-
-
Notifications
You must be signed in to change notification settings - Fork 438
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
.ci/merge-fixes.sh
: Obtain patches via URL, make customizable by repository variable
#36686
.ci/merge-fixes.sh
: Obtain patches via URL, make customizable by repository variable
#36686
Conversation
d259cac
to
b2925cb
Compare
b2925cb
to
89a1901
Compare
03691bc
to
39a76d3
Compare
39a76d3
to
35cdbd0
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM in general.
.ci/merge-fixes.sh
Outdated
@@ -2,7 +2,9 @@ | |||
# Merge open PRs from sagemath/sage labeled "blocker". | |||
REPO="sagemath/sage" | |||
GH="gh -R $REPO" | |||
PRs="$($GH pr list --label "p: blocker / 1" --json number --jq '.[].number')" | |||
mkdir -p upstream | |||
PRs="$($GH pr list --label "p: blocker / 1" --json number --jq '.[].number' | tee upstream/ci-fixes.txt)" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why not add --label "s: positive review"
? Multiple --label
arguments work just fine.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We can do that later when the label-warring phase has been outgrown
LGTM, but how is it tested? |
So far, only on its own CI run... |
320f0a9
to
23114e6
Compare
23114e6
to
28c86cb
Compare
.ci/merge-fixes.sh
: Obtain patches via URL.ci/merge-fixes.sh
: Obtain patches via URL, make customizable by repository secret
144a315
to
a886324
Compare
.ci/merge-fixes.sh
: Obtain patches via URL, make customizable by repository secret.ci/merge-fixes.sh
: Obtain patches via URL, make customizable by repository variable
a886324
to
01d9b73
Compare
I've reworked and generalized this a bit. Ready for review. |
Documentation preview for this PR (built with commit 49b7f74; changes) is ready! 🎉 |
See it in action here: https://github.com/kwankyu/sage/actions/runs/6887591924/job/18735031877 It seems working well. You can even watch it recovering from applying-patch failure. Comparing with the original (1) It is just slightly faster. (2) It fails in applying the blocker #36617 but the original |
Thanks for testing!
Note it is much faster in runs where patches apply directly
I think this failure would go away if we rebased #36617. |
Right. It was slow this time because of #36617 that failed to be applied directly.
Failing in applying non-rebased blockers may be a good thing, by the way. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM too.
I think so, too. It sort of enforces some restraint on what to do in these CI fixes. |
Thanks both for the review! |
…customizable by repository variable <!-- ^^^^^ Please provide a concise, informative and self-explanatory title. Don't put issue numbers in there, do this in the PR body below. For example, instead of "Fixes sagemath#1234" use "Introduce new method to calculate 1+1" --> <!-- Describe your changes here in detail --> Instead of using `gh pr checkout`, we obtain the CI fixes via their patch URLs. - This is faster because typically we do not have to unshallow the repo to apply the patches (seconds instead of ~2 minutes) - Fewer surprises when applied to a PR based on an older release - Conjecturally uses fewer API queries, helping avoid sagemath#36685 When a repository variable `SAGE_CI_FIXES_FROM_REPOSITORIES` is set in a fork, it is used instead of the hardcoded sagemath/sage as the source(s) of the CI fixes; this gives better control in decentralized development. When set to "none", this also makes it possible to see the "ground truth", addressing a concern raised in sagemath#36349 (comment). See comments at the top of `.ci/merge-fixes.sh` for details. Also improving the display of details of the merged PRs, as requested by @tornaria in sagemath#36696 (comment) Example runs: - https://github.com/sagemath/sage/actions/runs/6854931832/job/186389018 59?pr=36686 (here it recovers gracefully from failed merges due to the 10.2.rc1/10.2.rc2 tagging confusions) - https://github.com/mkoeppe/sage/actions/runs/6855410957/job/1864038039 8#step:3:8 (I have set `SAGE_CI_FIXES_FROM_REPOSITORIES=none` in my fork -- for testing this PR) <!-- Why is this change required? What problem does it solve? --> <!-- If this PR resolves an open issue, please link to it here. For example "Fixes sagemath#12345". --> <!-- If your change requires a documentation PR, please link it appropriately. --> ### 📝 Checklist <!-- Put an `x` in all the boxes that apply. --> <!-- If your change requires a documentation PR, please link it appropriately --> <!-- If you're unsure about any of these, don't hesitate to ask. We're here to help! --> <!-- Feel free to remove irrelevant items. --> - [x] The title is concise, informative, and self-explanatory. - [x] The description explains in detail what this PR is about. - [x] I have linked a relevant issue or discussion. - [ ] I have created tests covering the changes. - [ ] I have updated the documentation accordingly. ### ⌛ Dependencies <!-- List all open PRs that this PR logically depends on - sagemath#12345: short description why this is a dependency - sagemath#34567: ... --> <!-- If you're unsure about any of these, don't hesitate to ask. We're here to help! --> URL: sagemath#36686 Reported by: Matthias Köppe Reviewer(s): Gonzalo Tornaría, Kwankyu Lee, Matthias Köppe
It fails to apply #36372 now (and sometimes the jmol PR as well), although this PR is up-to-date with the newest dev branch. It's seems like an unnecessary constraint that blocker PR's need to be rebased (instead of merging the latest develop). Would be nice if this could be fixed as soon as possible. |
Instead of using
gh pr checkout
, we obtain the CI fixes via their patch URLs..ci/merge-fixes.sh
: GraphQL: API rate limit exceeded #36685When a repository variable
SAGE_CI_FIXES_FROM_REPOSITORIES
is set in a fork, it is used instead of the hardcoded sagemath/sage as the source(s) of the CI fixes; this gives better control in decentralized development. When set to "none", this also makes it possible to see the "ground truth", addressing a concern raised in #36349 (comment). See comments at the top of.ci/merge-fixes.sh
for details.Also improving the display of details of the merged PRs, as requested by @tornaria in #36696 (comment)
Example runs:
SAGE_CI_FIXES_FROM_REPOSITORIES=none
in my fork -- for testing this PR)📝 Checklist
⌛ Dependencies