fix: merge main (v0.161.0) into classification feature branch
Resolve all 40 merge conflicts from merging origin/main into the classification feature branch. Key resolutions: - Auto-generated files (BUILD_DATE, VERSION, etc.): use main's version - API tokens: take main's version (token expiry, reactivation, hard-delete) - Auth: take main's Dropbox credential sharing + token expiry checking - Config: take main's social_auth_dropbox_use_global_credentials option - Files API: take main's improved duplicate handling + rate limiting - Models: keep ClassificationRuleModel alongside main's new models - Mobile: take main's mature implementation - Templates/translations: take main's versions (device deletion, reactivation keys) - Migration: renumber 038_add_classification_rules → 039_add_classification_rules to chain after main's 038_add_api_token_expires_at - Requirements: take main's version (adds segno QR library) - Tests: take main's more complete token tests, keep classification imports
This commit is contained in:
+68
-11
@@ -13,7 +13,7 @@ plaintext is returned exactly once at creation time.
|
||||
import hashlib
|
||||
import logging
|
||||
import secrets
|
||||
from datetime import datetime, timezone
|
||||
from datetime import datetime, timedelta, timezone
|
||||
from typing import Annotated, Any
|
||||
|
||||
from fastapi import APIRouter, Depends, HTTPException, Request, status
|
||||
@@ -105,6 +105,7 @@ def _token_to_dict(t: ApiToken) -> dict[str, Any]:
|
||||
"last_used_ip": t.last_used_ip,
|
||||
"created_at": t.created_at,
|
||||
"revoked_at": t.revoked_at,
|
||||
"expires_at": t.expires_at,
|
||||
}
|
||||
|
||||
|
||||
@@ -117,6 +118,12 @@ class TokenCreate(BaseModel):
|
||||
"""Schema for creating a new API token."""
|
||||
|
||||
name: str = Field(..., min_length=1, max_length=255, description="Human-readable label for the token")
|
||||
expires_in_days: int | None = Field(
|
||||
default=None,
|
||||
ge=1,
|
||||
le=3650, # Maximum 10 years; keeps tokens from being effectively permanent while allowing long-lived CI/CD tokens.
|
||||
description="Optional lifetime in days. If omitted the token never expires.",
|
||||
)
|
||||
|
||||
|
||||
class TokenResponse(BaseModel):
|
||||
@@ -130,6 +137,7 @@ class TokenResponse(BaseModel):
|
||||
last_used_ip: str | None
|
||||
created_at: datetime | None
|
||||
revoked_at: datetime | None
|
||||
expires_at: datetime | None
|
||||
|
||||
model_config = {"from_attributes": True}
|
||||
|
||||
@@ -160,11 +168,16 @@ async def create_token(
|
||||
token_hash_value = hash_token(plaintext)
|
||||
prefix = plaintext[:12] # "de_" prefix + 9 random chars = 12 chars total
|
||||
|
||||
expires_at = None
|
||||
if body.expires_in_days is not None:
|
||||
expires_at = datetime.now(timezone.utc) + timedelta(days=body.expires_in_days)
|
||||
|
||||
db_token = ApiToken(
|
||||
owner_id=owner_id,
|
||||
name=body.name,
|
||||
token_hash=token_hash_value,
|
||||
token_prefix=prefix,
|
||||
expires_at=expires_at,
|
||||
)
|
||||
try:
|
||||
db.add(db_token)
|
||||
@@ -185,6 +198,7 @@ async def create_token(
|
||||
"last_used_ip": db_token.last_used_ip,
|
||||
"created_at": db_token.created_at,
|
||||
"revoked_at": db_token.revoked_at,
|
||||
"expires_at": db_token.expires_at,
|
||||
"token": plaintext,
|
||||
}
|
||||
|
||||
@@ -235,30 +249,73 @@ async def list_mobile_tokens(
|
||||
|
||||
|
||||
@router.delete("/{token_id}", status_code=status.HTTP_200_OK)
|
||||
async def revoke_token(
|
||||
async def revoke_or_delete_token(
|
||||
token_id: int,
|
||||
owner_id: CurrentOwner,
|
||||
db: DbSession,
|
||||
) -> dict[str, str]:
|
||||
"""Revoke (soft-delete) an API token.
|
||||
"""Revoke or permanently delete an API token.
|
||||
|
||||
The token row is kept for audit purposes but marked inactive with a
|
||||
``revoked_at`` timestamp.
|
||||
* **Active token** – soft-revoked: the row is kept for audit purposes
|
||||
but marked inactive with a ``revoked_at`` timestamp.
|
||||
* **Already-revoked token** – hard-deleted: the row is permanently
|
||||
removed from the database.
|
||||
"""
|
||||
db_token = db.query(ApiToken).filter(ApiToken.id == token_id, ApiToken.owner_id == owner_id).first()
|
||||
if not db_token:
|
||||
raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="Token not found")
|
||||
|
||||
if not db_token.is_active:
|
||||
raise HTTPException(status_code=status.HTTP_400_BAD_REQUEST, detail="Token is already revoked")
|
||||
if db_token.is_active:
|
||||
# Soft-revoke the active token.
|
||||
try:
|
||||
db_token.is_active = False
|
||||
db_token.revoked_at = datetime.now(timezone.utc)
|
||||
db.commit()
|
||||
except Exception:
|
||||
db.rollback()
|
||||
raise
|
||||
logger.info("API token revoked: id=%s owner=%s", token_id, owner_id)
|
||||
return {"detail": "Token revoked"}
|
||||
|
||||
# Hard-delete an already-revoked token.
|
||||
try:
|
||||
db_token.is_active = False
|
||||
db_token.revoked_at = datetime.now(timezone.utc)
|
||||
db.delete(db_token)
|
||||
db.commit()
|
||||
except Exception:
|
||||
db.rollback()
|
||||
raise
|
||||
logger.info("API token permanently deleted: id=%s owner=%s", token_id, owner_id)
|
||||
return {"detail": "Token deleted"}
|
||||
|
||||
logger.info("API token revoked: id=%s owner=%s", token_id, owner_id)
|
||||
return {"detail": "Token revoked"}
|
||||
|
||||
@router.post("/{token_id}/reactivate", status_code=status.HTTP_200_OK, response_model=TokenResponse)
|
||||
async def reactivate_token(
|
||||
token_id: int,
|
||||
owner_id: CurrentOwner,
|
||||
db: DbSession,
|
||||
) -> dict[str, Any]:
|
||||
"""Reactivate a previously revoked API token.
|
||||
|
||||
Clears the ``revoked_at`` timestamp and sets ``is_active`` back to
|
||||
``True``. The token can be used for authentication again immediately.
|
||||
If the token had an ``expires_at`` in the past the caller should
|
||||
consider re-creating a new token instead.
|
||||
"""
|
||||
db_token = db.query(ApiToken).filter(ApiToken.id == token_id, ApiToken.owner_id == owner_id).first()
|
||||
if not db_token:
|
||||
raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="Token not found")
|
||||
|
||||
if db_token.is_active:
|
||||
raise HTTPException(status_code=status.HTTP_400_BAD_REQUEST, detail="Token is already active")
|
||||
|
||||
try:
|
||||
db_token.is_active = True
|
||||
db_token.revoked_at = None
|
||||
db.commit()
|
||||
db.refresh(db_token)
|
||||
except Exception:
|
||||
db.rollback()
|
||||
raise
|
||||
|
||||
logger.info("API token reactivated: id=%s owner=%s", token_id, owner_id)
|
||||
return _token_to_dict(db_token)
|
||||
|
||||
@@ -21,6 +21,67 @@ _DEFAULT_REDIS_URL = "redis://localhost:6379/0"
|
||||
|
||||
router = APIRouter()
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Unauthenticated probe endpoints for Kubernetes liveness / readiness checks.
|
||||
# These intentionally skip authentication so that kubelet can reach them
|
||||
# without credentials. They live under /diagnostic/healthz/* so that the
|
||||
# existing authenticated /diagnostic/health endpoint is unaffected.
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
@router.get("/diagnostic/healthz/live")
|
||||
async def liveness_probe() -> JSONResponse:
|
||||
"""Lightweight liveness probe for Kubernetes.
|
||||
|
||||
Returns **200 OK** as long as the process is running. Kubernetes uses
|
||||
this to decide whether to *restart* the container — it should therefore
|
||||
be as cheap as possible and **never** check external dependencies.
|
||||
|
||||
**Authentication:** None (designed for kubelet probes).
|
||||
"""
|
||||
return JSONResponse(content={"status": "ok"}, status_code=200)
|
||||
|
||||
|
||||
@router.get("/diagnostic/healthz/ready")
|
||||
async def readiness_probe() -> JSONResponse:
|
||||
"""Readiness probe for Kubernetes.
|
||||
|
||||
Verifies that the application can serve traffic by checking the database
|
||||
and Redis. Kubernetes uses this to decide whether to *route traffic* to
|
||||
the pod.
|
||||
|
||||
Returns **200 OK** when all critical subsystems are reachable, or
|
||||
**503 Service Unavailable** when the database is down.
|
||||
|
||||
**Authentication:** None (designed for kubelet probes).
|
||||
"""
|
||||
checks: dict[str, dict[str, str]] = {}
|
||||
db_ok = False
|
||||
|
||||
# ── Database check ─────────────────────────────────────────────────
|
||||
try:
|
||||
with engine.connect() as conn:
|
||||
conn.execute(text("SELECT 1"))
|
||||
checks["database"] = {"status": "ok"}
|
||||
db_ok = True
|
||||
except Exception as exc:
|
||||
logger.warning("Readiness probe: database check failed: %s", exc)
|
||||
checks["database"] = {"status": "error", "detail": str(exc)}
|
||||
|
||||
# ── Redis check ────────────────────────────────────────────────────
|
||||
try:
|
||||
redis_url = settings.redis_url or _DEFAULT_REDIS_URL
|
||||
r = redis_lib.from_url(redis_url, socket_connect_timeout=2, socket_timeout=2)
|
||||
r.ping()
|
||||
checks["redis"] = {"status": "ok"}
|
||||
except Exception as exc:
|
||||
logger.warning("Readiness probe: Redis check failed: %s", exc)
|
||||
checks["redis"] = {"status": "error", "detail": str(exc)}
|
||||
|
||||
http_status = 503 if not db_ok else 200
|
||||
overall = "ready" if db_ok else "not_ready"
|
||||
return JSONResponse(content={"status": overall, "checks": checks}, status_code=http_status)
|
||||
|
||||
|
||||
@router.get("/diagnostic/health")
|
||||
@require_login
|
||||
|
||||
@@ -5,6 +5,7 @@ Dropbox API endpoints
|
||||
import logging
|
||||
import os
|
||||
from typing import Annotated, Optional
|
||||
from urllib.parse import quote
|
||||
|
||||
import httpx
|
||||
from fastapi import APIRouter, Depends, Form, HTTPException, Request, status
|
||||
@@ -23,6 +24,93 @@ logger = logging.getLogger(__name__)
|
||||
router = APIRouter()
|
||||
|
||||
|
||||
def _build_dropbox_redirect_uri(request: Request) -> str:
|
||||
"""Build the Dropbox OAuth callback redirect URI.
|
||||
|
||||
Uses ``PUBLIC_BASE_URL`` when configured (recommended for deployments behind
|
||||
a reverse proxy that doesn't forward ``X-Forwarded-Proto``). Falls back to
|
||||
deriving the URI from the incoming request's scheme and host headers.
|
||||
"""
|
||||
if settings.public_base_url:
|
||||
return settings.public_base_url.rstrip("/") + "/dropbox-callback"
|
||||
return f"{request.url.scheme}://{request.url.netloc}/dropbox-callback"
|
||||
|
||||
|
||||
@router.get("/dropbox/global-authorize-url")
|
||||
@require_login
|
||||
async def dropbox_global_authorize_url(request: Request):
|
||||
"""Return the Dropbox OAuth authorization URL using the global app credentials.
|
||||
|
||||
This endpoint is used when ``DROPBOX_ALLOW_GLOBAL_CREDENTIALS_FOR_INTEGRATIONS``
|
||||
is enabled so that users can authorize their personal Dropbox integration without
|
||||
needing to supply their own app key/secret. Only the public ``app_key`` is
|
||||
embedded in the URL; the ``app_secret`` is never sent to the browser.
|
||||
"""
|
||||
if not settings.dropbox_allow_global_credentials_for_integrations:
|
||||
raise HTTPException(
|
||||
status_code=status.HTTP_403_FORBIDDEN,
|
||||
detail="Global credentials for integrations are not enabled",
|
||||
)
|
||||
if not settings.dropbox_app_key or not settings.dropbox_app_secret:
|
||||
raise HTTPException(
|
||||
status_code=status.HTTP_503_SERVICE_UNAVAILABLE,
|
||||
detail="Global Dropbox credentials are not configured",
|
||||
)
|
||||
redirect_uri = _build_dropbox_redirect_uri(request)
|
||||
authorize_url = (
|
||||
"https://www.dropbox.com/oauth2/authorize"
|
||||
f"?client_id={settings.dropbox_app_key}"
|
||||
"&response_type=code"
|
||||
"&token_access_type=offline"
|
||||
f"&redirect_uri={quote(redirect_uri, safe='')}"
|
||||
)
|
||||
return {"authorize_url": authorize_url}
|
||||
|
||||
|
||||
@router.post("/dropbox/exchange-token-global")
|
||||
@require_login
|
||||
async def exchange_dropbox_token_global(
|
||||
request: Request,
|
||||
code: Annotated[str, Form(...)],
|
||||
redirect_uri: Annotated[str, Form(...)],
|
||||
):
|
||||
"""Exchange an authorization code using the global Dropbox app credentials.
|
||||
|
||||
Used when ``DROPBOX_ALLOW_GLOBAL_CREDENTIALS_FOR_INTEGRATIONS`` is enabled so
|
||||
that the ``app_secret`` is never exposed to the browser. Only the OAuth code
|
||||
and redirect URI need to be supplied by the client.
|
||||
"""
|
||||
if not settings.dropbox_allow_global_credentials_for_integrations:
|
||||
raise HTTPException(
|
||||
status_code=status.HTTP_403_FORBIDDEN,
|
||||
detail="Global credentials for integrations are not enabled",
|
||||
)
|
||||
if not settings.dropbox_app_key or not settings.dropbox_app_secret:
|
||||
raise HTTPException(
|
||||
status_code=status.HTTP_503_SERVICE_UNAVAILABLE,
|
||||
detail="Global Dropbox credentials are not configured",
|
||||
)
|
||||
|
||||
token_url = "https://api.dropboxapi.com/oauth2/token"
|
||||
payload = {
|
||||
"client_id": settings.dropbox_app_key,
|
||||
"client_secret": settings.dropbox_app_secret,
|
||||
"code": code,
|
||||
"redirect_uri": redirect_uri,
|
||||
"grant_type": "authorization_code",
|
||||
}
|
||||
|
||||
token_data = exchange_oauth_token(provider_name="Dropbox", token_url=token_url, payload=payload)
|
||||
|
||||
return {
|
||||
"refresh_token": token_data["refresh_token"],
|
||||
"access_token": token_data["access_token"],
|
||||
"expires_in": token_data.get("expires_in", 14400),
|
||||
# Return the public app_key so the callback can store it in the integration
|
||||
"app_key": settings.dropbox_app_key,
|
||||
}
|
||||
|
||||
|
||||
@router.post("/dropbox/exchange-token")
|
||||
@require_login
|
||||
async def exchange_dropbox_token(
|
||||
|
||||
+64
-22
@@ -20,6 +20,7 @@ from sqlalchemy.orm import Session
|
||||
from app.auth import require_login
|
||||
from app.config import settings
|
||||
from app.database import get_db
|
||||
from app.middleware.upload_rate_limit import require_upload_rate_limit
|
||||
from app.models import FileProcessingStep, FileRecord, ProcessingLog
|
||||
from app.tasks.convert_to_pdf import convert_to_pdf
|
||||
from app.tasks.process_document import process_document
|
||||
@@ -346,7 +347,9 @@ def bulk_delete_files(request: Request, file_ids: List[int], db: DbSession):
|
||||
|
||||
try:
|
||||
# Find all file records
|
||||
file_records = db.query(FileRecord).filter(FileRecord.id.in_(file_ids)).all()
|
||||
query = db.query(FileRecord).filter(FileRecord.id.in_(file_ids))
|
||||
query = apply_owner_filter(query, request)
|
||||
file_records = query.all()
|
||||
|
||||
if not file_records:
|
||||
raise HTTPException(status_code=404, detail="No files found with the provided IDs")
|
||||
@@ -385,7 +388,9 @@ def bulk_reprocess_files(request: Request, file_ids: List[int], db: DbSession):
|
||||
"""
|
||||
try:
|
||||
# Find all file records
|
||||
file_records = db.query(FileRecord).filter(FileRecord.id.in_(file_ids)).all()
|
||||
query = db.query(FileRecord).filter(FileRecord.id.in_(file_ids))
|
||||
query = apply_owner_filter(query, request)
|
||||
file_records = query.all()
|
||||
|
||||
if not file_records:
|
||||
raise HTTPException(status_code=404, detail="No files found with the provided IDs")
|
||||
@@ -457,7 +462,9 @@ def bulk_reprocess_files_cloud_ocr(request: Request, file_ids: List[int], db: Db
|
||||
Useful for re-running OCR on files with poor text quality or missing OCR text.
|
||||
"""
|
||||
try:
|
||||
file_records = db.query(FileRecord).filter(FileRecord.id.in_(file_ids)).all()
|
||||
query = db.query(FileRecord).filter(FileRecord.id.in_(file_ids))
|
||||
query = apply_owner_filter(query, request)
|
||||
file_records = query.all()
|
||||
|
||||
if not file_records:
|
||||
raise HTTPException(status_code=404, detail="No files found with the provided IDs")
|
||||
@@ -537,7 +544,9 @@ def bulk_download_files(request: Request, file_ids: List[int], db: DbSession):
|
||||
Files not found on disk are silently skipped.
|
||||
"""
|
||||
try:
|
||||
file_records = db.query(FileRecord).filter(FileRecord.id.in_(file_ids)).all()
|
||||
query = db.query(FileRecord).filter(FileRecord.id.in_(file_ids))
|
||||
query = apply_owner_filter(query, request)
|
||||
file_records = query.all()
|
||||
|
||||
if not file_records:
|
||||
raise HTTPException(status_code=404, detail="No files found with the provided IDs")
|
||||
@@ -619,7 +628,9 @@ def reprocess_single_file(request: Request, file_id: int, db: DbSession):
|
||||
"""
|
||||
try:
|
||||
# Find the file record
|
||||
file_record = db.query(FileRecord).filter(FileRecord.id == file_id).first()
|
||||
query = db.query(FileRecord).filter(FileRecord.id == file_id)
|
||||
query = apply_owner_filter(query, request)
|
||||
file_record = query.first()
|
||||
|
||||
if not file_record:
|
||||
raise HTTPException(status_code=404, detail=f"File with ID {file_id} not found")
|
||||
@@ -675,7 +686,9 @@ def reprocess_with_cloud_ocr(request: Request, file_id: int, db: DbSession):
|
||||
"""
|
||||
try:
|
||||
# Find the file record
|
||||
file_record = db.query(FileRecord).filter(FileRecord.id == file_id).first()
|
||||
query = db.query(FileRecord).filter(FileRecord.id == file_id)
|
||||
query = apply_owner_filter(query, request)
|
||||
file_record = query.first()
|
||||
|
||||
if not file_record:
|
||||
raise HTTPException(status_code=404, detail=f"File with ID {file_id} not found")
|
||||
@@ -938,7 +951,9 @@ def retry_subtask(
|
||||
"""
|
||||
try:
|
||||
# Find the file record
|
||||
file_record = db.query(FileRecord).filter(FileRecord.id == file_id).first()
|
||||
query = db.query(FileRecord).filter(FileRecord.id == file_id)
|
||||
query = apply_owner_filter(query, request)
|
||||
file_record = query.first()
|
||||
|
||||
if not file_record:
|
||||
raise HTTPException(status_code=404, detail=f"File with ID {file_id} not found")
|
||||
@@ -1080,7 +1095,9 @@ def get_file_preview(
|
||||
|
||||
try:
|
||||
# Find the file record
|
||||
file_record = db.query(FileRecord).filter(FileRecord.id == file_id).first()
|
||||
query = db.query(FileRecord).filter(FileRecord.id == file_id)
|
||||
query = apply_owner_filter(query, request)
|
||||
file_record = query.first()
|
||||
|
||||
if not file_record:
|
||||
raise HTTPException(status_code=404, detail=f"File with ID {file_id} not found")
|
||||
@@ -1160,7 +1177,9 @@ def download_file(
|
||||
|
||||
try:
|
||||
# Find the file record
|
||||
file_record = db.query(FileRecord).filter(FileRecord.id == file_id).first()
|
||||
query = db.query(FileRecord).filter(FileRecord.id == file_id)
|
||||
query = apply_owner_filter(query, request)
|
||||
file_record = query.first()
|
||||
|
||||
if not file_record:
|
||||
raise HTTPException(status_code=404, detail=f"File with ID {file_id} not found")
|
||||
@@ -1249,7 +1268,12 @@ async def _save_upload_file_chunks(file: UploadFile, target_path: str, max_size:
|
||||
|
||||
|
||||
def _check_for_exact_duplicate(db: DbSession, target_path: str, safe_filename: str) -> dict | None:
|
||||
"""Check for an exact duplicate of the uploaded file and return a warning if found."""
|
||||
"""Check for an exact duplicate of the uploaded file.
|
||||
|
||||
Returns a dict with duplicate info when the file's SHA-256 hash matches an
|
||||
already-processed document, or ``None`` when no duplicate is found (or
|
||||
deduplication is disabled).
|
||||
"""
|
||||
if not settings.enable_deduplication:
|
||||
return None
|
||||
|
||||
@@ -1268,8 +1292,8 @@ def _check_for_exact_duplicate(db: DbSession, target_path: str, safe_filename: s
|
||||
"original_file_id": existing.id,
|
||||
"original_filename": existing.original_filename,
|
||||
"message": (
|
||||
"This file appears to be an exact duplicate of an already-processed document. "
|
||||
"It will still be queued but will be flagged as a duplicate."
|
||||
"This file is an exact duplicate of an already-processed document. "
|
||||
"It has not been queued for processing again."
|
||||
),
|
||||
}
|
||||
except Exception as e:
|
||||
@@ -1280,7 +1304,12 @@ def _check_for_exact_duplicate(db: DbSession, target_path: str, safe_filename: s
|
||||
|
||||
@router.post("/ui-upload")
|
||||
@require_login
|
||||
async def ui_upload(request: Request, db: DbSession, file: UploadFile = File(...)):
|
||||
async def ui_upload(
|
||||
request: Request,
|
||||
db: DbSession,
|
||||
file: UploadFile = File(...),
|
||||
_rate_ok: None = Depends(require_upload_rate_limit),
|
||||
):
|
||||
"""Endpoint to accept a user-uploaded file and enqueue it for processing."""
|
||||
workdir = settings.workdir
|
||||
|
||||
@@ -1366,6 +1395,25 @@ async def ui_upload(request: Request, db: DbSession, file: UploadFile = File(...
|
||||
logger.info(f"Saved uploaded file '{safe_filename}' as '{target_filename}'")
|
||||
file_size = written_size
|
||||
|
||||
# ── Early duplicate rejection ──────────────────────────────────────────
|
||||
# Check for exact duplicates (same SHA-256 hash) BEFORE enqueuing a
|
||||
# processing task. When deduplication is enabled and the file already
|
||||
# exists, we skip processing entirely, clean up the temp file, and
|
||||
# return the existing file's information to the caller.
|
||||
exact_duplicate = _check_for_exact_duplicate(db, target_path, safe_filename)
|
||||
if exact_duplicate:
|
||||
# Remove the just-saved temp file — it's a duplicate.
|
||||
try:
|
||||
os.remove(target_path)
|
||||
except OSError:
|
||||
pass
|
||||
return {
|
||||
"status": "duplicate",
|
||||
"original_filename": safe_filename,
|
||||
"stored_filename": target_filename,
|
||||
"duplicate_of": exact_duplicate,
|
||||
}
|
||||
|
||||
# Determine if the file is a PDF or needs conversion
|
||||
mime_type, _ = mimetypes.guess_type(target_path)
|
||||
file_ext = os.path.splitext(target_path)[1].lower()
|
||||
@@ -1429,6 +1477,8 @@ async def ui_upload(request: Request, db: DbSession, file: UploadFile = File(...
|
||||
".tif",
|
||||
".webp",
|
||||
".svg",
|
||||
".heic",
|
||||
".heif",
|
||||
}:
|
||||
# If it's an image, convert to PDF first
|
||||
task = convert_to_pdf.delay(target_path, original_filename=safe_filename, owner_id=upload_owner_id)
|
||||
@@ -1442,20 +1492,12 @@ async def ui_upload(request: Request, db: DbSession, file: UploadFile = File(...
|
||||
logger.warning(f"Unsupported MIME type {mime_type} for {target_path}, attempting conversion")
|
||||
task = convert_to_pdf.delay(target_path, original_filename=safe_filename, owner_id=upload_owner_id)
|
||||
|
||||
# Check for exact duplicates (same SHA-256 hash) before returning.
|
||||
# This gives the caller an immediate warning without waiting for the pipeline.
|
||||
# Only performed when deduplication is enabled in settings.
|
||||
exact_duplicate_warning = _check_for_exact_duplicate(db, target_path, safe_filename)
|
||||
|
||||
response: dict = {
|
||||
return {
|
||||
"task_id": task.id,
|
||||
"status": "queued",
|
||||
"original_filename": safe_filename,
|
||||
"stored_filename": target_filename,
|
||||
}
|
||||
if exact_duplicate_warning:
|
||||
response["duplicate_warning"] = exact_duplicate_warning
|
||||
return response
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
@@ -32,6 +32,21 @@ from app.utils.encryption import decrypt_value, encrypt_value
|
||||
from app.utils.subscription import get_tier, get_user_tier_id
|
||||
from app.utils.user_scope import get_current_owner_id
|
||||
|
||||
# Optional Dropbox SDK — imported at module level so tests can patch it cleanly.
|
||||
try:
|
||||
import dropbox as dbx_lib
|
||||
from dropbox.exceptions import AuthError as _DropboxAuthError
|
||||
from dropbox.exceptions import BadInputError as _DropboxBadInputError
|
||||
except ImportError: # pragma: no cover
|
||||
dbx_lib = None # type: ignore[assignment]
|
||||
|
||||
class _DropboxAuthError(Exception): # type: ignore[no-redef]
|
||||
"""Stub — only used when the dropbox package is missing."""
|
||||
|
||||
class _DropboxBadInputError(Exception): # type: ignore[no-redef]
|
||||
"""Stub — only used when the dropbox package is missing."""
|
||||
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
router = APIRouter(prefix="/integrations", tags=["integrations"])
|
||||
|
||||
@@ -550,6 +565,47 @@ def _test_s3_connection(config: dict[str, Any] | None, credentials: dict[str, An
|
||||
return {"success": False, "message": "S3 connection failed"}
|
||||
|
||||
|
||||
def _test_dropbox_connection(config: dict[str, Any] | None, credentials: dict[str, Any] | None) -> dict[str, Any]:
|
||||
"""Test a Dropbox connection by verifying OAuth credentials via the Dropbox API."""
|
||||
if dbx_lib is None:
|
||||
return {"success": False, "message": "dropbox package is not installed"} # pragma: no cover
|
||||
|
||||
creds = credentials or {}
|
||||
app_key = creds.get("app_key", "")
|
||||
app_secret = creds.get("app_secret", "")
|
||||
refresh_token = creds.get("refresh_token", "")
|
||||
|
||||
if not refresh_token:
|
||||
return {"success": False, "message": "Missing required credential: refresh_token"}
|
||||
if not app_key or not app_secret:
|
||||
return {"success": False, "message": "Missing required credentials: app_key and app_secret"}
|
||||
|
||||
try:
|
||||
dbx = dbx_lib.Dropbox(
|
||||
app_key=app_key,
|
||||
app_secret=app_secret,
|
||||
oauth2_refresh_token=refresh_token,
|
||||
)
|
||||
account = dbx.users_get_current_account()
|
||||
display_name = getattr(account, "name", None)
|
||||
name_str = ""
|
||||
if display_name:
|
||||
name_str = f" ({getattr(display_name, 'display_name', '') or ''})"
|
||||
return {"success": True, "message": f"Dropbox connection successful{name_str}"}
|
||||
except _DropboxAuthError as exc:
|
||||
logger.warning("Dropbox auth error: %s", exc)
|
||||
return {
|
||||
"success": False,
|
||||
"message": "Dropbox authentication failed — check app_key, app_secret, and refresh_token",
|
||||
}
|
||||
except _DropboxBadInputError as exc:
|
||||
logger.warning("Dropbox bad input error: %s", exc)
|
||||
return {"success": False, "message": "Dropbox connection failed — invalid credentials format"}
|
||||
except Exception as exc: # noqa: BLE001
|
||||
logger.warning("Dropbox connection error: %s", exc)
|
||||
return {"success": False, "message": "Dropbox connection failed — check credentials and network connectivity"}
|
||||
|
||||
|
||||
def _test_webdav_connection(config: dict[str, Any] | None, credentials: dict[str, Any] | None) -> dict[str, Any]:
|
||||
"""Test a WebDAV/Nextcloud connection by issuing an HTTP PROPFIND."""
|
||||
import urllib.request
|
||||
@@ -596,6 +652,7 @@ def _test_webdav_connection(config: dict[str, Any] | None, credentials: dict[str
|
||||
|
||||
|
||||
_CONNECTION_TESTERS: dict[str, Any] = {
|
||||
IntegrationType.DROPBOX: _test_dropbox_connection,
|
||||
IntegrationType.IMAP: _test_imap_connection,
|
||||
IntegrationType.S3: _test_s3_connection,
|
||||
IntegrationType.WEBDAV: _test_webdav_connection,
|
||||
|
||||
+23
-8
@@ -120,6 +120,7 @@ class WhoAmIResponse(BaseModel):
|
||||
email: str | None
|
||||
avatar_url: str | None
|
||||
is_admin: bool
|
||||
preferred_language: str | None
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
@@ -273,31 +274,44 @@ async def list_devices(
|
||||
return [_device_to_response(d) for d in devices]
|
||||
|
||||
|
||||
@router.delete("/devices/{device_id}", status_code=status.HTTP_204_NO_CONTENT)
|
||||
@router.delete("/devices/{device_id}", status_code=status.HTTP_200_OK)
|
||||
@require_login
|
||||
async def deactivate_device(
|
||||
request: Request,
|
||||
device_id: int,
|
||||
owner_id: CurrentOwner,
|
||||
db: DbSession,
|
||||
) -> None:
|
||||
"""Deactivate a push-notification device registration.
|
||||
) -> dict[str, str]:
|
||||
"""Deactivate or permanently delete a push-notification device registration.
|
||||
|
||||
The device record is kept for audit purposes but will no longer receive
|
||||
push notifications.
|
||||
* **Active device** – soft-deactivated: the record is kept for audit
|
||||
purposes but will no longer receive push notifications.
|
||||
* **Already-inactive device** – hard-deleted: the record is permanently
|
||||
removed from the database.
|
||||
"""
|
||||
device = db.get(MobileDevice, device_id)
|
||||
if not device or device.owner_id != owner_id:
|
||||
raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="Device not found")
|
||||
|
||||
device.is_active = False
|
||||
if device.is_active:
|
||||
device.is_active = False
|
||||
try:
|
||||
db.commit()
|
||||
except Exception:
|
||||
db.rollback()
|
||||
raise
|
||||
logger.info("Mobile device deactivated: id=%s owner=%s", device_id, owner_id)
|
||||
return {"detail": "Device deactivated"}
|
||||
|
||||
# Hard-delete an already-inactive device.
|
||||
try:
|
||||
db.delete(device)
|
||||
db.commit()
|
||||
except Exception:
|
||||
db.rollback()
|
||||
raise
|
||||
|
||||
logger.info("Mobile device deactivated: id=%s owner=%s", device_id, owner_id)
|
||||
logger.info("Mobile device permanently deleted: id=%s owner=%s", device_id, owner_id)
|
||||
return {"detail": "Device deleted"}
|
||||
|
||||
|
||||
@router.get("/whoami", response_model=WhoAmIResponse)
|
||||
@@ -344,4 +358,5 @@ async def whoami(
|
||||
"email": email,
|
||||
"avatar_url": avatar_url,
|
||||
"is_admin": is_admin,
|
||||
"preferred_language": profile.preferred_language if profile else None,
|
||||
}
|
||||
|
||||
@@ -19,10 +19,13 @@ Security properties:
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import base64
|
||||
import io
|
||||
import logging
|
||||
from datetime import datetime
|
||||
from typing import Annotated, Any
|
||||
|
||||
import segno
|
||||
from fastapi import APIRouter, Depends, HTTPException, Request, status
|
||||
from pydantic import BaseModel, Field
|
||||
from sqlalchemy.orm import Session
|
||||
@@ -72,6 +75,7 @@ class CreateChallengeResponse(BaseModel):
|
||||
expires_at: datetime
|
||||
ttl_seconds: int = Field(description="Seconds until the challenge expires (use for client-side countdown).")
|
||||
qr_payload: str = Field(description="The string to encode in the QR code.")
|
||||
qr_code_svg: str = Field(description="Base64-encoded SVG data URI of the QR code, ready for use in an <img> src.")
|
||||
|
||||
|
||||
class ChallengeStatusResponse(BaseModel):
|
||||
@@ -106,6 +110,29 @@ class ClaimChallengeResponse(BaseModel):
|
||||
created_at: datetime
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Helpers
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
# QR code rendering parameters
|
||||
_QR_ERROR_LEVEL = "M" # Medium error correction (~15% recovery); sufficient for on-screen display
|
||||
_QR_SCALE = 4 # Each QR module is rendered as 4×4 SVG pixels
|
||||
|
||||
|
||||
def _generate_qr_svg(payload: str) -> str:
|
||||
"""Generate a QR code for *payload* and return it as a base64 SVG data URI.
|
||||
|
||||
Using ``segno`` (pure-Python, no Pillow dependency) and SVG output so the
|
||||
QR code scales crisply at any resolution without requiring a canvas or any
|
||||
client-side JavaScript library.
|
||||
"""
|
||||
qr = segno.make(payload, error=_QR_ERROR_LEVEL)
|
||||
buf = io.BytesIO()
|
||||
qr.save(buf, kind="svg", scale=_QR_SCALE, xmldecl=False, svgclass=None, lineclass=None, omitsize=True)
|
||||
svg_bytes = buf.getvalue()
|
||||
return "data:image/svg+xml;base64," + base64.b64encode(svg_bytes).decode("ascii")
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Endpoints
|
||||
# ---------------------------------------------------------------------------
|
||||
@@ -143,6 +170,7 @@ async def create_challenge(
|
||||
"expires_at": challenge.expires_at,
|
||||
"ttl_seconds": ttl_seconds,
|
||||
"qr_payload": qr_payload,
|
||||
"qr_code_svg": _generate_qr_svg(qr_payload),
|
||||
}
|
||||
|
||||
|
||||
|
||||
@@ -11,11 +11,12 @@ from typing import Optional
|
||||
|
||||
import aiofiles
|
||||
import httpx
|
||||
from fastapi import APIRouter, HTTPException, Request
|
||||
from fastapi import APIRouter, Depends, HTTPException, Request
|
||||
from pydantic import BaseModel, HttpUrl, field_validator
|
||||
|
||||
from app.auth import require_login
|
||||
from app.config import settings
|
||||
from app.middleware.upload_rate_limit import require_upload_rate_limit
|
||||
from app.tasks.process_document import process_document
|
||||
from app.utils.allowed_types import ALLOWED_MIME_TYPES
|
||||
from app.utils.filename_utils import sanitize_filename
|
||||
@@ -107,7 +108,11 @@ def validate_file_type(content_type: str, filename: str) -> bool:
|
||||
|
||||
@router.post("/process-url")
|
||||
@require_login
|
||||
async def process_url(request: Request, url_request: URLUploadRequest):
|
||||
async def process_url(
|
||||
request: Request,
|
||||
url_request: URLUploadRequest,
|
||||
_rate_ok: None = Depends(require_upload_rate_limit),
|
||||
):
|
||||
"""
|
||||
Download a file from a URL and enqueue it for processing.
|
||||
|
||||
|
||||
+20
-3
@@ -103,11 +103,18 @@ if AUTH_ENABLED and settings.social_auth_apple_enabled:
|
||||
logger.warning("SOCIAL_AUTH_APPLE_ENABLED=true but client ID/team ID not configured")
|
||||
|
||||
if AUTH_ENABLED and settings.social_auth_dropbox_enabled:
|
||||
if settings.social_auth_dropbox_client_id and settings.social_auth_dropbox_client_secret:
|
||||
# Determine which credentials to use for Dropbox social login
|
||||
_dropbox_client_id = settings.social_auth_dropbox_client_id
|
||||
_dropbox_client_secret = settings.social_auth_dropbox_client_secret
|
||||
if settings.social_auth_dropbox_use_global_credentials and not _dropbox_client_id:
|
||||
_dropbox_client_id = settings.dropbox_app_key
|
||||
_dropbox_client_secret = settings.dropbox_app_secret
|
||||
|
||||
if _dropbox_client_id and _dropbox_client_secret:
|
||||
oauth.register(
|
||||
name="dropbox",
|
||||
client_id=settings.social_auth_dropbox_client_id,
|
||||
client_secret=settings.social_auth_dropbox_client_secret,
|
||||
client_id=_dropbox_client_id,
|
||||
client_secret=_dropbox_client_secret,
|
||||
authorize_url="https://www.dropbox.com/oauth2/authorize",
|
||||
access_token_url="https://api.dropboxapi.com/oauth2/token",
|
||||
userinfo_endpoint="https://api.dropboxapi.com/2/users/get_current_account",
|
||||
@@ -186,6 +193,16 @@ def _resolve_bearer_user(request: Request, db: Session) -> dict | None:
|
||||
logger.debug("[AUTH] _resolve_bearer_user: no active API token matched the provided hash")
|
||||
return None
|
||||
|
||||
# Reject tokens that have passed their optional expiry.
|
||||
if db_token.expires_at is not None:
|
||||
now_utc = datetime.now(timezone.utc)
|
||||
expires_aware = db_token.expires_at
|
||||
if expires_aware.tzinfo is None:
|
||||
expires_aware = expires_aware.replace(tzinfo=timezone.utc)
|
||||
if now_utc > expires_aware:
|
||||
logger.debug("[AUTH] _resolve_bearer_user: API token id=%s has expired", db_token.id)
|
||||
return None
|
||||
|
||||
logger.debug(
|
||||
"[AUTH] _resolve_bearer_user: matched API token id=%s owner=%s",
|
||||
db_token.id,
|
||||
|
||||
@@ -13,6 +13,24 @@ class Settings(BaseSettings):
|
||||
|
||||
database_url: str
|
||||
redis_url: str
|
||||
|
||||
# Database connection-pool tuning (ignored for SQLite, which uses NullPool).
|
||||
db_pool_size: int = Field(
|
||||
default=10,
|
||||
description="Number of persistent connections kept in the pool per worker process.",
|
||||
)
|
||||
db_max_overflow: int = Field(
|
||||
default=20,
|
||||
description="Additional connections allowed beyond db_pool_size under burst load.",
|
||||
)
|
||||
db_pool_timeout: int = Field(
|
||||
default=30,
|
||||
description="Seconds to wait for a connection from the pool before raising a TimeoutError.",
|
||||
)
|
||||
db_pool_recycle: int = Field(
|
||||
default=1800,
|
||||
description="Recycle (close and reopen) connections after this many seconds to avoid stale connections.",
|
||||
)
|
||||
openai_api_key: str
|
||||
openai_base_url: str = "https://api.openai.com/v1" # Default to OpenAI's endpoint
|
||||
openai_model: str = "gpt-4o-mini" # Default model
|
||||
@@ -102,6 +120,16 @@ class Settings(BaseSettings):
|
||||
dropbox_app_secret: Optional[str] = None
|
||||
dropbox_folder: Optional[str] = None
|
||||
dropbox_refresh_token: Optional[str] = None
|
||||
dropbox_allow_global_credentials_for_integrations: bool = Field(
|
||||
default=False,
|
||||
description=(
|
||||
"When True, users may authorize their personal Dropbox integrations using the global "
|
||||
"DROPBOX_APP_KEY / DROPBOX_APP_SECRET credentials configured by the admin, without "
|
||||
"needing to create their own Dropbox app. The Dropbox OAuth flow is initiated "
|
||||
"server-side so the app secret is never exposed to the browser. "
|
||||
"Default: False (each user must supply their own app credentials)."
|
||||
),
|
||||
)
|
||||
|
||||
# Making Nextcloud optional
|
||||
nextcloud_enabled: bool = Field(
|
||||
@@ -165,6 +193,16 @@ class Settings(BaseSettings):
|
||||
google_docai_processor_id: Optional[str] = None
|
||||
google_docai_location: str = "us" # Processor location, e.g. "us" or "eu"
|
||||
external_hostname: str = "localhost" # Default to localhost
|
||||
public_base_url: Optional[str] = Field(
|
||||
default=None,
|
||||
description=(
|
||||
"The full public base URL of the application, including scheme "
|
||||
"(e.g., 'https://docuelevate.example.com'). "
|
||||
"When set, this overrides the auto-detected URL for OAuth redirect URIs. "
|
||||
"This is required when the application is behind a reverse proxy that does "
|
||||
"not forward X-Forwarded-Proto headers correctly."
|
||||
),
|
||||
)
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Document Translation Settings
|
||||
@@ -299,6 +337,16 @@ class Settings(BaseSettings):
|
||||
social_auth_dropbox_enabled: bool = False
|
||||
social_auth_dropbox_client_id: Optional[str] = None
|
||||
social_auth_dropbox_client_secret: Optional[str] = None
|
||||
social_auth_dropbox_use_global_credentials: bool = Field(
|
||||
default=False,
|
||||
description=(
|
||||
"When True, Dropbox social login uses the global DROPBOX_APP_KEY / DROPBOX_APP_SECRET "
|
||||
"credentials (the storage integration credentials) instead of requiring separate "
|
||||
"SOCIAL_AUTH_DROPBOX_CLIENT_ID / SOCIAL_AUTH_DROPBOX_CLIENT_SECRET values. "
|
||||
"Requires SOCIAL_AUTH_DROPBOX_ENABLED=True and the global Dropbox app credentials to be set. "
|
||||
"Default: False."
|
||||
),
|
||||
)
|
||||
|
||||
# Local user signup
|
||||
allow_local_signup: bool = Field(
|
||||
@@ -1115,6 +1163,20 @@ class Settings(BaseSettings):
|
||||
),
|
||||
)
|
||||
|
||||
# Per-user upload rate limiting (health-aware, Redis-backed sliding window)
|
||||
upload_rate_limit_per_user: int = Field(
|
||||
default=20,
|
||||
description=(
|
||||
"Maximum number of file uploads allowed per user within the sliding window. "
|
||||
"The effective limit may be reduced dynamically when the system is under heavy load "
|
||||
"(high queue depth or CPU usage). Set to 0 to disable per-user upload rate limiting."
|
||||
),
|
||||
)
|
||||
upload_rate_limit_window: int = Field(
|
||||
default=60,
|
||||
description="Sliding window size in seconds for per-user upload rate limiting (default: 60).",
|
||||
)
|
||||
|
||||
# Rate Limiting Configuration (see SECURITY_AUDIT.md and docs/API.md)
|
||||
# Protects against DoS attacks and API abuse
|
||||
rate_limiting_enabled: bool = Field(
|
||||
|
||||
+31
-2
@@ -10,6 +10,7 @@ from typing import Any
|
||||
from sqlalchemy import create_engine, exc
|
||||
from sqlalchemy.engine.url import make_url
|
||||
from sqlalchemy.orm import Session, declarative_base, sessionmaker
|
||||
from sqlalchemy.pool import NullPool, QueuePool
|
||||
|
||||
from app.config import settings
|
||||
|
||||
@@ -17,9 +18,37 @@ logger = logging.getLogger(__name__)
|
||||
|
||||
Base = declarative_base()
|
||||
|
||||
# Parse the DATABASE_URL
|
||||
# ---------------------------------------------------------------------------
|
||||
# Engine construction
|
||||
# ---------------------------------------------------------------------------
|
||||
DB_URL = settings.database_url
|
||||
engine = create_engine(DB_URL, connect_args={"check_same_thread": False})
|
||||
_parsed_url = make_url(DB_URL)
|
||||
|
||||
_connect_args: dict[str, Any] = {}
|
||||
_engine_kwargs: dict[str, Any] = {
|
||||
"pool_pre_ping": True, # detect stale / dropped connections before use
|
||||
}
|
||||
|
||||
if _parsed_url.get_backend_name() == "sqlite":
|
||||
# SQLite does not benefit from connection pooling and is prone to
|
||||
# QueuePool exhaustion under concurrent access. NullPool opens a fresh
|
||||
# connection for each request and closes it immediately afterwards,
|
||||
# completely avoiding the "QueuePool limit reached" TimeoutError.
|
||||
_connect_args["check_same_thread"] = False
|
||||
_engine_kwargs["poolclass"] = NullPool
|
||||
else:
|
||||
# PostgreSQL / MySQL — use a bounded QueuePool with configurable limits.
|
||||
_engine_kwargs["poolclass"] = QueuePool
|
||||
_engine_kwargs.update(
|
||||
{
|
||||
"pool_size": settings.db_pool_size,
|
||||
"max_overflow": settings.db_max_overflow,
|
||||
"pool_timeout": settings.db_pool_timeout,
|
||||
"pool_recycle": settings.db_pool_recycle,
|
||||
}
|
||||
)
|
||||
|
||||
engine = create_engine(DB_URL, connect_args=_connect_args, **_engine_kwargs)
|
||||
SessionLocal = sessionmaker(autocommit=False, autoflush=False, bind=engine)
|
||||
|
||||
|
||||
|
||||
@@ -20,6 +20,9 @@ How it works:
|
||||
|
||||
Exempt paths (CSRF is not checked even for state-changing methods):
|
||||
- ``/oauth-callback`` – OAuth 2.0 callback; protected by the ``state`` parameter.
|
||||
- ``/api/qr-auth/claim`` – Called by the unauthenticated mobile app; the
|
||||
cryptographically-random, single-use challenge token provides equivalent
|
||||
protection.
|
||||
"""
|
||||
|
||||
import logging
|
||||
@@ -39,6 +42,10 @@ CSRF_PROTECTED_METHODS = {"POST", "PUT", "DELETE", "PATCH"}
|
||||
# their own replay-protection mechanism).
|
||||
CSRF_EXEMPT_PATHS = {
|
||||
"/oauth-callback",
|
||||
# The mobile app calls this endpoint without a browser session/CSRF token.
|
||||
# The cryptographically-random, single-use challenge token already provides
|
||||
# equivalent protection against cross-site request forgery.
|
||||
"/api/qr-auth/claim",
|
||||
}
|
||||
|
||||
|
||||
|
||||
@@ -0,0 +1,290 @@
|
||||
"""Per-user, health-aware upload rate limiter for DocuElevate.
|
||||
|
||||
This module provides a FastAPI dependency that enforces per-user upload rate
|
||||
limits using a Redis-backed sliding window counter. The effective limit is
|
||||
dynamically reduced when the system is under heavy load (high Celery queue
|
||||
depth or elevated CPU load average), ensuring the server remains responsive
|
||||
to all users even during bulk-upload scenarios.
|
||||
|
||||
Usage in an endpoint::
|
||||
|
||||
from app.middleware.upload_rate_limit import require_upload_rate_limit
|
||||
|
||||
@router.post("/ui-upload")
|
||||
@require_login
|
||||
async def ui_upload(
|
||||
request: Request,
|
||||
_rate_ok: None = Depends(require_upload_rate_limit),
|
||||
...
|
||||
):
|
||||
...
|
||||
|
||||
See ``docs/ConfigurationGuide.md`` for the configuration options
|
||||
(``UPLOAD_RATE_LIMIT_PER_USER``, ``UPLOAD_RATE_LIMIT_WINDOW``).
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import logging
|
||||
import os
|
||||
import time
|
||||
from typing import Any
|
||||
|
||||
import redis
|
||||
from fastapi import HTTPException, Request, status
|
||||
|
||||
from app.config import settings
|
||||
from app.utils.user_scope import get_current_owner_id
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Redis key prefix
|
||||
# ---------------------------------------------------------------------------
|
||||
_KEY_PREFIX = "docuelevate:upload_rate"
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Health-check queue names (Celery defaults used by DocuElevate)
|
||||
# ---------------------------------------------------------------------------
|
||||
_CELERY_QUEUES = ("document_processor", "default", "celery")
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Singleton Redis client (lazy-initialised; fail-open when unavailable)
|
||||
# ---------------------------------------------------------------------------
|
||||
_redis_client: redis.Redis | None = None
|
||||
|
||||
|
||||
def _get_redis() -> redis.Redis | None:
|
||||
"""Return a shared Redis client, or *None* when Redis is unavailable."""
|
||||
global _redis_client
|
||||
if _redis_client is not None:
|
||||
return _redis_client
|
||||
try:
|
||||
_redis_client = redis.Redis.from_url(
|
||||
settings.redis_url,
|
||||
decode_responses=True,
|
||||
socket_connect_timeout=2,
|
||||
socket_timeout=2,
|
||||
)
|
||||
# Quick connectivity check – raises on failure.
|
||||
_redis_client.ping()
|
||||
return _redis_client
|
||||
except Exception: # noqa: BLE001
|
||||
logger.debug("Redis unavailable for upload rate limiter – falling back to allow-all", exc_info=True)
|
||||
_redis_client = None
|
||||
return None
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Health metrics helpers
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
def _get_queue_depth(r: redis.Redis) -> int:
|
||||
"""Return the total number of pending tasks across all Celery queues."""
|
||||
total = 0
|
||||
for queue_name in _CELERY_QUEUES:
|
||||
try:
|
||||
total += r.llen(queue_name)
|
||||
except Exception: # noqa: BLE001, S110
|
||||
logger.debug("Could not read queue length for %r", queue_name, exc_info=True)
|
||||
return total
|
||||
|
||||
|
||||
def _get_cpu_load_ratio() -> float:
|
||||
"""Return the 1-minute load average divided by the number of CPU cores.
|
||||
|
||||
Returns ``0.0`` on platforms that do not support :func:`os.getloadavg`
|
||||
(e.g. Windows) so that the limiter never penalises on those systems.
|
||||
"""
|
||||
try:
|
||||
load_1m = os.getloadavg()[0]
|
||||
cpu_count = os.cpu_count() or 1
|
||||
return load_1m / cpu_count
|
||||
except (OSError, AttributeError):
|
||||
return 0.0
|
||||
|
||||
|
||||
def compute_effective_limit(
|
||||
base_limit: int,
|
||||
queue_depth: int = 0,
|
||||
cpu_load_ratio: float = 0.0,
|
||||
) -> tuple[int, float, str]:
|
||||
"""Compute the effective upload rate limit based on system health.
|
||||
|
||||
The function applies a *reduction factor* (``0.0 < factor ≤ 1.0``) to the
|
||||
configured base limit. Both queue depth and CPU load contribute
|
||||
independently; the lowest factor wins.
|
||||
|
||||
Args:
|
||||
base_limit: The configured maximum uploads per window.
|
||||
queue_depth: Total pending tasks in Celery queues.
|
||||
cpu_load_ratio: 1-minute load average divided by CPU count.
|
||||
|
||||
Returns:
|
||||
A 3-tuple of ``(effective_limit, factor, reason)`` where *reason*
|
||||
is a human-readable tag for logging.
|
||||
"""
|
||||
factor = 1.0
|
||||
reason = "normal"
|
||||
|
||||
# --- Queue-depth thresholds ---
|
||||
if queue_depth > 200:
|
||||
factor, reason = min(factor, 0.10), f"critical_queue({queue_depth})"
|
||||
elif queue_depth > 100:
|
||||
factor, reason = min(factor, 0.25), f"high_queue({queue_depth})"
|
||||
elif queue_depth > 50:
|
||||
factor, reason = min(factor, 0.50), f"moderate_queue({queue_depth})"
|
||||
|
||||
# --- CPU-load thresholds ---
|
||||
if cpu_load_ratio > 3.0:
|
||||
new_factor = 0.10
|
||||
if new_factor < factor:
|
||||
factor, reason = new_factor, f"critical_cpu({cpu_load_ratio:.1f})"
|
||||
elif cpu_load_ratio > 2.0:
|
||||
new_factor = 0.25
|
||||
if new_factor < factor:
|
||||
factor, reason = new_factor, f"high_cpu({cpu_load_ratio:.1f})"
|
||||
elif cpu_load_ratio > 1.5:
|
||||
new_factor = 0.50
|
||||
if new_factor < factor:
|
||||
factor, reason = new_factor, f"moderate_cpu({cpu_load_ratio:.1f})"
|
||||
|
||||
effective = max(1, int(base_limit * factor))
|
||||
return effective, factor, reason
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Core sliding-window check (Redis sorted set)
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
def _check_and_record(
|
||||
r: redis.Redis,
|
||||
user_id: str,
|
||||
window: int,
|
||||
effective_limit: int,
|
||||
) -> dict[str, Any] | None:
|
||||
"""Atomically check the user's upload count and record the new upload.
|
||||
|
||||
Uses a Redis sorted set where each member is a unique timestamp-based ID
|
||||
and the score is the Unix timestamp. Entries older than *window* seconds
|
||||
are pruned on every call so the set never grows unbounded.
|
||||
|
||||
Returns:
|
||||
``None`` if the request is allowed, or a ``dict`` with ``count``,
|
||||
``limit``, and ``retry_after`` if the limit is exceeded.
|
||||
"""
|
||||
key = f"{_KEY_PREFIX}:{user_id}"
|
||||
now = time.time()
|
||||
window_start = now - window
|
||||
|
||||
pipe = r.pipeline(transaction=True)
|
||||
# 1. Remove entries outside the window
|
||||
pipe.zremrangebyscore(key, "-inf", window_start)
|
||||
# 2. Count current entries
|
||||
pipe.zcard(key)
|
||||
# 3. Retrieve the oldest entry's score (to compute retry_after)
|
||||
pipe.zrange(key, 0, 0, withscores=True)
|
||||
results = pipe.execute()
|
||||
|
||||
current_count: int = results[1]
|
||||
oldest_entries: list = results[2]
|
||||
|
||||
if current_count >= effective_limit:
|
||||
# Compute how long until the oldest entry expires from the window.
|
||||
if oldest_entries:
|
||||
oldest_score = oldest_entries[0][1]
|
||||
retry_after = max(1, int((oldest_score + window) - now))
|
||||
else:
|
||||
retry_after = max(1, window // 2)
|
||||
return {
|
||||
"count": current_count,
|
||||
"limit": effective_limit,
|
||||
"retry_after": retry_after,
|
||||
}
|
||||
|
||||
# 4. Record this upload (unique member = timestamp with random suffix)
|
||||
member = f"{now}:{os.urandom(4).hex()}"
|
||||
pipe2 = r.pipeline(transaction=True)
|
||||
pipe2.zadd(key, {member: now})
|
||||
pipe2.expire(key, window + 60) # TTL slightly longer than window
|
||||
pipe2.execute()
|
||||
|
||||
return None
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# FastAPI dependency
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
async def require_upload_rate_limit(request: Request) -> None:
|
||||
"""FastAPI dependency that enforces per-user upload rate limits.
|
||||
|
||||
The dependency is designed to **fail open**: if Redis is unavailable the
|
||||
request is allowed through so that uploads are never blocked by a
|
||||
monitoring outage.
|
||||
|
||||
Raises:
|
||||
HTTPException: 429 Too Many Requests when the per-user upload limit
|
||||
is exceeded. The ``Retry-After`` header indicates how many
|
||||
seconds the client should wait before retrying.
|
||||
"""
|
||||
r = _get_redis()
|
||||
if r is None:
|
||||
# Redis unavailable – fail open.
|
||||
return
|
||||
|
||||
# Identify the user (owner_id for multi-user, IP fallback).
|
||||
user_id = get_current_owner_id(request)
|
||||
if not user_id:
|
||||
user_id = f"ip:{request.client.host}" if request.client else "ip:unknown"
|
||||
|
||||
base_limit: int = settings.upload_rate_limit_per_user
|
||||
window: int = settings.upload_rate_limit_window
|
||||
|
||||
# Gather health metrics and compute effective limit.
|
||||
try:
|
||||
queue_depth = _get_queue_depth(r)
|
||||
except Exception: # noqa: BLE001
|
||||
queue_depth = 0
|
||||
|
||||
cpu_load_ratio = _get_cpu_load_ratio()
|
||||
effective_limit, factor, health_reason = compute_effective_limit(base_limit, queue_depth, cpu_load_ratio)
|
||||
|
||||
# Sliding-window check.
|
||||
try:
|
||||
rejection = _check_and_record(r, user_id, window, effective_limit)
|
||||
except Exception as exc: # noqa: BLE001
|
||||
logger.warning("Upload rate-limit check failed (allowing request): %s", exc)
|
||||
return
|
||||
|
||||
if rejection is not None:
|
||||
retry_after = rejection["retry_after"]
|
||||
logger.warning(
|
||||
"Upload rate limit exceeded: user=%s count=%d/%d window=%ds health=%s retry_after=%ds",
|
||||
user_id,
|
||||
rejection["count"],
|
||||
rejection["limit"],
|
||||
window,
|
||||
health_reason,
|
||||
retry_after,
|
||||
)
|
||||
raise HTTPException(
|
||||
status_code=status.HTTP_429_TOO_MANY_REQUESTS,
|
||||
detail=(
|
||||
f"Upload rate limit exceeded ({rejection['count']}/{rejection['limit']} "
|
||||
f"in {window}s). Retry after {retry_after}s."
|
||||
),
|
||||
headers={"Retry-After": str(retry_after)},
|
||||
)
|
||||
|
||||
if factor < 1.0:
|
||||
logger.info(
|
||||
"Upload allowed with reduced limit: user=%s effective=%d/%d health=%s",
|
||||
user_id,
|
||||
effective_limit,
|
||||
base_limit,
|
||||
health_reason,
|
||||
)
|
||||
@@ -786,6 +786,9 @@ class ApiToken(Base):
|
||||
created_at = Column(DateTime(timezone=True), server_default=func.now())
|
||||
revoked_at = Column(DateTime(timezone=True), nullable=True)
|
||||
|
||||
# Optional expiry: if set, the token is rejected after this timestamp.
|
||||
expires_at = Column(DateTime(timezone=True), nullable=True)
|
||||
|
||||
|
||||
class SharedLink(Base):
|
||||
"""Shareable, time-limited or view-limited document link.
|
||||
|
||||
@@ -205,7 +205,7 @@ def convert_to_pdf(
|
||||
".pdf", # PDF (already in PDF format but can be processed)
|
||||
}
|
||||
|
||||
IMAGE_EXTENSIONS = {".jpg", ".jpeg", ".png", ".gif", ".bmp", ".tiff", ".tif", ".webp", ".svg"}
|
||||
IMAGE_EXTENSIONS = {".jpg", ".jpeg", ".png", ".gif", ".bmp", ".tiff", ".tif", ".webp", ".svg", ".heic", ".heif"}
|
||||
|
||||
HTML_EXTENSIONS = {".html", ".htm"}
|
||||
|
||||
|
||||
@@ -68,6 +68,8 @@ IMAGE_MIME_TYPES: set[str] = {
|
||||
"image/tiff",
|
||||
"image/webp",
|
||||
"image/svg+xml",
|
||||
"image/heic",
|
||||
"image/heif",
|
||||
}
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
@@ -124,6 +126,8 @@ ALLOWED_EXTENSIONS: set[str] = {
|
||||
".tif",
|
||||
".webp",
|
||||
".svg",
|
||||
".heic",
|
||||
".heif",
|
||||
# Web
|
||||
".html",
|
||||
".htm",
|
||||
@@ -234,7 +238,7 @@ FILE_TYPE_CATEGORIES: dict[str, dict] = {
|
||||
},
|
||||
"images": {
|
||||
"label": "Images",
|
||||
"description": "Image files (.jpg, .png, .gif, .bmp, .tiff, .webp, .svg)",
|
||||
"description": "Image files (.jpg, .png, .gif, .bmp, .tiff, .webp, .svg, .heic, .heif)",
|
||||
"mime_types": frozenset(
|
||||
{
|
||||
"image/jpeg",
|
||||
@@ -245,6 +249,8 @@ FILE_TYPE_CATEGORIES: dict[str, dict] = {
|
||||
"image/tiff",
|
||||
"image/webp",
|
||||
"image/svg+xml",
|
||||
"image/heic",
|
||||
"image/heif",
|
||||
}
|
||||
),
|
||||
"extensions": frozenset(
|
||||
@@ -258,6 +264,8 @@ FILE_TYPE_CATEGORIES: dict[str, dict] = {
|
||||
".tif",
|
||||
".webp",
|
||||
".svg",
|
||||
".heic",
|
||||
".heif",
|
||||
}
|
||||
),
|
||||
},
|
||||
|
||||
@@ -39,6 +39,50 @@ SETTING_METADATA = {
|
||||
"required": True,
|
||||
"restart_required": True,
|
||||
},
|
||||
"db_pool_size": {
|
||||
"category": "Core",
|
||||
"description": (
|
||||
"Number of persistent database connections kept in the pool per worker process. "
|
||||
"Ignored for SQLite (which uses NullPool). Default: 10."
|
||||
),
|
||||
"type": "integer",
|
||||
"sensitive": False,
|
||||
"required": False,
|
||||
"restart_required": True,
|
||||
},
|
||||
"db_max_overflow": {
|
||||
"category": "Core",
|
||||
"description": (
|
||||
"Additional database connections allowed beyond db_pool_size under burst load. "
|
||||
"Ignored for SQLite. Default: 20."
|
||||
),
|
||||
"type": "integer",
|
||||
"sensitive": False,
|
||||
"required": False,
|
||||
"restart_required": True,
|
||||
},
|
||||
"db_pool_timeout": {
|
||||
"category": "Core",
|
||||
"description": (
|
||||
"Seconds to wait for a database connection from the pool before raising a TimeoutError. "
|
||||
"Ignored for SQLite. Default: 30."
|
||||
),
|
||||
"type": "integer",
|
||||
"sensitive": False,
|
||||
"required": False,
|
||||
"restart_required": True,
|
||||
},
|
||||
"db_pool_recycle": {
|
||||
"category": "Core",
|
||||
"description": (
|
||||
"Recycle (close and reopen) database connections after this many seconds "
|
||||
"to avoid stale connections. Ignored for SQLite. Default: 1800."
|
||||
),
|
||||
"type": "integer",
|
||||
"sensitive": False,
|
||||
"required": False,
|
||||
"restart_required": True,
|
||||
},
|
||||
"workdir": {
|
||||
"category": "Core",
|
||||
"description": "Working directory for file storage and processing",
|
||||
@@ -55,6 +99,18 @@ SETTING_METADATA = {
|
||||
"required": True, # Required for OAuth redirects and external URLs
|
||||
"restart_required": True,
|
||||
},
|
||||
"public_base_url": {
|
||||
"category": "Core",
|
||||
"description": (
|
||||
"Full public base URL including scheme (e.g., https://docuelevate.example.com). "
|
||||
"When set, overrides auto-detected URLs for OAuth redirect URIs. "
|
||||
"Required when behind a reverse proxy that does not forward X-Forwarded-Proto."
|
||||
),
|
||||
"type": "string",
|
||||
"sensitive": False,
|
||||
"required": False,
|
||||
"restart_required": True,
|
||||
},
|
||||
"debug": {
|
||||
"category": "Core",
|
||||
"description": "Enable debug mode for verbose logging",
|
||||
@@ -326,6 +382,19 @@ SETTING_METADATA = {
|
||||
"required": False,
|
||||
"restart_required": True,
|
||||
},
|
||||
"social_auth_dropbox_use_global_credentials": {
|
||||
"category": "Social Login",
|
||||
"description": (
|
||||
"When True, Dropbox social login uses the global DROPBOX_APP_KEY / DROPBOX_APP_SECRET "
|
||||
"credentials instead of requiring separate SOCIAL_AUTH_DROPBOX_CLIENT_ID / "
|
||||
"SOCIAL_AUTH_DROPBOX_CLIENT_SECRET values. "
|
||||
"Requires SOCIAL_AUTH_DROPBOX_ENABLED=True and global Dropbox credentials to be set."
|
||||
),
|
||||
"type": "boolean",
|
||||
"sensitive": False,
|
||||
"required": False,
|
||||
"restart_required": True,
|
||||
},
|
||||
"social_auth_dropbox_enabled": {
|
||||
"category": "Social Login",
|
||||
"description": (
|
||||
@@ -719,6 +788,18 @@ SETTING_METADATA = {
|
||||
"required": False,
|
||||
"restart_required": False,
|
||||
},
|
||||
"dropbox_allow_global_credentials_for_integrations": {
|
||||
"category": "Storage Providers",
|
||||
"description": (
|
||||
"When True, users may authorize their personal Dropbox integrations using the global "
|
||||
"DROPBOX_APP_KEY / DROPBOX_APP_SECRET credentials configured by the admin, without "
|
||||
"needing to create their own Dropbox app."
|
||||
),
|
||||
"type": "boolean",
|
||||
"sensitive": False,
|
||||
"required": False,
|
||||
"restart_required": False,
|
||||
},
|
||||
# Storage Providers - Nextcloud
|
||||
"nextcloud_enabled": {
|
||||
"category": "Storage Providers",
|
||||
@@ -2557,6 +2638,27 @@ SETTING_METADATA = {
|
||||
"required": False,
|
||||
"restart_required": False,
|
||||
},
|
||||
# Per-user upload rate limiting
|
||||
"upload_rate_limit_per_user": {
|
||||
"category": "Security",
|
||||
"description": (
|
||||
"Maximum number of uploads a single user may submit within upload_rate_limit_window seconds. "
|
||||
"The health-aware limiter may reduce this dynamically under high Redis queue depth or CPU load. "
|
||||
"Default: 20."
|
||||
),
|
||||
"type": "integer",
|
||||
"sensitive": False,
|
||||
"required": False,
|
||||
"restart_required": False,
|
||||
},
|
||||
"upload_rate_limit_window": {
|
||||
"category": "Security",
|
||||
"description": ("Sliding window in seconds over which upload_rate_limit_per_user is enforced. Default: 60."),
|
||||
"type": "integer",
|
||||
"sensitive": False,
|
||||
"required": False,
|
||||
"restart_required": False,
|
||||
},
|
||||
# Rate Limiting
|
||||
"rate_limiting_enabled": {
|
||||
"category": "Security",
|
||||
|
||||
+27
-1
@@ -14,6 +14,19 @@ from app.views.base import APIRouter, Depends, get_db, require_login, settings,
|
||||
router = APIRouter()
|
||||
|
||||
|
||||
def _get_dropbox_callback_url(request: Request) -> str:
|
||||
"""Return the Dropbox OAuth callback URL.
|
||||
|
||||
Uses ``PUBLIC_BASE_URL`` when configured so that the redirect URI displayed
|
||||
to the user (and registered in the Dropbox developer console) matches the
|
||||
one used in the OAuth authorization request. Falls back to deriving the URL
|
||||
from the incoming request when ``PUBLIC_BASE_URL`` is not set.
|
||||
"""
|
||||
if settings.public_base_url:
|
||||
return settings.public_base_url.rstrip("/") + "/dropbox-callback"
|
||||
return f"{request.url.scheme}://{request.url.netloc}/dropbox-callback"
|
||||
|
||||
|
||||
@router.get("/dropbox-setup")
|
||||
@require_login
|
||||
async def dropbox_setup_page(
|
||||
@@ -30,6 +43,8 @@ async def dropbox_setup_page(
|
||||
path from the integration's existing config is pre-populated; global
|
||||
admin credentials are never exposed in this mode.
|
||||
"""
|
||||
callback_url = _get_dropbox_callback_url(request)
|
||||
|
||||
if integration_id is not None:
|
||||
owner_id = get_current_owner_id(request)
|
||||
integration = (
|
||||
@@ -46,6 +61,12 @@ async def dropbox_setup_page(
|
||||
cfg = {}
|
||||
# Support both "folder" (DROPBOX destination) and "folder_path" (WATCH_FOLDER source)
|
||||
folder_path = cfg.get("folder", cfg.get("folder_path", ""))
|
||||
# Determine if global credentials are available for users to reuse
|
||||
global_creds_available = bool(
|
||||
settings.dropbox_allow_global_credentials_for_integrations
|
||||
and settings.dropbox_app_key
|
||||
and settings.dropbox_app_secret
|
||||
)
|
||||
return templates.TemplateResponse(
|
||||
"dropbox.html",
|
||||
{
|
||||
@@ -56,9 +77,12 @@ async def dropbox_setup_page(
|
||||
"integration_name": integration.name,
|
||||
"integration_type": integration.integration_type,
|
||||
"folder_path": folder_path,
|
||||
"app_key_value": "",
|
||||
# Only expose the public app key (not the secret) when global creds are allowed
|
||||
"app_key_value": settings.dropbox_app_key if global_creds_available else "",
|
||||
"app_secret_value": "",
|
||||
"refresh_token_value": "",
|
||||
"global_creds_available": global_creds_available,
|
||||
"callback_url": callback_url,
|
||||
},
|
||||
)
|
||||
|
||||
@@ -78,6 +102,7 @@ async def dropbox_setup_page(
|
||||
"integration_id": integration_id,
|
||||
"integration_name": None,
|
||||
"integration_type": None,
|
||||
"callback_url": callback_url,
|
||||
},
|
||||
)
|
||||
|
||||
@@ -108,5 +133,6 @@ async def dropbox_callback(request: Request, code: str = None, error: str = None
|
||||
"app_key_value": "", # The callback will prioritize sessionStorage values
|
||||
"app_secret_value": "", # The callback will prioritize sessionStorage values
|
||||
"folder_path": "", # The callback will prioritize sessionStorage values
|
||||
"callback_url": _get_dropbox_callback_url(request),
|
||||
},
|
||||
)
|
||||
|
||||
Reference in New Issue
Block a user