feat(qwen): rerank 50 candidates and return five

scripts/test-retrieval-server.py

@@ -205,6 +205,51 @@ class RetrievalServerTests(unittest.TestCase):
         self.assertEqual(MODULE.reciprocal_rank_fusion([1, 2, 3], [4, 2, 3]), [2, 3, 1, 4])
         self.assertEqual(MODULE.reciprocal_rank_fusion([], [4, 2]), [4, 2])
 
+    def test_hybrid_candidates_and_document_metadata(self):
+        with TemporaryDirectory() as temporary:
+            root = Path(temporary)
+            for number in range(110):
+                (root / f'{number:03}.md').write_text(f'# Symbol{number}\ncontent {number}', encoding='utf-8', newline='\n')
+            index = MODULE.Index(root, root / 'data')
+            documents = []
+
+            def encode(values, **kwargs):
+                if kwargs.get('prompt_name') != 'query':
+                    documents.extend(values)
+                return [embedding() for _ in values]
+
+            pairs = []
+            index.encoder = lambda: SimpleNamespace(encode=encode, tokenizer=tokenize)
+            index.reranker = lambda: SimpleNamespace(predict=lambda values, **kwargs:
+                pairs.extend(values) or [float('109.md' in document) for _, document in values])
+            numpy = SimpleNamespace(dot=lambda a, b: 1.0,
+                                    frombuffer=lambda blob, **kwargs: array('f', blob), float32=None)
+            original_fusion = MODULE.reciprocal_rank_fusion
+            rankings = []
+
+            def fuse(*values):
+                rankings.extend(values)
+                return original_fusion(*values)
+
+            try:
+                with patch.dict('sys.modules', numpy=numpy), patch.object(MODULE, 'reciprocal_rank_fusion', fuse):
+                    results = index.search('109.md Symbol109 content', 20)
+                self.assertEqual([len(ranking) for ranking in rankings], [50, 50])
+                self.assertEqual(len(pairs), 50)
+                self.assertEqual(len(results), 5)
+                self.assertEqual(results[0]['path'], '109.md')
+                self.assertEqual(results[0]['symbol'], 'Symbol109')
+                self.assertIn('Path: 109.md\nSymbol: Symbol109\n\n# Symbol109\ncontent 109', documents)
+                self.assertTrue(all(document in documents for _, document in pairs))
+                with patch.dict('sys.modules', numpy=numpy):
+                    self.assertEqual(len(index.search('" OR * () : -')), 5)
+                    self.assertEqual(len(index.search('***')), 5)
+                    (root / '109.md').unlink()
+                    index.search('Symbol109')
+                self.assertEqual(index.db.execute('select count(*) from lexical where lexical match ?', ('"Symbol109"',)).fetchone()[0], 0)
+            finally:
+                index.db.close()
+
     def test_old_cache_is_preserved_but_not_reused(self):
         with TemporaryDirectory() as temporary:
             root = Path(temporary)

shared/retrieval/server.py

@@ -22,8 +22,8 @@ CHUNK_TOKENS = 400
 OVERLAP_TOKENS = 64
 DENSE_CANDIDATES = 50
 LEXICAL_CANDIDATES = 50
-RERANK_CANDIDATES = 20
-FINAL_RESULTS = 20
+RERANK_CANDIDATES = 50
+FINAL_RESULTS = 5
 RRF_K = 60
 INDEX_VERSION = 2
 RETRIEVAL_INSTRUCTION = "Given a codebase question, retrieve relevant code and documentation that answer the question."
@@ -216,7 +216,7 @@ class Index:
                 self.delete_path(relative)
                 self.db.execute("delete from files where path = ?", (relative,))
 
-    def search(self, query: str, top_k: int = 5):
+    def search(self, query: str, top_k: int = FINAL_RESULTS):
         if not query.strip():
             raise ValueError("query must not be empty")
         with self.lock:
@@ -258,7 +258,7 @@ def main():
     mcp = FastMCP("my-ai-qwen3-retrieval")
 
     @mcp.tool()
-    def semantic_search(query: str, top_k: int = 5) -> list[dict]:
+    def semantic_search(query: str, top_k: int = FINAL_RESULTS) -> list[dict]:
         """Search repository concepts, behavior, and implementation patterns first with Qwen3 embedding and reranking. Verify returned passages against current files; use direct reads for known paths."""
         return index.search(query, top_k)