Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

update-pinnings fixes #667

Merged
merged 5 commits into from
Jul 2, 2020
Merged

update-pinnings fixes #667

merged 5 commits into from
Jul 2, 2020

Conversation

mbargull
Copy link
Member

@mbargull mbargull commented Jul 2, 2020

Some fixes for update-pinnings.

IMPORTANT: The change from https://github.com/bioconda/bioconda-utils/pull/662/files#diff-2aed786d397cbb062fb58d1ff86a6950R26 caused different build strings for r-base-dependent packages because it triggered the logic from conda/conda-build#3895. I hadn't thought of that before unfortunately.
Thankfully the last Bioconductor bulk work happened before that change. That means it should not affect terribly many recipes.

mbargull added 5 commits July 2, 2020 17:24
E.g., "--skip-variants python r_base" should skip recipes that build
different variants for python or r-base.
With bypass_env_check we get invalid build string like in
"r-dimsum=1.1.3=r>=36_0". conda-build special-cases python/r-base/...
deps but with bypass_env_check it cannot resolve the actual versions.
In most cases the recipe then needs manual editing anyway because it
should use build/skip rather than constrain versions in requirements.
@@ -541,6 +541,9 @@ def dag(recipe_folder, config, packages="*", format='gml', hide_singletons=False
help="""Bump package build numbers even if the only applicable pinning
change is the python version. This is generally required unless you plan
on building everything.""")
@arg('--skip-variants',
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I added this as a convenience things so we can update pinnings for Python/R packages separately.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

However, those skips happen after the dependency graph is created, ergo packages that are dependencies of others may not be bumped.


State = update_pinnings.State

for status, recip in zip(utils.parallel_iter(needs_bump, dag, "Processing..."), dag):
for status, recip in utils.parallel_iter(needs_bump, dag, "Processing..."):
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The order of utils.parallel_iter(...) and iter(dag) is different, apparently, which means we got "random" misassigned build bumps.

@mbargull
Copy link
Member Author

mbargull commented Jul 2, 2020

I'm going ahead with the merge and will tag yet another release so we get the r_base: [4.0.*] change live.
If there are any issues with the update-pinnings, we can address them later since it is just a "on-demand" command but does not affect daily business.

@mbargull mbargull merged commit 8205224 into master Jul 2, 2020
@mbargull mbargull deleted the update-pinnings-fixes branch July 2, 2020 19:20
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant