Ok. I ran into a segmentation fault with both gfxtest and worldtest with the patch applied. I only had time to make sure I didn't have it without the patch. I'll try to look into it soon. It may be something different with my setup, or I may have munged the patch. If you can try world test that may help also.
Tyler On Feb 8, 2008 4:24 PM, Rian Shelley <[EMAIL PROTECTED]> wrote: > Its almost bedtime already for me. We are a few timezones off :D. > > I've attached the modified gfxtest.cc that I used to test the new > code. You can do what you want with it. > > Let me know if there is anything else that needs worked on! > > > On Feb 8, 2008 2:45 PM, Tyler Nielsen <[EMAIL PROTECTED]> wrote: > > Well, I'm not going to complain when he's doing so much more, faster > than I > > have been. My plan is to get this stuff commited tonight after I put > the > > kids to bed. (Roughly 8 hours from now). If you have a usage example, > it > > will help speed up my testing. But anything you want to do before I > commit > > is fine. If you want me to hold off until you make one more pass that's > > fine also. > > > > Tyler > > > > > > > > On Feb 8, 2008 2:31 PM, Rian Shelley <[EMAIL PROTECTED]> wrote: > > > > > > > > > > > > > > > On Feb 8, 2008 2:05 PM, Kai Sterker <[EMAIL PROTECTED]> wrote: > > > > On Feb 8, 2008 5:41 PM, Rian Shelley <[EMAIL PROTECTED]> wrote: > > > > > > > > > I believe that I have successfully implemented all of your > > suggestions: > > > > > > > > That was fast :-). > > > It doesnt take much time, I just have to sit down and do it > > > > > > > > > > > > > > > All of the underscores in the method names should be gone > > > > > > > > Oops. Should have been the other way round (i.e. conditional_purge > or > > > > set_max_mem). Sorry, if I confused you. > > > > > > I guess I'm not used to using them, so I was going through and > > > correcting all of my usual style mistakes :D > > > > > > > > > > > > > > > > > > One final thing that could be done now, is generating the Python > > > > wrappers for the new cache class. That is usually achieved by > > > > including new classes in the corresponding module under > > > > src/py-wrappers/adonthell. In our case, that would be adding the > line > > > > '%include "gfx/surfacecacher.h"' to the end of py_gfx.i. SWIG will > > > > then take care of the rest. > > > > > > I've done a lot of C++ and a lot of Python, but this will be the first > > > time I've ever done both... > > > > > > > > > > > > > > With the wrapper in place, a test script could be written (see > > > > examples in the test/ directory) that would serve two purposes: > > > > * act as a usage example > > > > * allow (some) regression testing in case of code changes > > > > > > I've actually already written one. I just didn't include it in the zip > > file. > > > > > > > > > > > > > > Since Tyler will integrate your code with his own, I'm not sure how > > > > much sense it makes to do that right now, as things might change a > > > > bit. But if you want to get more familiar with that aspect of the > > > > engine, you could give it a shot. > > > > > > I'll wait and see what he thinks. > > > > > > > > > > > > > > > > > > > > > _______________________________________________ > > > Adonthell-devel mailing list > > > Adonthell-devel@nongnu.org > > > http://lists.nongnu.org/mailman/listinfo/adonthell-devel > > > > > > > > > _______________________________________________ > > Adonthell-devel mailing list > > Adonthell-devel@nongnu.org > > http://lists.nongnu.org/mailman/listinfo/adonthell-devel > > > > > > _______________________________________________ > Adonthell-devel mailing list > Adonthell-devel@nongnu.org > http://lists.nongnu.org/mailman/listinfo/adonthell-devel > >
_______________________________________________ Adonthell-devel mailing list Adonthell-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/adonthell-devel
