Announcements

:wave: Hello friends, a couple of new data and Library Importer features for ya' ## :new: My Library Importer improvements We're adding two new features to the Library Importer system today ### ComicK importer We now natively support [ComicK export files](/my/library/import/comick) in the library importer. It should work just like all the other importers, if not, report it in [Inbox](https://mangabaka.org/forum/mangabaka/inbox) ### :magic_wand: Universal Importer Okay, not going to lie, we will never support _all_ the semi-random formats of JSON and CSV exports that exists out there, it's a never ending stream of new formats coming and going, and frankly, I'm tired of manually adding new format support every couple of months - so I'm done adding more importers to MangaBaka ... .... because I probably (_hopefully_!) don't have to, thanks to the new :sparkles: [Universal importer](/my/library/import/universal) :sparkles: #### What the heck is an universal importer ?!  _Excellent question!_ The Universal Importer can read (almost) any CSV or JSON format file, and allow _you_ to map the columns (CSV) and fields (JSON) to specific MangaBaka fields the importer system uses.   And not only that, you can also map specific values (like custom lists and reading status) to MangaBaka importer values, giving you complete control of the entire process. The system auto-saves your mapping configuration, but it can also be exported and imported, so if your backup is in a custom (or common) format you can reuse the mapping settings someone else made, which can be useful considering it's a relatively complex system compared to the first party importers we support. Share your configurations over in POST!73 Once everything is mapped, you will see the familiar `import` UI from the more specialized import pages like Anilist and MyAnimeList.  ### How it works 1. The system will accept any `.json` and `.csv` file, optionally `.gz` compressed 2. It will sniff for UTF-16 markers, and fall back to UTF-8 if none is found and then read th file 3. For CSV it will try to sniff the separator used, `,`, `;` or `\t` 4. For JSON it will check that the document is an array of identical (key wise) objects - JSON files must be an array of flat objects, we do not support nested keys at this time 5. The system will then sample ~50 rows and try to guess what the value is for each and auto-map them for your review - including date formats. 6. It will apply the full mapping to the file on the client, check things look good (e.g. `reading status` have a reasonable amount of unique values, like 15) etc. 7. The mapped and parsed data will then be shipped to our backend for further inspection, validation, enrichment, etc. and you get to review the result before importing the data. The system is capped at 50.000 entries per file, which is ~20% more than the largest libraries we have today, so think we should be okay in that regard.**bold text** ### What's next? For importing, I think we're pretty much done, we will rely on the community to come up with the needed Universal Importer configurations for systems we don't support. Share your configurations in POST!73 5 The next thing I want to make Exporters for everything we support Importing. I'm strong believer in open data, moving in and out (and between) systems with ease and without lock-in, so expect some more exporters in the coming weeks, as well as a `Universal Exporter` similar to the importer. ## :new: `Original Language` series field As part of the work towards a fully integrated Publisher system, that is a prerequisite for a bunch of other stuff in the future, we've added a new `Original Language` Series field to the API and Database Dumps. You can not currently filter on it, and it's not visible on the Series cards yet, but the data is there to read and consume for just about all series but 0.4% the system couldn't somewhat reliably detect or decide it for. The field is also available in the Series Submission system as of today.