Topic: ANSI Releases of C++ Compilers


Author: rotherme@bcstec.ca.boeing.com (Pete Rothermel)
Date: Wed, 21 Apr 1993 21:59:45 GMT
Raw View
I'm wondering what the compiler vendors are going to do
after the final release of the ANSI C++ standard. Will
they make a new release that is "ANSI compliant"? How about
a compiler switch that flags the now ANSI portions of code as
warnings. Let's hear from you vendors. What are you planning?

Pete R





Author: bright@nazgul.UUCP (Walter Bright)
Date: 23 Apr 93 07:40:35 GMT
Raw View
In article <C5ut3n.2qn@bcstec.ca.boeing.com> rotherme@bcstec.ca.boeing.com (Pete Rothermel) writes:
/I'm wondering what the compiler vendors are going to do
/after the final release of the ANSI C++ standard. Will
/they make a new release that is "ANSI compliant"? How about
/a compiler switch that flags the now ANSI portions of code as
/warnings. Let's hear from you vendors. What are you planning?

We'll probably do the same thing we do for C. We provide a -A (ANSI compile)
switch, which is primarilly useful for running ANSI test suites. The default
mode will reflect the way the language is actually used by the majority of
our customers. Our customers pay the bills, not the ANSI committee, so we set
up the compiler the way they want it.

(An example of this are trigraphs, which I have yet to see in a program that is
not some sort of test suite. Processing trigraphs significantly slows
compilation, so by default they are not done. Another example are // comments
in C.)