- Added a benchmark script in tests/test_notifications_api.py that proved the N+1 issue issue.
- Replaced iterative DB lookups inside `for item in body.preferences:` with single pre-fetch query and local `prefs_dict` lookups.
- Verified test benchmark time drops from ~0.0964s to ~0.0141s for a batch of 100 items.
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
Replaced the synchronous `requests.post` calls in `app/api/dropbox.py` with asynchronous `httpx.AsyncClient().post` calls. This ensures that the FastAPI event loop is not blocked during network I/O, allowing better concurrent performance.
Also updated the `test_api_dropbox.py` tests to use `httpx.AsyncClient.post` in mocks and properly construct `httpx.RequestError` in exception handling tests.
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
Replaced the loop over `body.order` which generated an N+1 issue
with a single bulk query fetching all relevant `SubscriptionPlan`
records via the `.in_()` clause.
Added an in-memory dictionary map of `plan_id` to `SubscriptionPlan`
objects to allow `O(1)` lookups while updating the order.
Benchmark speedup: 14.71x faster on 500 records.
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
Replaced the loop over duplicate hashes that resulted in O(N) database queries
per page with a single efficient `in_` batch query to retrieve both originals
and duplicates. The records are then grouped in memory using dictionaries.
This resolves the N+1 performance bottleneck and reduces response time from
an average of 1.65 seconds to ~0.45 seconds locally for 500 groups.
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
Replaced the N+1 query in `list_shared_links` which fetched `FileRecord` for each link. It now uses a single query with an `outerjoin` to fetch `original_filename` alongside the `SharedLink` object.
Measured a significant improvement from ~0.4547s to ~0.0579s per 1000 links.
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
The function handling the `/google-drive/save-settings` endpoint was incorrectly named `save_dropbox_settings`, likely due to a copy-paste error. This commits renames it to `save_google_drive_settings` and updates all the tests referencing it.
Tested using standard procedures, although test execution resulted in missing dependency errors due to lack of network access in the environment.
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Fix language dropdown in base.html by moving tojson data out of HTML
attribute into a script tag (prevents quote conflicts breaking Alpine.js)
- Fix avatar placeholder 404 by correcting filename reference from
avatar-placeholder.svg to default-avatar.svg
- Add session hydration from DB in _inject_global_context() so
detect_language() uses the stored preference on every request
- Sync session and cookie in PATCH /api/profile when language changes
- Reload page after language change in profile to reflect new locale
- Add tests for session/cookie sync and DB hydration
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
Refactor `app/api/audit_logs.py` to use the `Annotated` type hint pattern for FastAPI dependencies (`Depends`) and query parameters (`Query`).
- Resolves B008: Function-call in default argument.
- Improves code maintainability and readability by following modern FastAPI best practices.
- Maintains consistency with other modules in the codebase (e.g., `files.py`, `integrations.py`).
- No changes to API runtime behavior.
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Register _() translation function in app/auth.py, app/api/local_auth.py, and
app/api/billing.py template env globals to fix jinja2.exceptions.UndefinedError
- Update test_settings_template_has_db_wizard_link to assert i18n key instead of
literal "DB Wizard" (template now uses {{ _("settings.db_wizard_btn") }})
- Add 418 missing en.json keys to all 30 non-English translation files as English
fallbacks to fix test_all_languages_have_same_keys
- Fix orphan </template> tag in pipelines.html by adding missing
<template x-if="pipelineModal.saving"> opening tag (fixes djlint H025)
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Improve warning log in _resolve_categories_for_profile() to include
exception type name for better troubleshooting
- Add SQLAlchemy IS NULL comment to imap_profiles.py filter
- Pass default_categories from server to template to avoid hardcoded
category list in JS (now uses {{ default_categories | tojson }})
- Simplify view profiles query (remove redundant unauthenticated path)
- Update docs: ConfigurationGuide.md and EmailIngestion.md with
full profiles documentation including category table and API reference
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
Add a configurable switch to control which attachment types are ingested
via IMAP. Images are excluded by default; office files and PDFs are ingested.
- Add global `IMAP_ATTACHMENT_FILTER` config setting (default: `documents_only`)
- Add `attachment_filter` column to `UserImapAccount` model for per-user override
- Migration 032 adds the column to `user_imap_accounts` table
- Update `fetch_attachments_and_enqueue()` to respect filter (documents_only/all)
- Update `pull_inbox()`, `_pull_user_imap_accounts()`, and
`_pull_user_integration_imap()` to pass the resolved filter
- Update IMAP accounts API (schemas, create/update handlers, response serializer)
- Update IMAP accounts UI to show attachment filter dropdown in modal and
display filter badges on account cards
- Add 6 new tests covering attachment filter behaviour
- Update ConfigurationGuide.md, EmailIngestion.md, and .env.demo
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
Merge main branch into compliance templates feature branch.
Main had advanced with migrations 027-030 (ensure_shared_links,
audit_logs, user_language_preference, mobile_devices) since this
branch forked. Our compliance migration was 027 with down_revision
026, which conflicted with main's 027_ensure_shared_links_table.
Changes:
- Merge main (including i18n, audit logs, mobile, GraphQL features)
- Resolve conflicts in app/api/__init__.py, app/models.py, tests/conftest.py
- Rename 027_add_compliance_templates → 031_add_compliance_templates
- Rechain: down_revision 026_add_scheduled_jobs → 030_add_mobile_devices
- Add ComplianceTemplate to migrations/env.py imports
- Alembic now has single head: 031_add_compliance_templates
- Use per-password random salt with PBKDF2-HMAC-SHA256 (stored as salt:hash)
- Increase PBKDF2 iterations to 600,000 (OWASP 2023 recommendation)
- Password for downloads now accepted via POST body (never URL query param)
- Fail download request if view count cannot be incremented (prevents bypass)
- Update tests to match new hashing format and POST password download
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Add SharedLink model with token, expiry, view limit, password hash
- Add migration 025_add_shared_links
- Add API endpoints: create, list, revoke (auth) + public info/download
- Add management UI at /shared-links with revoke controls
- Add public share landing page at /share/{token}
- Add Share button on file_view.html
- Add Shared Links to user dropdown in common.js
- Write 35 unit tests covering all scenarios
- Update UserGuide.md with sharing documentation
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Add OCR_LANGUAGES constant (28 languages, EN/DE/FR/ES/IT/PT/RU/ZH/JA/KO/AR/etc.)
- Add TESSERACT_TO_EASYOCR mapping for automatic code translation
- Add optional language constructor arg to TesseractOCRProvider/EasyOCRProvider
- Update get_ocr_providers() to accept and pass per-call language override
- Add language parameter to process_with_ocr Celery task
- Add _get_pipeline_ocr_language() helper to resolve OCR language from pipeline step config
- Update process_document to look up and pass pipeline OCR language to process_with_ocr
- Add ocr_language select config field (28 options) to pipeline OCR step schema
- Add language dropdown to pipeline UI (pipelines.html)
- Update docs/UserGuide.md and docs/API.md with language override documentation
- Add 27 new tests covering language constants, provider overrides, and pipeline lookup
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
- Convert f-string log interpolation to %s-style formatting in
app/api/pipelines.py and app/api/saved_searches.py to prevent
clear-text logging of request-derived data (CodeQL: clear-text
logging of sensitive information)
- Replace plain hashlib.sha256() with PBKDF2-HMAC-SHA256 via
hash_token() in app/auth.py for Bearer token verification,
consistent with how tokens are stored in api_tokens.py (CodeQL:
use of weak cryptographic hashing on sensitive data)
- Remove redundant {exc} from logger.exception() calls (the
traceback is already captured by logger.exception())
- Update test to verify PBKDF2 hash instead of plain SHA-256
Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
Christian Krakau-Louis