app/core/qdrant_points.py aktualisiert

app/core/qdrant_points.py

@@ -1,5 +1,5 @@
 from __future__ import annotations
-import os
+import uuid
 from typing import List, Tuple
 from qdrant_client.http import models as rest
 
@@ -8,39 +8,59 @@ def _names(prefix: str) -> Tuple[str, str, str]:
     return f"{prefix}_notes", f"{prefix}_chunks", f"{prefix}_edges"
 
 
-def points_for_note(prefix: str, note_payload: dict, note_vec: List[float] | None, dim: int) -> Tuple[str, List[rest.PointStruct]]:
+def _to_uuid(stable_key: str) -> str:
     """
-    Liefert (collection_name, [PointStruct]) für die Notes-Collection.
+    Erzeuge eine stabile UUIDv5 aus einem stabilen String-Key (z. B. note_id, chunk_id, edge_id).
-    Falls kein Note-Embedding übergeben wurde, wird ein Nullvektor der Länge `dim` verwendet.
+    Wir verwenden NAMESPACE_URL, damit die UUIDs deterministisch sind.
-    Hintergrund: Die Notes-Collection ist in ensure_collections mit Vektor-Dimension angelegt.
+    """
+    return str(uuid.uuid5(uuid.NAMESPACE_URL, stable_key))
+
+
+def points_for_note(
+    prefix: str,
+    note_payload: dict,
+    note_vec: List[float] | None,
+    dim: int,
+) -> Tuple[str, List[rest.PointStruct]]:
+    """
+    (collection_name, [PointStruct]) für die Notes-Collection.
+    Falls kein Note-Embedding vorhanden -> Nullvektor der Länge `dim`.
     """
     notes_col, _, _ = _names(prefix)
     vector = note_vec if note_vec is not None else [0.0] * int(dim)
-    pt = rest.PointStruct(id=note_payload["note_id"], vector=vector, payload=note_payload)
+    # Qdrant-Point-ID MUSS int oder UUID sein -> aus note_id eine UUIDv5 machen
+    point_id = _to_uuid(note_payload["note_id"])
+    pt = rest.PointStruct(id=point_id, vector=vector, payload=note_payload)
     return notes_col, [pt]
 
 
-def points_for_chunks(prefix: str, chunk_payloads: List[dict], vectors: List[List[float]]) -> Tuple[str, List[rest.PointStruct]]:
+def points_for_chunks(
+    prefix: str,
+    chunk_payloads: List[dict],
+    vectors: List[List[float]],
+) -> Tuple[str, List[rest.PointStruct]]:
     """
-    Liefert (collection_name, [PointStruct]) für die Chunks-Collection.
+    (collection_name, [PointStruct]) für die Chunks-Collection.
-    Erwartet für jeden Chunk einen Embedding-Vektor (oder Nullvektor, wenn --skip-embed).
+    Erwartet pro Chunk einen Vektor (oder Nullvektor, wenn --skip-embed).
     """
     _, chunks_col, _ = _names(prefix)
     points: List[rest.PointStruct] = []
     for pl, vec in zip(chunk_payloads, vectors):
-        points.append(rest.PointStruct(id=pl["chunk_id"], vector=vec, payload=pl))
+        point_id = _to_uuid(pl["chunk_id"])
+        points.append(rest.PointStruct(id=point_id, vector=vec, payload=pl))
     return chunks_col, points
 
 
 def points_for_edges(prefix: str, edge_payloads: List[dict]) -> Tuple[str, List[rest.PointStruct]]:
     """
-    Liefert (collection_name, [PointStruct]) für die Edges-Collection.
+    (collection_name, [PointStruct]) für die Edges-Collection.
-    Edges-Collection ist VEKTORENLOS angelegt → nur Payload.
+    Edges-Collection ist ohne Vektor angelegt -> nur Payload + UUID-IDs.
     """
     _, _, edges_col = _names(prefix)
     points: List[rest.PointStruct] = []
     for pl in edge_payloads:
-        points.append(rest.PointStruct(id=pl["edge_id"], payload=pl))
+        point_id = _to_uuid(pl["edge_id"])
+        points.append(rest.PointStruct(id=point_id, payload=pl))
     return edges_col, points