Word to PDF
Convert Word to PDF: the text of a .docx file on A4 pages, without its formatting.
Your files stay on this device. Nothing is uploaded. How to check
How to use Word to PDF
Key features of Word to PDF
Every paragraph in the document body is carried over in order, including text inside table cells, set in Helvetica 12 pt on A4 pages.
Fonts, bold and italic, headings, images, tables, headers, footers and page layout are not reproduced. Long paragraphs are broken into lines of 75 characters.
The file is unpacked and read by JavaScript in your browser. Nothing is sent to a server.
About Word to PDF
Word to PDF opens the .docx package with JSZip, reads every paragraph's text from word/document.xml, and typesets it with pdf-lib in Helvetica on A4 pages with a 50 point margin. Paragraphs are separated by a small gap, and a new page starts when the current one is full.
It is enough to get the words of a report, a letter or meeting notes into a PDF you can read or print, without installing anything. It is not a layout engine: long paragraphs are cut into 75-character lines, so a word can be split at a line end, and headers, footers, footnotes and images are left out. For a PDF that looks like the original, export from Word or LibreOffice.
When to use Word to PDF
Someone emails you a .docx and you are on a Chromebook, a phone or a locked-down computer with no Office app. Convert it here and read the text in any PDF viewer. The result is plain 12 pt Helvetica on A4, which is enough to read a letter, a policy or a set of minutes.
You have a Word draft and a PDF of the signed version and want to know what changed in the wording. Convert the draft here, then run both files through PDF Compare, which works on the text layer and ignores formatting anyway. The same route works for PDF Highlight and PDF Redact, both of which need real, searchable text, which this output has.
You need a record of what a document said at a point in time, for example the wording of terms and conditions before an update, where the layout is irrelevant. The PDF holds the paragraphs in order and can be locked with a password in Protect PDF afterwards.
For a plain .txt file, Text to PDF is the better choice: it wraps lines at word boundaries and offers three font sizes. For the opposite direction, PDF to Word turns a PDF's text back into a .docx. If the document is mostly images, numeric tables or a form, neither tool reproduces it; the PDF export in Word or LibreOffice does.
Tips and troubleshooting
- The message This file is not a valid .docx document means the file is not a zipped Word package: an .odt, .rtf or .pages file renamed to .docx, or a download that did not finish. Open it in the original app and save it as .docx again.
- Tabs and manual line breaks inside a paragraph are stored as separate elements, not as text, so they disappear: the words on either side of a tab run together, and a line break inside a paragraph is lost. Put such content in separate paragraphs if the break matters.
- Bullets and list numbers are formatting rather than text, so a numbered list comes out as plain lines without its numbers. Type the numbers into the text if they must survive.
- Tracked changes are resolved as if every change had been accepted: inserted text is included, deleted text is skipped, and comments are left out. Reject changes in Word first if you need the earlier wording.
- A character the built-in Helvetica font cannot draw, such as Chinese, Arabic or Devanagari script or an emoji, stops the conversion with an error that mentions WinAnsi, and no PDF is produced. Accented Latin letters, curly quotes and the euro sign are fine.
- The download takes the document's own name with a .pdf extension. There is no page size, margin or font option: every page is A4 with a 50 point margin, and about 40 lines of 12 pt text fit on one.
- The 150 MB limit is rarely reached by text; a .docx that large is full of images, which are not read anyway. Long documents take longer and produce more pages.
- If your document is a web page or an HTML export rather than a .docx file, HTML to PDF renders it as it looks in a browser, which this text-only converter does not.
How Word to PDF works in your browser
The file is read into memory as an ArrayBuffer and its first four bytes are checked: the signature D0 CF 11 E0 marks an old OLE compound file, which is the .doc case that is refused. Otherwise JSZip opens the package in memory and returns word/document.xml as a string, which the browser's own DOMParser turns into an XML tree. The script walks every w:p element in order and joins the text of its own w:t runs into one paragraph, skipping paragraphs that end up empty. Word stores a text box twice, as a drawing and as an older VML copy, so only the first copy is read, and its text follows the paragraph it is anchored to; text moved with tracked changes is read only at its new position.
pdf-lib then creates the PDF: an A4 page, the standard Helvetica font (one of the 14 fonts every PDF viewer supplies, so no font file is embedded), and one drawText call per 75-character line with 18 points between lines and a 4 point gap after each paragraph, starting a new page when the cursor reaches 60 points from the bottom. The result is saved to bytes and wrapped in a Blob for the Download File button. Both libraries come from cdnjs.cloudflare.com when the page opens, and neither the .docx nor the PDF leaves your browser.
Frequently asked questions
Will my formatting, images and tables be kept?
No. Only the text is converted, in reading order. Fonts, headings, images and table layout are not reproduced; table cell text comes through as plain lines. Export from Word if you need an exact copy.
Are .doc files supported?
No. The older binary .doc format cannot be read here, and the tool stops with a message that says so. Open the file in Word or LibreOffice, save it as .docx and convert that.
Why does the tool say no text was found?
The document body has no paragraph text, for example a file made only of images or drawings. There is nothing for this converter to lay out, so no PDF is produced.
Are headers, footers and footnotes included?
No. They are stored in separate parts of the .docx package and only the main body, word/document.xml, is read. Endnotes and comments are left out for the same reason.
What happens to tracked changes?
The PDF reads as if every change had been accepted: inserted text is included and deleted text is left out, because deletions are stored as a different element. Reject the changes in Word first if you need the earlier wording.
Can it convert a document written in Chinese, Arabic or another non-Latin script?
No. The built-in Helvetica font covers Latin letters, accents and common symbols only, and a character outside that set stops the conversion with an error that mentions WinAnsi. Export such documents from Word or LibreOffice instead.
Last reviewed