299fac3904
- Implement kordoc /parse endpoint (HWP/HWPX/PDF via kordoc lib, text files direct read, images flagged for OCR) - Add queue consumer with APScheduler (1min interval, stage chaining extract→classify→embed, stale item recovery, retry logic) - Add extract worker (kordoc HTTP call + direct text read) - Add classify worker (Qwen3.5 AI classification with think-tag stripping and robust JSON extraction from AI responses) - Add embed worker (GPU server nomic-embed-text, graceful failure) - Add DEVONthink migration script with folder mapping for 16 DBs, dry-run mode, batch commits, and idempotent file_path UNIQUE - Enhance ai/client.py with strip_thinking() and parse_json_response() Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
108 lines
3.8 KiB
Python
108 lines
3.8 KiB
Python
"""AI 추상화 레이어 — 통합 클라이언트. 기본값은 항상 Qwen3.5."""
|
|
|
|
import json
|
|
import re
|
|
from pathlib import Path
|
|
|
|
import httpx
|
|
|
|
from core.config import settings
|
|
|
|
|
|
def strip_thinking(text: str) -> str:
|
|
"""Qwen3.5의 <think>...</think> 블록 제거"""
|
|
return re.sub(r"<think>.*?</think>", "", text, flags=re.DOTALL).strip()
|
|
|
|
|
|
def parse_json_response(raw: str) -> dict | None:
|
|
"""AI 응답에서 JSON 객체 추출 (think 태그, 코드블록 등 제거)"""
|
|
cleaned = strip_thinking(raw)
|
|
# 코드블록 내부 JSON 추출
|
|
code_match = re.search(r"```(?:json)?\s*(\{.*?\})\s*```", cleaned, re.DOTALL)
|
|
if code_match:
|
|
cleaned = code_match.group(1)
|
|
# 마지막 유효 JSON 객체 찾기
|
|
matches = list(re.finditer(r"\{[^{}]*(?:\{[^{}]*\}[^{}]*)*\}", cleaned, re.DOTALL))
|
|
for m in reversed(matches):
|
|
try:
|
|
return json.loads(m.group())
|
|
except json.JSONDecodeError:
|
|
continue
|
|
# 최후 시도: 전체 텍스트를 JSON으로
|
|
try:
|
|
return json.loads(cleaned)
|
|
except json.JSONDecodeError:
|
|
return None
|
|
|
|
# 프롬프트 로딩
|
|
PROMPTS_DIR = Path(__file__).parent.parent / "prompts"
|
|
|
|
|
|
def _load_prompt(name: str) -> str:
|
|
return (PROMPTS_DIR / name).read_text(encoding="utf-8")
|
|
|
|
|
|
CLASSIFY_PROMPT = _load_prompt("classify.txt") if (PROMPTS_DIR / "classify.txt").exists() else ""
|
|
|
|
|
|
class AIClient:
|
|
"""AI Gateway를 통한 통합 클라이언트. 기본값은 항상 Qwen3.5."""
|
|
|
|
def __init__(self):
|
|
self.ai = settings.ai
|
|
self._http = httpx.AsyncClient(timeout=120)
|
|
|
|
async def classify(self, text: str) -> dict:
|
|
"""문서 분류 — 항상 primary(Qwen3.5) 사용"""
|
|
prompt = CLASSIFY_PROMPT.replace("{document_text}", text)
|
|
response = await self._call_chat(self.ai.primary, prompt)
|
|
return response
|
|
|
|
async def summarize(self, text: str, force_premium: bool = False) -> str:
|
|
"""문서 요약 — 기본 Qwen3.5, 장문이거나 명시적 요청 시만 Claude"""
|
|
model = self.ai.primary
|
|
if force_premium or len(text) > 15000:
|
|
model = self.ai.premium
|
|
return await self._call_chat(model, f"다음 문서를 500자 이내로 요약해주세요:\n\n{text}")
|
|
|
|
async def embed(self, text: str) -> list[float]:
|
|
"""벡터 임베딩 — GPU 서버 전용"""
|
|
response = await self._http.post(
|
|
self.ai.embedding.endpoint,
|
|
json={"model": self.ai.embedding.model, "prompt": text},
|
|
)
|
|
response.raise_for_status()
|
|
return response.json()["embedding"]
|
|
|
|
async def ocr(self, image_bytes: bytes) -> str:
|
|
"""이미지 OCR — GPU 서버 전용"""
|
|
# TODO: Qwen2.5-VL-7B 비전 모델 호출 구현
|
|
raise NotImplementedError("OCR는 Phase 1에서 구현")
|
|
|
|
async def _call_chat(self, model_config, prompt: str) -> str:
|
|
"""OpenAI 호환 API 호출 + 자동 폴백"""
|
|
try:
|
|
return await self._request(model_config, prompt)
|
|
except (httpx.TimeoutException, httpx.ConnectError):
|
|
if model_config == self.ai.primary:
|
|
return await self._request(self.ai.fallback, prompt)
|
|
raise
|
|
|
|
async def _request(self, model_config, prompt: str) -> str:
|
|
"""단일 모델 API 호출"""
|
|
response = await self._http.post(
|
|
model_config.endpoint,
|
|
json={
|
|
"model": model_config.model,
|
|
"messages": [{"role": "user", "content": prompt}],
|
|
"max_tokens": model_config.max_tokens,
|
|
},
|
|
timeout=model_config.timeout,
|
|
)
|
|
response.raise_for_status()
|
|
data = response.json()
|
|
return data["choices"][0]["message"]["content"]
|
|
|
|
async def close(self):
|
|
await self._http.aclose()
|