ANSI C-compliant compiler (info req)

Eric Giguere giguere at aries5.uucp
Thu May 18 14:20:26 AEST 1989


In article <10281 at smoke.BRL.MIL> gwyn at brl.arpa (Doug Gwyn) writes:
>In article <198 at maytag.waterloo.edu> giguere at aries5.waterloo.edu (Eric Giguere) writes:
>>technically though, there are NO ANSI-conformant compilers.
>
>There are sure a lot of C compilers that attempt to be ANSI conforming,
>most of them currently in Beta test.  Remember that there is a delay
>between initial development of a product and its commercial availability.

The point I was making here was that NO ANSI Standard yet exists, hence
no ANSI-conformant compiler can exist.  The Standard is still in the
Proposed state, though very close (finally) to being approved and made
an official ANSI Standard.... whether the ISO will approve it is another
matter but I think ANSI is all most of the North American implementors
really care about anyhow...

>Validation of compilers that claim Standard conformance is another issue.
>I think there are efforts underway, and there certainly are test suites,
>but I don't have accurate current information about compiler validation.
>Validation is an issue because it would be the simplest way to demonstrate
>compliance with procurement specifications that require Standard conformance.

There are definitely some validation suites out there... I've had experience
with the Plum Hall suite.  The suites are quite expensive to license and
must constantly be updated because the Drafts have been changing (and bugs
keep being found in the suites).  I haven't heard that ANSI will be 
adopting any suite for use in certification tests, but I know the British
are looking at using one suite for official validation.

As I see it the real headache in the next few years will be getting people
to convert to ANSI-style compilers.  I cringe whenever I'm forced to use
`cc'... I miss the prototypes and the new-style declarations.  I think it's
a good idea to encourage people to use any conversion tools available to
convert old C into new C... as long as the new C code doesn't break, of
course.

Eric Giguere                                  268 Phillip St #CL-46
For the curious: it's French ("jee-gair")     Waterloo, Ontario  N2L 6G9
Bitnet  : GIGUERE at WATCSG                   (519) 746-6565
Internet: giguere at aries5.UWaterloo.ca         "Nothing but urges from HELL!!"



More information about the Comp.lang.c mailing list