On Wed, May 23, 2001 at 11:58:51AM -0400, Bruce Momjian wrote:
> > > I don't see the problem here. My assumption is that the comment is not
> > > part of the define, right?
> >
> > Well, that's the question. ANSI C requires comments to be replaced by
> > whitespace before preprocessor commands are detected/executed, but there
> > was an awful lot of variation in preprocessor behavior before ANSI.
> > I suspect there are still preprocessors out there that might misbehave
> > on this input --- for example, by leaving the text "* end-of-string */"
> > present in the preprocessor output. Now we still go to considerable
> > lengths to support not-quite-ANSI preprocessors. I don't like the idea
> > that all the work done by configure and c.h in that direction might be
> > wasted because of pgindent carelessness.
>
> I agree, but in a certain sense, we would have found those compilers
> already. This is not new behavour as far as I know, and clearly this
> would throw a compiler error.
This is good news!
Maybe this process can be formalized. That is, each official release
migh contain a source file with various "modern" constructs which we
suspect might break old compilers.
A comment block at the top requests that any breakage be reported.
A configure option would allow a user to avoid compiling it, and a
comment in the file would explain how to use the option. After a
major release, any modern construct that caused no trouble in the
last release is considered OK to use.
This process makes it easy to leave behind obsolete language
restrictions: if you wonder if it's OK now to use a feature that once
broke some crufty platform, drop it in modern.c and forget about it.
After the next release, you know the answer.
Nathan Myers
ncm@zembu.com
