app/core/qdrant_points.py aktualisiert

app/core/qdrant_points.py

@@ -4,24 +4,16 @@
 Name:        app/core/qdrant_points.py
 Version:     v1.2.0 (2025-09-05)
 
-Kurzbeschreibung:
+Kurzbeschreibung
-  Erzeugt Qdrant-Points für Notes, Chunks, Edges.
+  Qdrant-Points für Notes/Chunks/Edges.
-  - Edges: 1D-Dummy-Vektor (Workaround).
+  NEU: Edge-ID berücksichtigt optional 'occ' (Occurrence je Match):
-  - **NEU**: Edge-ID berücksichtigt optional 'occ' (Occurrence-Index),
+       "{kind}:{src}->{tgt}#{seq}@{occ}"
-    um Mehrfach-Vorkommen von Wikilinks eindeutig zu machen.
 
-Aufruf:
+Aufruf
-  from app.core.qdrant_points import (
+  from app.core.qdrant_points import points_for_note, points_for_chunks, points_for_edges, upsert_batch
-      points_for_note, points_for_chunks, points_for_edges, upsert_batch
-  )
 
-Kompatibilität:
+Changelog
-  - Bestehende Edge-IDs bleiben gültig; neue Edges können ein '@{occ}'-Suffix in der ID haben.
+  v1.2.0: Edge-ID mit '@occ' falls vorhanden.
-  - Idempotenz weiterhin über UUIDv5 auf stabilem Key.
-
-Changelog:
-  v1.2.0: Edge-ID-Logik erweitert: "{kind}:{src}->{tgt}#{seq}@{occ}" wenn 'occ' vorhanden.
-  v1.1.x: Stabilisierung Note-/Chunk-Upsert.
 """
 
 from __future__ import annotations
@@ -29,23 +21,13 @@ import uuid
 from typing import List, Tuple
 from qdrant_client.http import models as rest
 
-
 def _names(prefix: str) -> Tuple[str, str, str]:
     return f"{prefix}_notes", f"{prefix}_chunks", f"{prefix}_edges"
 
-
 def _to_uuid(stable_key: str) -> str:
-    """Stabile UUIDv5 aus einem String-Key (deterministisch)."""
     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]]:
-def points_for_note(
-    prefix: str,
-    note_payload: dict,
-    note_vec: List[float] | None,
-    dim: int,
-) -> Tuple[str, List[rest.PointStruct]]:
-    """Notes-Collection: falls kein Note-Embedding -> Nullvektor der Länge dim."""
     notes_col, _, _ = _names(prefix)
     vector = note_vec if note_vec is not None else [0.0] * int(dim)
     raw_note_id = note_payload.get("note_id") or note_payload.get("id") or "missing-note-id"
@@ -53,18 +35,7 @@ def points_for_note(
     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]]:
-    """
-    Chunks-Collection: erwartet pro Chunk einen Vektor.
-    Robustheit:
-      - Fehlt 'chunk_id', nutze 'id', sonst baue '${note_id}#${i}' (1-basiert).
-      - Schreibe die abgeleitete ID zurück in die Payload (pl['chunk_id']).
-    """
     _, chunks_col, _ = _names(prefix)
     points: List[rest.PointStruct] = []
     for i, (pl, vec) in enumerate(zip(chunk_payloads, vectors), start=1):
@@ -77,13 +48,12 @@ def points_for_chunks(
         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]]:
     """
-    Edges-Collection mit 1D-Dummy-Vektor.
+    Edges-Collection mit 1D-Dummy-Vektor (keine Vektorsuche auf Edges).
-    ID-Bildung:
+    Stable-ID:
-      - Basis: "{kind}:{src}->{tgt}#{seq}"
+      Basis: "{kind}:{src}->{tgt}#{seq}"
-      - **Neu**: Falls 'occ' in Payload, erweitern zu "{...}@{occ}"
+      Neu:   Falls 'occ' vorhanden -> "{…}@{occ}"
     """
     _, _, edges_col = _names(prefix)
     points: List[rest.PointStruct] = []
@@ -94,7 +64,7 @@ def points_for_edges(prefix: str, edge_payloads: List[dict]) -> Tuple[str, List[
             s = pl.get("source_id", "unknown-src")
             t = pl.get("target_id", "unknown-tgt")
             seq = pl.get("seq") or pl.get("order") or ""
-            occ = pl.get("occ")  # optionaler Occurrence-Index
+            occ = pl.get("occ")
             base = f"{kind}:{s}->{t}#{seq}"
             edge_id = f"{base}@{occ}" if occ is not None else base
             pl["edge_id"] = edge_id
@@ -102,7 +72,6 @@ def points_for_edges(prefix: str, edge_payloads: List[dict]) -> Tuple[str, List[
         points.append(rest.PointStruct(id=point_id, vector=[0.0], payload=pl))
     return edges_col, points
 
-
 def upsert_batch(client, collection: str, points: List[rest.PointStruct]) -> None:
     if not points:
         return