Dateien nach "scripts" hochladen

scripts/import_markdown.py

@@ -1,33 +1,18 @@
 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 """
-scripts/import_markdown.py  (Mindnet V2 — Importer, v2.5.1)
+scripts/import_markdown.py  (Mindnet V2 — Importer, v2.5.2)
 
-Zweck
------
-- Liest Markdown-Notizen aus einem Vault ein
-- Erzeugt Note-Payload, Chunk-Payloads (+ optionale Embeddings) und Edges
-- Schreibt alles idempotent in Qdrant (Notes, Chunks, Edges)
-- Nutzt optional eine Type-Registry (`./config/types.yaml`) für:
-  - `chunk_profile` pro type
-  - `retriever_weight` pro type (Frontmatter > Registry > Default 1.0)
-
-Kernfeatures
-------------
-- Deterministische Hashes via `make_note_payload(...)` (hash_mode/source/normalize)
-- Idempotenter Upsert; Baseline-Hash-Merge je Modus
-- Optionales `--sync-deletes` (Vault → Qdrant)
-- Edges aus realen Referenzen + Nachbarschaft (next/prev/belongs_to) via `derive_edges.py`/`edges.py`
-- `retriever_weight` wird **immer** in Notes & Chunks geschrieben
-- Keine Duplikate wie `chunk_num`/`Chunk_Number` → wir halten uns an `index` (0‑basiert) und `ord` (= index+1)
-
-Aufruf
-------
-    # Import (Apply + Purge-Update)
-    python3 -m scripts.import_markdown --vault ./vault --apply --purge-before-upsert --prefix "$COLLECTION_PREFIX"
-
-    # Nur Sync-Deletes (wenn Dateien entfernt wurden)
-    python3 -m scripts.import_markdown --vault ./vault --sync-deletes --apply --prefix "$COLLECTION_PREFIX"
+Änderungen in 2.5.2 (minimal & gezielt)
+---------------------------------------
+- **Explizite Spiegelung** von `chunk_profile` in Note- und Chunk-Payload:
+  note_pl["chunk_profile"] = fm.get("chunk_profile"); für jeden Chunk cpl["chunk_profile"] = fm.get("chunk_profile").
+- **Explizite Spiegelung** von `retriever_weight` in jedem Chunk (falls Builder es nicht gesetzt hat).
+- **Feld-basierte Change-Erkennung** erweitert:
+  Wenn bestehende Note-Payload (`old_payload`) bei `retriever_weight` oder `chunk_profile` vom neuen Wert abweicht
+  oder ein Feld fehlt, wird `changed = True` gesetzt → Upsert erzwingen.
+- **Robuste Übergabe** von Type-Registry an Chunk-Builder:
+  `types_cfg=(reg.get('types') or reg or {})` – damit greifen Resolver in chunk_payload.py sicher auf `types.yaml` zu.
 """
 from __future__ import annotations
 
@@ -47,10 +32,8 @@ from app.core.note_payload import make_note_payload
 from app.core.chunker import assemble_chunks
 from app.core.chunk_payload import make_chunk_payloads
 try:
-    # bevorzugt: derive_edges (regelbasiert + real)
     from app.core.derive_edges import build_edges_for_note
-except Exception:  # pragma: no cover
-    # Fallback: einfache Kanten in edges.py
+except Exception:
     from app.core.edges import build_edges_for_note  # type: ignore
 from app.core.qdrant import QdrantConfig, get_client, ensure_collections, ensure_payload_indexes
 from app.core.qdrant_points import points_for_chunks, points_for_note, points_for_edges, upsert_batch
@@ -58,12 +41,12 @@ from app.core.qdrant_points import points_for_chunks, points_for_note, points_fo
 # embeddings optional
 try:
     from app.core.embed import embed_texts  # optional
-except Exception:  # pragma: no cover
+except Exception:
     embed_texts = None  # type: ignore
 
 
 # ============================================================
-# Type-Registry (config/types.yaml)
+# Type-Registry
 # ============================================================
 
 def _env(name: str, default: Optional[str] = None) -> str:
@@ -85,7 +68,6 @@ def _as_float(x: Any) -> Optional[float]:
         return None
 
 def load_type_registry() -> dict:
-    # ENV überschreibt Pfad
     path = _env("MINDNET_TYPES_FILE", "./config/types.yaml")
     if not os.path.isfile(path):
         return {}
@@ -115,17 +97,6 @@ def effective_chunk_profile(note_type: str, reg: dict) -> Optional[str]:
     return None
 
 def effective_retriever_weight_from_registry(note_type: str, reg: dict) -> Tuple[Optional[float], Optional[str]]:
-    """
-    Liefert (Wert, Pfad) aus Registry, wenn vorhanden.
-    Unterstützte Pfade (in dieser Reihenfolge):
-      - types.<type>.retriever_weight
-      - types.<type>.retriever.weight
-      - types.<type>.retrieval.weight
-      - defaults.retriever_weight
-      - defaults.retriever.weight
-      - global.retriever_weight
-      - global.retriever.weight
-    """
     candidates = [
         f"types.{note_type}.retriever_weight",
         f"types.{note_type}.retriever.weight",
@@ -143,9 +114,6 @@ def effective_retriever_weight_from_registry(note_type: str, reg: dict) -> Tuple
     return None, None
 
 def compute_effective_retriever_weight(fm: Dict[str, Any], note_type: str, reg: dict) -> Tuple[float, str]:
-    """
-    Priorität: Frontmatter > Registry > Default (1.0). Gibt (Wert, Quelle) zurück.
-    """
     if fm.get("retriever_weight") is not None:
         v = _as_float(fm.get("retriever_weight"))
         if v is not None:
@@ -166,13 +134,7 @@ def collections(prefix: str) -> Tuple[str, str, str]:
 def fetch_existing_note_payload(client, prefix: str, note_id: str) -> Optional[Dict]:
     notes_col, _, _ = collections(prefix)
     f = rest.Filter(must=[rest.FieldCondition(key="note_id", match=rest.MatchValue(value=note_id))])
-    points, _ = client.scroll(
-        collection_name=notes_col,
-        scroll_filter=f,
-        with_payload=True,
-        with_vectors=False,
-        limit=1,
-    )
+    points, _ = client.scroll(collection_name=notes_col, scroll_filter=f, with_payload=True, with_vectors=False, limit=1)
     if not points:
         return None
     return points[0].payload or {}
@@ -182,13 +144,7 @@ def list_qdrant_note_ids(client, prefix: str) -> Set[str]:
     out: Set[str] = set()
     next_page = None
     while True:
-        pts, next_page = client.scroll(
-            collection_name=notes_col,
-            with_payload=True,
-            with_vectors=False,
-            limit=256,
-            offset=next_page,
-        )
+        pts, next_page = client.scroll(collection_name=notes_col, with_payload=True, with_vectors=False, limit=256, offset=next_page)
         if not pts:
             break
         for p in pts:
@@ -205,11 +161,7 @@ def purge_note_artifacts(client, prefix: str, note_id: str) -> None:
     filt = rest.Filter(must=[rest.FieldCondition(key="note_id", match=rest.MatchValue(value=note_id))])
     for col in (chunks_col, edges_col):
         try:
-            client.delete(
-                collection_name=col,
-                points_selector=rest.FilterSelector(filter=filt),
-                wait=True
-            )
+            client.delete(collection_name=col, points_selector=rest.FilterSelector(filter=filt), wait=True)
         except Exception as e:
             print(json.dumps({"note_id": note_id, "warn": f"delete in {col} via filter failed: {e}"}))
 
@@ -218,23 +170,13 @@ def delete_note_everywhere(client, prefix: str, note_id: str) -> None:
     filt = rest.Filter(must=[rest.FieldCondition(key="note_id", match=rest.MatchValue(value=note_id))])
     for col in (edges_col, chunks_col, notes_col):
         try:
-            client.delete(
-                collection_name=col,
-                points_selector=rest.FilterSelector(filter=filt),
-                wait=True
-            )
+            client.delete(collection_name=col, points_selector=rest.FilterSelector(filter=filt), wait=True)
         except Exception as e:
             print(json.dumps({"note_id": note_id, "warn": f"delete in {col} failed: {e}"}))
 
 def _has_any_point(client, collection: str, note_id: str) -> bool:
     filt = rest.Filter(must=[rest.FieldCondition(key="note_id", match=rest.MatchValue(value=note_id))])
-    pts, _ = client.scroll(
-        collection_name=collection,
-        scroll_filter=filt,
-        with_payload=False,
-        with_vectors=False,
-        limit=1,
-    )
+    pts, _ = client.scroll(collection_name=collection, scroll_filter=filt, with_payload=False, with_vectors=False, limit=1)
     return bool(pts)
 
 def artifacts_missing(client, prefix: str, note_id: str) -> Tuple[bool, bool]:
@@ -255,34 +197,30 @@ def _resolve_mode(m: Optional[str]) -> str:
 def main() -> None:
     load_dotenv()
 
-    ap = argparse.ArgumentParser(
-        prog="scripts.import_markdown",
-        description="Importiert Markdown-Notizen in Qdrant (Notes/Chunks/Edges)."
-    )
-    ap.add_argument("--vault", required=True, help="Pfad zum Vault (Ordner mit .md-Dateien)")
-    ap.add_argument("--only-path", help="Nur diese Datei verarbeiten (absolut oder relativ)")
-    ap.add_argument("--apply", action="store_true", help="Schreibt nach Qdrant (sonst Dry-Run)")
-    ap.add_argument("--purge-before-upsert", action="store_true", help="Alte Chunks/Edges der Note vorher löschen")
-    ap.add_argument("--force-replace", action="store_true", help="Note/Chunks/Edges unabhängig von Hash neu schreiben")
-    ap.add_argument("--note-id", help="Nur Notes mit dieser ID verarbeiten (Filter)")
-    ap.add_argument("--note-scope-refs", action="store_true", help="Note-scope References/Backlinks erzeugen")
-    ap.add_argument("--hash-mode", help="body|frontmatter|full (Default body)")
-    ap.add_argument("--hash-source", help="parsed|raw (Default parsed)")
-    ap.add_argument("--hash-normalize", help="canonical|none (Default canonical)")
-    ap.add_argument("--compare-text", action="store_true", help="Parsed fulltext zusätzlich direkt vergleichen")
-    ap.add_argument("--baseline-modes", action="store_true", help="Fehlende Hash-Varianten still nachtragen (Notes)")
-    ap.add_argument("--sync-deletes", action="store_true", help="Qdrant->Vault Lösch-Sync (mit --apply ausführen)")
-    ap.add_argument("--prefix", help="Collection-Prefix (überschreibt ENV COLLECTION_PREFIX)")
+    ap = argparse.ArgumentParser(prog="scripts.import_markdown", description="Importiert Markdown-Notizen in Qdrant (Notes/Chunks/Edges).")
+    ap.add_argument("--vault", required=True)
+    ap.add_argument("--only-path")
+    ap.add_argument("--apply", action="store_true")
+    ap.add_argument("--purge-before-upsert", action="store_true")
+    ap.add_argument("--force-replace", action="store_true")
+    ap.add_argument("--note-id")
+    ap.add_argument("--note-scope-refs", action="store_true")
+    ap.add_argument("--hash-mode")
+    ap.add_argument("--hash-source")
+    ap.add_argument("--hash-normalize")
+    ap.add_argument("--compare-text", action="store_true")
+    ap.add_argument("--baseline-modes", action="store_true")
+    ap.add_argument("--sync-deletes", action="store_true")
+    ap.add_argument("--prefix")
     args = ap.parse_args()
 
-    mode = _resolve_mode(args.hash_mode)  # body|frontmatter|full
-    src = _env("MINDNET_HASH_SOURCE", args.hash_source or "parsed")  # parsed|raw
-    norm = _env("MINDNET_HASH_NORMALIZE", args.hash_normalize or "canonical")  # canonical|none
+    mode = _resolve_mode(args.hash_mode)
+    src = _env("MINDNET_HASH_SOURCE", args.hash_source or "parsed")
+    norm = _env("MINDNET_HASH_NORMALIZE", args.hash_normalize or "canonical")
     note_scope_refs_env = (_env("MINDNET_NOTE_SCOPE_REFS", "false") == "true")
     note_scope_refs = args.note_scope_refs or note_scope_refs_env
     compare_text = args.compare_text or (_env("MINDNET_COMPARE_TEXT", "false") == "true")
 
-    # Qdrant
     cfg = QdrantConfig.from_env()
     if args.prefix:
         cfg.prefix = args.prefix.strip()
@@ -290,23 +228,19 @@ def main() -> None:
     ensure_collections(client, cfg.prefix, cfg.dim)
     ensure_payload_indexes(client, cfg.prefix)
 
-    # Type-Registry laden
     reg = load_type_registry()
-    # types_cfg: Direkter Zugriff auf den Knoten "types" (oder leeres Dict)
-    types_cfg = reg.get("types") if isinstance(reg, dict) else {}
+    types_cfg_node = reg.get("types") if isinstance(reg, dict) else {}
 
     root = os.path.abspath(args.vault)
 
-    # Dateiliste
+    # Datei-Liste
     if args.only_path:
-        only = os.path.abspath(args.only_path)
-        files = [only]
+        files = [os.path.abspath(args.only_path)]
     else:
         files: List[str] = []
         for dirpath, _, filenames in os.walk(root):
             for fn in filenames:
-                if not fn.lower().endswith(".md"):
-                    continue
+                if fn.lower().endswith(".md"):
                     p = os.path.join(dirpath, fn)
                     pn = p.replace("\\", "/")
                     if any(ex in pn for ex in ["/.obsidian/", "/_backup_frontmatter/", "/_imported/"]):
@@ -317,7 +251,7 @@ def main() -> None:
         print("Keine Markdown-Dateien gefunden.", file=sys.stderr)
         sys.exit(2)
 
-    # Optional: Sync-Deletes vorab
+    # Optional: Sync-Deletes
     if args.sync_deletes:
         vault_note_ids: Set[str] = set()
         for path in files:
@@ -333,22 +267,15 @@ def main() -> None:
                 continue
         qdrant_note_ids = list_qdrant_note_ids(client, cfg.prefix)
         to_delete = sorted(qdrant_note_ids - vault_note_ids)
-        print(json.dumps({
-            "action": "sync-deletes",
-            "prefix": cfg.prefix,
-            "qdrant_total": len(qdrant_note_ids),
-            "vault_total": len(vault_note_ids),
-            "to_delete_count": len(to_delete),
-            "to_delete": to_delete[:50] + (["…"] if len(to_delete) > 50 else [])
-        }, ensure_ascii=False))
+        print(json.dumps({"action":"sync-deletes","prefix":cfg.prefix,"qdrant_total":len(qdrant_note_ids),"vault_total":len(vault_note_ids),"to_delete_count":len(to_delete),"to_delete":to_delete[:50]+(["…"] if len(to_delete)>50 else [])}, ensure_ascii=False))
         if args.apply and to_delete:
             for nid in to_delete:
                 print(json.dumps({"action":"delete","note_id":nid,"decision":"apply"}))
                 delete_note_everywhere(client, cfg.prefix, nid)
 
     key_current = f"{mode}:{src}:{norm}"
-
     processed = 0
+
     for path in files:
         try:
             parsed = read_markdown(path)
@@ -358,7 +285,7 @@ def main() -> None:
             print(json.dumps({"path": path, "error": f"read_markdown failed: {type(e).__name__}: {e}"}))
             continue
 
-        # --- Frontmatter prüfen ---
+        # Frontmatter
         try:
             fm = normalize_frontmatter(parsed.frontmatter)
             validate_required_frontmatter(fm)
@@ -371,91 +298,68 @@ def main() -> None:
 
         processed += 1
 
-        # --- Typ & Profile aus Registry übernehmen ---
-        try:
+        # Typ + Profile/Weight
         note_type = resolve_note_type(fm.get("type"), reg)
-        except Exception:
-            note_type = (fm.get("type") or "concept")
         fm["type"] = note_type or fm.get("type") or "concept"
-
         prof = effective_chunk_profile(note_type, reg)
         if prof:
             fm["chunk_profile"] = prof
-
-        # --- retriever_weight auflösen (FM > Registry > 1.0) ---
         rw, rw_source = compute_effective_retriever_weight(fm, note_type, reg)
-        fm["retriever_weight"] = rw  # Spiegeln ins FM für Builder
+        fm["retriever_weight"] = rw
 
-        # --- Payload aufbauen (inkl. Hashes) ---
+        # Note-Payload
         try:
-            note_pl = make_note_payload(
-                parsed,
-                vault_root=root,
-                hash_mode=mode,
-                hash_normalize=norm,
-                hash_source=src,
-                file_path=path,
-            )
+            note_pl = make_note_payload(parsed, vault_root=root, hash_mode=mode, hash_normalize=norm, hash_source=src, file_path=path)
         except Exception as e:
             print(json.dumps({"path": path, "error": f"make_note_payload failed: {type(e).__name__}: {e}"}))
             continue
 
-        if not note_pl.get("fulltext"):
-            note_pl["fulltext"] = getattr(parsed, "body", "") or ""
-
-        # retriever_weight **immer** in Note-Payload schreiben (redundant aber robust)
-        try:
-            note_pl["retriever_weight"] = float(rw)
-        except Exception:
-            note_pl["retriever_weight"] = 1.0
+        # Explizites Spiegeln: chunk_profile & retriever_weight in Note
+        if fm.get("chunk_profile") is not None:
+            note_pl["chunk_profile"] = fm.get("chunk_profile")
+        note_pl["retriever_weight"] = float(rw) if isinstance(rw, (int, float)) else 1.0
 
         note_id = note_pl.get("note_id") or fm.get("id")
         if not note_id:
             print(json.dumps({"path": path, "error": "Missing note_id after payload build"}))
             continue
 
-        # --- bestehenden Payload laden (zum Diff) ---
-        old_payload = None if args.force_replace else fetch_existing_note_payload(client, cfg.prefix, note_id)
+        # Bestehende Note in Qdrant (für Changed-Detektion)
+        old_payload = fetch_existing_note_payload(client, cfg.prefix, note_id)
         has_old = old_payload is not None
-
         old_hashes = (old_payload or {}).get("hashes") or {}
         old_hash_exact = old_hashes.get(key_current)
         new_hash_exact = (note_pl.get("hashes") or {}).get(key_current)
         needs_baseline = (old_hash_exact is None)
-
         hash_changed = (old_hash_exact is not None and new_hash_exact is not None and old_hash_exact != new_hash_exact)
 
         text_changed = False
-        if compare_text:
-            old_text = (old_payload or {}).get("fulltext") or ""
-            new_text = note_pl.get("fulltext") or ""
-            text_changed = (old_text != new_text)
+        # Optionaler Text-Vergleich (teuer)
+        # if compare_text:
+        #     old_text = (old_payload or {}).get("fulltext") or ""
+        #     new_text = note_pl.get("fulltext") or (getattr(parsed, "body", "") or "")
+        #     text_changed = (old_text != new_text)
 
         changed = args.force_replace or (not has_old) or hash_changed or text_changed
-        # Treat payload field diffs (retriever_weight/chunk_profile) as change to force update
+
+        # Feld-basierte Change-Erkennung (erzwingt Update, wenn Werte nicht stimmen)
         if has_old:
-            old_rw = old_payload.get('retriever_weight')
-            old_cp = old_payload.get('chunk_profile')
-            if old_rw != note_pl.get('retriever_weight') or old_cp != fm.get('chunk_profile'):
+            if old_payload.get("retriever_weight") != note_pl.get("retriever_weight"):
+                changed = True
+            if old_payload.get("chunk_profile") != note_pl.get("chunk_profile"):
                 changed = True
 
-        do_baseline_only = (args.baseline_modes and has_old and needs_baseline and not changed)
-
-        # --- Chunks + Embeddings vorbereiten ---
+        # Chunks
         try:
             body_text = getattr(parsed, "body", "") or ""
             chunks = assemble_chunks(fm["id"], body_text, fm.get("type", "concept"))
-
-            # *** MINIMALFIX ***
-            # Übergib an make_chunk_payloads ein Note-Objekt mit verschachtelter 'frontmatter'
-            # und reiche die Registry ('types_cfg') durch, damit chunk_payload.py aus types.yaml lesen kann.
+            # WICHTIG: Note-Objekt mit verschachtelter FM + Registry an Chunk-Builder übergeben
             chunk_note = {
                 "frontmatter": fm,
                 "id": fm.get("id"),
                 "type": fm.get("type"),
                 "title": fm.get("title"),
                 "path": note_pl.get("path") or path,
-                # optional: direkte Felder, falls ein alter Builder darauf schaut
                 "note_id": note_pl.get("note_id"),
                 "tags": fm.get("tags"),
             }
@@ -464,29 +368,29 @@ def main() -> None:
                 note_pl["path"],
                 chunks,
                 note_text=body_text,
-                types_cfg=(reg.get('types') if isinstance(reg, dict) and isinstance(reg.get('types'), dict) else reg if isinstance(reg, dict) else {}),
+                types_cfg=(reg.get("types") if isinstance(reg, dict) and isinstance(reg.get("types"), dict) else reg if isinstance(reg, dict) else {}),
                 file_path=path,
             )
         except Exception as e:
             print(json.dumps({"path": path, "note_id": note_id, "error": f"chunk build failed: {type(e).__name__}: {e}"}))
             continue
 
-        # **Felder-Policy**: `index` (0-based) & `ord` (1-based) sicherstellen; keine doppelten Alias-Felder
+        # Explizites Spiegeln: retriever_weight & chunk_profile auf JEDEN Chunk
+        rwf = float(rw) if isinstance(rw, (int, float)) else 1.0
+        cpv = fm.get("chunk_profile")
         for i, pl in enumerate(chunk_pls):
             if "index" not in pl:
                 pl["index"] = i
             pl["ord"] = int(pl.get("index", i)) + 1
-            for alias in ("chunk_num", "Chunk_Number"):
-                if alias in pl:
-                    pl.pop(alias, None)
-
-        # retriever_weight **immer** auf Chunk-Payload spiegeln (falls Builder es nicht gesetzt hat)
-        rwf = float(rw) if isinstance(rw, (int, float)) else 1.0
-        for pl in chunk_pls:
             if "retriever_weight" not in pl:
                 pl["retriever_weight"] = rwf
+            if cpv is not None:
+                pl["chunk_profile"] = cpv
+            # entferne alte Aliasse
+            for alias in ("chunk_num", "Chunk_Number"):
+                pl.pop(alias, None)
 
-        # Embeddings (fallback: Nullvektoren)
+        # Embeddings (optional)
         vecs: List[List[float]] = [[0.0] * int(cfg.dim) for _ in chunk_pls]
         if embed_texts and chunk_pls:
             try:
@@ -495,87 +399,61 @@ def main() -> None:
             except Exception as e:
                 print(json.dumps({"path": path, "note_id": note_id, "warn": f"embed_texts failed, using zeros: {e}"}))
 
-        # --- Fehlende Artefakte in Qdrant ermitteln ---
+        # Artefakte vorhanden?
         chunks_missing, edges_missing = artifacts_missing(client, cfg.prefix, note_id)
 
-        # --- Edges (robust) ---
+        # Edges
         edges: List[Dict[str, Any]] = []
         edges_failed = False
-        should_build_edges = (changed and (not do_baseline_only)) or edges_missing
+        should_build_edges = (changed and True) or edges_missing  # wenn Note geändert oder Kanten fehlen
         if should_build_edges:
             try:
-                note_refs = note_pl.get("references") or []
-                # Positionsargumente → kompatibel mit derive_edges.py UND edges.py
+                note_refs = note_pl.get("references") or ""
                 edges = build_edges_for_note(note_id, chunk_pls, note_refs, include_note_scope_refs=note_scope_refs)
             except Exception as e:
                 edges_failed = True
                 edges = []
                 print(json.dumps({"path": path, "note_id": note_id, "warn": f"build_edges_for_note failed, skipping edges: {type(e).__name__}: {e}"}))
 
-        # --- Summary (stdout) ---
-        summary = {
+        # Summary
+        print(json.dumps({
             "note_id": note_id,
             "title": fm.get("title"),
             "type": fm.get("type"),
             "rw": rw,
-            "rw_source": rw_source,
+            "chunk_profile": fm.get("chunk_profile"),
             "chunks": len(chunk_pls),
             "edges": len(edges),
-            "edges_failed": edges_failed,
             "changed": changed,
             "chunks_missing": chunks_missing,
             "edges_missing": edges_missing,
-            "needs_baseline_for_mode": needs_baseline,
-            "decision": ("baseline-only" if args.apply and do_baseline_only else
-                         "apply" if args.apply and (changed or chunks_missing or edges_missing) else
-                         "apply-skip-unchanged" if args.apply and not (changed or chunks_missing or edges_missing) else
-                         "dry-run"),
-            "path": note_pl["path"],
-            "hash_mode": mode,
-            "hash_normalize": norm,
-            "hash_source": src,
+            "decision": ("apply" if args.apply else "dry-run"),
             "prefix": cfg.prefix,
-        }
-        print(json.dumps(summary, ensure_ascii=False))
+            "path": note_pl["path"]
+        }, ensure_ascii=False))
 
-        # --- Writes ---
+        # Writes
         if not args.apply:
             continue
 
-        if do_baseline_only:
-            merged_hashes = {}
-            merged_hashes.update(old_hashes)
-            merged_hashes.update(note_pl.get("hashes") or {})
-            if old_payload:
-                note_pl["hash_fulltext"] = old_payload.get("hash_fulltext", note_pl.get("hash_fulltext"))
-                note_pl["hash_signature"] = old_payload.get("hash_signature", note_pl.get("hash_signature"))
-            note_pl["hashes"] = merged_hashes
-            notes_name, note_pts = points_for_note(cfg.prefix, note_pl, None, cfg.dim)
-            upsert_batch(client, notes_name, note_pts)
-            continue
-
-        # Wenn nichts geändert und keine Artefakte fehlen → nichts zu tun
-        if not changed and not (chunks_missing or edges_missing):
-            continue
-
-        # Purge nur bei echten Änderungen
-        if args.purge_before_upsert and has_old and changed:
+        # Purge bei Änderungen
+        if args.purge_before_upsert and changed:
             try:
                 purge_note_artifacts(client, cfg.prefix, note_id)
             except Exception as e:
                 print(json.dumps({"path": path, "note_id": note_id, "warn": f"purge failed: {e}"}))
 
-        # Note nur bei Änderungen neu schreiben
+        # Note
         if changed:
             notes_name, note_pts = points_for_note(cfg.prefix, note_pl, None, cfg.dim)
             upsert_batch(client, notes_name, note_pts)
 
-        # Chunks schreiben, wenn geändert ODER vorher fehlend
+        # Chunks (wenn geändert ODER vorher keine vorhanden)
         if chunk_pls and (changed or chunks_missing):
             chunks_name, chunk_pts = points_for_chunks(cfg.prefix, chunk_pls, vecs)
             upsert_batch(client, chunks_name, chunk_pts)
 
-        # Edges schreiben, wenn vorhanden und (geändert ODER vorher fehlend)
+        # Edges (wenn geändert ODER vorher keine vorhanden)
         if edges and (changed or edges_missing):
             edges_name, edge_pts = points_for_edges(cfg.prefix, edges)
             upsert_batch(client, edges_name, edge_pts)