离线 · 无上传 · 浏览器本地完成Offline · no uploads · runs in the browser
Word 转 PDF——离线出 A4 图像，版式由浏览器决定Word to PDF offline — an A4 image, laid out by the browser
「Word 转 PDF」这件事在桌面上是 Word 自己做的：调用排版引擎，输出带文字层的 PDF。浏览器里这条链路走的是另一条路——mammoth 先把 .docx 读成语义 HTML，再整页栅格化、按 A4 切片。结果是：任何设备打开都长一样，但文字层没有，且版式与 Word 打印会有出入，因为排版由浏览器与系统字体决定，不由 Word 决定。On a desktop, Word makes the PDF itself: a layout engine, a text layer, fidelity. This browser route is a different animal — mammoth reads the .docx as semantic HTML, the page is rasterised whole and sliced onto A4. You get a file that looks the same on every device, and you give up the text layer, and the layout can drift from Word’s own print because the browser and your system fonts do the typesetting, not Word.
这条链路怎么走How the route runs
Word (.docx)Word (.docx) HTMLHTML PDFPDF
共 2 步，由扩展的路径搜索自动选出，中间产物不落盘。2 steps, chosen by the extension's own path search; intermediate results never touch the disk.
会保留什么What carries over
- 内容结构：标题、列表、表格、加粗斜体、内嵌图片按语义 HTML 渲染到页面上The structure: headings, lists, tables, emphasis and embedded images render onto the page from the semantic HTML
- A4 逐页切片，长文档自然分页；最长边 8192 px 上限兜底A4 slicing page by page, with an 8192 px longest-edge cap protecting very long documents
- 中间 HTML 经 DOMPurify 净化，文档内的可执行标记不进入产物The intermediate HTML passes through DOMPurify, so executable markup in the document never reaches the artifact
如实说明的限制What it does not do
- 没有文字层：PDF 里的文字选不中、检索不到，屏幕阅读器也读不到No text layer — nothing selectable, searchable, or readable by a screen reader
- 页边距、分页、字体与 Word 打印不同；缺字体时用系统回退，字形宽度一变，分页位置就跟着变Margins, breaks and fonts differ from Word’s print; a missing font falls back to a system face, and once glyph widths change, every page break after it moves
- 页眉页脚、目录域、批注、修订不进渲染；样式表意义上的 Word 主题不生效Headers and footers, TOC fields, comments and tracked changes do not enter the render, and the Word theme does not apply
- 要「可被检索的归档 PDF」请回到 Word 自己导出；这条适合的是分发观感For a searchable archival PDF let Word export it yourself; this route is for distributing an appearance
实操提示Working with it
- 先转成 HTML 预览，是判断分页与字体是否可接受的最快办法Converting to HTML first is the fastest way to judge whether the breaks and fonts are acceptable
- 批量最多 200 个文件；多个 PDF 一次打包下载Up to 200 files per batch; several PDFs can arrive in one ZIP
常见问题Questions people actually ask
为什么和我在 Word 里导出的 PDF 不一样？Why does it differ from Word’s own PDF?
因为排版者不同：这条链路是浏览器按语义 HTML 排版并用系统字体，Word 用自己的排版与主题字体。文字层与逐字保真请交给 Word。Because a different engine does the typesetting: here the browser lays out semantic HTML with your system fonts, while Word uses its own engine and theme fonts. Give Word the job when the text layer and character-exact layout are the point.
那这份 PDF 适合做什么？So what is this PDF good for?
发给不需要改的人：审稿预览、打印分发、把 Markdown 或 Word 内容固定成一个观感。合同归档与检索场景不适合。Sending it to people who will not edit it: a review preview, a print distribution, freezing document content into one appearance. It is the wrong artifact for contract archives or anything that has to be searched.
支持 .doc 吗？Is .doc supported?
不支持。97-2003 的二进制 .doc 不在支持的 14 种格式里，请先另存为 .docx。No. Binary .doc from 97-2003 is outside the 14 supported formats, so save it as .docx first.
不需要账号，也不需要联网：装好扩展后点图标，在新标签页里打开工作台，文件从磁盘直接选。No account and no network: install the extension, click its icon, and the workbench opens in a new tab reading straight off your disk.
从 Chrome 应用商店安装Install from the Chrome Web Store 安装步骤Install steps GitHub 源码Source on GitHub