import { load } from "cheerio";

export interface PageImage { url: string; source: string; }

export function httpUrl(value: string, base?: string): string {
  const url = new URL(value, base);
  if (!["http:", "https:"].includes(url.protocol)) throw new Error("Only HTTP(S) URLs are supported.");
  url.hash = "";
  return url.href;
}

// Preserve Product.image array order; never align unrelated DOM and JSON-LD arrays by index.
export function selectPageImages(html: string, pageUrl: string): PageImage[] {
  const $ = load(html);
  const nodes: Record<string, unknown>[] = [];
  function visit(value: unknown): void {
    if (Array.isArray(value)) { value.forEach(visit); return; }
    if (!value || typeof value !== "object") return;
    const node = value as Record<string, unknown>;
    nodes.push(node);
    Object.values(node).forEach(visit);
  }
  $('script[type="application/ld+json"]').each((_, element) => {
    try { visit(JSON.parse($(element).text())); } catch { /* Skip malformed JSON-LD. */ }
  });
  const products = nodes.filter(node => [node["@type"]].flat().includes("Product"));
  const canonical = (value: unknown): string | undefined => {
    try { return typeof value === "string" ? httpUrl(value, pageUrl).replace(/\/$/, "") : undefined; }
    catch { return undefined; }
  };
  const currentPage = canonical(pageUrl);
  const product = products.find(node => canonical(node.url) === currentPage || canonical(node["@id"]) === currentPage)
    ?? products[0];
  const result: PageImage[] = [];
  const seen = new Set<string>();
  function add(value: unknown, source: string): void {
    if (result.length >= 64) return;
    if (value && typeof value === "object") {
      const obj = value as Record<string, unknown>;
      value = obj.contentUrl ?? obj.url;
    }
    if (typeof value !== "string" || !value.trim()) return;
    try {
      const url = httpUrl(value, pageUrl);
      if (!seen.has(url)) { seen.add(url); result.push({ url, source }); }
    } catch { /* Skip data URLs and unsupported schemes. */ }
  }
  if (product) {
    const images = Array.isArray(product.image) ? product.image : [product.image];
    images.forEach((image, index) => add(image, `Product.image[${index + 1}]`));
  }
  if (result.length) return result;
  $('meta[property="og:image"], meta[name="twitter:image"]').each((_, element) => {
    add($(element).attr("content"), "page metadata image");
  });
  if (result.length) return result;
  $('main img, article img').each((_, element) => {
    add($(element).attr("data-src") ?? $(element).attr("src"), "main/article img");
  });
  return result;
}

export async function readLimited(response: Response, limit: number): Promise<Buffer> {
  if (!response.ok || !response.body) {
    await response.body?.cancel();
    throw new Error(`HTTP ${response.status}: ${response.url}`);
  }
  const reader = response.body.getReader();
  const chunks: Buffer[] = [];
  let size = 0;
  try {
    while (true) {
      const { done, value } = await reader.read();
      if (done) break;
      size += value.length;
      if (size > limit) throw new Error(`Response exceeds ${limit} bytes.`);
      chunks.push(Buffer.from(value));
    }
  } finally {
    await reader.cancel();
    reader.releaseLock();
  }
  if (!size) throw new Error("Empty response body.");
  return Buffer.concat(chunks);
}

import { load } from "cheerio";

export interface PageImage { url: string; source: string; }

export function httpUrl(value: string, base?: string): string {
  const url = new URL(value, base);
  if (!["http:", "https:"].includes(url.protocol)) throw new Error("Only HTTP(S) URLs are supported.");
  url.hash = "";
  return url.href;
}

// Preserve Product.image array order; never align unrelated DOM and JSON-LD arrays by index.
export function selectPageImages(html: string, pageUrl: string): PageImage[] {
  const $ = load(html);
  const nodes: Record<string, unknown>[] = [];
  function visit(value: unknown): void {
    if (Array.isArray(value)) { value.forEach(visit); return; }
    if (!value || typeof value !== "object") return;
    const node = value as Record<string, unknown>;
    nodes.push(node);
    Object.values(node).forEach(visit);
  }
  $('script[type="application/ld+json"]').each((_, element) => {
    try { visit(JSON.parse($(element).text())); } catch { /* Skip malformed JSON-LD. */ }
  });
  const products = nodes.filter(node => [node["@type"]].flat().includes("Product"));
  const canonical = (value: unknown): string | undefined => {
    try { return typeof value === "string" ? httpUrl(value, pageUrl).replace(/\/$/, "") : undefined; }
    catch { return undefined; }
  };
  const currentPage = canonical(pageUrl);
  const product = products.find(node => canonical(node.url) === currentPage || canonical(node["@id"]) === currentPage)
    ?? products[0];
  const result: PageImage[] = [];
  const seen = new Set<string>();
  function add(value: unknown, source: string): void {
    if (result.length >= 64) return;
    if (value && typeof value === "object") {
      const obj = value as Record<string, unknown>;
      value = obj.contentUrl ?? obj.url;
    }
    if (typeof value !== "string" || !value.trim()) return;
    try {
      const url = httpUrl(value, pageUrl);
      if (!seen.has(url)) { seen.add(url); result.push({ url, source }); }
    } catch { /* Skip data URLs and unsupported schemes. */ }
  }
  if (product) {
    const images = Array.isArray(product.image) ? product.image : [product.image];
    images.forEach((image, index) => add(image, `Product.image[${index + 1}]`));
  }
  if (result.length) return result;
  $('meta[property="og:image"], meta[name="twitter:image"]').each((_, element) => {
    add($(element).attr("content"), "page metadata image");
  });
  if (result.length) return result;
  $('main img, article img').each((_, element) => {
    add($(element).attr("data-src") ?? $(element).attr("src"), "main/article img");
  });
  return result;
}

export async function readLimited(response: Response, limit: number): Promise<Buffer> {
  if (!response.ok || !response.body) {
    await response.body?.cancel();
    throw new Error(`HTTP ${response.status}: ${response.url}`);
  }
  const reader = response.body.getReader();
  const chunks: Buffer[] = [];
  let size = 0;
  try {
    while (true) {
      const { done, value } = await reader.read();
      if (done) break;
      size += value.length;
      if (size > limit) throw new Error(`Response exceeds ${limit} bytes.`);
      chunks.push(Buffer.from(value));
    }
  } finally {
    await reader.cancel();
    reader.releaseLock();
  }
  if (!size) throw new Error("Empty response body.");
  return Buffer.concat(chunks);
}