hummingbird-me / kitsu-server Goto Github PK

Few, if any, chapters will have an official description and so relies on users to create them. Volumes on the other hand, will usually have descriptions created by the publisher.

Similarly to the above and it seems that the DB has an image field for volumes but isn't exposed via GQL.

• Gravatar
• Libravatar

Type SiteLink is missing a name field in the GraphQL API. This can be included in the JSON API like this https://kitsu.io/api/edge/users/1/profile-links?include=profileLinkSite.

Sometimes, I'd like to be able to permalink to a specific version of a post, like I can on StackOverflow (https://biology.stackexchange.com/revisions/108774/5).

Currently you can get a Volume type via the chapter. It would also make sense to have the volumes on the root of the Manga type the same as chapters.

With the advent of ActivityPub powering content delivery across PixelFed and Mastodon as their user counts skyrocket, I see no disadvantage to being the first anime tracker to support ActivityPub.

https://fediverse.party/en/fediverse and danbooru/danbooru#5446 (comment) provide some additional information.

Hey, the import has been broken for a while and I think you are aware of it anyway, but I thought I'd just make an issue so people can keep track of it :)

I've been planning to use it as a middleman format to export from other anime and manga websites.

It appears that searchMangaByTitle returns R18 even without an authorisation token.

query minMangaByTitle($title: String!, $limit: Int, $after: String) {
  searchMangaByTitle(first: $limit, title: $title, after: $after) {
    nodes {
      id
      titles {
        canonical
      }
      sfw
      ageRating
      }
    }
}

{
  "title": "innocent",
  "limit": 2,
  "after": ""
}

Result:

{
  "data": {
    "searchMangaByTitle": {
      "nodes": [
        {
          "id": "25489",
          "titles": {
            "canonical": "Innocent"
          },
          "sfw": true,
          "ageRating": "R"
        },
        {
          "id": "13942",
          "titles": {
            "canonical": "Innocent"
          },
          "sfw": false,
          "ageRating": "R18"
        }
      ]
    }
  }
}

findMangaById does appear to filter as 13942 returns null. I also checked searchAnimeByTitle which IS filtering out R18. I also checked the search with the JSON:API and that DOES filter too.

Apologies if this is already know.

I should imagine one for findEpisodeById would be useful too.

I can't feasibly keep re-enabling web notifications every time I reinstall my browser(/OS) or switch to a new device, so in lieu of the work necessary to setup and maintain e-mail notifications, could you instead provide an Atom feed? Atom is the standardized successor to RSS (it's basically identical except that it works - see bookwyrm-social/bookwyrm#3221 (comment) for a comprehensive explanation).

This issue is causing me a fair bit of problems when querying for episodes.

Hi,

I have some questions (perhaps an FAQ might be a good idea) about self-hosting. I may be interested in self-hosting my own instance of Kitsu - in theory, this should be possible, right? Is there any documentation for how one would go about doing this? I see there's documentation for contributing to this project, but I don't see any wiki or anything for self-hosting.

For example, I couldn't help but notice it looks like there isn't any DB seeder for like a complete list of anime, so how would that work? How would you populate the database with a full list of anime?

Also, for self-hosting something like this, would you just run the API (kitsu-server), then download the front-end (kitsu-web) and just point the front-end to your own API?

Cheers.