On Mon, 1 Jul 2002 20:11:11 -0500 Mark Evans <[EMAIL PROTECTED]> wrote:
ME> I am interested in possibly reviving the integrated Python ME> scripting in Mahogany. Glad to hear it! ME> I'm a complete newcomer to this project and before diving in, wish to ME> ask what issues I might face. Let me assure that you'll find all the help you need here. ME> Could someone tell me what are the problems associated with such an ME> effort. This depends on what you want to do. To be honest, I'd much prefer minimal program, at least initially, as over ambitious projects have an annoying tendency to never get completed and although it's up to you to decide, I'd propose the following: 1. clean up the existing interfgace files, check that they still work 2. add the missing functions, wrap the new methods which are not currently wrapped and so on ME> I would be interested in using Python 2.2.1 (the latest). There should be no particular problems with this, although it would be nice if we could keep support for the older versions as well -- at least because then I'd be able to test it myself. ME> I have embedded the Python interpreter before, How did you do it? Would you already know about SWIG by chance (look at http://www.swig.org/ if you don't)? ME> As I understand the Python scripting capability, there was/is Python ME> code interfacing to the internal C++ object model of Mahogany. Yes, and the glue between the 2 is provided by SWIG. ME> While I'm on the subject, I'll just toss in a possibly helpful link to ME> my favorite C++/Python interfacing tool, the Boost Python Library. ME> http://www.boost.org/libs/python/doc/index.html I've heard a lot of good about boost Python module but I'm not sure if it really makes sense to break everything and start anew. If you're a boost expert and are completely unfamiliar with SWIG it might, but otherwise I'd rather keep the existing infrastructure. ME> Another semi-related question -- What about a wxPython port of Mahogany? ME> Does anyone think that would be worthwhile? Worthwhile -- maybe. Doable -- hardly. Mahogany contains a lot of code and even though big parts of it would profit from a rewrite, it would be a lot of work and I don't believe it can be done without massive development effort. I'd estimate that only a third of Mahogany is about GUI (and so might be easily ported to wxPython), the rest is not. And I've never heard of anybody porting a C++ program of this size to Python before. And although there are a lot of advantaged in using Python, there are quite a few problems as well. Regards, VZ ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ Mahogany-Developers mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mahogany-developers
