import { buildUrl, isEnglishContent, NON_ENGLISH_RE, JUNIOR_TITLE_RE, JAVA_TITLE_RE } from "../fetch.ts";
import type { SearchConfig } from "../config.ts";

const HEADERS = {
  "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36",
  "Accept-Language": "en-US,en;q=0.9",
  Accept: "text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8",
};

function clean(raw: string): string {
  return raw.replace(/<[^>]+>/g, "").replace(/&amp;/g, "&").replace(/&lt;/g, "<").replace(/&gt;/g, ">").replace(/&quot;/g, '"').replace(/&#39;/g, "'").replace(/\s+/g, " ").trim();
}

// parseJobs without Switzerland location guard
function parseJobsGlobal(html: string): { id: string; title: string; company: string; location: string }[] {
  const jobs = [];
  const idRe = /data-entity-urn="urn:li:jobPosting:(\d+)"/g;
  const positions: { id: string; start: number }[] = [];
  let m;
  while ((m = idRe.exec(html)) !== null) positions.push({ id: m[1], start: m.index });

  for (let i = 0; i < positions.length; i++) {
    const { id, start } = positions[i];
    const card = html.slice(start, Math.min(positions[i + 1]?.start ?? Infinity, start + 3000));
    const titleM = card.match(/<h3 class="base-search-card__title">\s*([\s\S]*?)\s*<\/h3>/);
    if (!titleM) continue;
    const companyM = card.match(/<h4 class="base-search-card__subtitle">[\s\S]{0,300}?<a[^>]*>\s*([\s\S]*?)\s*<\/a>/);
    const locM = card.match(/job-search-card__location[^>]*>\s*([^<]+)/);
    jobs.push({
      id,
      title: clean(titleM[1]),
      company: companyM ? clean(companyM[1]) : "",
      location: locM ? locM[1].trim() : "",
    });
  }
  return jobs;
}

async function fetchGlobal(config: SearchConfig): Promise<{ id: string; title: string; company: string; location: string }[]> {
  const baseUrl = buildUrl(config);
  const all = [];
  const seen = new Set<string>();

  for (let page = 0; page < 6; page++) {
    if (page > 0) await Bun.sleep(400);
    const url = `${baseUrl}&start=${page * 25}`;
    process.stdout.write(`  page ${page + 1} … `);
    const res = await fetch(url, { headers: HEADERS });
    if (!res.ok) { console.error(`failed (${res.status})`); break; }
    const html = await res.text();
    const jobs = parseJobsGlobal(html)
      .filter(j => !NON_ENGLISH_RE.test(j.title))
      .filter(j => !JUNIOR_TITLE_RE.test(j.title))
      .filter(j => !JAVA_TITLE_RE.test(j.title));

    let newCount = 0;
    for (const j of jobs) {
      if (!seen.has(j.id)) { seen.add(j.id); all.push(j); newCount++; }
    }
    console.log(`${jobs.length} jobs, ${newCount} new`);
    if (newCount === 0) break;
  }
  return all;
}

const TECH_LEAD_RE = /\b(tech\s*lead|technical\s*lead|lead\s*(engineer|developer|backend|full.?stack)|staff\s*(engineer|developer)|principal\s*(engineer|developer)|engineering\s*lead|head\s*of\s*eng)\b/i;

const searches: SearchConfig[] = [
  {
    label: "Global Remote — Rails Tech Lead",
    keywords: "Ruby on Rails Tech Lead OR Rails Lead Engineer OR Rails Staff Engineer OR Rails Principal Engineer OR Ruby Lead Developer",
    workType: ["remote"],
    timePosted: "month",
  },
  {
    label: "Global Remote — Ruby Lead (broader)",
    keywords: "Ruby on Rails Lead OR Ruby on Rails Staff OR Ruby on Rails Principal OR Ruby Tech Lead",
    workType: ["remote"],
    timePosted: "month",
  },
];

type Result = { title: string; company: string; location: string; id: string; link: string };
const results: Result[] = [];
const seen = new Set<string>();

for (const config of searches) {
  console.log(`\n[${config.label}]`);
  const jobs = await fetchGlobal(config);

  for (const job of jobs) {
    if (seen.has(job.id)) continue;
    if (!TECH_LEAD_RE.test(job.title)) continue;
    seen.add(job.id);

    process.stdout.write(`  checking: ${job.title} @ ${job.company} (${job.location}) … `);
    const english = await isEnglishContent(job.id);
    if (!english) { console.log("dropped (non-English)"); continue; }
    console.log("ok");
    results.push({ title: job.title, company: job.company, location: job.location, id: job.id, link: `https://www.linkedin.com/jobs/view/${job.id}/` });
  }
}

console.log(`\n${results.length} global remote Rails lead jobs.\n`);
for (const r of results) console.log(`${r.title} | ${r.company} | ${r.location} | ${r.link}`);