huayang/frontend/scripts/sync-fastimg-assets.mjs

import fs from 'node:fs'
import path from 'node:path'
import crypto from 'node:crypto'
import { fileURLToPath } from 'node:url'

const __dirname = path.dirname(fileURLToPath(import.meta.url))
const rootDir = path.resolve(__dirname, '..')
const publicDir = path.join(rootDir, 'public')
const publicAssetsDir = path.join(publicDir, 'assets')

/**
 * 本地化外链静态资源（与 App.vue 中 rewriteLegacyStylesheetText 规则对齐）：
 * - 归档 omo-oss + thefastimg/thefastfile
 * - 商城 presale.jhtsoft.com
 * - 阿里 cdn *.alicdn.com
 *
 * 用法：npm run sync:fastimg * 仅重写已下载：npm run sync:fastimg -- --rewrite-only
 */

const CDN_URL_RE =
  /https?:\/\/omo-oss-(?:image|file)\d*\.thefast(?:img|file)\.com\/[^\s"'<>)\]}]+/gi
const PRESALE_URL_RE = /https?:\/\/(?:www\.)?presale\.jhtsoft\.com\/[^\s"'<>)\]}]+/gi
const ALICDN_URL_RE = /https?:\/\/[^\s"'<>)\]}]*\.alicdn\.com\/[^\s"'<>)\]}]+/gi

const DEFAULT_HEADERS = {
  'User-Agent':
    'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
  Referer: 'https://www.haihofoods.com/',
  Accept: '*/*'
}

function headersForUrl(url) {
  let u
  try {
    u = new URL(url)
  } catch {
    return DEFAULT_HEADERS
  }
  if (/presale\.jhtsoft\.com$/i.test(u.hostname)) {
    return {
      ...DEFAULT_HEADERS,
      Referer: 'https://haihofoods.cutmall.cn/',
      Origin: 'https://haihofoods.cutmall.cn'
    }
  }
  if (/\.alicdn\.com$/i.test(u.hostname)) {
    return { ...DEFAULT_HEADERS, Referer: 'https://www.taobao.com/' }
  }
  return DEFAULT_HEADERS
}

function walkFiles(dir, extensions) {
  const out = []
  if (!fs.existsSync(dir)) return out
  for (const name of fs.readdirSync(dir)) {
    const full = path.join(dir, name)
    const st = fs.statSync(full)
    if (st.isDirectory()) out.push(...walkFiles(full, extensions))
    else if (extensions.some((e) => name.endsWith(e))) out.push(full)
  }
  return out
}

function normalizeCollectedUrl(raw) {
  let u = raw.trim().replace(/[,;]+$/, '')
  while (/[.,;:]$/.test(u)) u = u.slice(0, -1)
  return u
}

const CDN_HOST_OK = /^omo-oss-(?:image|file)\d*\.thefast(?:img|file)\.com$/i
const PRESALE_HOST_OK = /^(?:www\.)?presale\.jhtsoft\.com$/i
const ALICDN_HOST_OK = /\.alicdn\.com$/i

function sanitizeGenericAssetUrl(raw, hostOk) {
  let s = normalizeCollectedUrl(raw)
  const cut = s.search(/&quot;|&#\d+;|[\s"'<>]/)
  if (cut !== -1) s = s.slice(0, cut)
  while (/[.,;:]$/.test(s)) s = s.slice(0, -1)
  try {
    const u = new URL(s)
    if (typeof hostOk === 'function') {
      if (!hostOk(u.hostname)) return null
    } else if (!hostOk.test(u.hostname)) {
      return null
    }
    const last = decodeURIComponent(u.pathname.split('/').filter(Boolean).pop() || '')
    if (!/\.(png|jpe?g|gif|webp|svg|ico|ttf|woff2?|eot)$/i.test(last)) return null
    u.hash = ''
    return u.href
  } catch {
    return null
  }
}

function sanitizeCdnArchivedUrl(raw) {
  return sanitizeGenericAssetUrl(raw, CDN_HOST_OK)
}

function sanitizePresaleUrl(raw) {
  return sanitizeGenericAssetUrl(raw, PRESALE_HOST_OK)
}

function sanitizeAlicdnUrl(raw) {
  return sanitizeGenericAssetUrl(raw, (h) => ALICDN_HOST_OK.test(h))
}

function pathnameHasAsset(pathname) {
  const parts = pathname.split('/').filter(Boolean)
  return parts.length > 0
}

function dedupeByPath(urls) {
  const pick = new Map()
  for (const url of urls) {
    let u
    try {
      u = new URL(url)
    } catch {
      continue
    }
    const key = `${u.hostname.toLowerCase()}${u.pathname}${u.search}`
    const cur = pick.get(key)
    if (!cur || (cur.startsWith('http:') && url.startsWith('https:'))) pick.set(key, url)
  }
  return [...pick.values()]
}

function pathKey(url) {
  const u = new URL(url)
  return `${u.hostname.toLowerCase()}${u.pathname}${u.search}`
}

function suggestedFilename(url) {
  const u = new URL(url)
  const last = decodeURIComponent(u.pathname.split('/').filter(Boolean).pop() || '')
  if (!last) return null
  const safe = last.replace(/[^a-zA-Z0-9._-]+/g, '_')
  return safe || null
}

function uniqueDestPath(preferredName, occupied) {
  let name = preferredName
  let n = 0
  while (occupied.has(name)) {
    n += 1
    const ext = path.extname(preferredName)
    const base = ext ? preferredName.slice(0, -ext.length) : preferredName
    name = `${base}-${n}${ext || ''}`
  }
  occupied.add(name)
  return name
}

async function downloadOne(url, destFsPath) {
  const res = await fetch(url, { redirect: 'follow', headers: headersForUrl(url) })
  if (!res.ok) {
    throw new Error(`HTTP ${res.status} ${res.statusText} — ${url}`)
  }
  const buf = Buffer.from(await res.arrayBuffer())
  if (!buf.length) {
    throw new Error(`Empty body — ${url}`)
  }
  fs.mkdirSync(path.dirname(destFsPath), { recursive: true })
  fs.writeFileSync(destFsPath, buf)
}

function collectFromRegex(text, re, sanitize) {
  const cleanedSet = new Set()
  re.lastIndex = 0
  let m
  while ((m = re.exec(text)) !== null) {
    const cleaned = sanitize(m[0])
    if (!cleaned) continue
    try {
      if (!pathnameHasAsset(new URL(cleaned).pathname)) continue
    } catch {
      continue
    }
    cleanedSet.add(cleaned)
  }
  return cleanedSet
}

function replaceBothProtocols(text, fromUrl, toPath) {
  let out = text.split(fromUrl).join(toPath)
  try {
    const u = new URL(fromUrl)
    const altProto = u.protocol === 'https:' ? 'http:' : 'https:'
    const alt = `${altProto}//${u.host}${u.pathname}${u.search}`
    if (alt !== fromUrl) out = out.split(alt).join(toPath)
  } catch {
    /* ignore */
  }
  return out
}

async function main() {
  const dryRun = process.argv.includes('--dry-run')
  const skipDownload = process.argv.includes('--rewrite-only')

  const sourceFiles = [
    ...walkFiles(publicDir, ['.css']),
    ...walkFiles(path.join(rootDir, 'src', 'generated'), ['.html', '.txt', '.json'])
  ]

  const allFound = new Set()
  for (const file of sourceFiles) {
    const text = fs.readFileSync(file, 'utf8')
    for (const s of collectFromRegex(text, CDN_URL_RE, sanitizeCdnArchivedUrl)) allFound.add(s)
    for (const s of collectFromRegex(text, PRESALE_URL_RE, sanitizePresaleUrl)) allFound.add(s)
    for (const s of collectFromRegex(text, ALICDN_URL_RE, sanitizeAlicdnUrl)) allFound.add(s)
  }

  const fetchUrls = dedupeByPath([...allFound]).sort()
  console.log(
    `Found ${allFound.size} remote asset URL hits (${fetchUrls.length} unique URLs to fetch).`
  )

  const keyToPublicPath = new Map()
  const occupiedNames = new Set(
    fs.existsSync(publicAssetsDir)
      ? fs.readdirSync(publicAssetsDir).filter((n) => fs.statSync(path.join(publicAssetsDir, n)).isFile())
      : []
  )

  function isRasterExt(name) {
    return /\.(png|jpe?g|gif|webp)$/i.test(name)
  }

  for (const url of fetchUrls) {
    const key = pathKey(url)
    let fname = suggestedFilename(url)
    if (!fname) {
      const h = crypto.createHash('sha256').update(url).digest('hex').slice(0, 16)
      fname = `asset-${h}.bin`
    }
    fname = uniqueDestPath(fname, occupiedNames)
    const publicPath = `/assets/${fname}`

    if (dryRun) {
      keyToPublicPath.set(key, publicPath)
      continue
    }
    if (skipDownload) {
      const fsPath = path.join(publicAssetsDir, fname)
      if (!fs.existsSync(fsPath)) {
        console.warn(`[skip-download] missing file, run without --rewrite-only: ${fname}`)
      } else {
        keyToPublicPath.set(key, publicPath)
      }
      continue
    }

    const dest = path.join(publicAssetsDir, fname)
    if (fs.existsSync(dest) && fs.statSync(dest).size > 0) {
      console.log(`OK (exists) ${fname}`)
      keyToPublicPath.set(key, publicPath)
      continue
    }

    process.stdout.write(`GET ${url.slice(0, 80)}… → ${fname} `)
    try {
      await downloadOne(url, dest)
      console.log('ok')
      keyToPublicPath.set(key, publicPath)
    } catch (e) {
      console.log(`FAIL (${e.message})`)
      if (isRasterExt(fname)) {
        console.warn(`  → use /assets/no-img.svg for ${key}`)
        keyToPublicPath.set(key, '/assets/no-img.svg')
      } else {
        console.warn(`  → skip (font/other); URL left unchanged in sources`)
      }
    }
  }

  if (dryRun) {
    console.log('Dry run: no files written.')
    return
  }

  const urlToPublicPath = new Map()
  for (const u of allFound) {
    try {
      if (!pathnameHasAsset(new URL(u).pathname)) continue
      const pub = keyToPublicPath.get(pathKey(u))
      if (pub) urlToPublicPath.set(u, pub)
    } catch {
      /* ignore */
    }
  }

  let touched = 0
  for (const file of sourceFiles) {
    let text = fs.readFileSync(file, 'utf8')
    const original = text
    const replacements = [...urlToPublicPath.entries()].sort((a, b) => b[0].length - a[0].length)
    for (const [absUrl, pubPath] of replacements) {
      text = replaceBothProtocols(text, absUrl, pubPath)
    }
    if (text !== original) {
      fs.writeFileSync(file, text, 'utf8')
      touched += 1
      console.log(`Rewrote ${path.relative(rootDir, file)}`)
    }
  }
  console.log(`Done. Updated ${touched} files.`)
}

main().catch((e) => {
  console.error(e)
  process.exit(1)
})