How to convert PDF to TEX?

Message

nahpets · Post by **nahpets** » Sat Jul 10, 2004 7:42 pm

I have a PDF ebook that isn't well formatted. I want to extract all the text into a .tex file and generate a new, nicely formatted PDF document using Latex. I've tried doing this two ways:

First
- pdftotext to get a .txt file with all the text
- enscript to generate a rtf file
- unrtf to get an .tex file

This method doesn't really work because I loose all the formatting and breaks. I basically get one long paragraph in Latex.

Second
- pdftohtml generates one big HTML file with the proper formatting in HTML.
- I open the HTML in ooffice and "save as" .txt file
- enscript and unrtf like before to get a .tex file

The problem with this method is that ooffice sometimes doesn't want to open the whole document because I guess it's too big. Does anyone know of a better way to get what I want?

boglin · Post by **boglin** » Sat Jul 10, 2004 7:52 pm

what about tweaking the second method? Instead of using OO to convert html 2 txt then txt 2 latex, use html2latex.

nahpets · Post by **nahpets** » Sun Jul 11, 2004 10:01 am

what ebuild does "html2latex" belong to? I found "html2text"...

boglin · Post by **boglin** » Fri Jul 16, 2004 6:12 pm

ermm, my bad: html2latex DNE. Sorry.

MacMasta · Post by **MacMasta** » Sat Jul 17, 2004 12:03 am

You meant 'latex2html', which is most helpful.

I don't think there'll be much of a way to maintain the formatting; you'll need to do most of that yourself, once you get the paragraph...

~Mac~

nahpets · Post by **nahpets** » Sat Jul 17, 2004 6:55 am

Yeah, I figured as much. Doing it myself will take a long time, so I think I'll give up on it.

machinelou · Post by **machinelou** » Mon Jul 26, 2004 4:34 pm

Wouldn't it be reletively simple to write a bash or perl script to extract the text from the html file for your second method? That way you wouldn't need to use OpenOffice. Maybe there's a step or something I'm missing. Or, why not use pdftotext to extract

I just stumbled across this thread looking for someway to convert my pdf files into a more open format because I often find myself wishing it were possible to catalog and search through my collection of academic papers (which are currently stored on my disk as pdf files) without opening each one manually in acrobat or something. Pdftotext and pdftohtml just might be the ticket, but maybe a tool already exists to search a bunch of pdf files?

nahpets · Post by **nahpets** » Tue Jul 27, 2004 4:09 am

If you just want to search PDF files for keywords, I guess you can use pdftotext since formatting doesn't matter. In my case, I wanted to preserve the formating. The only tool I found which preserves the formatting was pdftohtml. I could have written a bash/python/c/java program to parse the HTML to give me text, but it wasn't worth the effort.

slartibartfasz · Post by **slartibartfasz** » Tue Jul 27, 2004 7:09 am

machinelou wrote:[...] but maybe a tool already exists to search a bunch of pdf files?

htdig (http://www.htdig.org) allows indexing of pdf files via pdftotext.

nahpets · Post by **nahpets** » Tue Aug 03, 2004 6:24 am

machinelou wrote: I just stumbled across this thread looking for someway to convert my pdf files into a more open format because I often find myself wishing it were possible to catalog and search through my collection of academic papers (which are currently stored on my disk as pdf files) without opening each one manually in acrobat or something. Pdftotext and pdftohtml just might be the ticket, but maybe a tool already exists to search a bunch of pdf files?

this may be what you're looking for:
http://multivalent.sourceforge.net/

How to convert PDF to TEX?

How to convert PDF to TEX?

html2latex

I'm retarded