cxyz/src/manhwasusu/chapters.js

const { parse } = require("node-html-parser");

const { CHAPTER_PATHNAME_PATTERN, MANHWASUSU_HOSTS } = require("./constants.js");
const { fetchManhwaSusuHtml } = require("./io.js");

/**
 * @typedef {object} ChapterLink
 * @property {string} url - Absolute chapter URL.
 * @property {string} title - Chapter label (trimmed).
 */

/**
 * @param {string} pathname - URL pathname.
 * @returns {boolean} Whether this path is a ManhwaSusu chapter reader URL.
 */
function isChapterPathname(pathname) {
  const p = pathname.replace(/\/+$/u, "");
  return CHAPTER_PATHNAME_PATTERN.test(p);
}

/**
 * @param {string} url - Any http(s) URL.
 * @returns {boolean} True when hostname matches known ManhwaSusu hosts (hint only).
 */
function isLikelyManhwaSusuSite(url) {
  try {
    const { hostname } = new URL(url);
    return MANHWASUSU_HOSTS.some((h) => h === hostname.toLowerCase());
  } catch {
    return false;
  }
}

/**
 * Skip duplicate “First / Last chapter” buttons; the full list still contains the same targets.
 * @param {import("node-html-parser").HTMLElement} anchor - `<a>`.
 * @returns {boolean} True when the link text is only “First chapter” or “Last chapter”.
 */
function isFirstLastNavOnlyLink(anchor) {
  const t = (anchor.textContent || "").trim().replace(/\s+/gu, " ");
  return /^first chapter$/iu.test(t) || /^last chapter$/iu.test(t);
}

/**
 * Parses chapter anchors from a series / listing HTML document.
 * @param {string} html - Raw HTML.
 * @param {string} pageUrl - Base URL for resolving relative `href`s.
 * @returns {ChapterLink[]} Unique links in **document order** (site lists newest first).
 */
function extractChapterLinksFromListingHtml(html, pageUrl) {
  const root = parse(html);
  const base = new URL(pageUrl);

  /** @type {Map<string, ChapterLink>} */
  const byUrl = new Map();

  for (const a of root.querySelectorAll("a[href]")) {
    const href = (a.getAttribute("href") || "").trim();
    if (href.length === 0 || href.startsWith("#")) {
      continue;
    }

    let abs;
    try {
      abs = new URL(href, base).href;
    } catch {
      continue;
    }

    const { pathname } = new URL(abs);
    if (!isChapterPathname(pathname)) {
      continue;
    }

    if (isFirstLastNavOnlyLink(a)) {
      continue;
    }

    if (byUrl.has(abs)) {
      continue;
    }

    let title = "";
    const pEl = a.querySelector("p");
    if (pEl) {
      title = (pEl.textContent || "").trim().replace(/\s+/gu, " ");
    }
    if (title.length === 0) {
      title = (a.textContent || "").trim().replace(/\s+/gu, " ");
    }
    title = title
      .replace(/\s+\d+\s*(min|hour|day|week|mth|month|year)s?\s+ago$/iu, "")
      .trim();

    const fallback = pathname.split("/").filter(Boolean).pop() || "chapter";
    byUrl.set(abs, { url: abs, title: title.length > 0 ? title : fallback });
  }

  return [...byUrl.values()];
}

/**
 * @param {string} html - Series listing HTML.
 * @returns {string|null} Comic title when found in markup.
 */
function extractComicTitleFromListingHtml(html) {
  const root = parse(html);

  const stripSiteSuffix = (raw) =>
    raw.replace(/\s*-\s*ManhwaSusu\s*$/iu, "").trim().replace(/\s+/gu, " ");

  const og = root.querySelector('meta[property="og:title"]');
  if (og) {
    const t = stripSiteSuffix(og.getAttribute("content") || "");
    if (t.length > 0) {
      return t;
    }
  }

  const titleEl = root.querySelector("title");
  if (titleEl) {
    const t = stripSiteSuffix(titleEl.textContent || "");
    if (t.length > 0) {
      return t;
    }
  }

  for (const h1 of root.querySelectorAll("h1")) {
    const t = (h1.textContent || "").trim().replace(/\s+/gu, " ");
    if (
      t.length > 0 &&
      !/^ManhwaSusu$/iu.test(t) &&
      !/chapter\s*list/iu.test(t)
    ) {
      return t;
    }
  }

  return null;
}

/**
 * @param {string} pageUrl - Series or chapter URL under `/read/{slug}/`.
 * @returns {string} Series slug segment, or `"comic"`.
 */
function seriesSlugFromListingUrl(pageUrl) {
  try {
    const { pathname } = new URL(pageUrl);
    const parts = pathname.split("/").filter(Boolean);
    const readIdx = parts.indexOf("read");
    if (readIdx !== -1 && parts[readIdx + 1] !== undefined) {
      return parts[readIdx + 1];
    }
  } catch {
    /* invalid URL */
  }
  return "comic";
}

/**
 * @param {string} listingPageUrl - e.g. `https://manhwasusu.com/read/{slug}/`
 * @returns {Promise<{ chapters: ChapterLink[], comicTitle: string }>} Chapter rows and series title.
 */
async function fetchChapterLinks(listingPageUrl) {
  const html = await fetchManhwaSusuHtml(listingPageUrl);
  const chapters = extractChapterLinksFromListingHtml(html, listingPageUrl);
  const comicTitle =
    extractComicTitleFromListingHtml(html) || seriesSlugFromListingUrl(listingPageUrl);
  return { chapters, comicTitle };
}

module.exports = {
  extractChapterLinksFromListingHtml,
  extractComicTitleFromListingHtml,
  fetchChapterLinks,
  isChapterPathname,
  isFirstLastNavOnlyLink,
  isLikelyManhwaSusuSite,
  seriesSlugFromListingUrl,
};