/**
 * Previews any page no other provider claims, for `rich-content-service.ts`,
 * from whatever metadata its HTML carries. Its reading of a page,
 * {@link genericPreview}, is also the base of the Bandcamp, SoundCloud and
 * YouTube channel previews.
 *
 * @module
 */

import type { RichContent } from "../../../shared/dump.ts";
import type { RichContentProvider } from "./provider.ts";
import { siteHost } from "../../../shared/url.ts";
import {
  absoluteHttpUrl,
  normalizeCssColor,
  type Page,
  parsePage,
} from "../html.ts";
import { fetchPage } from "../fetch.ts";

/** Returns the path of `url` without its trailing slashes, `/` for the root. */
function pathOf(url: string): string {
  return new URL(url).pathname.replace(/\/+$/, "") || "/";
}

/**
 * Strips a trailing ` | Site`, ` – Site` or the like from a `<title>`, which
 * names the page and then its site; the site is in the preview's badge.
 */
function withoutSiteSuffix(title: string, siteName: string): string {
  const escaped = siteName.replace(/[.*+?^${}()|[\]\\]/g, "\\$&");
  const suffix = new RegExp(`\\s+[|–—\\-·•]\\s+${escaped}\\s*$`, "i");
  return title.replace(suffix, "") || title;
}

/**
 * Returns the preview of the page `url` that `page` read, served from
 * `finalUrl` after redirects. Each field falls back from OG tags to Twitter
 * tags, JSON-LD and the markup itself. The icon and brand color feed the
 * placeholder the frontend draws when there is no thumbnail.
 */
export function genericPreview(
  page: Page,
  url: string,
  finalUrl: string,
): RichContent {
  // An og:url naming the home page from another page means the OG tags are
  // the site's defaults, not this page's: their title and image are skipped.
  // Any other og:url is the page's canonical address, often with a slug the
  // link left out, and its tags are the page's.
  const ogUrl = page.meta("og:url");
  let useOg = true;
  try {
    useOg = !ogUrl || pathOf(new URL(ogUrl, finalUrl).href) !== "/" ||
      pathOf(finalUrl) === "/";
  } catch { /* an og:url that doesn't parse names no other page */ }

  const ogSiteName = page.meta("og:site_name");
  const pageTitle = page.title && ogSiteName
    ? withoutSiteSuffix(page.title, ogSiteName)
    : page.title;
  const title = (useOg ? page.meta("og:title") : undefined) ??
    page.meta("twitter:title") ??
    page.jsonLd.title ??
    pageTitle;

  const description = page.meta("og:description") ??
    page.meta("twitter:description") ??
    page.jsonLd.description ??
    page.meta("description");

  // Never the favicon: a 16×16 icon cropped into a thumbnail looks broken,
  // and the placeholder drawn instead shows the icon properly. A tag naming
  // no usable URL falls through to the next.
  const thumbnailUrl =
    (useOg ? absoluteHttpUrl(page.meta("og:image"), finalUrl) : undefined) ??
      absoluteHttpUrl(page.meta("twitter:image"), finalUrl) ??
      page.jsonLd.thumbnailUrl ??
      page.largeImage ??
      page.firstContentImage;

  // `/favicon.ico` is a guess. When it doesn't load, the placeholder shows
  // the site's initial.
  const faviconUrl = page.bestIcon ?? `${new URL(finalUrl).origin}/favicon.ico`;

  const accentColor = normalizeCssColor(
    page.themeColor ??
      page.meta("msapplication-TileColor") ??
      page.maskIconColor,
  );

  return {
    type: "generic",
    url,
    title,
    description,
    thumbnailUrl,
    faviconUrl,
    accentColor,
    siteName: ogSiteName ?? siteHost(url),
  };
}

/** Matches every URL, so it comes last in the service's list. */
export const genericProvider: RichContentProvider = {
  name: "generic",

  matches(_url: string): boolean {
    return true;
  },

  async fetch(url: string): Promise<RichContent> {
    const fetched = await fetchPage(url);
    if (fetched.html === undefined) return { type: "generic", url };
    return genericPreview(
      parsePage(fetched.html, fetched.url),
      url,
      fetched.url,
    );
  },
};