Unreleased

scraper will fail when there are too many errors while retrieving xblocks
move all inline Javascript to standalone files to comply with some CSP
no more retry for login + HTTP errors
display HTTP error responses or save them in a temporary file for analysis
save API response for course xblocks in temporary file for debugging
pin markupsafe until we upgrade Jinja2
fix issue with video that do not need to be re-encoded and were marked as missing
README.md: replace course in sample command by one under a Creative Commons license

1.0.1

fixed recursive paths and URLs in html_processor.py
fixed usage on older browsers (without ES6 support)
added multithreading support
fixed missing videos due to youtube_dl error
added support for convertion in high quality
use dynamic versions for opensans font
fixed mobile navigation menu
fixed video encoding in low quality in WebM
added support for internationalization
have proper language in ZIM metadata
updated zimscraperlib

1.0.0

new structure for the project
using fixed version of dependencies
wiki and forums now optional
using zimscraperlib for downloading and optimizing videos
refactored Dockerfile
added s3 based optimization cache
removed javascript dependencies from repository
added support for webm on systems without native support
fixed small favicons
using pylibzim for creating ZIMs
fixed internal links
using params instead of instance configs
using course layout from instance
added videojs based audio player to solve compatibility issues
fixed problem answer fetching