What does this tool do?

It takes a PDF apart: you get the images embedded in it, its text, its embedded fonts (converted to TrueType), its metadata (title, author, the program that created it, page size, …) and the files attached to it (e.g. the XML data of an electronic invoice).

Why do the extracted images look different than in the PDF?

You get the images as they are stored in the PDF. Cropping, rotation, scaling or color changes the PDF applies when it shows them are not applied. Transparency is kept (PNG). Tiny images (lines, dots) are left out, and an image used several times comes only once.

Why is some text missing?

Text that is part of an image (e.g. in a scanned page) is not text for a PDF. This tool does no text recognition (OCR).

Is it legal to extract fonts from a PDF?

Extracting is legal, but to use an extracted font you need the permission of its creator. The main use is to find out which fonts a PDF uses.

Why are letters missing in an extracted font?

A PDF often contains only the letters it uses (a subset of the font). If the PDF has no letter "P", the extracted font has none either. PDFs also contain no font metric files (.afm, .pfm) and often no hinting, so a text set in an extracted font may look worse than with the original font.

Limits

You can upload PDFs of up to 30 MB. At most 300 images are extracted. For a PDF protected with a password, enter it before you choose the file.