Repository navigation
Fix nightly: hermetic stale-level evidence tests, ground-truth installs [full] - #4912
Merged
Merged
Conversation
…truth installs [full] (#4911) The stale-level tests overlaid CASE on the committed report, whose other cases are marked stale after a refresh; pin the others current. The Ground Truth workflow now installs the full extra that feature_inventory --check needs. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AUWeQX7v4QBdErFbghEEEX
Owner
Author
|
No description provided. |
squid-protocol
marked this pull request as ready for review
October 11, 2026 17:28
squid-protocol
enabled auto-merge (squash)
October 11, 2026 17:28
Contributor
This was referenced Oct 11, 2026
Evidence refresh: re-measure translation every night (the reports' survey froze on 2026-10-07)
#4863
Closed
squid-protocol
added a commit
that referenced
this pull request
Oct 11, 2026
…nframe Ground Truth nightly) (#4920) test_feature_inventory_cli_check (feature_inventory.py --all-burned --check) runs only in the nightly Mainframe Ground Truth job, which could not reach it until #4912 installed [full]; once it ran, two fixtures were stale. - CBSA: bisected to #4885 exactly (fixture current at its parent, stale at it). CRECUST went from refused (COPY CEEIGZCT) to parsed: the copybook gap is gone, the two "unknown" CICS commands (EXEC CICS text inside a DISPLAY literal, misread before #4885's literal fix) are gone, and its statement / data-type / SQL counts are now the real parse (IF 503->478, DISPLAY 156->145, ADD folded into ARITH, GET/WRITE +1). - DBB Mortgage: data_types EDITED 59 -> 58 (same three programs). Stale already before #4885, i.e. drift since the fixture's creation (#4821); not attributed to one commit. Claude-Session: https://claude.ai/code/session_01AUWeQX7v4QBdErFbghEEEX Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
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.
Closes #4911. Unblocks bot PR #4880.
(A) tests/tools/test_evidence_report.py stale_fixture overlaid CASE on the committed carddemo report, whose other cases the refresh marks stale after a harness change (levels_stale 3 not 1, etc.). The fixture now pins every other case current. The refresh data was consistent; the tests were coupled to it. Verified on a scratch checkout of auto/evidence-refresh-full: the 3 failing tests pass.
(B) mainframe-ground-truth.yml installs .[full] (#4911).
🤖 Generated with Claude Code
https://claude.ai/code/session_01AUWeQX7v4QBdErFbghEEEX