@chenchaolong/plugin-trade-compliance-workbench 1.0.133 → 1.0.134

package/dist/lib/remote-components/trade-compliance-workbench/preview-source.mjs

@@ -3,6 +3,8 @@ import * as mammoth from 'mammoth'
 import * as XLSX from 'xlsx'
 
 const MAX_PDF_PAGES = 8
+const MAX_PREVIEW_FILE_BYTES = 20 * 1024 * 1024
+const MAX_RENDER_PIXELS = 6_000_000
 
 let workerUrl = ''
 function pdfWorkerUrl() {
@@ -11,40 +13,67 @@ function pdfWorkerUrl() {
 }
 GlobalWorkerOptions.workerSrc = pdfWorkerUrl()
 window.TradeComplianceSourcePreview = {
-  async renderPdf(url) {
-    const response = await fetch(url); if (!response.ok) throw new Error(`PDF 文件读取失败（${response.status}）`)
-    const document = await getDocument({ data: new Uint8Array(await response.arrayBuffer()) }).promise
+  async renderPdf(url, { signal } = {}) {
+    const bytes = await readResponseBytes(url, signal, 'PDF')
+    const loadingTask = getDocument({ data: bytes })
+    const document = await loadingTask.promise
     const pageCount = document.numPages
     const pageNumbers = selectPages(pageCount, MAX_PDF_PAGES)
     const pages = []
+    let completed = false
     try {
       for (const pageNumber of pageNumbers) {
+        throwIfAborted(signal)
         const page = await document.getPage(pageNumber)
         const base = page.getViewport({ scale: 1 })
-        const viewport = page.getViewport({ scale: Math.min(1.15, 900 / Math.max(1, base.width)) })
+        const scale = Math.min(1.15, 900 / Math.max(1, base.width), Math.sqrt(MAX_RENDER_PIXELS / Math.max(1, base.width * base.height)))
+        const viewport = page.getViewport({ scale })
         const canvas = window.document.createElement('canvas')
         canvas.width = Math.ceil(viewport.width)
         canvas.height = Math.ceil(viewport.height)
-        await page.render({ canvas, canvasContext: canvas.getContext('2d'), viewport }).promise
-        pages.push(await canvasToObjectUrl(canvas))
-        page.cleanup?.()
+        try {
+          await page.render({ canvas, canvasContext: canvas.getContext('2d'), viewport }).promise
+          throwIfAborted(signal)
+          pages.push(await canvasToObjectUrl(canvas))
+        } finally {
+          canvas.width = 0
+          canvas.height = 0
+          page.cleanup?.()
+        }
       }
+      completed = true
       return { pages, pageNumbers, pageCount }
     } finally {
+      if (!completed) revokePageUrls(pages)
       await document.destroy()
     }
   },
-  async readSpreadsheet(url) {
-    const response = await fetch(url); if (!response.ok) throw new Error(`表格文件读取失败（${response.status}）`)
-    const workbook = XLSX.read(await response.arrayBuffer(), { type: 'array', cellDates: true, raw: false, sheetRows: 200 })
+  async readSpreadsheet(url, { signal } = {}) {
+    const bytes = await readResponseBytes(url, signal, '表格')
+    const workbook = XLSX.read(bytes, { type: 'array', cellDates: true, raw: false, sheetRows: 200 })
+    throwIfAborted(signal)
     return workbook.SheetNames.map(name => ({ name, rows: XLSX.utils.sheet_to_json(workbook.Sheets[name], { header: 1, defval: '', raw: false }).slice(0, 200).map(row => Array.isArray(row) ? row.slice(0, 30).map(cell => String(cell ?? '')) : []) })).filter(sheet => sheet.rows.some(row => row.some(Boolean)))
   },
-  async convertDocx(url) {
-    const response = await fetch(url); if (!response.ok) throw new Error(`Word 文件读取失败（${response.status}）`)
-    const result = await mammoth.convertToHtml({ arrayBuffer: await response.arrayBuffer() }); return result.value.replace(/<script[\s\S]*?<\/script>/gi, '').replace(/<style[\s\S]*?<\/style>/gi, '').replace(/\son\w+=(?:"[^"]*"|'[^']*')/gi, '')
+  async convertDocx(url, { signal } = {}) {
+    const bytes = await readResponseBytes(url, signal, 'Word')
+    const result = await mammoth.convertToHtml({ arrayBuffer: bytes })
+    throwIfAborted(signal)
+    return result.value.replace(/<script[\s\S]*?<\/script>/gi, '').replace(/<style[\s\S]*?<\/style>/gi, '').replace(/\son\w+=(?:"[^"]*"|'[^']*')/gi, '')
   }
 }
 function selectPages(pageCount, maxPages) { if (pageCount <= maxPages) return Array.from({ length: pageCount }, (_, index) => index + 1); return [...Array.from({ length: maxPages - 2 }, (_, index) => index + 1), pageCount - 1, pageCount] }
+async function readResponseBytes(url, signal, label) {
+  throwIfAborted(signal)
+  const response = await fetch(url, { signal })
+  if (!response.ok) throw new Error(`${label}文件读取失败（${response.status}）`)
+  const length = Number(response.headers.get('content-length'))
+  if (Number.isFinite(length) && length > MAX_PREVIEW_FILE_BYTES) throw new Error(`${label}文件超过在线预览限制（20 MB），请下载原文件查看`)
+  const bytes = new Uint8Array(await response.arrayBuffer())
+  if (bytes.byteLength > MAX_PREVIEW_FILE_BYTES) throw new Error(`${label}文件超过在线预览限制（20 MB），请下载原文件查看`)
+  return bytes
+}
+function throwIfAborted(signal) { if (signal?.aborted) throw new DOMException('预览已取消', 'AbortError') }
+function revokePageUrls(pages) { for (const pageUrl of pages) if (pageUrl.startsWith('blob:')) URL.revokeObjectURL(pageUrl) }
 async function canvasToObjectUrl(canvas) {
   const blob = await new Promise((resolve, reject) => canvas.toBlob(value => value ? resolve(value) : reject(new Error('PDF 页面转码失败')), 'image/png'))
   return URL.createObjectURL(blob)