Files
service-finder/backend/app/services/document_service.py

135 lines
5.6 KiB
Python
Executable File

# /opt/docker/dev/service_finder/backend/app/services/document_service.py
import os
import logging
import asyncio
from PIL import Image
from uuid import uuid4
from datetime import datetime, timezone
from fastapi import UploadFile, BackgroundTasks, HTTPException
from sqlalchemy.ext.asyncio import AsyncSession
from sqlalchemy import select, func, and_
from app.models.document import Document
from app.models.identity import User
from app.services.config_service import config # 2.0 Dinamikus beállítások
from app.workers.ocr.robot_1_ocr_processor import OCRRobot # Robot 1 hívása
logger = logging.getLogger("Document-Service-2.0")
class DocumentService:
"""
Document Service 2.0 - Admin-vezérelt Pipeline.
Feladata: Tárolás, Optimalizálás, Kvótamanagement és Robot Trigger.
"""
@staticmethod
async def process_upload(
db: AsyncSession,
user_id: int,
file: UploadFile,
parent_type: str, # pl. "asset", "organization", "transfer"
parent_id: str,
doc_type: str, # pl. "invoice", "registration_card", "sale_contract"
background_tasks: BackgroundTasks
):
try:
# --- 1. ADMIN KVÓTA ELLENŐRZÉS ---
user_stmt = select(User).where(User.id == user_id)
user = (await db.execute(user_stmt)).scalar_one()
# Lekérjük a csomagnak megfelelő havi limitet (pl. Free: 1, Premium: 10)
limits = await config.get_setting(db, "ocr_monthly_limit", default={"free": 1, "premium": 10, "vip": 100})
user_role = user.role.value if hasattr(user.role, 'value') else str(user.role)
allowed_ocr = limits.get(user_role, 1)
# Megnézzük a havi felhasználást
now = datetime.now(timezone.utc)
start_of_month = now.replace(day=1, hour=0, minute=0, second=0, microsecond=0)
count_stmt = select(func.count(Document.id)).where(
and_(
Document.user_id == user_id,
Document.created_at >= start_of_month
)
)
used_count = (await db.execute(count_stmt)).scalar()
if used_count >= allowed_ocr:
raise HTTPException(
status_code=403,
detail=f"Havi dokumentum limit túllépve ({allowed_ocr}). Válts csomagot a folytatáshoz!"
)
# --- 2. DINAMIKUS TÁROLÁS ÉS OPTIMALIZÁLÁS ---
file_uuid = str(uuid4())
nas_base = await config.get_setting(db, "storage_nas_path", default="/mnt/nas/app_data")
# Útvonal sablonok az adminból (Pl. "vault/{parent_type}/{parent_id}")
vault_dir = os.path.join(nas_base, parent_type, parent_id, "vault")
thumb_dir = os.path.join(getattr(config, "STATIC_DIR", "static"), "previews", parent_type, parent_id)
os.makedirs(vault_dir, exist_ok=True)
os.makedirs(thumb_dir, exist_ok=True)
content = await file.read()
temp_path = f"/tmp/{file_uuid}_{file.filename}"
with open(temp_path, "wb") as f: f.write(content)
# Kép feldolgozása PIL-lel
img = Image.open(temp_path)
# Thumbnail generálás (SSD/Static területre)
thumb_filename = f"{file_uuid}_thumb.webp"
thumb_path = os.path.join(thumb_dir, thumb_filename)
thumb_img = img.copy()
thumb_img.thumbnail((300, 300))
thumb_img.save(thumb_path, "WEBP", quality=80)
# Optimalizált eredeti (NAS / Vault területre)
max_width = await config.get_setting(db, "img_max_width", default=1600)
vault_filename = f"{file_uuid}.webp"
vault_path = os.path.join(vault_dir, vault_filename)
if img.width > max_width:
ratio = max_width / img.width
img = img.resize((max_width, int(img.height * ratio)), Image.Resampling.LANCZOS)
img.save(vault_path, "WEBP", quality=85)
# --- 3. MENTÉS ---
new_doc = Document(
id=uuid4(),
user_id=user_id,
parent_type=parent_type,
parent_id=parent_id,
doc_type=doc_type,
original_name=file.filename,
file_hash=file_uuid,
file_ext="webp",
mime_type="image/webp",
file_size=os.path.getsize(vault_path),
has_thumbnail=True,
thumbnail_path=f"/static/previews/{parent_type}/{parent_id}/{thumb_filename}",
status="uploaded"
)
db.add(new_doc)
await db.flush()
# --- 4. ROBOT TRIGGER (OCR AUTOMATIZMUS) ---
# Megnézzük, hogy ez a típus (pl. invoice) igényel-e automatikus OCR-t
auto_ocr_types = await config.get_setting(db, "ocr_auto_trigger_types", default=["invoice", "registration_card", "sale_contract"])
if doc_type in auto_ocr_types:
# Robot 1 (OCR) sorba állítása háttérfolyamatként
background_tasks.add_task(OCRRobot.process_document, db, new_doc.id)
new_doc.status = "processing"
logger.info(f"🤖 Robot 1 (OCR) riasztva: {new_doc.id}")
await db.commit()
os.remove(temp_path)
return new_doc
except Exception as e:
if 'temp_path' in locals() and os.path.exists(temp_path): os.remove(temp_path)
logger.error(f"Document Upload Error: {e}")
raise e