v0.7.8 - New Filtering System

Another major feature has been reworked as I lay down those foundational changes you may hear me talk about. Kavita now has a brand new filtering system to enable users to really poke and prod their media. This was a huge undertaking ensuring a flexible system could be built and allow new features to be added upon it.

This new filtering allows for you to not only reproduce what was already there, but expand and AND your statements together. I originally sought out building a recursive system, but unfortunately I kept hitting difficulties in coding the UI, that I decided to simplify. I can always enhance later if needed. You will noticed many different combinations and new fields. I've added quite a few new fields, but if you have creative ideas, please do raise a feature request for them. I'm interested in expanding the filters even further to allow you to craft some unique filters.

In this update, I also spruced up the OPDS implementations thanks to some nice feature hub requests that got me thinking. I personally don't use OPDS, but I think for those that do, you'll notice it's a much better experience. No more formats at the end of the series names (as we put that in the description) and series descriptions are actually sending (they have been hooked up, but weren't being set, oops). While I was fixing those, I added new feeds for want to read, to make the experience the best it can be. If you've heard about the OPDS2 work, Kavita will be adding it as well. I've been involved in the discussions with Readium about the spec along with Komga, Codex, and Panels.

Lastly, localization has seen major progress. The community has really stepped up and as of me writing these release notes, Kavita is fully translated to Korean, Chinese (Simplified), and basically Italian (1 word...). Many other languages have over 80% strings translated. I've also fixed a ton of missed strings throughout the app. Thank you so much to those that have added their own languages. It's really cool seeing the application translated.

That wraps the big features. As I've been mentioning many times in these release notes, my focus for the v0.7.x releases are major foundational changes that I need so I can start the mobile apps for Kavita. I've unfortunately been struck by the Return to Office plague and due to this, have much less time to work on Kavita, so updates may slow (plus I have yet another holiday this month).

Added

• Added: From Reading list, you can now click on a character tag and open a pre-filtered search screen.
• Added: OPDS feeds now include a new section for Want to Read series
• Added: Admins now have the ability to change the cover image size. The options are: Default (320x455), Medium (640x909), Large (900x1277), Extra Large (1265x1795) (Thanks @Ark1369)
• Added: User day breakdown stat graph is now visible on a user stats page
• Added: Added an id for komf userscript to help it inject into Kavita's UI without relying on strings, given localization.
• Added: You can now filter by Path, which is the highest folder for a Series.
• Added: You can now filter by filepath, which is the full path of a file, including extension. This is much slower than Path, but still useable.
• Added: You can now filter by Series Summary (to help you find series with missing metadata)
• Added: New filter comparison called Must Contains which does just that, ensures that the values provided are in every series.
• Added: Added ability to delete single collections from card actions.

Changed

• Changed: Revamped how metadata filtering works to allow custom combinations, searching for information on nested properties, etc.
• Changed: OPDS Series names will no longer show (Format) in the series name. It will be displayed in the summary field instead along with the summary if one is set.
• Changed: If the language tag isn't properly set, then series detail wont show it anymore.
• Changed: Changed the HTTP Cache for all-languages API
• Changed: API key and OPDS url are now hidden by default. You can still copy from them or you can unhide if you want to manually copy.
• Changed: Merged images (collections and reading lists) now work differently. If there is only 1 image to select, it will use that image. If there are 2, they will be side by side, 3 will be two on top and one centered below, 4 will the the existing. This will ensure there is always a cover to view.
• Changed: Cover image on reading list detail page will now show as it's always available. Collection page still has the restraint of requiring a summary to show the cover image.
• Changed: Folder watching will now appropriately ignore changes that occur in blacklisted folders
• Changed: Rating tooltip and stars are bigger on mobile
• Changed: (Kavita+) Scrobble token validity checks are removed before writing a scrobble event to the DB
• Changed: (Kavita+) Added a lot more logging to Scrobbling operations to help debug issues, in case users are having them.
• Changed: (Performance) Optimized the license check code so that users without any license entered would still take advantage of the cache layer.
• Changed: (Performance) 10x speed improvement for checking if a library allowed scrobbling when opening up series detail page
• Changed: (Kavita+) Reviews are now mixed to show a better spectrum. You will always get the first and last 2 and a selection from the middle. This should help you get a fuller picture if the series is a good read for you.
• Changed: The Epub reader now has much better handling of highlighting when text is underneath the tap to paginate area. You can now highlight under it naturally. (Thanks @gmahomarf for driving this)
• Changed: (Performance) Removed single DB calls in many APIs.
• Changed: Updated transloco library which fixes older iOS browsers not being able to load Kavita
• Changed: Don't fully fail series metadata update if we can't cleanup due to another DB operation happening. This should help komf users that are hitting the api in short succession from having high CPU usage or long response times
• Changed: New languages for localization

Fixed

• Fixed: Fixed a bug where OPDS Feeds for Libraries, Collections, and Reading lists were not sending the image back in the XML
• Fixed: Fixed a bug where Series summary wasn't being sent over OPDS
• Fixed: Fixed a lot of localization keys
• Fixed: Fixed up Kavita rating on Series detail page not starting at the same place as other sections
• Fixed: Fixed reading list page being laggy when scrolling
• Fixed: Fixed a potential bug where series name could be taken from a chapter that isn't the first ordered (very unlikely) for epubs.
• Fixed: Fixed a bug where Volume 1.5 could be selected for series-level metadata over Volume 1.
• Fixed: Fixed a bug where expanding chapters on list item view would adjust spacing in a weird way
• Fixed: Fixed an annoying, harmless DB error when reconnecting signalR event streams, that said the entity has changed since.
• Fixed: Fixed paper background on epub reader not loading under base url
• Fixed: Makes Kavita and the entrypoint script executable, which removes the need to do that on every startup of the container, additionally is one step closer (still needs other changes) to potentially make the docker image run as rootless.
• Fixed: Fixed a bug where a user adds their own favicon and the check fails due to not stripping www.

API

• Changed: (API) The following APIs have a v2 counterpart with the existing being deprecated. I will support them for a few releases until enough of user base updates their installs:
  series/on-deck
  series/recently-added
  series/recently-updated
  series/ (post one)
  series/all
  want-to-read/
  reader/bookmarks

Known Issues

• Server Stats tab is broken. It requires selecting then refreshing your browser