Convert TXT to PDF

How plain text becomes pages

Text has no pages, no margins and no line breaks that mean anything beyond where the author happened to press return, so all of that is manufactured here. Paragraphs are wrapped to the width of the page using the real advance widths of the font being written, not an estimated character count — which is what keeps a wrapped line from running into the margin when it contains wide letters. A single word longer than the line, a URL or a hash, is broken across lines by measurement rather than left to overflow. Tabs are expanded to spaces before anything is measured, so what is wrapped is what is drawn.

A full Unicode font is embedded in the PDF rather than referenced, which is why the result opens identically on a machine that has never seen it, and why accented, Cyrillic and Greek text comes out right instead of as boxes. It also means a text file of a few kilobytes produces a PDF of a few hundred, because the font travels with it — the floor is the font, not the text, and a longer document is barely larger.

Scripts that need shaping are refused rather than mangled. Arabic letters change shape depending on their neighbours, Bengali and Devanagari form conjuncts that are single glyphs, and Chinese needs tens of thousands of characters that no general-purpose embedded font carries. Writing those code points one after another produces text that is not merely ugly but genuinely incorrect — so the page lists exactly which characters it cannot draw and stops, rather than handing back a document full of blanks you might not notice until someone else opens it.

The output carries real, selectable, searchable text — not a picture of text — so the PDF can be quoted from and indexed. Line breaks in the source are honoured as paragraph breaks; everything else about the layout is chosen here, because there was nothing in the file to read it from. A Markdown file converts as what it literally is, hashes and asterisks included; Markdown to PDF is the one that renders it.

When you want something else

For text that needs to become a designed document — headings, page numbers, a table of contents, footnotes — Pandoc is the tool: pandoc notes.txt -o notes.pdf is the short version, and with a template it produces typeset output that no plain-text converter can approach. It is free.

For complex scripts, or for CJK, the honest answer is that this page cannot help and a converter carrying the right fonts must. LibreOffice will open a text file and export it — soffice --headless --convert-to pdf notes.txt — using the fonts installed on the machine, with full shaping, which is exactly what is missing here.