feat: 3-mode inspection with tabbed UI + batch upload

- Add batch inspection backend (multipart upload, SSE streaming, MongoDB)
- Add tabbed UI (single page / site crawling / batch upload) on home and history pages
- Add batch inspection progress, result pages with 2-panel layout
- Rename "사이트 전체" to "사이트 크롤링" across codebase
- Add python-multipart dependency for file upload
- Consolidate nginx SSE location for all inspection types

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
jungwoo choi
2026-02-13 19:15:27 +09:00
parent 9bb844c5e1
commit 8326c84be9
32 changed files with 3700 additions and 61 deletions

View File

@ -26,6 +26,10 @@ class Settings(BaseSettings):
SITE_MAX_DEPTH: int = 2
SITE_CONCURRENCY: int = 8
# Batch inspection
BATCH_MAX_URLS: int = 200
BATCH_CONCURRENCY: int = 8
# Application
PROJECT_NAME: str = "Web Inspector API"

View File

@ -29,6 +29,11 @@ async def connect_db() -> None:
await _db.site_inspections.create_index([("domain", 1), ("created_at", -1)])
await _db.site_inspections.create_index([("created_at", -1)])
# Create indexes - batch_inspections
await _db.batch_inspections.create_index("batch_inspection_id", unique=True)
await _db.batch_inspections.create_index([("name", 1), ("created_at", -1)])
await _db.batch_inspections.create_index([("created_at", -1)])
# Verify connection
await _client.admin.command("ping")
logger.info("MongoDB connected successfully: %s", settings.DB_NAME)

View File

@ -10,7 +10,7 @@ from fastapi.middleware.cors import CORSMiddleware
from app.core.database import connect_db, close_db
from app.core.redis import connect_redis, close_redis
from app.routers import health, inspections, reports, site_inspections
from app.routers import health, inspections, reports, site_inspections, batch_inspections
# Configure logging
logging.basicConfig(
@ -57,3 +57,4 @@ app.include_router(health.router, prefix="/api", tags=["Health"])
app.include_router(inspections.router, prefix="/api", tags=["Inspections"])
app.include_router(reports.router, prefix="/api", tags=["Reports"])
app.include_router(site_inspections.router, prefix="/api", tags=["Site Inspections"])
app.include_router(batch_inspections.router, prefix="/api", tags=["Batch Inspections"])

View File

@ -0,0 +1,86 @@
"""
Pydantic models for batch inspection request/response validation.
Batch inspection allows inspecting multiple URLs from a file upload
without any crawling phase - URLs are inspected directly.
"""
from pydantic import BaseModel, Field
from typing import Optional
from datetime import datetime
from enum import Enum
from app.models.site_schemas import AggregateScores, PageStatus
# --- Enums ---
class BatchInspectionStatus(str, Enum):
"""Batch inspection status (no crawling phase)."""
INSPECTING = "inspecting"
COMPLETED = "completed"
ERROR = "error"
# --- Core Data Models ---
class BatchPage(BaseModel):
"""배치 검사 개별 페이지 (크롤링 없이 직접 지정된 URL)."""
url: str
depth: int = 0
parent_url: Optional[str] = None
inspection_id: Optional[str] = None
status: PageStatus = PageStatus.PENDING
title: Optional[str] = None
overall_score: Optional[int] = None
grade: Optional[str] = None
class BatchInspectionConfig(BaseModel):
"""배치 검사 설정."""
concurrency: int = 4
# --- Response Models ---
class StartBatchInspectionResponse(BaseModel):
"""배치 검사 시작 응답."""
batch_inspection_id: str
status: str = "inspecting"
name: str
total_urls: int
stream_url: str
class BatchInspectionResult(BaseModel):
"""배치 검사 전체 결과."""
batch_inspection_id: str
name: str
status: BatchInspectionStatus
created_at: datetime
completed_at: Optional[datetime] = None
config: BatchInspectionConfig
source_urls: list[str] = []
discovered_pages: list[BatchPage] = []
aggregate_scores: Optional[AggregateScores] = None
class BatchInspectionListItem(BaseModel):
"""배치 검사 목록 항목 (요약)."""
batch_inspection_id: str
name: str
status: BatchInspectionStatus
created_at: datetime
total_urls: int = 0
pages_inspected: int = 0
overall_score: Optional[int] = None
grade: Optional[str] = None
class BatchInspectionPaginatedResponse(BaseModel):
"""배치 검사 목록 페이지네이션 응답."""
items: list[BatchInspectionListItem]
total: int
page: int
limit: int
total_pages: int

View File

@ -0,0 +1,286 @@
"""
Batch inspections router.
Handles batch inspection lifecycle:
- Start batch inspection (multipart file upload + inspect all URLs)
- SSE stream for real-time progress
- Get batch inspection result
- List batch inspections (history)
IMPORTANT: Static paths (/batch-inspections) must be registered BEFORE
dynamic paths (/batch-inspections/{id}) to avoid routing conflicts.
"""
import json
import logging
from urllib.parse import urlparse
from fastapi import APIRouter, File, Form, HTTPException, Query, UploadFile
from sse_starlette.sse import EventSourceResponse
from app.core.config import get_settings
from app.core.database import get_db
from app.core.redis import get_redis
from app.models.batch_schemas import StartBatchInspectionResponse
from app.services.batch_inspection_service import BatchInspectionService
logger = logging.getLogger(__name__)
router = APIRouter()
# Maximum upload file size: 1MB
MAX_FILE_SIZE = 1 * 1024 * 1024 # 1MB
def _get_service() -> BatchInspectionService:
"""Get BatchInspectionService instance."""
db = get_db()
redis = get_redis()
return BatchInspectionService(db=db, redis=redis)
def _validate_url(url: str) -> bool:
"""Validate that a URL has http or https scheme."""
try:
parsed = urlparse(url)
return parsed.scheme in ("http", "https") and bool(parsed.netloc)
except Exception:
return False
def _parse_url_file(content: str) -> list[str]:
"""
Parse URL file content into a list of valid URLs.
- One URL per line
- Skip empty lines
- Skip lines starting with # (comments)
- Strip whitespace
"""
urls = []
for line in content.splitlines():
line = line.strip()
if not line or line.startswith("#"):
continue
urls.append(line)
return urls
# ============================================================
# POST /api/batch-inspections -- Start batch inspection
# ============================================================
@router.post("/batch-inspections", status_code=202)
async def start_batch_inspection(
file: UploadFile = File(...),
name: str = Form(...),
concurrency: int = Form(default=4),
):
"""
Start a new batch inspection from an uploaded URL file.
Returns 202 Accepted with batch_inspection_id immediately.
Inspection runs asynchronously in the background.
File format: .txt, one URL per line, max 1MB, max 200 URLs.
Lines starting with # are treated as comments and ignored.
"""
settings = get_settings()
# Validate file extension
if file.filename and not file.filename.lower().endswith(".txt"):
raise HTTPException(
status_code=422,
detail="텍스트 파일(.txt)만 업로드 가능합니다",
)
# Validate file size
content_bytes = await file.read()
if len(content_bytes) > MAX_FILE_SIZE:
raise HTTPException(
status_code=422,
detail="파일 크기가 1MB를 초과합니다",
)
# Parse file content
try:
content = content_bytes.decode("utf-8")
except UnicodeDecodeError:
raise HTTPException(
status_code=422,
detail="파일 인코딩이 올바르지 않습니다 (UTF-8만 지원)",
)
urls = _parse_url_file(content)
if not urls:
raise HTTPException(
status_code=422,
detail="파일에 유효한 URL이 없습니다",
)
# Validate URL count
if len(urls) > settings.BATCH_MAX_URLS:
raise HTTPException(
status_code=422,
detail=f"URL 수가 최대 허용치({settings.BATCH_MAX_URLS}개)를 초과합니다 (입력: {len(urls)}개)",
)
# Validate each URL
invalid_urls = [url for url in urls if not _validate_url(url)]
if invalid_urls:
# Show first 5 invalid URLs
sample = invalid_urls[:5]
detail = f"유효하지 않은 URL이 포함되어 있습니다: {', '.join(sample)}"
if len(invalid_urls) > 5:
detail += f"{len(invalid_urls) - 5}"
raise HTTPException(
status_code=422,
detail=detail,
)
# Validate concurrency
concurrency = max(1, min(concurrency, settings.BATCH_CONCURRENCY))
# Validate name
name = name.strip()
if not name:
raise HTTPException(
status_code=422,
detail="배치 검사 이름을 입력해주세요",
)
service = _get_service()
try:
batch_inspection_id = await service.start_batch_inspection(
name=name,
urls=urls,
concurrency=concurrency,
)
except Exception as e:
logger.error("Failed to start batch inspection: %s", str(e))
raise HTTPException(
status_code=400,
detail="배치 검사를 시작할 수 없습니다",
)
return StartBatchInspectionResponse(
batch_inspection_id=batch_inspection_id,
status="inspecting",
name=name,
total_urls=len(urls),
stream_url=f"/api/batch-inspections/{batch_inspection_id}/stream",
)
# ============================================================
# GET /api/batch-inspections -- List batch inspections (history)
# IMPORTANT: This MUST be before /{batch_inspection_id} routes
# ============================================================
@router.get("/batch-inspections")
async def list_batch_inspections(
page: int = Query(default=1, ge=1),
limit: int = Query(default=20, ge=1, le=100),
name: str = Query(default=None, description="이름 검색 필터"),
):
"""Get paginated batch inspection history."""
service = _get_service()
result = await service.get_batch_inspection_list(
page=page,
limit=limit,
name_filter=name,
)
return result
# ============================================================
# GET /api/batch-inspections/{batch_inspection_id}/stream -- SSE
# ============================================================
@router.get("/batch-inspections/{batch_inspection_id}/stream")
async def batch_inspection_stream(batch_inspection_id: str):
"""
Stream batch inspection progress via Server-Sent Events.
Events:
- connected: { batch_inspection_id, message }
- page_start: { batch_inspection_id, page_url, page_index }
- page_complete: { batch_inspection_id, page_url, inspection_id, overall_score, grade }
- aggregate_update: { batch_inspection_id, pages_inspected, pages_total, overall_score, grade }
- complete: { batch_inspection_id, status: "completed", aggregate_scores: {...} }
- error: { batch_inspection_id, message }
"""
async def event_generator():
redis = get_redis()
pubsub = redis.pubsub()
channel = f"batch-inspection:{batch_inspection_id}:events"
await pubsub.subscribe(channel)
try:
# Send initial connected event
yield {
"event": "connected",
"data": json.dumps({
"batch_inspection_id": batch_inspection_id,
"message": "SSE 연결 완료",
}, ensure_ascii=False),
}
# Listen for Pub/Sub messages
async for message in pubsub.listen():
if message["type"] == "message":
event_data = json.loads(message["data"])
event_type = event_data.pop("event_type", "progress")
yield {
"event": event_type,
"data": json.dumps(event_data, ensure_ascii=False),
}
# End stream on complete or error
if event_type in ("complete", "error"):
break
except Exception as e:
logger.error(
"SSE stream error for batch %s: %s",
batch_inspection_id, str(e),
)
yield {
"event": "error",
"data": json.dumps({
"batch_inspection_id": batch_inspection_id,
"status": "error",
"message": "스트리밍 중 오류가 발생했습니다",
}, ensure_ascii=False),
}
finally:
await pubsub.unsubscribe(channel)
await pubsub.aclose()
return EventSourceResponse(
event_generator(),
media_type="text/event-stream",
)
# ============================================================
# GET /api/batch-inspections/{batch_inspection_id} -- Get result
# ============================================================
@router.get("/batch-inspections/{batch_inspection_id}")
async def get_batch_inspection(batch_inspection_id: str):
"""Get batch inspection result by ID."""
service = _get_service()
result = await service.get_batch_inspection(batch_inspection_id)
if result is None:
raise HTTPException(
status_code=404,
detail="배치 검사 결과를 찾을 수 없습니다",
)
# Remove MongoDB _id field if present
result.pop("_id", None)
return result

View File

@ -0,0 +1,537 @@
"""
Batch inspection orchestration service.
Manages batch inspection lifecycle without crawling:
1. Accept a list of URLs (from uploaded file)
2. Parallel inspection of each URL (semaphore-controlled)
3. Aggregate score computation
4. Progress tracking via Redis Pub/Sub (SSE events)
5. Result storage in MongoDB (batch_inspections collection)
"""
import asyncio
import json
import logging
import uuid
from datetime import datetime, timezone
from typing import Optional
from motor.motor_asyncio import AsyncIOMotorDatabase
from redis.asyncio import Redis
from app.core.config import get_settings
from app.models.schemas import calculate_grade
from app.services.inspection_service import InspectionService
logger = logging.getLogger(__name__)
# Redis key TTLs
BATCH_RESULT_CACHE_TTL = 3600 # 1 hour
class BatchInspectionService:
"""Batch inspection orchestration service."""
def __init__(self, db: AsyncIOMotorDatabase, redis: Redis):
self.db = db
self.redis = redis
self.inspection_service = InspectionService(db=db, redis=redis)
# ------------------------------------------------------------------
# Public API
# ------------------------------------------------------------------
async def start_batch_inspection(
self,
name: str,
urls: list[str],
concurrency: int = 4,
) -> str:
"""
Start a batch inspection.
1. Generate batch_inspection_id
2. Create initial MongoDB document with status "inspecting"
3. Build discovered_pages array (depth=0, parent_url=None)
4. Launch background inspect-all task
5. Return batch_inspection_id
"""
settings = get_settings()
# Clamp concurrency to server-side limit
concurrency = min(concurrency, settings.BATCH_CONCURRENCY)
batch_inspection_id = str(uuid.uuid4())
# Build discovered_pages documents
discovered_pages = []
for url in urls:
discovered_pages.append({
"url": url,
"depth": 0,
"parent_url": None,
"inspection_id": None,
"status": "pending",
"title": None,
"overall_score": None,
"grade": None,
})
# Create initial document
doc = {
"batch_inspection_id": batch_inspection_id,
"name": name,
"status": "inspecting",
"created_at": datetime.now(timezone.utc),
"completed_at": None,
"config": {
"concurrency": concurrency,
},
"source_urls": urls,
"discovered_pages": discovered_pages,
"aggregate_scores": None,
}
await self.db.batch_inspections.insert_one(doc)
logger.info(
"Batch inspection started: id=%s, name=%s, total_urls=%d, concurrency=%d",
batch_inspection_id, name, len(urls), concurrency,
)
# Launch background task
asyncio.create_task(
self._inspect_all(batch_inspection_id, urls, concurrency)
)
return batch_inspection_id
async def get_batch_inspection(self, batch_inspection_id: str) -> Optional[dict]:
"""Get batch inspection result by ID (cache-first)."""
# Try Redis cache first
cache_key = f"batch-inspection:result:{batch_inspection_id}"
cached = await self.redis.get(cache_key)
if cached:
return json.loads(cached)
# Fetch from MongoDB
doc = await self.db.batch_inspections.find_one(
{"batch_inspection_id": batch_inspection_id},
{"_id": 0},
)
if doc:
# Only cache completed results
if doc.get("status") in ("completed", "error"):
await self.redis.set(
cache_key,
json.dumps(doc, ensure_ascii=False, default=str),
ex=BATCH_RESULT_CACHE_TTL,
)
return doc
return None
async def get_batch_inspection_list(
self,
page: int = 1,
limit: int = 20,
name_filter: Optional[str] = None,
) -> dict:
"""Get paginated list of batch inspections."""
limit = min(limit, 100)
skip = (page - 1) * limit
# Build query
query = {}
if name_filter:
query["name"] = {"$regex": name_filter, "$options": "i"}
total = await self.db.batch_inspections.count_documents(query)
cursor = self.db.batch_inspections.find(
query,
{
"_id": 0,
"batch_inspection_id": 1,
"name": 1,
"status": 1,
"created_at": 1,
"discovered_pages": 1,
"aggregate_scores": 1,
"source_urls": 1,
},
).sort("created_at", -1).skip(skip).limit(limit)
items = []
async for doc in cursor:
pages = doc.get("discovered_pages", [])
total_urls = len(doc.get("source_urls", []))
pages_inspected = sum(
1 for p in pages if p.get("status") == "completed"
)
agg = doc.get("aggregate_scores")
items.append({
"batch_inspection_id": doc.get("batch_inspection_id"),
"name": doc.get("name"),
"status": doc.get("status"),
"created_at": doc.get("created_at"),
"total_urls": total_urls,
"pages_inspected": pages_inspected,
"overall_score": agg.get("overall_score") if agg else None,
"grade": agg.get("grade") if agg else None,
})
total_pages = max(1, -(-total // limit))
return {
"items": items,
"total": total,
"page": page,
"limit": limit,
"total_pages": total_pages,
}
# ------------------------------------------------------------------
# Background task: Inspect All URLs
# ------------------------------------------------------------------
async def _inspect_all(
self,
batch_inspection_id: str,
urls: list[str],
concurrency: int = 4,
) -> None:
"""
Background task that inspects all URLs in parallel.
No crawling phase - URLs are inspected directly.
"""
try:
logger.info(
"Batch inspection started: %s, urls=%d",
batch_inspection_id, len(urls),
)
semaphore = asyncio.Semaphore(concurrency)
tasks = [
self._inspect_page_with_semaphore(
semaphore=semaphore,
batch_inspection_id=batch_inspection_id,
page_url=url,
page_index=idx,
total_pages=len(urls),
)
for idx, url in enumerate(urls)
]
await asyncio.gather(*tasks, return_exceptions=True)
# ==============================
# Finalize: Compute aggregates
# ==============================
aggregate_scores = await self._compute_and_store_aggregates(batch_inspection_id)
# Mark as completed
await self.db.batch_inspections.update_one(
{"batch_inspection_id": batch_inspection_id},
{
"$set": {
"status": "completed",
"completed_at": datetime.now(timezone.utc),
}
},
)
# Publish complete event
await self._publish_batch_event(batch_inspection_id, {
"event_type": "complete",
"batch_inspection_id": batch_inspection_id,
"status": "completed",
"aggregate_scores": aggregate_scores,
})
logger.info("Batch inspection completed: %s", batch_inspection_id)
except Exception as e:
logger.error(
"Batch inspection %s failed: %s",
batch_inspection_id, str(e), exc_info=True,
)
await self.db.batch_inspections.update_one(
{"batch_inspection_id": batch_inspection_id},
{
"$set": {
"status": "error",
"completed_at": datetime.now(timezone.utc),
}
},
)
await self._publish_batch_event(batch_inspection_id, {
"event_type": "error",
"batch_inspection_id": batch_inspection_id,
"status": "error",
"message": f"배치 검사 중 오류가 발생했습니다: {str(e)[:200]}",
})
async def _inspect_page_with_semaphore(
self,
semaphore: asyncio.Semaphore,
batch_inspection_id: str,
page_url: str,
page_index: int,
total_pages: int,
) -> None:
"""Inspect a single page with semaphore-controlled concurrency."""
async with semaphore:
await self._inspect_single_page(
batch_inspection_id=batch_inspection_id,
page_url=page_url,
page_index=page_index,
total_pages=total_pages,
)
async def _inspect_single_page(
self,
batch_inspection_id: str,
page_url: str,
page_index: int,
total_pages: int,
) -> None:
"""Run inspection for a single page in the batch."""
inspection_id = str(uuid.uuid4())
# Publish page_start event
await self._publish_batch_event(batch_inspection_id, {
"event_type": "page_start",
"batch_inspection_id": batch_inspection_id,
"page_url": page_url,
"page_index": page_index,
})
# Mark page as inspecting in MongoDB
await self.db.batch_inspections.update_one(
{
"batch_inspection_id": batch_inspection_id,
"discovered_pages.url": page_url,
},
{
"$set": {
"discovered_pages.$.status": "inspecting",
"discovered_pages.$.inspection_id": inspection_id,
}
},
)
try:
# Progress callback for per-page SSE updates
async def page_progress_callback(category: str, progress: int, current_step: str):
await self._publish_batch_event(batch_inspection_id, {
"event_type": "page_progress",
"batch_inspection_id": batch_inspection_id,
"page_url": page_url,
"page_index": page_index,
"category": category,
"progress": progress,
"current_step": current_step,
})
# Run the inspection
_, result = await self.inspection_service.run_inspection_inline(
url=page_url,
inspection_id=inspection_id,
progress_callback=page_progress_callback,
)
overall_score = result.get("overall_score", 0)
grade = result.get("grade", "F")
# Update page status in MongoDB
await self.db.batch_inspections.update_one(
{
"batch_inspection_id": batch_inspection_id,
"discovered_pages.url": page_url,
},
{
"$set": {
"discovered_pages.$.status": "completed",
"discovered_pages.$.overall_score": overall_score,
"discovered_pages.$.grade": grade,
}
},
)
# Publish page_complete event
await self._publish_batch_event(batch_inspection_id, {
"event_type": "page_complete",
"batch_inspection_id": batch_inspection_id,
"page_url": page_url,
"inspection_id": inspection_id,
"overall_score": overall_score,
"grade": grade,
})
# Compute and publish aggregate update
aggregate_scores = await self._compute_and_store_aggregates(batch_inspection_id)
await self._publish_batch_event(batch_inspection_id, {
"event_type": "aggregate_update",
"batch_inspection_id": batch_inspection_id,
"pages_inspected": aggregate_scores.get("pages_inspected", 0),
"pages_total": aggregate_scores.get("pages_total", total_pages),
"overall_score": aggregate_scores.get("overall_score", 0),
"grade": aggregate_scores.get("grade", "F"),
})
logger.info(
"Page inspection completed: batch=%s, page=%s, score=%d",
batch_inspection_id, page_url, overall_score,
)
except Exception as e:
logger.error(
"Page inspection failed: batch=%s, page=%s, error=%s",
batch_inspection_id, page_url, str(e),
)
# Mark page as error
await self.db.batch_inspections.update_one(
{
"batch_inspection_id": batch_inspection_id,
"discovered_pages.url": page_url,
},
{
"$set": {
"discovered_pages.$.status": "error",
}
},
)
# Publish page error (non-fatal, continue with other pages)
await self._publish_batch_event(batch_inspection_id, {
"event_type": "page_complete",
"batch_inspection_id": batch_inspection_id,
"page_url": page_url,
"inspection_id": None,
"overall_score": 0,
"grade": "F",
"error": str(e)[:200],
})
# ------------------------------------------------------------------
# Aggregate computation
# ------------------------------------------------------------------
async def _compute_and_store_aggregates(self, batch_inspection_id: str) -> dict:
"""
Compute aggregate scores from all completed page inspections.
Fetches each completed page's full inspection result from the
inspections collection, averages category scores, and stores
the aggregate in the batch_inspections document.
Returns the aggregate_scores dict.
"""
doc = await self.db.batch_inspections.find_one(
{"batch_inspection_id": batch_inspection_id},
)
if not doc:
return {}
pages = doc.get("discovered_pages", [])
total_pages = len(pages)
# Collect inspection IDs for completed pages
completed_ids = [
p["inspection_id"]
for p in pages
if p.get("status") == "completed" and p.get("inspection_id")
]
if not completed_ids:
aggregate = {
"overall_score": 0,
"grade": "F",
"html_css": 0,
"accessibility": 0,
"seo": 0,
"performance_security": 0,
"total_issues": 0,
"pages_inspected": 0,
"pages_total": total_pages,
}
await self._store_aggregates(batch_inspection_id, aggregate)
return aggregate
# Fetch all completed inspection results
cursor = self.db.inspections.find(
{"inspection_id": {"$in": completed_ids}},
{
"_id": 0,
"overall_score": 1,
"categories.html_css.score": 1,
"categories.accessibility.score": 1,
"categories.seo.score": 1,
"categories.performance_security.score": 1,
"summary.total_issues": 1,
},
)
scores_overall = []
scores_html_css = []
scores_accessibility = []
scores_seo = []
scores_perf = []
total_issues = 0
async for insp in cursor:
scores_overall.append(insp.get("overall_score", 0))
cats = insp.get("categories", {})
scores_html_css.append(cats.get("html_css", {}).get("score", 0))
scores_accessibility.append(cats.get("accessibility", {}).get("score", 0))
scores_seo.append(cats.get("seo", {}).get("score", 0))
scores_perf.append(cats.get("performance_security", {}).get("score", 0))
total_issues += insp.get("summary", {}).get("total_issues", 0)
pages_inspected = len(scores_overall)
def safe_avg(values: list[int]) -> int:
return round(sum(values) / len(values)) if values else 0
overall_score = safe_avg(scores_overall)
grade = calculate_grade(overall_score)
aggregate = {
"overall_score": overall_score,
"grade": grade,
"html_css": safe_avg(scores_html_css),
"accessibility": safe_avg(scores_accessibility),
"seo": safe_avg(scores_seo),
"performance_security": safe_avg(scores_perf),
"total_issues": total_issues,
"pages_inspected": pages_inspected,
"pages_total": total_pages,
}
await self._store_aggregates(batch_inspection_id, aggregate)
return aggregate
async def _store_aggregates(self, batch_inspection_id: str, aggregate: dict) -> None:
"""Store aggregate scores in MongoDB."""
await self.db.batch_inspections.update_one(
{"batch_inspection_id": batch_inspection_id},
{"$set": {"aggregate_scores": aggregate}},
)
# ------------------------------------------------------------------
# SSE event publishing
# ------------------------------------------------------------------
async def _publish_batch_event(self, batch_inspection_id: str, event_data: dict) -> None:
"""Publish an SSE event for batch inspection via Redis Pub/Sub."""
channel = f"batch-inspection:{batch_inspection_id}:events"
await self.redis.publish(
channel,
json.dumps(event_data, ensure_ascii=False, default=str),
)

View File

@ -33,5 +33,8 @@ Jinja2>=3.1.0
# Rules (YAML)
PyYAML>=6.0.0
# Multipart (file upload)
python-multipart>=0.0.7
# Utilities
python-slugify>=8.0.0