v0.11.0

0.11.0 (2020-01-06)

Bug Fixes

• web reader: remove webp as it's not supported in Safari (6770107)
• add support for jpeg2000/jbig2 formats (227975a), closes #50

Features

• handle archives without images (70a2da5), closes #56
• web reader: rtl option (98efa9b), closes #53
• display version in UI (4085f1f), closes #42
• rescan library (30208a2), closes #38
• api: reAnalyze library (fa65e94)
• api: reAnalyze series (e80451f)
• rest api: ability to re-analyze a book (8e81356), closes #51
• rest api: don't return hidden files for directory listings (a478d90)
• webui: add menu option to analyze book from browsing view (64f542d), closes #51
• webui: reAnalyze library (b599b72), closes #51
• webui: reAnalyze series (b997561)

v0.10.1

0.10.1 (2020-01-01)

Bug Fixes

• webui: remove CDN usage for icons and fonts (c88a27c), closes #45
• webui: show all books when browsing series (85ca99d)
• zip extractor: better handling of exotic charsets (0254d7d), closes #41

v0.10.0

Features

• webui: added Web Reader (#28)
• webui: display all books instead of only books in ready state
• webui: add 'Date updated' sort criteria for Series
• webui: add 'File size' sort criteria for Books
• api: handle the HTTP cache properly for dynamic resources: thumbnails and pages (#27)
• api: hide full path to non-admin users (for libraries, series and books)
• scanner: add configuration key to force the last modified time of directories (#37)

Fixes

• webui: rework dashboard sliders to be more touch-friendly
• webui: better display on mobile and small screens
• remove regeneration of missing thumbnails at startup (don't remember why I added it in the first place ¯_(ツ)_/¯ )

v0.9.1

This release is focused on performance enhancements.

Changes

• Hibernate lazy loading was not working because of Kotlin closed by default classes
• add Hibernate caches: second level, collections, query
• enhance code path to reduce number of database queries
• fix a bug where Series thumbnail would be retrieved by the first book in the collection, instead of the first by number
• remove (unused) bi-directional OneToOne relationship between Book and BookMetadata to reduce database fetch

v0.9.0

Features

• Web Interface enhancements (closes #18)
  • browse Libraries, Series, and Books
  • dashboard with recently added Series and Books
  • search Series and Books
  • remove browser pop-up for basic auth, use a dedicated login screen instead
  • http cache for static assets

Changes

• retrieve real ip in audit logs (for example if behind a reverse proxy)
• remove Humio metrics
• libraries are sorted by name in the API by default

Deprecation

• endpoints of the form /series/{seriesId}/books/{bookId}/**, use /books/{bookId}/** instead

Known issues

• UI is extremely slow when browsing libraries/series with many items (75+) on Chrome Android

v0.8.1

Changes

• add metrics exporter for InfluxDB (disabled by default)

v0.8.0

Changes

• change docker base image to adoptopenjdk 11
• retrieve file size of books when scanning. Add file size in REST API and OPDS.

Bug fixes

• Swagger is not showing Kotlin nullable types as optional (fixes #15)
• compare file modification time at millisecond. With JDK 9+ on Windows, time precision of the underlying clock would go over 6 digits, which is the precision of the timestamps in database, which would lead to loss of precision on saved timestamps, and failing comparisons of modification times at every library scan
• return only books in ready state via OPDS, else it would throw an error 500 because metadata is not ready
• prevent circular loop in the error resolver, which would complain in logs

v0.7.1

Bug fixes

• could not delete a library if it was specifically shared with any user

Changes

• add build and git info in /actuator/info endpoint
• add humio metrics exporter (disabled, need to be enabled by configuration)

v0.7.0

⚠️ Breaking changes

• admin and user users are deprecated and replaced by the User Management feature

Features

• User management:
  • Automatic creation of an admin account at startup if no user exist in database, outputting the login and a random password in the logs
  • Ability to add/remove user accounts
  • Ability to manage roles: Admin/User
  • Ability to manage access to shared libraries per user
  • Ability for a user to change his/her password

Changes

• OPDS entry links to file now include the full filename. This helps some (badly implemented) OPDS clients to correctly see the files.
• logs are written to disk, and available via the /actuator/logfile endpoint

Bug fixes

• fix OPDS link for page streaming where an incorrect url escape in the zero_based query parameter would generate an error for the first page, shift all pages, and prevent the last page to be streamed

v0.6.1

Fixes:

• Thumbnail media type was incorrect for OPDS feed
• /series endpoint library filter can accept a list instead of a single value
• /filesystem endpoint now sorts directories with case insensitive