Files
xiaoxia-saas/apps/api/app/api/routes/upload.py
T
xiaoxia 4c3dcfbf0f
Deploy / Deploy Staging (push) Failing after 1s
Deploy / Build Production Runtime Images (push) Has been skipped
Deploy / Deploy Production (push) Has been skipped
Deploy / Production Browser E2E (push) Has been skipped
CI/CD Pipeline / Validate Code Quality And Tests (push) Has been cancelled
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Failing after 1m0s
Tests / test (pull_request) Failing after 1m0s
Tests / lint (pull_request) Failing after 30s
fix(P2-5): add server-side file type validation with magic
2026-06-26 17:54:03 +08:00

246 lines
9.3 KiB
Python

from typing import Any
from uuid import uuid4
from app.auth import AuthenticatedUser, get_current_user
from app.config import get_settings
from app.core.celery_app import celery_app
from app.core.storage import OSSStorageService, get_storage_service
from app.dependencies import (
get_asset_library_repository,
get_ingest_job_repository,
get_project_repository,
get_workspace_member_repository,
)
from app.schemas.upload import (
DirectUploadCompleteRequest,
DirectUploadCompleteResponse,
DirectUploadPrepareRequest,
DirectUploadPrepareResponse,
UploadAssetResponse,
)
from fastapi import APIRouter, Depends, File, Form, HTTPException, UploadFile, status
from packages.application import GetProjectUseCase, SubmitIngestJobCommand, SubmitIngestJobUseCase
from packages.ports.workspace_member_repository import WorkspaceMemberRepository
router = APIRouter()
# Allowed MIME types for file uploads (P2-5: Server-side validation)
ALLOWED_MIME_TYPES = {
"image/jpeg", "image/png", "image/gif", "image/webp",
"video/mp4", "video/quicktime", "video/x-msvideo", "video/webm",
"audio/mpeg", "audio/wav", "audio/ogg", "audio/mp3",
}
def _require_workspace_member(
workspace_id: str,
authenticated_user: AuthenticatedUser,
workspace_member_repository: WorkspaceMemberRepository,
) -> None:
member = workspace_member_repository.find_by_workspace_and_user(workspace_id, authenticated_user.user.id)
if member is None:
raise HTTPException(status_code=status.HTTP_403_FORBIDDEN, detail="Workspace access denied")
def _require_project_and_library(
workspace_id: str,
project_id: str,
library_id: str,
project_repository: Any,
asset_library_repository: Any,
) -> None:
project = GetProjectUseCase(project_repository).execute(project_id)
if project is None or project.workspace_id != workspace_id:
raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="Project not found")
libraries = asset_library_repository.list_by_project(project_id)
if not any(item.id == library_id and item.workspace_id == workspace_id for item in libraries):
raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="Asset library not found")
def _submit_ingest_job(
workspace_id: str,
project_id: str,
library_id: str,
storage_key: str,
ingest_job_repository: Any,
) -> Any:
use_case = SubmitIngestJobUseCase(ingest_job_repository)
job = use_case.execute(
SubmitIngestJobCommand(
workspace_id=workspace_id,
project_id=project_id,
library_id=library_id,
storage_key=storage_key,
)
)
celery_app.send_task("worker.ingest_asset", args=[job.id])
return job
def _validate_file_type(content: bytes, filename: str) -> str:
"""
P2-5: Server-side file type validation using python-magic.
Args:
content: File content bytes
filename: Original filename
Returns:
Detected MIME type
Raises:
HTTPException: If file type is not allowed
"""
try:
import magic
detected_mime = magic.from_buffer(content, mime=True)
except ImportError:
# Fallback: infer from extension if magic is not available
import mimetypes
detected_mime = mimetypes.guess_type(filename)[0] or "application/octet-stream"
if detected_mime not in ALLOWED_MIME_TYPES:
raise HTTPException(
status_code=status.HTTP_400_BAD_REQUEST,
detail=f"Unsupported file type: {detected_mime}. Allowed types: {', '.join(sorted(ALLOWED_MIME_TYPES))}"
)
return detected_mime
@router.post("/direct/prepare", response_model=DirectUploadPrepareResponse)
async def prepare_direct_upload(
request: DirectUploadPrepareRequest,
authenticated_user: AuthenticatedUser = Depends(get_current_user),
project_repository: Any = Depends(get_project_repository),
asset_library_repository: Any = Depends(get_asset_library_repository),
workspace_member_repository: WorkspaceMemberRepository = Depends(get_workspace_member_repository),
storage_service: OSSStorageService = Depends(get_storage_service),
) -> DirectUploadPrepareResponse:
"""Prepare direct upload to OSS"""
settings = get_settings()
max_size_bytes = settings.OSS_DIRECT_UPLOAD_MAX_MB * 1024 * 1024
if request.file_size > max_size_bytes:
raise HTTPException(
status_code=status.HTTP_413_REQUEST_ENTITY_TOO_LARGE,
detail=f"File exceeds upload limit ({settings.OSS_DIRECT_UPLOAD_MAX_MB}MB)",
)
_require_workspace_member(request.workspace_id, authenticated_user, workspace_member_repository)
_require_project_and_library(
request.workspace_id,
request.project_id,
request.library_id,
project_repository,
asset_library_repository,
)
file_id = uuid4().hex[:8]
safe_filename = request.filename.replace("/", "_").replace("\\", "_")
storage_key = f"uploads/{file_id}/{safe_filename}"
try:
payload = storage_service.create_direct_upload_post(
storage_key=storage_key,
content_type=request.content_type or "application/octet-stream",
max_size_bytes=max_size_bytes,
expires_seconds=settings.OSS_DIRECT_UPLOAD_EXPIRE_SECONDS,
)
except RuntimeError as error:
raise HTTPException(status_code=status.HTTP_503_SERVICE_UNAVAILABLE, detail=str(error)) from error
return DirectUploadPrepareResponse(
upload_url=str(payload["url"]),
method=str(payload["method"]),
storage_key=str(payload["storage_key"]),
expires_at=str(payload["expires_at"]),
fields={str(key): str(value) for key, value in dict(payload["fields"]).items()},
max_size_bytes=max_size_bytes,
)
@router.post("/direct/complete", response_model=DirectUploadCompleteResponse)
async def complete_direct_upload(
request: DirectUploadCompleteRequest,
authenticated_user: AuthenticatedUser = Depends(get_current_user),
ingest_job_repository: Any = Depends(get_ingest_job_repository),
project_repository: Any = Depends(get_project_repository),
asset_library_repository: Any = Depends(get_asset_library_repository),
workspace_member_repository: WorkspaceMemberRepository = Depends(get_workspace_member_repository),
storage_service: OSSStorageService = Depends(get_storage_service),
) -> DirectUploadCompleteResponse:
"""Complete direct upload"""
_require_workspace_member(request.workspace_id, authenticated_user, workspace_member_repository)
_require_project_and_library(
request.workspace_id,
request.project_id,
request.library_id,
project_repository,
asset_library_repository,
)
normalized_key = storage_service._normalize_storage_key(request.storage_key)
if not normalized_key.startswith("uploads/"):
raise HTTPException(status_code=status.HTTP_400_BAD_REQUEST, detail="Invalid upload key")
if not storage_service.file_exists(normalized_key):
raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="Uploaded file not found")
job = _submit_ingest_job(
workspace_id=request.workspace_id,
project_id=request.project_id,
library_id=request.library_id,
storage_key=normalized_key,
ingest_job_repository=ingest_job_repository,
)
return DirectUploadCompleteResponse(storage_key=normalized_key, ingest_job_id=job.id)
@router.post("", response_model=UploadAssetResponse)
async def upload_asset(
file: UploadFile = File(..., description="文件上传"),
workspace_id: str = Form(..., description="工作空间 ID"),
project_id: str = Form(..., description="项目 ID"),
library_id: str = Form(..., description="资产库 ID"),
authenticated_user: AuthenticatedUser = Depends(get_current_user),
ingest_job_repository: Any = Depends(get_ingest_job_repository),
project_repository: Any = Depends(get_project_repository),
asset_library_repository: Any = Depends(get_asset_library_repository),
workspace_member_repository: WorkspaceMemberRepository = Depends(get_workspace_member_repository),
storage_service: OSSStorageService = Depends(get_storage_service),
) -> UploadAssetResponse:
"""Upload asset file"""
_require_workspace_member(workspace_id, authenticated_user, workspace_member_repository)
_require_project_and_library(workspace_id, project_id, library_id, project_repository, asset_library_repository)
# Read file content for validation (P2-5)
file_content = await file.read()
# P2-5: Server-side file type validation
detected_mime = _validate_file_type(file_content, file.filename)
# Reset file pointer for upload
await file.seek(0)
file_id = uuid4().hex[:8]
storage_key = f"uploads/{file_id}/{file.filename}"
file_url = storage_service.upload_file(
file.file,
storage_key,
content_type=file.content_type or detected_mime,
)
job = _submit_ingest_job(
workspace_id=workspace_id,
project_id=project_id,
library_id=library_id,
storage_key=storage_key,
ingest_job_repository=ingest_job_repository,
)
return UploadAssetResponse(
storage_key=storage_key,
ingest_job_id=job.id,
url=file_url,
)