Hitonabi
47f7a85510
Die Ampel-Nachruestung deckte 317/337 vorbestehende ruff-Verstoesse im ganzen Repo auf. Aufgeraeumt: - ruff.toml: intentionale Muster als Projekt-Politik ausgenommen (BLE001 blind-except, S110/S112 try-except-pass/continue, PLW1510 subprocess-best-effort, B008 FastAPI- Depends/File-Idiom, EXE001 Shebang, + wenige Stil-Regeln). __init__.py-Re-Exports geschuetzt (F401). - ruff --fix: 128 mechanische (Import-Sortierung, PEP585/604-Annotationen, tote Imports, ueberfluessige noqa) auto-behoben. - 12 echte Reste von Hand: PERF402/102, PLC3002 (Lambda->walrus), ISC004 (String-Concat geklammert), F841/RUF059 (ungenutzte Vars), PIE810 (startswith-Tuple), UP031 (f-string), UP035 (veraltete typing-Imports). Ergebnis: 'ruff check .' = 0, 'compileall' grün. Kein Verhaltenswechsel (nur Stil/Modernisierung). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
93 lines
2.7 KiB
Python
93 lines
2.7 KiB
Python
"""Memory-Endpoints (geteiltes Gedächtnis). LAN-only, kein Token in 2.0-Phase 3."""
|
|
|
|
import time
|
|
|
|
from fastapi import APIRouter, HTTPException
|
|
from pydantic import BaseModel
|
|
from services import memory
|
|
from services.voice_metrics import park, record_stage
|
|
|
|
router = APIRouter(prefix="/api")
|
|
|
|
|
|
class MemIn(BaseModel):
|
|
content: str
|
|
category: str = "knowledge"
|
|
source: str = "manual"
|
|
|
|
|
|
class MemUp(BaseModel):
|
|
content: str | None = None
|
|
category: str | None = None
|
|
|
|
|
|
class DedupeIn(BaseModel):
|
|
apply: bool = False
|
|
threshold: float = 0.85
|
|
|
|
|
|
class LearnIn(BaseModel):
|
|
text: str | None = None
|
|
messages: list[dict] | None = None
|
|
source: str = "auto"
|
|
category: str = "knowledge"
|
|
|
|
|
|
@router.get("/memory/export")
|
|
def export() -> dict:
|
|
return memory.export_text()
|
|
|
|
|
|
@router.get("/memory/graph")
|
|
def graph(min_score: float = 0.45, top_k: int = 3) -> dict:
|
|
"""Fakten als Ähnlichkeits-Graph (Knoten + semantische Kanten) für die Visualisierung."""
|
|
return memory.graph(min_score=min_score, top_k=top_k)
|
|
|
|
|
|
@router.post("/memory/learn", status_code=201)
|
|
def learn(body: LearnIn) -> dict:
|
|
"""Auto-Lernen: Gesprächs-Turns/Text durchreichen → Mem0 extrahiert Fakten selbst."""
|
|
return memory.learn(text=body.text, messages=body.messages,
|
|
source=body.source, category=body.category)
|
|
|
|
|
|
@router.post("/memory/dedupe")
|
|
def dedupe(body: DedupeIn) -> dict:
|
|
return memory.dedupe(apply=body.apply, threshold=body.threshold)
|
|
|
|
|
|
@router.get("/memory")
|
|
def list_mem(q: str = "", category: str = "") -> list[dict]:
|
|
# Semantischer Retrieve (q gesetzt) = u.a. Hermes' Mem0-Prefetch VOR jedem Turn. Dauer messen
|
|
# + parken, damit der laufende Voice-Chat-Turn sie als Unter-Detail seiner Hirn-Zeit einsammelt.
|
|
if q:
|
|
_t0 = time.perf_counter()
|
|
res = memory.list_memories(q=q, category=category)
|
|
_ms = (time.perf_counter() - _t0) * 1000.0
|
|
record_stage("memory_retrieve", _ms)
|
|
park("retrieve", _ms)
|
|
return res
|
|
return memory.list_memories(q=q, category=category)
|
|
|
|
|
|
@router.post("/memory", status_code=201)
|
|
def add(body: MemIn) -> dict:
|
|
if body.category not in memory.CATEGORIES:
|
|
raise HTTPException(400, f"Kategorie '{body.category}' unbekannt.")
|
|
return memory.add_memory(body.content, body.category, body.source)
|
|
|
|
|
|
@router.put("/memory/{mid}")
|
|
def update(mid: str, body: MemUp) -> dict:
|
|
res = memory.update_memory(mid, content=body.content, category=body.category)
|
|
if not res:
|
|
raise HTTPException(404, "Eintrag nicht gefunden")
|
|
return res
|
|
|
|
|
|
@router.delete("/memory/{mid}")
|
|
def delete(mid: str) -> dict:
|
|
if not memory.delete_memory(mid):
|
|
raise HTTPException(404, "Eintrag nicht gefunden")
|
|
return {"ok": True}
|