Alan W. Irwin wrote: > On 2009-07-01 02:17+0300 Dmitri Gribenko wrote: > > Thanks Terrence and Dmitri for your ideas about how to solve this issue. > > I believe Alban will not be able to comment in a timely manner because he is > probably out of e-mail contact right now. Given that situation, I suggest > you both in consultation make the casting changes you think are right and > send me the patch. If I find that examples/python/pyqt4_test.py (in the > installed examples tree) suddenly starts working (rather than delivering a > blank GUI result) and your patch shows no regressions for any of our > extensive tests of the qt devices or for qt_example itself, then I would be > happy to commit your patch. Of course, I would be also be happy to make any > further changes beyond that commit that Alban thinks are necessary when he > is back in contact. > > Alan > __________________________ > Alan W. Irwin > > Astronomical research affiliation with Department of Physics and Astronomy, > University of Victoria (astrowww.phys.uvic.ca). > > Programming affiliations with the FreeEOS equation-of-state implementation > for stellar interiors (freeeos.sf.net); PLplot scientific plotting software > package (plplot.org); the libLASi project (unifont.org/lasi); the Loads of > Linux Links project (loll.sf.net); and the Linux Brochure Project > (lbproject.sf.net). > __________________________ > > Linux-powered Science > __________________________ > > ------------------------------------------------------------------------------ > _______________________________________________ > Plplot-devel mailing list > Plplot-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/plplot-devel > Hello all,
Actually, this is one of the few days I am in contact, so: - I must have misunderstood something about dynamic casts because I believed that QtPLDriver* widget=(QtPLDriver*)(pls->dev) would have created a "pure" QtPLDriver* object, and thus the member functions called after this would have been e.g. QtPLDriver::drawLine rather than e.g. QtRasterDevice::drawLine (no call of the virtual functions). This was the meaning of my "cryptic comment". - There is a simple and straightforward way to fix that without worrying about RTTI or whatever (well, at least for that): just define plD_whatever functions for every kind of driver, using simple C-casts. I'm afraid lots of these matters exceeded my knowledge in C++, but I'm interested by the explanations you provide! Alban ------------------------------------------------------------------------------ _______________________________________________ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
