app/core/type_registry.py aktualisiert

app/core/type_registry.py

@@ -1,79 +1,111 @@
-# -*- coding: utf-8 -*-
 """
-app/core/type_registry.py — mindnet · WP-03  (Version 1.0.0)
+type_registry.py v1.0.0
 
 Zweck:
-  - Lädt eine optionale Typen-Registry aus config/types.yaml|json.
-  - Liefert pro `type` eine Konfiguration (z. B. Chunk-Profile, Standard-Edges,
-    Retriever-Gewichtungen). Wird in import/chunk/edges integriert.
+  - Optionale, konfigurierbare Type-Registry laden (YAML/JSON), um pro "type"
+    (aus Frontmatter) Chunk-Profile, Default-Edges und optionale
+    Retriever-Gewichte bereitzustellen – ohne bestehende Funktionen zu brechen.
 
-Verhalten:
-  - Fehlt die Datei oder der Typ → es werden Defaults genutzt.
-  - Registry wird gecacht (lazy-load).
+Kompatibilität:
+  - Keine Abhängigkeiten von anderen Modulen.
+  - Keine harten Fehler, wenn Registry fehlt oder unvollständig ist.
+
+Nutzung:
+  from app.core.type_registry import (
+      load_type_registry, resolve_chunk_profile, get_edge_defaults_for_type,
+      get_retriever_weight_for_type
+  )
+
+Umgebungsvariablen:
+  TYPE_REGISTRY_PATH (default: "config/types.yaml")
+
+Format (Beispiel):
+  version: 1.0
+  types:
+    concept:
+      chunk_profile: long
+      edge_defaults: [references, related_to]
+      retriever_weight: 1.0
+    task:
+      chunk_profile: short
+      edge_defaults: [depends_on, belongs_to]
+      retriever_weight: 0.8
+    experience:
+      chunk_profile: medium
+      edge_defaults: [derived_from, inspired_by]
+      retriever_weight: 0.9
 """
 
-__version__ = "1.0.0"
-
+from __future__ import annotations
 import os
 import json
-from typing import Any, Dict
+from typing import Any, Dict, List, Optional
+
 try:
     import yaml  # type: ignore
-except Exception:
-    yaml = None  # optional
+except Exception:  # yaml ist optional; JSON wird ebenfalls unterstützt
+    yaml = None  # type: ignore
 
-_CACHE: Dict[str, Any] = {}
+_CACHE: Dict[str, Dict[str, Any]] = {}
 
-DEFAULT_REG = {
-    "version": "1.0",
-    "types": {
-        "concept": {
-            "chunk_profile": "medium",
-            "edge_defaults": ["references", "related_to"],
-            "retriever_weight": 1.0
-        },
-        "task": {
-            "chunk_profile": "short",
-            "edge_defaults": ["depends_on", "belongs_to"],
-            "retriever_weight": 0.8
-        },
-        "experience": {
-            "chunk_profile": "medium",
-            "edge_defaults": ["derived_from"],
-            "retriever_weight": 0.9
-        }
-    }
-}
-
-def load_type_registry(path: str = "config/types.yaml", silent: bool = False) -> Dict[str, Any]:
-    global _CACHE
-    if _CACHE:
-        return _CACHE
-
-    def load_yaml(p: str) -> Dict[str, Any]:
-        if yaml is None:
-            return {}
-        with open(p, "r", encoding="utf-8") as f:
-            return yaml.safe_load(f) or {}
-
-    def load_json(p: str) -> Dict[str, Any]:
-        with open(p, "r", encoding="utf-8") as f:
-            return json.load(f) or {}
-
-    reg: Dict[str, Any] = {}
-    if os.path.exists(path):
+def _safe_load_yaml_or_json(path: str) -> Dict[str, Any]:
+    if not os.path.exists(path):
+        return {}
+    try:
+        with open(path, "r", encoding="utf-8") as f:
+            data = f.read()
+    except Exception:
+        return {}
+    # YAML bevorzugen, wenn verfügbar und Datei nach YAML aussieht
+    if path.lower().endswith((".yaml", ".yml")) and yaml is not None:
         try:
-            if path.endswith(".yaml") or path.endswith(".yml"):
-                reg = load_yaml(path)
-            elif path.endswith(".json"):
-                reg = load_json(path)
-        except Exception as e:
-            if not silent:
-                print(f"[type_registry] WARN: failed to load {path}: {e}")
+            return yaml.safe_load(data) or {}
+        except Exception:
+            return {}
+    # JSON fallback
+    try:
+        return json.loads(data)
+    except Exception:
+        return {}
 
-    if not reg:
-        # Fallback
-        reg = DEFAULT_REG
+def load_type_registry(path: Optional[str] = None) -> Dict[str, Any]:
+    """
+    Lädt einmalig die Registry und cached sie.
+    Fehlt sie, wird ein leeres Dict geliefert (keine Fehler).
+    """
+    key = path or os.getenv("TYPE_REGISTRY_PATH", "config/types.yaml")
+    if key in _CACHE:
+        return _CACHE[key]
+    obj = _safe_load_yaml_or_json(key)
+    if not isinstance(obj, dict):
+        obj = {}
+    _CACHE[key] = obj
+    return obj
 
-    _CACHE = reg
-    return _CACHE
+def _types_map(reg: Dict[str, Any]) -> Dict[str, Any]:
+    return reg.get("types", {}) if isinstance(reg, dict) else {}
+
+def resolve_chunk_profile(note_type: str, default_profile: str = "default") -> str:
+    reg = load_type_registry()
+    tmap = _types_map(reg)
+    entry = tmap.get(note_type, {})
+    return str(entry.get("chunk_profile", default_profile))
+
+def get_edge_defaults_for_type(note_type: str) -> List[str]:
+    reg = load_type_registry()
+    tmap = _types_map(reg)
+    entry = tmap.get(note_type, {})
+    v = entry.get("edge_defaults", [])
+    if not isinstance(v, list):
+        return []
+    return [str(x) for x in v]
+
+def get_retriever_weight_for_type(note_type: str) -> Optional[float]:
+    reg = load_type_registry()
+    tmap = _types_map(reg)
+    entry = tmap.get(note_type, {})
+    v = entry.get("retriever_weight", None)
+    try:
+        return float(v) if v is not None else None
+    except Exception:
+        return None