🚨 Severity: HIGH
💡 Vulnerability: User-provided IMAP `host` in `_test_imap_connection` and `pull_inbox` was not validated against private IPs, creating an SSRF risk.
🎯 Impact: Attackers could abuse the endpoints to port-scan or interact with internal/private network services.
🔧 Fix: Integrated `is_private_ip` from `app.utils.network` to block connections resolving to private, loopback, link-local, or reserved IPs.
✅ Verification: Ran `test_imap_tasks.py` and `test_api_imap_accounts.py` successfully. Checked `ruff` output and diffs. Removed all scratch files from the commit.
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Add Privacy Policy, Terms of Service, and Imprint links to WelcomeScreen
and LoginScreen for GDPR/Apple compliance (pre-login access)
- Enable multiple image selection in photo library picker
- Add HEIC/HEIF image support to backend (allowed_types, convert_to_pdf, upload handler)
- Create FileDetailScreen with processing status and logs
- Add search bar to FilesScreen with debounced search
- Set up i18n with expo-localization (EN, DE, ES, FR, IT)
- Add language selector to ProfileScreen settings
- Add Imprint link to ProfileScreen legal section
- Update docs and tests
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Move duplicate check before task enqueue in ui_upload endpoint
- Clean up temp file and return status "duplicate" for exact duplicates
- Add URI-level dedup guard in mobile UploadScreen to prevent repeated uploads
- Improve ShareContext URI normalization (collapse slashes, decode percent-encoding)
- Guard +not-found.tsx effect against re-firing for the same pathname
- Update mobile UploadResponse type and handlers for duplicate status
- Update web frontend upload.js to show duplicate status
- Update API and Configuration docs
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
Merge origin/main into branch, resolving conflict in app/database.py.
Combined improvements from both branches:
- Keep pool_pre_ping=True and structured variable approach from feature branch
- Add explicit QueuePool import and poolclass assignment from main
Introduces a Redis-backed sliding-window rate limiter for upload
endpoints (/api/ui-upload, /api/process-url) that:
- Enforces per-user limits (default: 20 uploads / 60 s)
- Dynamically reduces limits under system stress (queue depth, CPU load)
- Returns 429 with Retry-After header when exceeded
- Fails open when Redis is unavailable
- Works with the existing client-side adaptive back-off
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- audit_logs.py: Remove _db_dep pattern that fails with latest FastAPI on
Python 3.11. Use clean DbSession = Annotated[Session, Depends(get_db)]
without default values.
- billing.py: Remove owner_id from log messages to fix CodeQL clear-text
logging of sensitive information alerts.
- files.py: Remove owner_id from log messages to fix CodeQL clear-text
logging of sensitive information alerts.
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
Extracted file chunk saving and duplicate detection logic into separate helper functions (`_save_upload_file_chunks` and `_check_for_exact_duplicate`) to improve readability and maintainability of the `ui_upload` endpoint in `app/api/files.py`.
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Extract _get_user_id into shared auth.get_current_user_id() used by both
pipelines API and the assign-pipeline endpoint in files API
- Fix aria-live attribute: use two separate static containers (polite/assertive)
instead of dynamic Alpine.js binding for correct screen reader announcements
- Fix migration comment to accurately describe batch-mode FK creation
- Remove redundant tags parameter from reorder endpoint decorator
- Rename _make_file test helper to _make_test_file_record for clarity
- Update docs/UserGuide.md and docs/API.md with full Pipelines reference
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Add Pipeline and PipelineStep models with user-specific ownership
- Add pipeline_id FK column to FileRecord
- Migration 017_add_pipelines (batch mode for SQLite FK compat)
- Pipeline CRUD API at /api/pipelines with step management endpoints
- Reorder steps PUT endpoint placed before parameterised {step_id} routes
- POST /api/files/{id}/assign-pipeline for per-file pipeline assignment
- Admin-only POST /api/pipelines/admin/system for system-level pipelines
- Management UI at /pipelines (Jinja2 + Alpine.js + Tailwind)
- Pipelines link added to desktop and mobile navigation
- 41 new tests in tests/test_api_pipelines.py (all passing)
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
Subscription quota is now checked before the file is written to disk,
so users who have exceeded their quota do not waste bandwidth or disk
I/O. The post-write cleanup path for quota rejections is no longer
needed and has been removed.
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Use shared _require_admin from admin_users in subscriptions API endpoint
- Remove unnecessary Alpine.js hidden-div workaround in pricing.html
- Replace fragile string replace for OCR page count with proper Jinja {:,} format
- Improve comment wording in upload quota cleanup code
- Extract _scalar_count() helper in subscription.py to reduce repetition
- Add aria-valuemin='0' to all progressbar elements in subscription/index templates
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Add Free / Starter / Professional / Business tiers with lifetime, daily, and monthly
file limits (app/utils/subscription.py)
- Add subscription_tier column to UserProfile model + migration 014
- Enforce quotas at upload time (HTTP 402 on violation) in /api/ui-upload
- New REST API: GET /api/subscriptions/tiers, /my, /platform (admin)
- New pages: /pricing (marketing, public) and /subscription (per-user status)
- Enhanced dashboard: SaaS stats (files today/month, OCR count, active users)
in multi-user mode; original single-user layout preserved
- Admin users page: show Plan badge, allow tier editing via dropdown
- Navigation: add Pricing link + subscription icon in user header
- Tests: 23 unit tests for subscription tier logic
- Docs: docs/SubscriptionTiers.md
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Add near_duplicate_threshold config setting (default 0.85)
- New GET /api/duplicates endpoint listing all exact-duplicate groups
- New GET /api/files/{id}/duplicates endpoint returning exact + near-duplicates
- POST /api/ui-upload now returns immediate exact-duplicate warning (respects ENABLE_DEDUPLICATION)
- New /duplicates management UI with Exact Duplicates tab and Near-Duplicate Finder tab
- Add Duplicates link in admin nav menu (desktop + mobile)
- Document new config options in ConfigurationGuide.md and .env.demo
- 20 new tests covering all acceptance criteria
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Change default per_page from 50 to 25
- Rename total_items → total, total_pages → pages in pagination response
- Add next/previous URL fields to pagination response
- Update view and template to use new field names
- Update tests and API docs
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Add date range (date_from/date_to), storage provider, and tags filters to GET /api/files
- Add SavedSearch model and migration (005_add_saved_searches)
- Add CRUD API endpoints for saved searches at /api/saved-searches
- Update files.html template with new filter controls and saved searches UI
- Update files view to pass new filter parameters to template
- Add comprehensive tests for all new functionality (26 tests)
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Change default `version` param from 'original' to 'processed' so
GET /api/files/{id}/download (no param) returns the processed file
- Update docstring to reflect new default
- Add tests: ?version=processed, default→processed, invalid→400
- Fix test_file_download_missing_mime_type to use explicit ?version=original
- Add File Download section to docs/API.md
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
The upload retry logic now checks file_record.processed_file_path first
(the GPT-suggested filename stored during finalization), before falling
back to legacy hash-based and original-filename-based path patterns.
This fixes the case where the processed file has a different name than
the original (e.g., '2023-10-01_Unknown.pdf' vs 'cable_graphic.pdf')
and the retry couldn't find the file on disk.
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
Both get_file_preview and download_file were ignoring the DB-stored
original_file_path and processed_file_path fields, instead relying on
local_filename (a temp path that may be gone) and guessing patterns for
the processed file. This caused "Processed file not found" and
potentially "Original file not found" in the /files/{id} view even when
the files existed at their stored paths.
- version=original: check original_file_path first, fall back to local_filename
- version=processed: check processed_file_path first, fall back to
hash/filename guessing patterns
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Update _compute_processing_flow to recognize process_with_ocr as the OCR
stage and remap legacy process_with_azure_document_intelligence log entries
for backward compatibility
- Normalize legacy OCR step name in _compute_step_summary log fallback
- Add process_with_ocr to REAL_MAIN_STEPS/REAL_STEPS in step_manager,
file_status, and file_queries (keeping legacy name for old DB entries)
- Update retry logic in api/files.py to retry failed OCR via process_with_ocr
(handles both step names as aliases)
- Fix process_document.py to log process_with_ocr as skipped (not azure step)
for the local text extraction path
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Add RequestSizeLimitMiddleware that checks Content-Length header
before request body is read: non-multipart requests capped at
MAX_REQUEST_BODY_SIZE (default 1 MB), multipart uploads capped at
MAX_UPLOAD_SIZE (default 1 GB). Returns HTTP 413 on violation.
- Register middleware in app/main.py
- Add max_request_body_size setting to app/config.py
- Fix ui_upload in files.py to check Content-Length early and read
in 64 KB chunks (bounded memory usage), removing the post-write
os.path.getsize check
- Document MAX_REQUEST_BODY_SIZE in .env.demo and ConfigurationGuide.md
- Mark SECURITY_AUDIT.md item #4 as resolved
- Add 9 tests in test_request_size_limit.py
- Update test_upload_file_too_large to use patch.object instead of
the now-unused os.path.getsize mock
Closes#173
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Add comprehensive logging to all pipeline retry branches (process_document, process_with_azure_document_intelligence, extract_metadata_with_gpt, embed_metadata_into_pdf)
- Add original_file_path as 3rd fallback for embed_metadata_into_pdf (checks: local_filename, processed_file_path, original_file_path, workdir/tmp fallback)
- Include all checked paths with existence status in 400 error responses for easier debugging
- Add enhanced logging to upload task retry path showing which processed file paths were checked
- Log successful file path when found
- Add comprehensive test suite covering new logging and fallback behavior
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Update _retry_pipeline_step to check for file in tmp, processed, and fallback locations
- Pass full path to extract_metadata_with_gpt instead of just basename
- Update extract_metadata_with_gpt to handle both basename and full path parameters
- Add test case for retrying when file is in processed directory
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Add original_file_path and processed_file_path columns to FileRecord model
- Create database migration for new fields
- Implement get_unique_filepath_with_counter() with -0001 suffix format
- Update process_document to save immutable copy to /workdir/original
- Add force_cloud_ocr parameter to process_document for forced OCR
- Update embed_metadata to use new collision handling
- Update metadata JSON to include file path references
- Add /files/{file_id}/reprocess-with-cloud-ocr API endpoint
- Update processed_file_path in database during embedding
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Extract duplicated PDF text extraction into _extract_text_from_pdf helper
- Clarify empty metadata dict comment for embed_metadata_into_pdf retry
- Make test assertion for file_id passing more explicit
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Add file_id parameter to process_document to skip duplicate hash check on reprocess
- Pass file_id from reprocess_single_file and bulk_reprocess_files endpoints
- Extend retry-subtask endpoint to support pipeline steps (process_document,
process_with_azure_document_intelligence, extract_metadata_with_gpt,
embed_metadata_into_pdf) in addition to upload tasks
- Add retry button for failed main pipeline steps in file detail UI
- Add comprehensive tests for reprocessing and pipeline step retry
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
google-labs-jules[bot]