharing passwords among core
Run_exports etc: https://github.com/conda-forge/conda-forge-pinning-feedstock/pull/58
Michael Sarahan to document and give demo on how to transition old recipes to new style - https://github.com/conda-forge/AnacondaRecipesSync - 5/1: Will wait for more people on the dev call before Mike talks about this
Build packages on C3I and upload to conda-forge
Make is missing from the base image for PowerPC internal to Anaconda. Fun times!
Mike is open to other people helping with this. If interested, reach out! Helping means trying recipes, debugging any issues, and resolving any merge conflicts that have happened since Mike pulled them in last. Moving target.
Packages that have been built https://anaconda.org/cf-cb3 - these may need more work regarding versions. The graph was computed with the versions, but probably should have ignored them. When a pin is older than a newer recipe, the upstream recipe gets missed as a real dependency because of the version mismatch.
switch to cb3
what kind of things are going to break when we change compilers?
(@Jonathan H had a great response that I hope he’ll clarify here:) tl;dr C++ ABI
Some details on libstdc++ dual ABI, https://gcc.gnu.org/onlinedocs/libstdc++/manual/using_dual_abi.html
Sprint in NYC June 18th through 20th for REST API for conda-forge graph and better inspection of CLI/imports/includes for conda-forge packages.
Making the agenda and notes public again.
John will see if we can make dropbox paper readable by the world
other options are to just post the notes somewhere public after the meeting
discuss numpy pinning (https://github.com/conda-forge/conda-forge-pinning-feedstock/pull/44)
Michael working on different scheme. numpy-base has all files. numpy is metapackage that implies only python API usage. numpy-devel is metapackage that implies C API linkage, and imposes run_exports
Governance, CoD, and NumFOCUS affiliation.
PyPA discussions at PyCON
Some general agreement by PyPA that pip/wheels shouldn’t be used for everything
No clear delineation as to where pip/wheels should stop and other tools begin
numfocus affiliation: https://github.com/numfocus/projects-director/blob/master/projects/Matplotlib.md
Scopatz said he was interested in pushing this forward :)
Idea of “emeritus” maintainers — basically, if someone wants to step away from the project then we can still call them out as having been core members in the past but are no longer contributing. Switching between emeritus and active is as simple as the emeritus member beginning to contribute again. This has not been an issue yet, but if we do fail to reach a vote threshold as outlined in the governance document then we will revisit the idea of emeritus maintainers