Content and chapters

Identifiers

You select the identifier of a title in your source. The app joins it with your source identifier as sourceID||contentID. Your identifier must not hold ||.

getContent(contentId)

This method returns one title. Only title and coverImage are necessary.

async getContent(contentId: string): Promise<Content> {
  return {
    title: "Series Name",
    coverImage: "https://example.org/cover.jpg",
    bannerImage: "https://example.org/banner.jpg",
    summary: "...",
    status: ContentStatus.ONGOING,
    contentType: ContentType.MANGA,
    readingMode: ReadingMode.VERTICAL,
    webUrl: `https://example.org/series/${contentId}`,
    additionalTitles: ["Alternate name"],
    genres: [{ id: "action", title: "Action" }],
    credits: [{ name: "Author Name", role: "Story" }],
    endpoints: { anilist: "12345", mal: "678" },
  };
}

The app also reads artworks, statistics, additionalDetails, properties, collections, links, characters and chapters. The last one lets you return the chapter list here, and not in a separate method.

status

Value	Meaning
0	Unknown
1	Ongoing
2	Completed
3	Cancelled
4	Hiatus

contentType

Value	Meaning
0	Manga
1	Manhua
2	Manhwa
3	Comic
4	Novel

readingMode

Value	Meaning
0	Paged Manga, right to left
1	Paged Comic, left to right
2	Paged Vertical
3	Vertical, one continuous strip

The readingMode value is the strongest signal. It wins over the tags and over the default of the content type. Set it when you know the correct mode. Leave it out when you do not, and let the app decide.

endpoints

This field maps a tracker name to the identifier of this title on that tracker. Two functions use it:

1. The link between the title and an external tracker.
2. Linked Titles. Two entries with the same value are the same series.

The keys must match the values in the endpoint array of your configuration.

The rule for novels

A title with contentType: ContentType.NOVEL declares that its chapters come from getChapterText. The app does not go back to getChapterPages for that title.

A source with novels and image titles together writes both methods. The contentType of each title selects the method.

getChapters(contentId)

This method returns an array of chapters, with the newest first.

async getChapters(contentId: string): Promise<Chapter[]> {
  return chapters.map((chapter, position) => ({
    id: chapter.id,
    index: position,
    number: chapter.number,
    volume: chapter.volume ?? undefined,
    title: chapter.title,
    language: "en",
    date: new Date(chapter.published_at),
    webUrl: `https://example.org/chapter/${chapter.id}`,
  }));
}

Field	Necessary	Notes
id	Yes	The app sends it to getChapterPages.
index	Yes	The position in the array. 0 is the top.
number	Yes	The number of the chapter.
volume	No	Leave it out if you do not know it.
isSpecial	No	For an extra chapter or a special chapter.
language	No
title	No
date	No
webUrl	No
coverImage	No
isLocked	No	The chapter exists but a person cannot read it.
providers	No	The credit for a translation group.

The index field and the number field do different work. The index field sets the order. The number field is the identity of the chapter.

The app compares the number field to find the chapters that a person did not read. If your source gives a chapter a different number between two calls, the app shows unread chapters that the person already read.

getChapterPages(contentId, chapterId)

This method returns the array itself. Do not put the array in an object.

async getChapterPages(contentId: string, chapterId: string): Promise<ChapterPage[]> {
  return pages.map((page) => ({ url: page.url }));
}

Each page gives one source of image data:

Field	Use
url	Get the image from this address. This is the normal case.
b64	Base64 image data, for a page that your code builds.
raw	Raw data.
context	Data that the app sends back to willRequestImage and redrawImage.

getChapterText(contentId, chapterId)

This method returns text, not images. Use it for a novel.

return {
  format: "html",
  body: "...",
  resources: [
    { href: "images/1.png", url: "https://example.org/1.png", mediaType: "image/png" },
  ],
};

The format field takes html or plainText. The app treats HTML as not trusted and cleans it before it draws it. The app takes plain text as it is.

The resources array maps a reference in body to an address. The href value is the stable reference in your text. The url value is request data only.

Related

• Pages
• Capabilities