Fix indexer: handle date serialization and pgvector format

services/ingestion-worker/indexer.py

@@ -7,6 +7,7 @@ from __future__ import annotations
 import hashlib
 import json
 import logging
+from datetime import date, datetime
 from typing import Any
 
 import asyncpg
@@ -21,6 +22,14 @@ def sha256(text: str) -> str:
     return hashlib.sha256(text.encode('utf-8')).hexdigest()
 
 
+class DateTimeEncoder(json.JSONEncoder):
+    """JSON encoder that handles date/datetime objects."""
+    def default(self, obj):
+        if isinstance(obj, (date, datetime)):
+            return obj.isoformat()
+        return super().default(obj)
+
+
 async def upsert_document(
     conn: asyncpg.Connection,
     doc: ParsedDocument,
@@ -56,7 +65,7 @@ async def upsert_document(
             doc.title,
             doc.content_raw,
             content_hash,
-            json.dumps(doc.frontmatter),
+            json.dumps(doc.frontmatter, cls=DateTimeEncoder),
             doc.tags,
             doc.aliases,
             doc.word_count,
@@ -71,12 +80,14 @@ async def upsert_document(
         # ---- Insert chunks + embeddings ----
         chunk_records = []
         for chunk, embedding in zip(chunks, embeddings):
+            # Convert embedding list to pgvector string format
+            embedding_str = '[' + ','.join(str(x) for x in embedding) + ']'
             chunk_records.append((
                 row['id'],
                 chunk.chunk_index,
                 chunk.content,
                 chunk.token_count,
-                embedding,
+                embedding_str,
                 json.dumps(chunk.metadata),
             ))