[verified] add upload size and MIME guards

This commit is contained in:
2026-06-08 02:15:05 +01:00
parent a6753d092f
commit f5e05376f6
6 changed files with 169 additions and 17 deletions
+4 -3
View File
@@ -12,6 +12,7 @@ from modules.auth.supabase_bearer import SupabaseBearer, verify_supabase_jwt_str
from modules.logger_tool import initialise_logger
from modules.database.supabase.utils.client import SupabaseServiceRoleClient
from modules.database.supabase.utils.storage import StorageAdmin
from modules.upload_validation import read_upload_bytes
from modules.document_processor import DocumentProcessor
from modules.queue_system import (
enqueue_tika_task, enqueue_docling_task, enqueue_split_map_task,
@@ -70,13 +71,13 @@ async def upload_file(
# Stage DB row to get file_id
staged_path = f"{cabinet_id}/staging/{uuid.uuid4()}"
name = _safe_filename(path or file.filename)
file_bytes = await file.read()
file_bytes, mime_type = await read_upload_bytes(file)
insert_res = client.supabase.table('files').insert({
'cabinet_id': cabinet_id,
'name': name,
'path': staged_path,
'bucket': bucket,
'mime_type': file.content_type,
'mime_type': mime_type,
'uploaded_by': user_id,
'size_bytes': len(file_bytes),
'source': 'classroomcopilot-web'
@@ -89,7 +90,7 @@ async def upload_file(
# Final storage path: bucket/cabinet_id/file_id/file
final_storage_path = f"{cabinet_id}/{file_id}/{name}"
try:
storage.upload_file(bucket, final_storage_path, file_bytes, file.content_type or 'application/octet-stream', upsert=True)
storage.upload_file(bucket, final_storage_path, file_bytes, mime_type, upsert=True)
except Exception as e:
# cleanup staged row
client.supabase.table('files').delete().eq('id', file_id).execute()
+3 -3
View File
@@ -19,6 +19,7 @@ from fastapi.responses import JSONResponse
from modules.auth.supabase_bearer import SupabaseBearer
from modules.database.supabase.utils.client import SupabaseServiceRoleClient
from modules.database.supabase.utils.storage import StorageAdmin
from modules.upload_validation import read_upload_bytes
from modules.logger_tool import initialise_logger
router = APIRouter()
@@ -54,10 +55,9 @@ async def upload_file(
if not user_id:
raise HTTPException(status_code=401, detail="User ID required")
# Read file content
file_bytes = await file.read()
# Validate MIME/type and read file content with a hard size limit.
file_bytes, mime_type = await read_upload_bytes(file)
file_size = len(file_bytes)
mime_type = file.content_type or 'application/octet-stream'
filename = file.filename or path
logger.info(f"📤 Simplified upload: {filename} ({file_size} bytes) for user {user_id}")
+2 -5
View File
@@ -28,6 +28,7 @@ from api.services.docling.regions import detect_response_regions_from_pdf
from modules.database.services.exam_projection import project_template, project_template_safe
from modules.database.supabase.utils.client import SupabaseServiceRoleClient
from modules.database.supabase.utils.storage import StorageAdmin
from modules.upload_validation import read_pdf_upload_bytes
from modules.logger_tool import initialise_logger
from routers.exam.dependencies import ExamContext, get_exam_context, lookup_exam_code
from routers.exam.schemas import (
@@ -164,11 +165,7 @@ async def _upload_template_source_file(
institute_id: str,
upload: UploadFile,
) -> str:
file_bytes = await upload.read()
if not file_bytes:
raise HTTPException(status_code=400, detail="Uploaded PDF is empty")
if upload.content_type and upload.content_type != "application/pdf":
raise HTTPException(status_code=400, detail="Uploaded file must be a PDF")
file_bytes = await read_pdf_upload_bytes(upload)
service = SupabaseServiceRoleClient()
storage = StorageAdmin()
+7 -6
View File
@@ -26,6 +26,7 @@ from fastapi.responses import JSONResponse
from modules.auth.supabase_bearer import SupabaseBearer
from modules.database.supabase.utils.client import SupabaseServiceRoleClient
from modules.database.supabase.utils.storage import StorageAdmin
from modules.upload_validation import read_upload_bytes
from modules.logger_tool import initialise_logger
router = APIRouter()
@@ -59,10 +60,9 @@ async def upload_single_file(
if not user_id:
raise HTTPException(status_code=401, detail="User ID required")
# Read file content
file_bytes = await file.read()
# Validate MIME/type and read file content with a hard size limit.
file_bytes, mime_type = await read_upload_bytes(file)
file_size = len(file_bytes)
mime_type = file.content_type or 'application/octet-stream'
filename = file.filename or path
logger.info(f"📤 Simple upload: {filename} ({file_size} bytes) for user {user_id}")
@@ -234,10 +234,9 @@ async def upload_directory(
# Process each file
for i, (file, relative_path) in enumerate(zip(files, relative_paths)):
try:
# Read file content
file_bytes = await file.read()
# Validate MIME/type and read file content with a hard size limit.
file_bytes, mime_type = await read_upload_bytes(file)
file_size = len(file_bytes)
mime_type = file.content_type or 'application/octet-stream'
filename = file.filename or f"file_{i}"
total_size += file_size
@@ -291,6 +290,8 @@ async def upload_directory(
logger.info(f"📄 Uploaded file {i+1}/{len(files)}: {relative_path}")
except HTTPException:
raise
except Exception as e:
logger.error(f"Failed to upload file {relative_path}: {e}")
# Continue with other files, don't fail entire upload