I didn't realize the c++11 standard could make it possible to get rid of moc. Defenately better this way around ;)
I can't wait to see this happening upstream for Qt5.
moc magic for signals/slots was already possible with C++98, as proven by Gtkmm already back in 2000.
http://libsigc.sourceforge.net/
But it required beefy template support, which would leave out some of the compilers supported by Qt (not only MSVC).
I think I'm not contradicting you, but I just wanted to state that I've been using libsigc with MSVC since 2002 or 2003.
I wanted to make a point about other compilers, as for many HNers only MSVC, gcc and clang do exist, while Qt supports many others.
Advanced template metaprogramming is far less common of a skill than opening a generated source file and reading source code.
Re:CopperSpice, a Modern C++ Fork of Qt
by stevejones on Tuesday June 09, 2015 @09:23AM
(#9685735)
C++ is a far more common skill than moc. I know C++, I can barely get my head around writing a Qt app.
Exactly , Even if TMP is much harder than MOC, TMP is standard , by forcing people to learn it they are not doing something wrong or ridiculous , but by forcing people to learn something specific to framework we are doing something wrong in my opinion {then why should we have standard at first place ? every one can create its own C++ }
Well, one of the aim of C++11 (and 14) is to make template metaprogramming much easier.
To the user it shouldn't matter wether you use a moc DSL or a template DSL. If you're a developer, you'll have to learn the innards of the moc on top of C++, which already contains templating anyway. When I first saw the signal/slot API, I wondered how they solved it and was dissappointed that they used a standalone tool for something that could've been solved with standard C++.
You mean they replaced moc magic with C++11?