Convert ODT to PDF

Every document conversion involves one program reading a file another program wrote, and most of the difficulty lives in that gap. This one does not have the gap. ODT is the OpenDocument format, LibreOffice is what produced it, and LibreOffice is what converts it — the same software reading a file it defined.

A DOCX arriving at LibreOffice has to be interpreted: styles resolved, layout rules mapped from one model onto another, Microsoft fonts substituted with metric-compatible stand-ins. It works well, but there are decisions being made. With ODT there are no decisions. The file already describes itself in the terms the converter uses.

Fonts. Our server carries the common open families — Liberation, DejaVu, Noto — plus metric-compatible versions of the usual Microsoft ones. A document set in any of those converts exactly.

A document using a font that exists only on your own machine will be substituted, and because the substitute has different letter widths, lines can break in different places. If your document uses something unusual, exporting from your own LibreOffice is the safer path.

If you have LibreOffice open, use File → Export as PDF. It is the same engine, it happens instantly, and nothing leaves your computer.

This page exists for when LibreOffice is not in front of you — a borrowed laptop, a phone, a locked-down work machine, or a file somebody sent you in a format you have no software for. That is a real situation and it is the one this is for.

Headers, footers, page numbering, tables, embedded images, hyperlinks, and a table of contents generated by LibreOffice keeps its internal links. Tracked changes render as the document displays them, so accept or reject them first if you want a clean copy.

This is one of the conversions that runs on our server rather than in your browser, because reading an office document properly means running an office suite. The file is uploaded over an encrypted connection, converted, and deleted the moment the job finishes. No copy is kept and nobody looks at it.

Most of this site never touches your files at all, and where that is not true we would rather point at it than let you assume.

An ODT is a ZIP. The text inside it is compressed, and an image used on five pages is stored once. A PDF has to record where every character sits on every page, and it embeds the fonts it used.

So a short text document often grows rather than shrinks. The trade is deliberate: the PDF carries everything needed to render itself identically anywhere, which is the entire reason for converting it.

Whether tracked changes are still waiting to be accepted, because they render as the document displays them. Whether there are comments in the margins that were not meant to travel. And whether fields such as the date or the file path hold the value you intended.

None of the three is prominent while you are editing, and all three are fixed in place the moment the PDF exists. Thirty seconds now beats a follow-up email later.

It is possible, and it is not a round trip. What returns is the appearance rather than the structure: paragraph styles, outline levels and tables are reconstructed from where things happen to sit on the page.

The result is readable and it is a different document. If editing is going to continue, keep the ODT — it is the file that knows what a heading is.

Drop up to 100 files at once and take the results as a ZIP. Each is converted independently, so one file that fails does not take the batch with it.

Because this is one of the conversions that runs on our server, two further numbers apply that a browser conversion does not have. Each document must be 25 MB or smaller, and the free allowance is 100 server conversions a day — so a hundred-file batch is a whole day's worth. Large sets go better in groups, spread out.