Martin v. Löwis a écrit : > Neil Cerutti schrieb: >> On 2007-02-23, I V <[EMAIL PROTECTED]> wrote: >>> While that's true, C++ compiler vendors, for example, take >>> backwards compatibility significantly less seriously, it seems >>> to me. >> Compiler vendors usually take care of their customers with >> compiler switches that enable backwards compatibility. > > That's a bold statement, after I V already gave two examples > (g++ and MSVC 2005) of compilers that broke backwards compatibility > without providing compiler switches to bring it back. These two > aren't "minor" compilers.
In C++ land, people are expected to fix their code and not use broken compilers in such situations. They have some kind of moral high ground because they do not break compatibility just for that but to be more standards compliant. And the fact that all the C++ code I've written with g++ recently compiled perfectly on Visual 2005 without a single change ( except fixing new warnings ) shows that standard compliance in C++ compilers is a good thing to have. That same code had some major and very annoying breakage with Visual 6. The fix? Updating that outdated compiler. -- http://mail.python.org/mailman/listinfo/python-list
