Challenging python coders: Who can create a gnucash-like GUI in python in a few weeks?

[Christian Stimming]

Guys,

you do realize that you're discussing something completely different, don't 
you? I proposed to experiment with a python-written GUI on top of the existing 
wrappers, which means they will re-use the existing C engine code (100k LOC) 
just as I did with C++/Qt. If you're discussing the possibilities of porting 
to mobile devices, please discuss this a bit more substantially and also 
please realize that basically you won't be able to re-use any code at all from 
the existing codebase.

Am Dienstag, 11. Januar 2011 schrieb John Ralls:
> That's somewhat conflicting: Android is Java: Portable, but looks crappy on
> everything else. Might require rewriting the backend in Java, too.

Android means you will be able to support a significant range of mobile 
devices but only if you stick to a pure java application. If you instead try 
to re-use the existing C engine code via JNI, you loose the device 
independence that you would buy by Java and instead you will have to compile 
for all device processors separately (via the suitable NDK = native 
development kit), not to speak of potentially missing dependencies like glib 
on the respective NDK. Also, the Java GUI code in android is substantially 
different from normal PC-based Java (no Swing or similar). Effectively, in our 
case any code sharing between an android application and the conventional PC-
based gnucash is doomed to fail. You will have to write a new application. 
That's why I agree http://gnc4a.rednus.co.uk is a good idea.

> Apple's
> iOS is Objective-C native, but I think that Qt supports it and I know that
> wxWidgets does, so the C backend with a C++ frontend will work.

Code sharing between an iOS application and any GPL project is also doomed to 
fail if you want to distribute the iOS app via Apple's market, see the recent 
VLC withdrawal from the App Store, http://planet.videolan.org/ . Additionally, 
again I don't think the dependencies of our C engine code (glib etc) are 
available on iOS, so also for iOS I don't believe any code re-use from the 
existing codebase is possible. You would have to write a new application as 
well.

So if you want to write something like gnucash on a mobile device, sure, you 
can go ahead. But my call for python coders was a different issue: How easy is 
it to write a python GUI on top of the existing python wrappers, re-using the 
existing C engine code, which is 100k LOC? That's what I would like to see so 
that we can decide the most suitable platform to implement major new features 
like a real multi-user gnucash or similar...

Best Regards,

Christian