Repository navigation
[FIX] Reuse shared initialization in AI recruiter integration test - #2954
Merged
Nina Chikanov (nina-msft) merged 6 commits intoOct 3, 2026
Merged
Nina Chikanov (nina-msft) merged 6 commits into
Nina Chikanov (nina-msft) merged 6 commits into
Conversation
Filter incompatible seed groups before sampling while preserving the default technique set and selected objective text. Keep notebook sources synchronized and cover control characters, Unicode, and mixed groups. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 7d65728c-a1fd-4ace-8f7e-a337353ff787
Rely on the existing notebook integration test, as requested in review. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 7d65728c-a1fd-4ace-8f7e-a337353ff787
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 7d65728c-a1fd-4ace-8f7e-a337353ff787
Remove the import-time persistent SQLite initialization that conflicts with the in-memory database established by integration conftest. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com> Copilot-Session: 7d65728c-a1fd-4ace-8f7e-a337353ff787
Nina Chikanov (nina-msft)
marked this pull request as ready for review
October 2, 2026 20:16
Roman Lutz (romanlutz)
approved these changes
Oct 2, 2026
Nina Chikanov (nina-msft)
enabled auto-merge
October 2, 2026 20:23
Nina Chikanov (nina-msft)
deleted the
nina-msft-azure-sql-integration-fix
branch
October 3, 2026 00:13
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
AI recruiter integration tests fail during collection because their module requests persistent SQLite after the shared integration
conftest.pyhas initialized in-memory SQLite. Remove the redundant module-level initialization and its unused imports so the test uses the shared setup.Why this started failing
The change that exposed this failure merged on September 30, 2026, at 6:57 PM Pacific, in #2832 ("Moving Memory Functions to Async").
The conflicting initialization calls were already present. Previously, SQLite's singleton silently reused the existing in-memory database despite the recruiter test requesting persistent SQLite. The new validation explicitly rejects that mismatch.
This is an older test-setup inconsistency exposed by that change, not a new environment-file conflict. The first pipeline run that encountered it has not been confirmed. No pipeline, environment-file, secret, or production initialization changes are needed.
Database setup after this fix
The test continues to use in-memory SQLite, initialized by
tests/integration/conftest.pybefore pytest imports the recruiter test:That creates and initializes
SQLiteMemory(db_path=":memory:"), then registers it withCentralMemory. The recruiter test's PyRIT targets, scorers, and workflow use that shared instance.This removes only the second, conflicting initialization, not database setup altogether. PyRIT's database contents are temporary for that pytest process, not saved to persistent SQLite or Azure SQL. The recruiter application's own Docker-side storage is unaffected.
No breaking changes.
Tests and Documentation
.venv\Scripts\python.exe -m pytest tests\integration\ai_recruiter --collect-only --doctest-modules -q --tb=short: reproduced the exact SQLite-switch error before the fix; passed with 1 test collected after the fix. The existing unregisteredintegrationmarker warning remains..venv\Scripts\python.exe -m ruff check --fix tests\integration\ai_recruiter\test_ai_recruiter.py: passed, correcting one blank line after the removal..venv\Scripts\python.exe -m ruff format --check tests\integration\ai_recruiter\test_ai_recruiter.py: passed.git diff --check: passed.N/A: no documentation or notebook changes.