Merged
Conversation
Why: Relative directory-prefixed glob filters (e.g. apps/web/**/*.tsx) were matched directly against absolute stored file paths, causing silent zero-result searches. What: - Added resolve_file_filter_pattern utility to normalize relative prefixed filters to absolute patterns under the indexed root. - Applied normalization in MCP, HTTP, and CLI search entrypoints before calling SearchService. - Added unit tests covering wildcard-only, relative-prefixed, and absolute filter patterns. Test: - uv run ruff check src tests (pass) - uv run pytest tests/unit/test_runtime_path_resolution.py -q -v --tb=short (pass) - uv run pytest tests/ -v --tb=short -q --durations=10 (fails in existing unrelated tests) - uv run mypy src --ignore-missing-imports --no-error-summary (fails with existing repository typing errors)
…lative-paths fix(search): resolve relative file_filter patterns against indexed root
Why: Index/search/update resolved .aci/index.db from CWD, which breaks cross-directory CLI usage and causes 'Path has not been indexed' errors. What: Added a project-scoped metadata DB path helper and wired index/search/update to initialize services with <repo>/.aci/index.db instead of a CWD-relative default. Added unit tests to verify the scoped path behavior in CLI flows. Test: uv run ruff check src tests (pass) Test: uv run pytest tests/unit/test_cli.py tests/unit/test_cli_metadata_db_path.py -q (pass) Test: uv run pytest tests/ -v --tb=short -q --durations=10 (interrupted; suite showed pre-existing failures while running) Test: uv run mypy src --ignore-missing-imports --no-error-summary (fails with many pre-existing typing issues)
…tion-path fix(cli): scope metadata db path to target repository
…e tests Why: Tests using TiktokenTokenizer fail when network is unavailable (tiktoken downloads its vocabulary from openaipublic.blob.core.windows.net on first use). This caused 14 unit test failures in sandboxed/offline environments. The ACI_TOKENIZER env var was also undocumented, leaving Ollama users with no guidance on how to resolve the BPE/WordPiece mismatch. What: - .env.example: add ACI_TOKENIZER with documentation for Ollama/BERT users - test_chunker_core.py: use CharacterTokenizer fixture (network-free) - test_chunker_splitter.py: use CharacterTokenizer fixture (network-free) - test_incremental_update_scopes_to_root.py: pass CharacterTokenizer-based chunker to IndexingService to avoid tiktoken download at test time Test: uv run pytest tests/unit/ -> 226 passed, 0 failed Co-authored-by: AperturePlus <146049978+AperturePlus@users.noreply.github.com>
fix(tokenizer): document ACI_TOKENIZER and fix offline test failures caused by tiktoken network dependency
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.
fix #14
fix #19
fix #20