Repository navigation
feat(failure-intel): group failures by root cause in the report - #149
Merged
Merged
Conversation
Failed tests are grouped by a signature (category, exception type, normalised message, top application frame). The Failures tab shows "N tests, same cause" above the flat list, and AI analysis makes one call per group. Closes #140 Co-Authored-By: Claude Sonnet 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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What & why
Part of Failure Intelligence (#138), builds on the classifier (#148). Forty tests failing on one outage should read as one cause, not forty.
FailureSignature: category + exception type + normalised message (UUIDs, hex ids, numbers, quoted values stripped) + top application frame (class.method, no line number; JDK, Selenium, test runner and framework frames skipped).failureSignature.AiFailureAnalyzer: one API call per signature; same-cause tests reuse the first analysis, concurrent ones wait on it. A failed call is not cached, so a later test retries.Closes #140
Verification
mvn test: 638 run, 0 failures (newFailureSignatureTestandFailureGroupsReportTest).Not covered
The one-call-per-group path in
AiFailureAnalyzerhas no automated test: the API URL is a constant and there is no HTTP seam today. Reviewed by reading only.🤖 Generated with Claude Code