0063df0a87
- Бэкенд: FastAPI + SQLAlchemy async + PostgreSQL, JWT-аутентификация, Qdrant multi-tenant векторное хранилище, Celery воркер - Фронтенд: React + Vite + TypeScript SPA (логин, регистрация, дашборд, поиск по коллекциям, настройки) - Инфраструктура: docker-compose, Helm-чарт, GitHub Actions CI, Prometheus/Grafana мониторинг - Embedding через OpenRouter (nvidia/llama-nemotron-embed-vl-1b-v2:free) - LLM: google/gemma-4-31b-it:free через OpenRouter
113 lines
3.6 KiB
Python
113 lines
3.6 KiB
Python
import uuid
|
|
import json
|
|
import csv
|
|
import io
|
|
|
|
from fastapi import APIRouter, Depends, UploadFile, File, HTTPException, status
|
|
from sqlalchemy import select
|
|
from sqlalchemy.ext.asyncio import AsyncSession
|
|
|
|
from app.auth.api_key import get_user_by_api_key
|
|
from app.auth.deps import get_current_user
|
|
from app.database import get_db
|
|
from app.models.collection import Collection
|
|
from app.models.schemas import TicketInput, TicketsUpload, TicketsUploadResponse
|
|
from app.models.user import User
|
|
from app.services.chunking import build_chunk
|
|
from app.services.embedding import embed_texts
|
|
from app.services.vector_store import upsert_chunks
|
|
|
|
router = APIRouter(prefix="/api/collections", tags=["tickets"])
|
|
|
|
|
|
async def process_tickets(
|
|
db: AsyncSession,
|
|
collection_id: uuid.UUID,
|
|
user: User,
|
|
tickets: list[TicketInput],
|
|
) -> TicketsUploadResponse:
|
|
result = await db.execute(
|
|
select(Collection).where(Collection.id == collection_id, Collection.user_id == user.id)
|
|
)
|
|
collection = result.scalar_one_or_none()
|
|
if not collection:
|
|
raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="Collection not found")
|
|
|
|
processed = 0
|
|
skipped = 0
|
|
chunks = []
|
|
|
|
for t in tickets:
|
|
ticket_dict = {
|
|
"ticket_id": t.ticket_id,
|
|
"category": t.category,
|
|
"description": t.description,
|
|
"client": t.client,
|
|
"messages": [{"role": m.role, "text": m.text} for m in t.messages],
|
|
}
|
|
chunk = build_chunk(ticket_dict)
|
|
if chunk:
|
|
chunk["tenant_id"] = str(user.id)
|
|
chunk["collection_id"] = str(collection_id)
|
|
chunks.append(chunk)
|
|
processed += 1
|
|
else:
|
|
skipped += 1
|
|
|
|
if chunks:
|
|
search_texts = [c["search_text"] for c in chunks]
|
|
vectors = await embed_texts(search_texts)
|
|
await upsert_chunks(str(user.id), chunks, vectors)
|
|
|
|
collection.ticket_count += processed
|
|
await db.commit()
|
|
|
|
return TicketsUploadResponse(
|
|
processed=processed,
|
|
skipped=skipped,
|
|
collection_ticket_count=collection.ticket_count,
|
|
)
|
|
|
|
|
|
@router.post("/{collection_id}/tickets", response_model=TicketsUploadResponse)
|
|
async def upload_tickets_json(
|
|
collection_id: uuid.UUID,
|
|
body: TicketsUpload,
|
|
user: User = Depends(get_user_by_api_key),
|
|
db: AsyncSession = Depends(get_db),
|
|
):
|
|
return await process_tickets(db, collection_id, user, body.tickets)
|
|
|
|
|
|
@router.post("/{collection_id}/tickets/upload", response_model=TicketsUploadResponse)
|
|
async def upload_tickets_file(
|
|
collection_id: uuid.UUID,
|
|
file: UploadFile = File(...),
|
|
user: User = Depends(get_current_user),
|
|
db: AsyncSession = Depends(get_db),
|
|
):
|
|
content = await file.read()
|
|
|
|
if file.filename.endswith(".json"):
|
|
data = json.loads(content)
|
|
tickets = [TicketInput(**t) for t in data] if isinstance(data, list) else [TicketInput(**data)]
|
|
elif file.filename.endswith(".csv"):
|
|
text = content.decode("utf-8-sig")
|
|
reader = csv.DictReader(io.StringIO(text))
|
|
tickets = []
|
|
for row in reader:
|
|
tickets.append(
|
|
TicketInput(
|
|
ticket_id=row.get("ticket_id", ""),
|
|
category=row.get("category", ""),
|
|
description=row.get("description", ""),
|
|
client=row.get("client", ""),
|
|
)
|
|
)
|
|
else:
|
|
raise HTTPException(
|
|
status_code=status.HTTP_400_BAD_REQUEST,
|
|
detail="Unsupported format. Use JSON or CSV.",
|
|
)
|
|
|
|
return await process_tickets(db, collection_id, user, tickets)
|