Paul A. Giannaros wrote: > On Fri, Oct 30, 2009 at 5:28 PM, Ville M. Vainio > <[email protected]<mailto:[email protected]>> wrote:
> Have you rejected the += and -= operator overloads for connecting and > disconnecting signals? No, I just thought that would be a bit separate matter and might better constitute a PSEP of it's own. > > Regarging the renaming "pyqtSignal => signal"; I think pyqt follows > > this convention to avoid (future?) naming clashes with Qt. Does pyside > > have the luxury of ignoring such a convention? > > And even if it's not to avoid future name clashes, it's a little > unintuitive having a pair of functions which differ only by case > (QtCore.signal vs QtCore.SIGNAL). "pyqtSignal" should clearly not be > the standard name though. I believe Both of these concerns are really valid. I proposed the naming scheme since any alternatives I could think of were even worse. Any prefixes such as "py" or "python" would seem redundant to me. Having two functions differing only by case is also really unfortunate. On the other hand, I'd tend to take Hugo's approach regarding the risk of having clashes with nativd Qt names: I doubt such C++ functions would be added in Qt 4, and Qt 5 (assuming such as beast is even planned - I have no clue about it) would probably be backwards-incompatible in any case, so we could then fix the situation as we please. Also, I think in practice we could live with the signal and SIGNAL functions: the ugly spelling relates to the legacy style and the nice one to the pythonic one. ;-) I'll add the concerns in the draft PSEP on Monday, in any case. If you have any ideas for a better scheme, shoot. Cheers, ma.
_______________________________________________ PySide mailing list [email protected] http://lists.openbossa.org/listinfo/pyside
