Repository navigation
Fix the two failing hypothesis stateful tests - #2349
Merged
Merged
Conversation
These errors were found by the nightly upstream-dev job. zarr >= 3.3's `add_array` recreates the drawn array in the store under test without the sharding codec the model store got, so an all-fill-value write elides a chunk object on one side only. Override it with the fix from zarr-developers/zarr-python#4288 until that reaches a release. `check_file_invariants` required orphan tx logs to equal the pruned-ancestor refs of every snapshot whose *file* is on disk. This is no longer true now that we keep expired tx logs.
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #2349 +/- ##
=======================================
Coverage 84.82% 84.82%
=======================================
Files 88 88
Lines 39444 39444
=======================================
Hits 33460 33460
Misses 5984 5984
Flags with carried forward coverage won't be shown. Click here to find out more. 🚀 New features to boost your workflow:
|
li-em
approved these changes
Aug 26, 2026
| return | ||
| fi | ||
| gh issue comment "$number" --body \ | ||
| "Failed again in tonight's run: $RUN_URL — the issue body has the current failures." |
This was referenced Aug 26, 2026
aldenks
pushed a commit
to aldenks/icechunk
that referenced
this pull request
Sep 15, 2026
N-Clerkx
pushed a commit
to wolkwork/icechunk
that referenced
this pull request
Sep 24, 2026
Follow-up on earth-mover#2349, also detected in nightly runs There are two fixes in this PR: - a test case where `reset_branch` revived commits after expiration ran, and model predicted one extra deletion. ``` state.initialize(spec_version=1) state.set_and_commit(ncommits=7) state.reset_branch(branch=branches_0, commit=commits_5) state.expire_snapshots(delete_expired_branches=True, delete_expired_tags=True) state.reset_branch(branch=branches_0, commit=commits_6) state.garbage_collect() ``` - an upstream zarr bug, will open a PR there: ``` a = zarr.create_array({}, shape=(4, 4), chunks=(2, 2), shards=(2, 4), dtype="int64") a.oindex[np.array([0, 1]), np.array([2, 3])] = np.zeros((2, 2), dtype="int64") # ValueError: shape mismatch: value array of shape (2,2) could not be # broadcast to indexing result of shape (4,) ``` There is a probe like `ZARR_RECREATES_ARRAYS` in earth-mover#2349 that can be dropped once it is fixed upstream.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
These errors were found by the nightly upstream-dev job.
zarr >= 3.3's
add_arrayrecreates the drawn array in the store under test without the sharding codec the model store got, so an all-fill-value write elides a chunk object on one side only. Override it with the fix from zarr-developers/zarr-python#4288 until that reaches a release.check_file_invariantsrequired orphan tx logs to equal the pruned-ancestor refs of every snapshot whose file is on disk. This is no longer true now that we keep expired tx logs.