PDF to Text

A PDF made by a word processor stores its words as text, along with instructions for where to draw each run of them. This tool reads that text layer out, page by page, and writes it to a .txt file.

Everything happens inside this page. Your file is read by JavaScript running on your own machine, and no part of it is sent anywhere.

A scan will give you nothing

If your PDF is a scan, there is no text layer. The words are shapes in a picture, and there is nothing to extract. The output will be empty and that is the correct result rather than a failure.

Making a scan searchable needs OCR, which looks at the shapes and works out what letters they are. It is a genuinely different operation with a genuinely different accuracy: the text layer here is exact because it was never guessed, and OCR output never is.

The quick way to tell which kind you have is to open the PDF and try to select a word. If the selection highlights the text, this tool will work. If it draws a rectangle over the picture, it will not.

Why the layout does not survive

PDF does not store paragraphs, columns or tables. It stores runs of characters and coordinates. A two column page is not two columns as far as the file is concerned, it is a set of text runs that happen to sit in two groups.

So extracted text comes out in the order the runs were written, which is usually reading order and sometimes is not. Two column layouts interleave. Tables lose their alignment. Headers and footers appear in the middle of the flow. None of that is recoverable without guessing, and this tool does not guess.

Line breaks are approximate

The format has no concept of a line ending either. What is used here is the renderer's record of where one run of text finished, which is the closest thing available and is right most of the time.

Expect some lines to break where the original wrapped rather than where a paragraph ended. For most uses, feeding text to something else, searching it, counting it, that is fine.

Page markers

Each page's text is preceded by a marker giving its number, so you can tell where a passage came from. It also gives you something to split on if you want the pages separately.