Protokoll #23911

{
    "event": "PreToolUse",
    "tool_name": "Read",
    "tool_input": {
        "file_path": "\/var\/www\/scripts\/pipeline\/embed.py",
        "offset": 90,
        "limit": 70
    }
}

{
    "tool_response": {
        "type": "text",
        "file": {
            "filePath": "\/var\/www\/scripts\/pipeline\/embed.py",
            "content": "\n        # Prepare payload\n        payload = {\n            \"document_id\": document_id,\n            \"document_title\": document_title,\n            \"chunk_index\": i,\n            \"content\": chunk[\"content\"][:1000],  # Truncate for payload\n            \"heading_path\": json.dumps(chunk.get(\"heading_path\", [])),\n            \"source_path\": source_path,\n        }\n\n        # Add any chunk metadata\n        if chunk.get(\"metadata\"):\n            for key, value in chunk[\"metadata\"].items():\n                if isinstance(value, (str, int, float, bool)):\n                    payload[key] = value\n\n        # Store in Qdrant\n        if store_in_qdrant(\"documents\", point_id, embedding, payload):\n            # Update chunk in database with Qdrant ID\n            db.update_chunk_qdrant_id(chunk.get(\"db_id\"), point_id)\n            embedded_count += 1\n            db.log(\"INFO\", f\"Embedded chunk {i}\/{len(chunks)}\", f\"doc={document_id}\")\n        else:\n            db.log(\"ERROR\", f\"Failed to store chunk {i} in Qdrant\")\n\n    return embedded_count\n\n\ndef embed_pending_chunks(limit=100):\n    \"\"\"Process chunks that haven't been embedded yet.\"\"\"\n    db.connect()\n\n    try:\n        chunks = db.get_chunks_for_embedding(limit)\n        db.log(\"INFO\", f\"Found {len(chunks)} chunks to embed\")\n\n        if not chunks:\n            return 0\n\n        embedded = 0\n        for chunk in chunks:\n            embedding = get_embedding(chunk[\"content\"])\n\n            if not embedding:\n                continue\n\n            point_id = str(uuid.uuid4())\n\n            # Get document info\n            cursor = db.execute(\"SELECT filename, source_path FROM documents WHERE id = %s\", (chunk[\"document_id\"],))\n            doc = cursor.fetchone()\n            cursor.close()\n\n            payload = {\n                \"document_id\": chunk[\"document_id\"],\n                \"document_title\": doc[\"filename\"] if doc else \"\",\n                \"chunk_id\": chunk[\"id\"],\n                \"content\": chunk[\"content\"][:1000],\n                \"source_path\": doc[\"source_path\"] if doc else \"\",\n            }\n\n            if store_in_qdrant(\"documents\", point_id, embedding, payload):\n                db.update_chunk_qdrant_id(chunk[\"id\"], point_id)\n                embedded += 1\n\n        db.log(\"INFO\", f\"Embedded {embedded}\/{len(chunks)} chunks\")\n        return embedded\n\n    except Exception as e:",
            "numLines": 70,
            "startLine": 90,
            "totalLines": 205
        }
    }
}