John J. Lee wrote: > Alternatively, if you feel rather adventurous, try compiling KPDF on > Windows with the free Qt3 port(s?) or Qt4, if that's arrived yet.
I think you might be slightly too optimistic about getting KPDF to run natively on Windows. The project you mention does have a page describing how to run KDE 3.4 on Windows, but I think that may be taking things further than the original poster intended: http://kde-cygwin.sourceforge.net/kde3/index.php I believe someone is working on a Qt 4 renderer for Poppler, a PDF rendering library based on xpdf, so it could potentially be used in an application on Windows: http://poppler.freedesktop.org/ > If you're successful there, you still need to figure out writing KPart > plugins in Python, which certainly used to be tricky, but maybe that's > a solved problem by now... Writing KPart plugins in Python (for embedding into applications) was never all _that_ tricky; it just required some messing around with shared libraries. As it turns out, you don't need to do this if you just want to control KPDF from a client application written in Python. There are two approaches you can take (for Windows users, this is where we go off-topic): 1. Use DCOP to control two running instances of KPDF. At the simplest level, you could use the popen2 module and the "dcop" tool to control what KPDF displays. This could be integrated into a client application to varying degrees, with the application getting the window manager to do things with window placement, but it would probably end up being a lot of effort for less than the desired result. http://developer.kde.org/documentation/tutorials/ 2. Use the kparts module from PyKDE to display KPDF KParts in the client application's GUI. You can still control the KParts via their DCOP interfaces, and you can call methods on them directly to, for example, open specific URLs or add user interface elements. http://www.riverbankcomputing.co.uk/pykde/ Since I discovered how easy it was to embed KPDF like this, it makes me want to experiment more with using KParts in Python. It's already trivial to use KHTML from Python, and KParts like these could be used to create specialised tools for documentation and visualisation. David -- http://mail.python.org/mailman/listinfo/python-list
