On Wednesday 08 December 2004 12:43 pm, Phil Thompson wrote: > > On Wed, 8 Dec 2004 17:04:06 -0000 (GMT), Phil Thompson > > > > <[EMAIL PROTECTED]> wrote: > >> Actually I (as the author of PyQt) have been thinking about it, but I > >> haven't yet got my MythTV setup going - so I'm happy to help. Why would > >> you need a custom C++ layer? > > > > Just figured there would need to be some code in myth to handle the > > loading of the py files as plugins, as well as to expose the myth C++ > > api to python. I've never written a binding from C++ to python > > before, so I have no idea of how much you can get away with in pure > > python - I'll defer to your knowledge if you say this isn't neccessary > > =) > > I'm not yet familiar with how myth is architected. Some work would be > needed to enable plugins to be written in Python. Similar work has been > done elsewhere - for example being able to write KParts using PyKDE - but > it would nice to formalise it. > > If there is a formal myth C++ API then it should be relatively easy to > automatically generate corresponding Python modules. That would allow > standalone applications to be written in pure Python.
There is a library (libmyth) that all the current plugins use to keep the same l&f as the rest of myth. Basically it's a bunch of convenience functions, some network protocol code, wrapped Qt widgets (better remote control, mainly), and a couple classes to handle stuff like pop-up dialogs and the menus, etc. Isaac _______________________________________________ mythtv-dev mailing list [EMAIL PROTECTED] http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev
