app/core/qdrant_points.py aktualisiert

app/core/qdrant_points.py

@@ -1,37 +1,50 @@
 from __future__ import annotations
-import os, datetime
-from typing import List, Dict, Tuple
-from qdrant_client import QdrantClient
+import os
+from typing import List, Tuple
 from qdrant_client.http import models as rest
-from .qdrant import QdrantConfig, get_client, _collection_names
 
-def ts_iso() -> str:
-    return datetime.datetime.utcnow().replace(microsecond=0).isoformat() + "Z"
 
-def points_for_chunks(prefix: str, chunk_payloads: List[Dict], vectors: List[List[float]]) -> Tuple[str, List[rest.PointStruct]]:
-    assert len(chunk_payloads) == len(vectors)
-    _, chunks, _ = _collection_names(prefix)
-    pts = []
+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]]:
+    """
+    Liefert (collection_name, [PointStruct]) für die Notes-Collection.
+    Falls kein Note-Embedding übergeben wurde, wird ein Nullvektor der Länge `dim` verwendet.
+    Hintergrund: Die Notes-Collection ist in ensure_collections mit Vektor-Dimension angelegt.
+    """
+    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)
+    return notes_col, [pt]
+
+
+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.
+    Erwartet für jeden Chunk einen Embedding-Vektor (oder Nullvektor, wenn --skip-embed).
+    """
+    _, chunks_col, _ = _names(prefix)
+    points: List[rest.PointStruct] = []
     for pl, vec in zip(chunk_payloads, vectors):
-        pts.append(rest.PointStruct(id=pl["id"], vector=vec, payload=pl))
-    return chunks, pts
+        points.append(rest.PointStruct(id=pl["chunk_id"], vector=vec, payload=pl))
+    return chunks_col, points
 
-def points_for_note(prefix: str, note_payload: Dict, vector: List[float] | None) -> Tuple[str, List[rest.PointStruct]]:
-    notes, _, _ = _collection_names(prefix)
-    if vector is None:
-        # Für Notizen erlauben wir auch Payload-only Upserts (Vektor leer)
-        return notes, [rest.PointStruct(id=note_payload["note_id"], payload=note_payload)]
-    return notes, [rest.PointStruct(id=note_payload["note_id"], vector=vector, payload=note_payload)]
 
-def points_for_edges(prefix: str, edges: List[Dict]) -> Tuple[str, List[rest.PointStruct]]:
-    _, _, edges_col = _collection_names(prefix)
-    pts = []
-    for e in edges:
-        pid = f"{e['src_id']}~{e['edge_type']}~{e['dst_id']}~{e.get('scope','note')}"
-        # 1-dim Dummy-Vektor, weil Collection einen Vektor erwartet
-        pts.append(rest.PointStruct(id=pid, vector=[0.0], payload=e))
-    return edges_col, pts
+def points_for_edges(prefix: str, edge_payloads: List[dict]) -> Tuple[str, List[rest.PointStruct]]:
+    """
+    Liefert (collection_name, [PointStruct]) für die Edges-Collection.
+    Edges-Collection ist VEKTORENLOS angelegt → nur Payload.
+    """
+    _, _, edges_col = _names(prefix)
+    points: List[rest.PointStruct] = []
+    for pl in edge_payloads:
+        points.append(rest.PointStruct(id=pl["edge_id"], payload=pl))
+    return edges_col, points
 
-def upsert_batch(client: QdrantClient, collection: str, points: List[rest.PointStruct]):
-    if not points: return
-    client.upsert(collection, points=points, wait=True)
+
+def upsert_batch(client, collection: str, points: List[rest.PointStruct]) -> None:
+    if not points:
+        return
+    client.upsert(collection_name=collection, points=points, wait=True)