Re: [RFC] libstdc++/6720 and libstdc++/6671

[Gabriel Dos Reis <gdr at codesourcery dot com>]

Zack Weinberg <zack@codesourcery.com> writes:

| On Thu, May 23, 2002 at 04:58:44PM +0200, Gabriel Dos Reis wrote:
| > Zack Weinberg <zack@codesourcery.com> writes:
| > 
| > [...]
| > 
| > | I am opposed to any change involving wiring a list of headers into the
| > | preprocessor, because it would only raise a debate about which headers
| > | should be on the list,
| > 
| > Since the list of required standard headers is quite well-defined by
| > the language definition, that point is moot therefore the opposition also.
| 
| You are considering only C++.

Yes, I'm considering only C++ because this is a C++ problem, raised by
implementing C++.  

| A change to the preprocessor has to
| consider C also;

If C is not concerned, there is no point in considering C.  I'll
restate it, right now this is a C++ specific problem.

| C89 and C99 both have similar wording (C99 7.1.2p3:
| 
| 	 If a file with the same name as one of the above < and >
| 	 delimited sequences, not provided as part of the
| 	 implementation, is placed in any of the standard places that
| 	 are searched for included source files, the behavior is
| 	 undefined.)

Yes, C has a different wording and has weaker requirement that C++ (I
already provided quotes, at the request of Mark).  Libstdc++-v3 isn't
implementing C, but C++ and it is a fact the two languages have very
very different requirement about standard headers and their content.
Let's tackle a problem at once.  If you want to provide the same
extension for other standard out of the scope of C++, yes that is
good, I'm for it.  But I don't think that should mean that we should
not implement ythe basic requirement of the C++ standard.

-- Gaby