feat(tasks): dynamic routing to user-specific destination integrations

- Add app/tasks/upload_to_user_integration.py: new Celery task that
  uploads a processed document to a specific UserIntegration using its
  own per-user config and Fernet-decrypted credentials. Supports all
  DESTINATION types: Dropbox, S3, Google Drive, OneDrive, WebDAV,
  Nextcloud, FTP, SFTP, Paperless-ngx, Email (SMTP), and Rclone.

- Extend app/tasks/send_to_all.py: add send_to_user_destinations task
  (queries active DESTINATION UserIntegrations for an owner and
  dispatches one upload_to_user_integration task per integration) and
  get_user_destination_count helper used by finalize_document_storage.

- Refactor app/tasks/finalize_document_storage.py: after processing,
  look up the document owner; if the owner has active DESTINATION
  integrations route exclusively to those (user-specific routing),
  otherwise fall back to the global send_to_all_destinations.

- Update tests/test_finalize_storage.py: add autouse fixture to prevent
  Redis hangs, update all existing tests with new mock parameters, add
  TestFinalizeDocumentStorageUserRouting class with four new tests that
  validate user-specific vs global routing decisions.

- Add tests/test_user_integration_upload.py: 14 new unit tests covering
  upload_to_user_integration (handler dispatch, error persistence,
  last_used_at update, credential decryption, skip for unknown types)
  and send_to_user_destinations / get_user_destination_count.

- Update docs/StorageArchitecture.md: document the user-specific
  destination routing feature, supported types, multiple-destination
  behaviour, and global fallback semantics.

Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
This commit is contained in:
copilot-swe-agent[bot]
2026-03-08 21:24:14 +00:00
parent 5d3fe88be9
commit 89bf07d2ef
6 changed files with 1579 additions and 60 deletions
+50 -17
View File
@@ -10,8 +10,13 @@ from app.database import SessionLocal
from app.models import FileRecord
from app.tasks.retry_config import BaseTaskWithRetry
# Import the aggregator task and validator
from app.tasks.send_to_all import get_configured_services_from_validator, send_to_all_destinations
# Import the aggregator tasks and validator
from app.tasks.send_to_all import (
get_configured_services_from_validator,
get_user_destination_count,
send_to_all_destinations,
send_to_user_destinations,
)
# Import database and logging utils from main
from app.utils import log_task_progress
@@ -26,8 +31,16 @@ logger = logging.getLogger(__name__)
def finalize_document_storage(self, original_file: str, processed_file: str, metadata: dict, file_id: int = None):
"""
Final storage step after embedding metadata.
We will now call 'send_to_all_destinations' to push the final PDF to Dropbox/Nextcloud/Paperless.
After uploading, send a notification about the processed file.
Routes the processed document to the appropriate destination(s):
1. If the document has an identified owner and that owner has active
DESTINATION UserIntegrations, the file is uploaded to each of those
integrations (user-specific routing).
2. Otherwise the file is forwarded to the globally-configured destinations
via :func:`send_to_all_destinations` (system-wide fallback).
After queuing uploads, optional PDF/A archival conversion and embedding
computation are triggered, and a completion notification is sent.
"""
task_id = self.request.id
logger.info(f"[{task_id}] Finalizing document storage for {processed_file}")
@@ -41,7 +54,8 @@ def finalize_document_storage(self, original_file: str, processed_file: str, met
file_id=file_id,
)
# Get file_id from database if not provided (fallback logic)
# 2. Resolve file_id and owner_id from the database
owner_id = None
if file_id is None:
with SessionLocal() as db:
# Only as a last resort, try to find by exact match on local_filename
@@ -49,32 +63,52 @@ def finalize_document_storage(self, original_file: str, processed_file: str, met
file_record = db.query(FileRecord).filter(FileRecord.local_filename == tmp_path).first()
if file_record:
file_id = file_record.id
owner_id = file_record.owner_id
else:
with SessionLocal() as db:
file_record = db.query(FileRecord).filter(FileRecord.id == file_id).first()
if file_record:
owner_id = file_record.owner_id
# 2. Determine Configured Destinations
# This is needed for the notification message later
# 3. Determine configured destinations for notification
configured_destinations = []
try:
configured_services = get_configured_services_from_validator()
# Get list of service names that are configured
for service_name, is_configured in configured_services.items():
if is_configured:
# Format service names for display
display_name = service_name.replace("_", " ").title()
configured_destinations.append(display_name)
except Exception as e:
logger.warning(f"[WARNING] Could not determine configured destinations: {e}")
configured_destinations = ["configured destinations"]
# 3. Queue Uploads
logger.info(f"[{task_id}] Queueing uploads to all destinations")
# 4. Queue Uploads — prefer user-specific destinations when available
log_task_progress(
task_id, "finalize_document_storage", "success", "Queuing uploads to destinations", file_id=file_id
)
# Note: send_to_all_destinations is asynchronous and queues upload tasks
send_to_all_destinations.delay(processed_file, True, file_id)
user_dest_count = 0
if owner_id:
try:
user_dest_count = get_user_destination_count(owner_id)
except Exception as e:
logger.warning("[%s] Could not query user destination count for owner=%s: %s", task_id, owner_id, e)
# 3a. Trigger PDF/A archival conversion if enabled (from feature branch)
if owner_id and user_dest_count > 0:
# User has configured their own destinations → use those exclusively
logger.info(
"[%s] Routing to %d user-specific destination(s) for owner=%s",
task_id,
user_dest_count,
owner_id,
)
send_to_user_destinations.delay(processed_file, owner_id, file_id)
else:
# No user-specific destinations → fall back to global configuration
logger.info("[%s] No user-specific destinations found; using global destinations", task_id)
send_to_all_destinations.delay(processed_file, True, file_id)
# 4a. Trigger PDF/A archival conversion if enabled
if settings.enable_pdfa_conversion:
try:
from app.tasks.convert_to_pdfa import convert_to_pdfa
@@ -84,7 +118,7 @@ def finalize_document_storage(self, original_file: str, processed_file: str, met
except Exception as e:
logger.warning(f"[{task_id}] Could not queue PDF/A conversion: {e}")
# 3b. Queue embedding computation (from main branch)
# 4b. Queue embedding computation
if file_id is not None:
try:
from app.tasks.compute_embedding import compute_document_embedding
@@ -94,9 +128,8 @@ def finalize_document_storage(self, original_file: str, processed_file: str, met
except Exception as e:
logger.warning(f"[{task_id}] Could not queue embedding task: {e}")
# 4. Send Notification
# 5. Send Notification
try:
# Get file information
file_size = os.path.getsize(processed_file) if os.path.exists(processed_file) else 0
filename = os.path.basename(processed_file)