// หน้าแรก > งานบริการ > KM องค์กรแห่งการเรียนรู้

const { execSync } = require("child_process");
const cheerio = require("cheerio");
const fs = require("fs");
const path = require("path");
const axios = require("axios").default;

const BASE = "https://ladsawai.go.th";
const OUT = path.join(process.cwd(), "KM องค์กรแห่งการเรียนรู้");
fs.mkdirSync(OUT, { recursive: true });

function curlHtml(url) {
  return execSync(
    `curl -L -s "${url}" -H "User-Agent: Mozilla/5.0" -H "Accept-Language: th-TH,th;q=0.9"`,
    { encoding: "utf8", maxBuffer: 30 * 1024 * 1024 }
  );
}

function absUrl(href) {
  if (!href) return null;
  if (href.startsWith("http")) return href;
  if (href.startsWith("/")) return BASE + href;
  return BASE + "/" + href;
}

function scrapeDetailImagesContent(detailUrl) {
  const html = curlHtml(detailUrl);
  const $ = cheerio.load(html);

  // ---------- images ----------
  const imgSet = new Set();

  $(".maingroup.gallery a[href]").each((_, a) => {
    const href = ($(a).attr("href") || "").trim();
    const full = absUrl(href);
    if (full) imgSet.add(full);
  });

  if (imgSet.size === 0) {
    $("a[href]").each((_, a) => {
      const href = ($(a).attr("href") || "").trim();
      const full = absUrl(href);
      if (full && /\.(jpg|jpeg|png|webp|gif)(\?|$)/i.test(full)) imgSet.add(full);
    });
  }

  // ---------- content ----------
  // ✅ เลือกกล่องที่ไม่ใช่ gallery และ "มีข้อความจริง"
  const candidates = $(".col-12.maingroup").not(".gallery");

  let bestBox = null;
  let bestScore = -1;

  candidates.each((_, el) => {
    const $el = $(el);

    // เอา text โดยตัดของไม่เกี่ยว (emoji img, script, style)
    const text = $el
      .clone()
      .find("img, script, style")
      .remove()
      .end()
      .text()
      .replace(/\s+/g, " ")
      .trim();

    const pCount = $el.find("p").length;
    const score = (text ? text.length : 0) + pCount * 50; // ให้ p มีน้ำหนักเพิ่ม

    if (score > bestScore) {
      bestScore = score;
      bestBox = $el;
    }
  });

  let content = "";
  if (bestBox && bestBox.length) {
    const lines = [];

    // วนตามลำดับจริงใน DOM: ทั้งหัวข้อ (h2) และเนื้อหา (p)
    bestBox.find("h2, p").each((_, el) => {
      const $node = $(el);
      const tag = ($node.prop("tagName") || "").toLowerCase();

      const t = $node
        .clone()
        .find("img") // ตัดรูป emoji ใน p/h2
        .remove()
        .end()
        .text()
        .replace(/\s+/g, " ")
        .trim();

      if (!t) return;

      // แยกหัวข้อให้เด่นขึ้นเล็กน้อย (ยังคงเป็น plain text)
      if (tag === "h2") lines.push(t);
      else lines.push(t);
    });

    content = lines.length
      ? lines.join("\n")
      : bestBox
          .clone()
          .find("img, script, style")
          .remove()
          .end()
          .text()
          .replace(/\s+/g, " ")
          .trim();
  }

  let mainImageUrl = ''
  try{
    const mainImageDiv = $(".imagestopic img[src]");
    const src = ($(mainImageDiv).attr("src") || "").trim();
    const full = absUrl(src);
    if (full) mainImageUrl = full;
  }
  catch(error){
    mainImageUrl = '' 
  }

  return { imgs: [...imgSet], text: content, mainImage: mainImageUrl };
}

function buildUrl(menuId, page) {
  return `${BASE}/public/list/data/index/menu/${menuId}/page/${page}`;
}

function detectTotalPages($) {
  let maxPage = 1;
  $("a").each((_, a) => {
    const t = $(a).text().trim();
    if (/^\d+$/.test(t)) maxPage = Math.max(maxPage, Number(t));
  });
  return maxPage;
}

function extractFileLinksFromDetail(detailUrl) {
    const html = curlHtml(detailUrl);
    const $ = cheerio.load(html);
  
    const files = [];
  
    $("a.uploadconfig_link").each((_, a) => {
      const el = $(a);
      const href = el.attr("href");
      const dataHref = el.attr("data-href");
      const fileUrl = absUrl(dataHref || href);
      if (!fileUrl) return;
  
      const text = el.text().replace(/\s+/g, " ").trim() || null
      let title = text
      let downloadCount = 0
      if(text && text.includes('ดาวน์โหลดแล้ว')){
          try {
              const splitList = text.split(' ดาวน์โหลดแล้ว ')
              title = splitList[0]
              downloadCount = parseInt(splitList[1].replace('ครั้ง', '').trim())
          } catch (error) {
              title = text
              downloadCount = 0
          }
      }
  
      files.push({
        text: title,
        url: fileUrl,
        downloadCount: downloadCount
      });
    });
  
    // fallback: ลิงก์ไฟล์แบบตรง ๆ
    $("a[href]").each((_, a) => {
      const href = $(a).attr("href");
      const u = absUrl(href);
      if (!u) return;
  
      if (/\.(pdf|doc|docx|xls|xlsx|ppt|pptx|zip|rar)(\?|$)/i.test(u)) {
        if (!files.some((f) => f.url === u)) {
          files.push({ text: $(a).text().trim() || null, url: u });
        }
      }
    });
  
    return files;
  }

// ✅ ยิง api /status/1/ เพื่อเอา path จริง
async function resolveRealFilePath(fileUrl) {
  try {
    // กันกรณีมี / ท้ายอยู่แล้ว
    const statusUrl = fileUrl.replace(/\/$/, "") + "/status/1/";
    const res = await axios.get(statusUrl, { timeout: 30000 });
    return res?.data?.path || null;
  } catch (e) {
    return null;
  }
}

// ✅ limit concurrency แบบง่าย (กันยิงหนักเกิน)
async function mapLimit(arr, limit, mapper) {
  const ret = [];
  let i = 0;

  async function worker() {
    while (i < arr.length) {
      const idx = i++;
      ret[idx] = await mapper(arr[idx], idx);
    }
  }

  const workers = Array.from({ length: Math.min(limit, arr.length) }, worker);
  await Promise.all(workers);
  return ret;
}

async function scrapeOnePage(menuId, page, saveHtml = false) {
    const url = buildUrl(menuId, page);
    const html = curlHtml(url);
  
    if (saveHtml) {
      fs.writeFileSync(
        path.join(OUT, `debug-menu-${menuId}-page-${page}.html`),
        html,
        "utf8"
      );
    }
  
    const $ = cheerio.load(html);
  
    // ✅ แปลง rows เป็น array ก่อน
    const rows = $(".row.data-row").toArray();
  
    // ✅ ประมวลผลแบบมี limit (เช่น 5 concurrent)
    const items = (await mapLimit(rows, 5, async (row) => {
      const el = $(row);
      const a = el.find("a.listdataconfig_link[href]").first();
      if (!a.length) return null;
  
      const title =
        a.find("label.font-weight").text().replace(/\s+/g, " ").trim() ||
        a.text().replace(/\s+/g, " ").trim();
  
      if (!title) return null;
  
      const detailUrl = absUrl(a.attr("href"));
      let files = [];
      let realPathFiles = []
  
      try {
        if (detailUrl) files = extractFileLinksFromDetail(detailUrl);
        for(let i = 0; i < files.length; i++){
          const file = files[i]
          let realPath = null;
          let fileObject = {
              fileName: file.text,
              fileUrl: file.url, // ไฟล์จากหน้า detail
              filePath: "", // ✅ ของจริงจาก api /status/1/
              downloadCount: file.downloadCount
          }
          try {
              const fileUrl = file?.url ? absUrl(file.url) : null;
              if (fileUrl) {
                realPath = await resolveRealFilePath(fileUrl);
                fileObject.filePath = `https://ladsawai.go.th/public/${realPath}`
              }
          } catch (error) {
              realPath = null;
          }
  
          realPathFiles.push(fileObject)
        }
      } catch (e) {
        files = [];
      }

      let detail = undefined
      if(files.length == 0){
        const { text, imgs, mainImage } = detailUrl ? scrapeDetailImagesContent(detailUrl) : [];

        detail = {
            img: imgs,
            content: text,
            mainImage: mainImage
        }
      }
  
      let detailHtmlPath = ''
      if (saveHtml) {
        try {
          // Extract id and menu from URL: /public/list/data/detail/id/{id}/menu/{menu}/page/{page}
          const urlMatch = detailUrl.match(/\/id\/(\d+)\/menu\/(\d+)/);
          const id = urlMatch ? urlMatch[1] : null;
          // const menu = urlMatch ? urlMatch[2] : null;
          
          const detailPageHtml = curlHtml(detailUrl);
  
          detailHtmlPath = `debug-menu-${menuId}-detail-${id}.html`
          fs.writeFileSync(
            path.join(OUT, detailHtmlPath),
            detailPageHtml,
            "utf8"
          );
  
        } catch (error) {
          console.error('error :', error)
          detailHtmlPath = ''
        }
      }
  
      return {
        title,
        detailUrl: detailUrl || null,
        files: realPathFiles,
        detail: detail,
        detailPageHtml: detailHtmlPath ?? undefined, // ไฟล์จากหน้า detail
        sourcePage: page,
        sourceUrl: url,
      };
    }))
      .filter(Boolean); // ตัด null ออก
  
    const output = {
      source: url,
      scrapedAt: new Date().toISOString(),
      menuId,
      page,
      count: items.length,
      items,
    };
  
    fs.writeFileSync(
      path.join(OUT, `menu-${menuId}-page-${page}.json`),
      JSON.stringify(output, null, 2),
      "utf8"
    );
  
    console.log(`✅ page ${page} -> items ${items.length}`);
    return { $, items };
}

(async function main() {
  const menuId = 1628;

  const first = await scrapeOnePage(menuId, 1, true);
  const totalPages = detectTotalPages(first.$);
  console.log("✅ totalPages =", totalPages);

  const all = [];
  const seen = new Set();

  function addItems(items) {
    for (const it of items) {
      const key = `${it.title}|${it.detailUrl || ""}|${it.filePath || ""}`;
      if (seen.has(key)) continue;
      seen.add(key);
      all.push(it);
    }
  }

  addItems(first.items);

  for (let p = 2; p <= totalPages; p++) {
    const { items } = await scrapeOnePage(menuId, p, false);
    addItems(items);
  }

  const merged = {
    menuId,
    totalPages,
    scrapedAt: new Date().toISOString(),
    totalItems: all.length,
    items: all,
  };

  const outAll = path.join(OUT, `menu-${menuId}-all.json`);
  fs.writeFileSync(outAll, JSON.stringify(merged, null, 2), "utf8");

  console.log("🎉 Saved all:", outAll);
  console.log("🎉 Total unique:", all.length);
})();