ISTM that both indicate an issue that should be closed unless more information that prove its validity are provided. Something like this is definitely useful to have, but perhaps we don’t need to distinguish between the two.
On bpo we currently have test needed and pending. There is some overlap between them and with the labels you suggest. I think pending could be mapped, but probably not test needed.
FTR Black and PyPI have S: awaiting response and S: needs repro, Jupyter has status:Needs info.
I think it would make sense to automatically close after a few days issues that have been marked with the aforementioned labels by a human, since they are not useful in their current state. I wouldn’t close old/stale issues just because they are old.
I would consider renaming type-compile-error to type-build in order to encompass all build related errors: configure not functioning as intended, makefile bugs, problems with build related tools (freeze, etc.), and of course compilation errors.
Note that we already have the Build and the Cross-Build components, and I was already planning to merge them. Should we also add type-compile-error and combine all three into type-build (or just build)?
Since renaming and removing labels can be easily done at any time after the migration, I decided to map most of the fields to labels during the migration, and we can then update the labels after the migration. Colors and descriptions can be updated too.
Merging label can technically be done after the migration too, by selecting all issues with label B, adding label A to all of them, and removing label B. Doing this after the migration however will generate two new events on each issue (addition of A and removal of B), and will update their “Last updated” date, making some searches more difficult. Because of this, doing it before the migration is better – the downside is that if we change our minds on a merge, it’s tricky to undo it.
Based on the feedback I received, I changed the following things:
Removed type-compile-error and type-performance and added a build and performance labels that can be combined with the other type-* labels
Temporarily renamed docs and tests to type-documentation and type-tests to match the existing labels in the CPython repo (these should be renamed)
Renamed type-enhancement to type-feature
Added labels for 3.7-3.11 (these can be removed afterwards)
Added release-blocker and deferred-blocker labels (these can also be removed)
Updated pending to map to pending instead of stale
Added labels for most components
Regarding the components, this is the full mapping:
Library (Lib)-> library
Interpreter Core-> interpreter-core
Extension Modules-> extension-modules
C API-> expert-C-API
2to3 (2.x to 3.x conversion tool)-> expert-2to3
Cross-Build-> build (easily searchable, not too useful)
Demos and Tools-> `` (only a few issues, not too useful)
Regular Expressions-> expert-regex
Argument Clinic-> expert-argument-clinic
FreeBSD-> `` (only a few issues, easily searchable)
Parser-> interpreter-core (only a few issues, easily searchable)
Distutils-> library (only a few issues, easily searchable)
FreeBSD and Demos and Tools have no corresponding labels, Cross-build and Build have been merged into build, Distutils has been included into library, Parser into interpreter-core.
I’ve already renamed some of the labels on the python/cpython repo:
type-bugfix → type-bug (this will replace type-behavior too)
type-enhancement → type-feature
type-documentation → docs
type-performance → performance
type-tests → tests
I also noticed that the python/cpython already has some stage-like labels: awaiting change review, awaiting changes, awaiting core review, awaiting merge, awaiting review. bpo has these stages test needed, needs patch, patch review, commit review, backport needed, resolved.
Should we map patch review and commit review to awaiting review?
The awaiting * labels don’t seem to be documented, so I’m not entirely sure what mapping (if any) would make more sense.
The reason I was suggesting this mapping is that patch review/commit review are currently used on bpo to indicate issues that have a proposed solution (either a PR or a patch) ready to be reviewed. If we add the awaiting review, it will be possible to filter for these issues with is:issue is:open label:"awaiting review".
In theory this should be possible by searching for is:issue is:open linked:pr, but this has two problems:
PR linking is not supported during the migration (I’m still looking for a solution)
Issues with a patch on bpo won’t be included in the search without a label
Moving forward it won’t be necessary to add this label to issues, since PRs will be linked properly, so it’s mostly a way to preserve some extra metadata of the migrated issues. If we decide against it, it will still be possible to search for issues awaiting for review on bpo even though the list will become outdated as more issues gets closed on GitHub.
commit review: A triager performed a patch review and it looks good. This signals to core developers the patch or pull request needs a quick once-over to make sure nothing was overlooked before committing it.
Either way there are only 21 issues with commit review, so it doesn’t make much difference if they are not included. There are 2199 issues with patch review and also 2869 with the patch keyword (added automatically for PRs).