Add graph expansion and cross-encoder reranking features

- Implemented GraphExpander to enhance search results with related symbols using precomputed neighbors.
- Added CrossEncoderReranker for second-stage search ranking, allowing for improved result scoring.
- Created migrations to establish necessary database tables for relationships and graph neighbors.
- Developed tests for graph expansion functionality, ensuring related results are populated correctly.
- Enhanced performance benchmarks for cross-encoder reranking latency and graph expansion overhead.
- Updated schema cleanup tests to reflect changes in versioning and deprecated fields.
- Added new test cases for Treesitter parser to validate relationship extraction with alias resolution.
This commit is contained in:
catlog22
2025-12-31 16:58:59 +08:00
parent 4bde13e83a
commit 31a45f1f30
27 changed files with 2566 additions and 97 deletions

View File

@@ -4,6 +4,11 @@ import sqlite3
from pathlib import Path
from typing import List, Dict, Any, Optional
from codexlens.config import Config
from codexlens.entities import SearchResult
from codexlens.search.graph_expander import GraphExpander
from codexlens.storage.path_mapper import PathMapper
class RelationshipEnricher:
"""Enriches search results with code graph relationships."""
@@ -148,3 +153,19 @@ class RelationshipEnricher:
def __exit__(self, exc_type, exc_val, exc_tb) -> None:
self.close()
class SearchEnrichmentPipeline:
"""Search post-processing pipeline (optional enrichments)."""
def __init__(self, mapper: PathMapper, *, config: Optional[Config] = None) -> None:
self._config = config
self._graph_expander = GraphExpander(mapper, config=config)
def expand_related_results(self, results: List[SearchResult]) -> List[SearchResult]:
"""Expand base results with related symbols when enabled in config."""
if self._config is None or not getattr(self._config, "enable_graph_expansion", False):
return []
depth = int(getattr(self._config, "graph_expansion_depth", 2) or 2)
return self._graph_expander.expand(results, depth=depth)