Helpers & Utilities

Additional SDK tools for common tasks: HTML parsing, pagination, cookies, and content rating. All functions are synchronous, no async/await needed.

HTML Parsing

The SDK provides a jQuery-style HTML parsing API backed by the host’s native HTML parser (SwiftSoup on iOS) via the WIT glyph:extension/html@0.1.0 interface. No JS HTML parser is bundled, parsing happens natively on the device, which is faster and keeps bundle sizes small. In tests and dev mode, a cheerio-based shim is used instead; the API is identical, so you don’t need to think about the difference.

import { load } from '@glyphmoe/sdk'

const $ = load(html)
const title = $('h1.title').text().trim()
const cover = $('img.cover').attr('src')
const chapters = $('ul li a')
  .map((_, el) => ({
    title: $(el).text().trim(),
    url: $(el).attr('href'),
  }))
  .get()

Querying within elements

Use $(el).find(selector) or the find() method to query within a matched element:

$('.novel-card').each((_, el) => {
  const title = $(el).find('.title').text().trim()
  const cover = $(el).find('img').attr('src')
  const url = $(el).attr('href')
})

Tips

• Always call .trim() on text content, since HTML often has whitespace
• Use .attr('href') for attributes, the API doesn’t resolve relative URLs
• Check for data-src or data-lazy attributes for lazy-loaded images
• Use .html() to get inner HTML (for chapter content), .text() for plain text
• Use .find() to query within a matched element’s subtree
• .map() uses (index, element) callback order and returns .get() for a plain array

Pagination

Automatically fetch all pages of a paginated endpoint:

import { fetchAllPages, get, load } from '@glyphmoe/sdk'

const allChapters = fetchAllPages((page, accumulated) => {
  const html = get(`${BASE}/chapters?page=${page}`)
  const $ = load(html)

  const items = $('li.chapter')
    .map((i, el) => ({
      id: $(el).find('a').attr('href')!,
      title: $(el).find('a').text().trim(),
      number: accumulated + i + 1,
      url: $(el).find('a').attr('href')!,
    }))
    .get()

  const hasNextPage = $('a.next').length > 0
  return { items, hasNextPage }
})

Parameters:

• fetcher(page, accumulated): Called for each page. accumulated is the total items collected so far (useful for numbering).
• startPage: First page number (default: 1)
• maxPages: Safety limit (default: 200)

Returns: Flattened array of all items from all pages.

Cookies

Read and write cookies for the current source via the WIT glyph:extension/host@0.1.0 interface. Per-source cookie isolation is enforced by the host, one extension can’t access another’s cookies. Both getCookies() and setCookie() are synchronous.

import { getCookies, setCookie } from '@glyphmoe/sdk'

// Read all cookies for a URL
const cookies = getCookies('https://example.com')
console.log(cookies['session_id']) // 'abc123'

// Set a cookie
setCookie('https://example.com', 'auth_token', 'xyz789')

Note: Cookies are stored in memory and lost when the app restarts. This is by design for privacy. If a site requires login, the user will need to re-authenticate each session.

Content Rating

Check the user’s content rating preference to filter content server-side. These also go through the WIT glyph:extension/host@0.1.0 interface and are synchronous.

import { getMaxContentRating, isRatingAllowed } from '@glyphmoe/sdk'

// Get the raw rating level
const rating = getMaxContentRating()
// Returns: 'everyone' | 'teen' | 'mature' | 'adult'

// Check if a specific rating is allowed
if (isRatingAllowed('mature')) {
  // Show mature content
}

Rating levels (from least to most restrictive):

Automatic Rating Detection

The app automatically detects content rating from novel tags. Extensions should include tags in their novels and discover items:

Tags that trigger adult: adult, smut, hentai, erotica, r-18, r18, pornographic

Tags that trigger mature: mature, gore, sexual content, ecchi, lemon, explicit

Tags that trigger teen: teen, shounen, shoujo, violence, suggestive

If none match, the novel is rated everyone.

Discover Section Helpers

Shorthand functions to define discover page sections:

import { featured, carousel, updates, genres } from '@glyphmoe/sdk'

getDiscoverSections() {
  return [
    featured('hot', 'Hot Right Now'),           // Large hero cards
    carousel('popular', 'Popular'),             // Horizontal cover scroll
    updates('latest', 'Latest Updates', 'Subtitle text'),  // Novel + chapter rows
    genres('genres', 'Genres'),                  // Tag grid
  ]
}

Each function takes (id, title, subtitle?) and returns a DiscoverSection object with the appropriate type.