How to extract the text from a PDF

Getting the text out of a PDF takes one command in PDF by Zaweba. Choose File > Export and save all the text as a .txt file. If you'd rather do it from a script or a batch file, the command line has --extract-text, which pulls the text out of a file, or just chosen pages of it, without opening a window.

Plain text is the most useful format there is for words. It opens in anything, it can be searched, pasted into another program, fed to a translation tool or kept as a lightweight copy of a long document.

From the menu

Open the PDF, choose File > Export and pick the text option. Name the file and save. Every page's text goes into one file, ready to open in Notepad, Jotter by Zaweba or any other editor.

Export > Accessible text is offered as well, alongside the ordinary text export. If the first doesn't give you what you need from a complicated page, try the other and compare the two files.

For a small amount of text, of course, you don't need to export at all. Select it on the page and press Ctrl+C, or Ctrl+Shift+C to keep the lines and gaps as they sit on the page.

From the command line

The command line is for when there are a lot of files, or when text extraction is one step of a bigger job. The form is ZawebaPDF.exe --extract-text file.pdf, and you can add --pages to take only some pages, written like 1-3,7, and --out to name the text file to write. If the PDF is protected, --password supplies its password.

The command also reports how it went through its exit code, so a script can tell what happened: 0 means done, 1 means it failed, 2 means the arguments were wrong, and 3 means a password was needed or the one given was wrong. Running ZawebaPDF.exe --cli-help lists all the commands.

A small batch file that loops over a folder of PDFs and calls --extract-text on each one is a quick way to turn an archive of reports into searchable text files.

When no text comes out

If you export and the file is empty, or nearly so, the PDF almost certainly has no real text in it. Scanned pages are pictures of words. You can check by trying to select a word with the mouse: if nothing highlights, there's nothing to extract.

A PDF can also be made with permissions that stop text being copied. PDF by Zaweba respects those settings, and when copying is refused it tells you the reason.

Making the text easier to use

Text taken from a PDF arrives line by line, because that's how a PDF stores it. For reading or searching that's fine. If you need flowing paragraphs to edit, export to a Word document instead, where lines are joined back into paragraphs and headings are recognised.

Watch for page headers, footers and page numbers repeated through the text, and remove them if they get in the way.

PDF by Zaweba is free software for Windows. It works offline with no account, so the text of private documents never passes through someone else's server.