Repository navigation
tests: evidence-report tests no longer depend on the committed reports' counts (Part of #4863) - #4869
Merged
Conversation
…s' counts (Part of #4863) The stale-level and Db2-ledger tests build their fixtures from the committed reports and asserted counts of those reports: exactly one CardDemo program at L3, and lgacus01 / LGSTSQ reached only through the genapp-lgacus01 case. Once the evidence refresh re-measures the translation (#4864), the reports move (COCRDLIC / COMEN01C / CORPT00C at L3; genapp-lgtestc1 LINKs LGACUS01 with a ledger entry), so the bot's refresh PR #4867 failed them. The stale test now asserts only CASE's stale level; the two Db2 fixtures can isolate DB2_CASE (every other ledger entry dropped). Passes on main's reports and on #4867's. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MhEx8PwSow3Pyfg6J4KZ97
squid-protocol
added
bug
Owner
Author
|
No description provided. |
squid-protocol
marked this pull request as ready for review
October 11, 2026 01:13
squid-protocol
enabled auto-merge (squash)
October 11, 2026 01:13
Owner
Author
|
No description provided. |
Contributor
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Part of #4863. Unblocks the bot's evidence refresh PR #4867.
What broke. #4864 made the nightly refresh re-measure translation, so its PR #4867 carries the reports as they really are today. Three tests in
test_evidence_report.pybuild their fixtures from the committed reports and asserted facts about the old data:test_a_harness_only_change_keeps_the_last_level_marked_staleasserted exactly 1 CardDemo program at L3. Today COCRDLIC, COMEN01C and CORPT00C are L3 too.test_a_db2_case_with_no_sweep_and_no_ledger_entry_is_not_runassumed LGACUS01 is reached only by its own case. Since 10/09 the front-end casegenapp-lgtestc1LINKs it and has a ledger entry, so it is honestly L2.test_a_linked_program_never_inherits_…assumedgenapp-lgacus01is LGSTSQ's chosen LINKing case. It was only chosen because the other GenApp Db2 cases had no ledger entries yet.Fix (tests only).
histogram_stale/levels_stale= 1 and theL3+: N (1 awaiting re-check)line).db2_fixture(..., isolate=True)drops every other case's ledger entry, so DB2_CASE is the only Db2 case with a verdict, as those two tests intend.Checked: the file passes on main's reports (58 passed, 1 skipped) and on #4867's reports (61 passed, 1 skipped; the skip is the advisory currentness check).
pr_gates.py --fast5/5.After merge, #4867 needs its branch updated (the shepherd does it for a behind, green PR, or
gh pr update-branch 4867).🤖 Generated with Claude Code
https://claude.ai/code/session_01MhEx8PwSow3Pyfg6J4KZ97