PDF to text

How text extraction works

Most PDFs made from Word, Google Docs, a web browser or accounting software contain a text layer: every character, its font and its exact position on the page. The tool reads that layer with PDF.js, the PDF engine built into Firefox, then rebuilds readable text from the positions. Characters are sorted into lines from top to bottom, gaps become spaces, and wide gaps become tabs so that table columns stay apart.

What you get, and what you don’t

To include the answers in a filled-in form, print the form to a PDF first: printing turns the field contents into page text. Magazine-style pages with three or more columns, text boxes and captions may come out with some lines in an unexpected order; check the result before relying on it.

Keep line breaks or join paragraphs?

A PDF has no idea of a paragraph; it only knows that one line ends and another begins lower down. Say a page reads "The parties agree to the follow-" on one line and "ing terms and conditions." on the next. Keep as in PDF gives you both lines exactly as printed, which is best for addresses, poems, code and anything where the layout matters. Join paragraphs gives you "The parties agree to the following terms and conditions." as one line, which is what you want when pasting into an email, a Word document or a translation tool. A blank line is kept wherever the PDF leaves extra space, so paragraphs stay separate.

Tips

• Only need a section? Choose pages and type a range such as 12-18 to read just those.
• Page separators make it easy to quote a page number later; turn them off for one continuous text.
• Comparing two versions of a contract? Compare PDFs shows the words that changed, side by side.
• Need the page as it looks, not just its words? Save it with PDF to JPG.

Everything happens in your browser, so the tool is safe for confidential material. To share only part of a long PDF as a PDF, use extract pages or split PDF.