| |||||||||||||
What Every C Programmer Should Know About Undefined Behavior #3/3What Every C Programmer Should Know About Undefined Behavior #3/3Posted May 23, 2011 3:25 UTC (Mon) by neilbrown (subscriber, #359)In reply to: What Every C Programmer Should Know About Undefined Behavior #3/3 by dgm Parent article: What Every C Programmer Should Know About Undefined Behavior #3/3
> Maybe it's time to try to reduce and simplify C++ instead of adding more and more features.
As the same time that Sun was working on Oak (which became Java), there was a team working on "Clarity" which was meant to be a simpler C++... I wonder what became of that (i.e. I cannot be bothered hunting to find an answer, I want someone to tell me :-)
(Log in to post comments)
What Every C Programmer Should Know About Undefined Behavior #3/3 Posted May 24, 2011 17:23 UTC (Tue) by cmccabe (guest, #60281) [Link]
It would be interesting if C++ had a "use strict" mode similar to Perl. Something that would try to eliminate the worst abuses. The problem is that there really is a lot less consensus on what "the worst abuses" are than in Perl. I have my own list, but I'm sure that posting it here would generate a flamewar (and probably lead to me educating a lot of people about obscure C++ trivia).
I know that you realize this (and clearly all the kernel developers do too), but a lot of developers don't realize that you can use C exactly like a "simpler C++." Instead of classes, you have functions which all take a pointer to a struct. Instead of private methods, you have functions which are static to a file. Instead of virtual functions, you have tables of function pointers.
I would also encourage anyone who is tired of C++ to check out Google Go. At least give it a try.
What Every C Programmer Should Know About Undefined Behavior #3/3 Posted Jun 5, 2011 11:07 UTC (Sun) by JanC_ (guest, #34940) [Link]
I'm sure somebody can write a 2400 page book to define a safe subset of C++ ;-)
|
|||||||||||||