> I admit I trusted info that are written on the wiki and did not take > (precious) time to check each entry. I agree with Jelle that having > wrong info is worse than having no info at all. I remember an old > discussion we had on the older pythonocc-ml, and someone said: "A wiki > is better than no wiki". I'm not sure of that anymore.
Spot on. Why on earth is there still no unit testing for wiki entries ;') > I'm afraid that Steve has to build his own OpenCascade 6.3.0 from > sources. I work with Ubuntu 8.04 (not 8.10) and never tried to build > pythonOCC on this platform. However, the problem of linux binary is > that > there are two many different distributions and packaging for Linux is > not so easy (I mean compared to Windows or MacOSX). > > The best we can do, in my opinion, is to provide a reliable build > system > that ensure pythonOCC can be built everywhere. Maybe I should go > back to > the scons script for a couple of days and try to improve it. That would be an interesting idea sure. Still, there is the effort of adapting the Scons script to different linux distributions, or does Scons ( in cmake fashion ) succeeds to abstract the platform to a large extent? -jelle _______________________________________________ Pythonocc-users mailing list Pythonocc-users@gna.org https://mail.gna.org/listinfo/pythonocc-users
