Extract Text (2)

Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic

   Hi There,

I was hoping someone could help me with a problem I'm having or at least point me in the right direction. I'm trying to extract the co-ordinates of every word on the page of PDF programatically. I've tried to use GetPageText(4), in conjunction with GetPageText(1) -- because GetPageText(4) gives extremely corrupted results. In some cases I've got it to work, but in others, words are merged into one another or letters in the wrong order making it hard to compare the two results.

Does anyone know of another package that could let me do this? The price of the product doesn't matter too much, as long as it's reliable.

Thank you

Author	Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic
tren Members Profile Find Members Posts Beginner Joined: 07 Feb 06 Location: Australia Status: Offline Points: 5	Post Options Post Reply Quote tren Report Post Thanks(0) Quote Reply Topic: Extract Text (2) Posted: 13 Jun 06 at 10:24PM
	Hi There, I was hoping someone could help me with a problem I'm having or at least point me in the right direction. I'm trying to extract the co-ordinates of every word on the page of PDF programatically. I've tried to use GetPageText(4), in conjunction with GetPageText(1) -- because GetPageText(4) gives extremely corrupted results. In some cases I've got it to work, but in others, words are merged into one another or letters in the wrong order making it hard to compare the two results. Does anyone know of another package that could let me do this? The price of the product doesn't matter too much, as long as it's reliable. Thank you

chicks Members Profile Find Members Posts Debenu Quick PDF Library Expert Joined: 29 Oct 05 Location: United States Status: Offline Points: 251	Post Options Post Reply Quote chicks Report Post Thanks(0) Quote Reply Posted: 13 Jun 06 at 11:14PM
	http://pdftohtml.sourceforge.net/ This generally does an excellent job, and it's free. Command-line only, unless you know C really well... Use the XML output option to get the text, font and positonal info. You will probably need to do an XSL transform to get the output in a really usable form - if you need help, I can share some examples.

tren Members Profile Find Members Posts Beginner Joined: 07 Feb 06 Location: Australia Status: Offline Points: 5	Post Options Post Reply Quote tren Report Post Thanks(0) Quote Reply Posted: 13 Jun 06 at 11:51PM
	Chicks, you're a legend. I've done a bit of XSLT in the past so should be ok with this. To get the absolute position of words, I imagine I'm going to have to calculate the width of each word on the line based on font/font size? Thanks again for the help -- btw quickpdf rules.

ECPVFR Members Profile Find Members Posts Beginner Joined: 17 May 06 Location: Germany Status: Offline Points: 11	Post Options Post Reply Quote ECPVFR Report Post Thanks(0) Quote Reply Posted: 12 Jul 06 at 1:27PM
	Hi tren, did You ever try 'ExtractFilePageText' with option '4'? This does the whole job extracting every piece of text and option '4' also extracts all included information (font, color, text size, position and the text).
	Best Regards, Volker