Re: Preprocessor keyword #echo is missing

[Gabriel Dos Reis <gdr at integrable-solutions dot net>]

Martin Sebor <sebor@roguewave.com> writes:

| Zack Weinberg wrote:
| > "J. Grant" <jg-web@jguk.org> writes:
| >
| >>>Frankly I do not consider the elimination of that extra text to be
| >>>sufficient justification for adding another directive.  You can do
| >>>gcc ... 2>&1 | sed -e 's/warning: #warning //' if it really bugs you.
| >>
| >>I am using this currently, an #echo solution would be a better way to
| >>resolve this though.
| > I honestly do not see why you think that.
| > Note that even #warning is an extension, and has been controversial.
| >
| >>OK, well would you consider that #echo would be a useful addition to
| >>the GCC preprocessor language?  I personally do and I can not believe
| >>I am the only one who sees this gap that needs filling.
| > Sorry, but I do not see a need.
| 
| A #echo at the compiler level rather than at the preprocessor level
| would, IMHO, be extremely useful in the context of metaprogramming.
| At the very least, it would provide a framework for the proposed
| implementation of the static assert feature:
| 
| http://anubis.dkuug.dk/jtc1/sc22/wg21/docs/papers/2002/n1381.htm

Hi Martin,

 As you know, I (and Joddy and Robert) reported on that paper at the
last Santa Cruz meeting.  The declaration form proposed in the paper
doesn't integrate into the language general rules.  Robert and I
agreed on pursing the idea from expression point of view.  But there
remains some corner cases to work out.  

At any rate I would not support a preprocessor extension to implement
that -- for reasons given the the refrenced paper.

-- Gaby