Turn a text file into a PDF

A plain text file has no pages. It is a run of characters with line breaks, and every program that opens one decides for itself how wide the lines may be and where one page ends.

This makes those decisions and writes the result out. Each line of the file becomes a line on the page, blank lines stay blank, and a new page starts when the next line would run past the bottom margin. The text is set in Helvetica at body size, on A4, with a margin all round.

What is kept and what is not

The order of the lines, the blank lines between them and the words themselves are kept exactly. What is not kept is anything a plain text file cannot carry in the first place: there is no bold, no heading size and no colour to preserve, because the file never held any.

Two things change on the way in. A tab becomes four spaces, and spaces at the end of a line are dropped, since nobody can see them and they only make a line wrap early. Runs of spaces inside a line collapse to one when the line is wrapped, so text that was aligned into columns with spaces will not stay aligned.

If your file is Markdown, use Markdown to PDF instead, which reads the headings and lists. A .txt file put through a Markdown reader can change meaning, because a line that starts with a hash sign or an asterisk is read as structure. This tool reads nothing as structure.

Text saved by other programs

A .txt file does not say which encoding it was saved in. Most today are UTF-8, but older Windows programs saved single-byte text and some editors write UTF-16. The file is checked for both before it is read, so an accented letter or a curly quote arrives as itself and not as a run of odd characters.

Characters Helvetica cannot draw

The PDF is set in Helvetica, which every reader has and which covers Latin text and not much else. A character it cannot draw is replaced, and the result tells you how many were replaced rather than letting them vanish. If the text is in Hindi or Arabic, the font for that script is loaded for the job. Chinese, Japanese and Korean are not covered: each such character becomes a question mark, and the count is shown. Hindi and Arabic display correctly, but some viewers rebuild the text for copying and searching from the shapes on the page, and letters that share a shape, such as the Arabic letters that differ only by their dots, can come out as the wrong one. The words are also stored in the file as you typed them, which readers that honour that use.

When a PDF is the better copy

A text file opens differently on every machine, and some phones open it in whatever app claims it. A PDF looks the same wherever it is opened, so it suits notes you want to attach to an email, a log you want to hand to someone who has no editor, or a draft you want to keep exactly as it was on the day you sent it.

Nothing leaves the device

The file is read and the PDF is written by a worker inside your browser. There is no upload, no account and no server copy, and no request goes anywhere while it happens. That matters for the files people most often print from a text editor: notes, logs, drafts and pasted emails.