[prev in list] [next in list] [prev in thread] [next in thread] 

List:       koffice-devel
Subject:    Re: pdf import in KWord
From:       Martin Pfeiffer <hubipete () gmx ! net>
Date:       2006-10-16 17:47:56
Message-ID: 200610161947.56565.hubipete () gmx ! net
[Download RAW message or body]

On Monday 16 October 2006 18:26, Cyrille Berger wrote:
> > Its on the TODO list for quite some time now, but nobody every actually
> > started work on it.
Yeah I came also up with that on IRC two weeks ago and I looked at the poppler 
sources and asked on #okular. The current state of my research:
- the private poppler core lib is hard to use ( as I heard on #okular )
- we would need to write a special binding because the core lib is private
- for abiword there exists a poppler import filter: http://jauco.nl/blog/
- they have written such a binding that translates pdf to a special xml which 
is not documented afaik
- using it would be a ( bad imho ) compromiss because we need to translate xml 
to odf
- imo the best approach would be to talk with abiword people to create a 
binding == outputdev( such as the existing qt4 ) used by any office suite as 
interface to poppler
- or second approach to use odf instead of that special xml

About a filter's quality:
- a problem are fonts: they are stored inside the pdf doc as vectors and we 
would need something to recognize them correctly as the wrong choice of fonts 
while importing often breaks the whole layout
- we can get the position of each character in the doc, which would require a 
layout recognition too. Paper about this is here: 
http://dbis.uni-trier.de/Mitarbeiter/klink_files/www/Postscript/DAS2000-FinalVersion.pdf

Cheers Martin
_______________________________________________
koffice-devel mailing list
koffice-devel@kde.org
https://mail.kde.org/mailman/listinfo/koffice-devel
[prev in list] [next in list] [prev in thread] [next in thread]