Determining "boxes of interest" on a PDF page

Question

I want to be able to determine the bounding box of areas of text, images and paths on a PDF page, similar to what is shown here:

http://www.windjack.com/products/screenshot/pdfcanscreenshot2.html

Looking at the PDF spec, I can see how to determine the bounding boxes of paths and images, but I can't see how to arrive at them for text. Do I have to calculate it by hand, determining the height and width of the glyphs from the font size, etc., or is there a more straightforward way?

Answer 1

You may be able to start with the solution to "How do I get character offset information from a pdf document?" That will give you x, y, width and height for characters and/or substrings in the document. From there, the harder part is to bound the groups of characters into spatially distinct regions. There's no guarantee that spatially grouped text on a page will be close to each other in the syntax of the file format...