#!/usr/bin/env node
// cse.js — cari di Google Programmable Search Engine (element API, pure JS)

const CX = '006516753008110874046:1ugcdt3vo7z' // download/direct-link search
const UA = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/126.0.0.0 Safari/537.36'
const BASE = 'https://cse.google.com'

let jar = {} // cookie jar: name -> value

async function get(url, referer) {
    const r = await fetch(url, {
        headers: {
            'User-Agent': UA,
            'Accept': '*/*',
            'Accept-Language': 'en-US,en;q=0.9',
            ...(referer ? { 'Referer': referer } : {}),
            ...(Object.keys(jar).length ? { 'Cookie': Object.entries(jar).map(([k, v]) => `${k}=${v}`).join('; ') } : {})
        }
    })
    const raw = await r.text()
    // simpan cookie baru dari response
    let sc
    if (typeof r.headers.getSetCookie === 'function') sc = r.headers.getSetCookie()
    else if (r.headers.get('set-cookie')) sc = [r.headers.get('set-cookie')]
    if (sc) for (const c of sc) {
        const [pair] = c.split(';')
        const eq = pair.indexOf('=')
        if (eq > 0) jar[pair.slice(0, eq).trim()] = pair.slice(eq + 1).trim()
    }
    return { status: r.status, raw }
}

// ambil sca_esv dari halaman utama
async function mainPage(cx) {
    const { raw } = await get(`${BASE}/cse?cx=${encodeURIComponent(cx)}`)
    const m = /sca_esv\\x3d([a-f0-9]+)/.exec(raw) || /sca_esv=([a-f0-9]+)/.exec(raw)
    return m ? m[1] : null
}

// ambil cse_token dari cse.js
async function cseTok(sca, cx) {
    const { raw } = await get(`${BASE}/cse.js?sca_esv=${sca}&hpg=1&cx=${encodeURIComponent(cx)}`, `${BASE}/cse?cx=${encodeURIComponent(cx)}`)
    const m = /"cse_token":\s*"([^"]+)"/.exec(raw)
    if (!m) throw new Error('cse_token tidak ditemukan')
    return m[1]
}

// ambil hasil
async function search(cx, q, start, num) {
    const sca = await mainPage(cx)
    const tok = await cseTok(sca, cx)
    const cb = `google.search.cse.api${Date.now() % 100000}`
    const params = new URLSearchParams({
        rsz: 'filtered_cse', num: String(num), hl: 'en', source: 'gcsc',
        cselibv: '93e57cb47c499c53', cx, q, safe: 'off', cse_tok: tok,
        filter: '1', sort: '', exp: 'cc,esbfa', fexp: '121991397,121991398,121877338,121877336',
        oq: q, gs_l: 'partner-generic.3...3113.3113.0.3141.1.1.0.0.0.0.0.0..0.0.csems,nrl=10...0.....34.partner-generic..1.0.0.',
        cseclient: 'hosted-page-client', callback: cb,
        rurl: `${BASE}/cse?cx=${encodeURIComponent(cx)}`
    })
    if (start) params.set('start', String(start))
    const { status, raw } = await get(`${BASE}/cse/element/v1?${params}`, `${BASE}/cse?cx=${encodeURIComponent(cx)}`)
    if (status !== 200) throw new Error(`HTTP ${status}`)
    const i = raw.indexOf('(')
    const j = raw.lastIndexOf(')')
    return JSON.parse(raw.slice(i + 1, j))
}

function clean(s) {
    if (s == null) return ''
    return String(s).replace(/<[^>]+>/g, '').replace(/&amp;/g, '&').replace(/&lt;/g, '<').replace(/&gt;/g, '>').replace(/&quot;/g, '"').replace(/&#39;/g, "'").trim()
}

function fmt(d) {
    return (d.results || []).map(r => ({
        title: clean(r.title),
        link: r.unescapedUrl || r.url || '',
        domain: r.visibleUrl || '',
        snippet: clean(r.content)
    }))
}

async function run() {
    const args = process.argv.slice(2)
    const asJson = args.includes('--json')
    const outIdx = args.indexOf('--out')
    const outFile = outIdx >= 0 ? (args[outIdx + 1] && !args[outIdx + 1].startsWith('--') ? args[outIdx + 1] : 'output/cse.json') : null
    let n = 10, start = 0
    const ni = args.indexOf('--n')
    if (ni >= 0) n = parseInt(args[ni + 1], 10) || 10
    const pi = args.indexOf('--page')
    if (pi >= 0) start = (parseInt(args[pi + 1], 10) || 1) * 10
    const q = args.find(a => !a.startsWith('--'))
    if (!q) {
        console.error('usage: node cse.js <kata> [--n 30] [--page 2] [--json] [--out f]')
        process.exit(1)
    }

    // kunjungan pertama ke cse.google.com untuk dapat NID
    await get(`${BASE}/cse?cx=${encodeURIComponent(CX)}`)

    const all = []
    let total = 0
    for (let s = start; s < start + n && all.length < n; s += 10) {
        const d = await search(CX, q, s, Math.min(10, n - all.length))
        if (s === start) total = parseInt((d.cursor || {}).estimatedResultCount, 10) || 0
        const items = fmt(d)
        if (!items.length) break
        all.push(...items)
        if (items.length < 10) break
    }

    const out = { query: q, total, shown: all.length, results: all }
    if (asJson) {
        console.log(JSON.stringify(out, null, 2))
    } else {
        console.log(`${q}: ${total} hasil (${all.length} ditampilkan)`)
        all.forEach((r, i) => {
            console.log(`\n[${i + 1}] ${r.title}`)
            if (r.domain) console.log(`    ${r.domain}`)
            console.log(`    ${r.link}`)
            if (r.snippet) console.log(`    ${r.snippet.slice(0, 160)}`)
        })
    }
    if (outFile) {
        const { mkdir, writeFile } = await import('node:fs/promises')
        const { dirname } = await import('node:path')
        await mkdir(dirname(outFile), { recursive: true })
        await writeFile(outFile, JSON.stringify(out, null, 2))
        console.error(`\nsaved: ${outFile}`)
    }
}

run().catch(e => { console.error('error:', e.message); process.exit(1) })