0.2.13
0.2.13 - 2022-06-02
This release is focused on performance improvements and easier scaling. It:
- Disables multi-threaded tesseract code. This makes it easier to scale doctor in a k8s environment due to at most one CPU being used per conversion.
- Sets the number of gunicorn workers to 1 by default. This makes it so that scaling is can be moved to k8s instead of gunicorn.
- Tells tesseract not to look for white text on black backgrounds. This is just a simple performance tweak.
- Upgrades to PyPDF2 version 2.0.0.
0.2.12 - 2022-05-19
Features:
- Add an even better encoding for extract_from_html
0.2.11 - 2022-05-12
Features:
- Add even better encoding for extract_from_html
- Add better error message
0.2.10 - 2022-05-02
Features:
- Adds better encoding for extract_from_html
- Bump seal-rookery to 2.2.1
- Update seal-rookery call
0.2.9 - 2022-04-19
Features:
- Fix for mime type detection for weird PDF failures
- Test for broken PDFs
0.2.8 - 2022-04-14
Features:
- Drop m1 specific docker builds.
- Return 406's when validation of forms fails
- Add tests for incomplete post requests to the server.
- Reduce build installs and build install time.
0.2.7 - 2022-04-12
Features:
- Bump seal-rookey to speed up builds.
- Add m1 build in Makefile.
0.2.6 - 2022-04-12
Fixes:
- Add additional workers and worker resets to the gunicorn configuration. The default is now four workers, and additional ones can be created with the DOCTOR_WORKERS env.
0.2.5 - 2022-03-24
Features:
- Add two new endpoints
- Extensions from blob
- Mime type from blob
Changes:
- Drop NGINX
- Combine installation
0.2.4 - 2022-03-23
Features:
- Refactor document/extract/ endpoint to return json and drop cookies
Changes:
- Fix dockerfile update-seals
- Drop cookie support and use JSON responses when necessary
- Update tests
- Update heartbeat to match disclosure endpoint
0.2.3 - 2022-03-22
Features:
- Update type of response object
- Drop json response success = False if invalid form and just return Bad Request
Changes:
0.2.2 - 2022-03-21
Features:
- Split audio conversion into two steps: first convert to mp3 and a second method to fetch audio duration..
Changes:
- Update readme.
- Bump version to 0.2.2
- Update tests for new endpoint.
0.2.1 - 2022-03-18
Features:
- Update nginx config for longer timeouts
Changes:
- Update nginx config for longer timeouts
- Bump python version for linting
- Fix typo in DEVELOPING.md
0.2.0 - 2022-03-16
Features:
- Greatly improved documentation
- Improved speed
Changes:
- Overhauled the entire codebase
- Dropped seal-rookery image
- Switched to Django and gunicorn from uWSGI and Flask
- Completed api tests
- Added Makefile for building and pushing
- Updated NGINX config
- Added DEVELOPING.md
- Added composefile for testing with or without docker networking
- Removed financial disclosures (coming soon as a separate project).
- General improvements and cleanup.
- Add support for multiple architectures. (linux/amd64,linux/arm64)
- Added changelog
0.1.0 - 2021-11-08
0.0.36 - 2021-05-11
0.0.36 - 2021-03-17