Re: [Patch] Fix ac_c99_complex configury

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

Paolo Carlini <pcarlini@suse.de> writes:

| Gabriel Dos Reis wrote:
| 
| >Old iostream members refer to the classic iostreams,
| >i.e. <iostream.h>.  Thereis no doubt about it :-)
| >  
| >
| Agreed, you are using the smiley, but I'd like to see a bit of
| substantive evidence for this interpretation.

Just look around.
What else do think "old iostreams members" mean in the compatibility
section? 

| >The reason for removing complex.h is that it leads to real clashes in
| >semantics with respect to C99 and we don't really need it.  I have not
| >seen similar reasons for other cases.  That is a sound rationale:  If it
| >won't break don't fix it.  <complex.h> breaks; others don't pause
| >maintainer problems we know of.  
| >
| Then, maybe we should find a way to solve the problems themselves. For
| instance, AFAIK, ICC/Dinkum used to provide a set of backward/ headers,
| among which <complex.h> and <algobase.h> and <hash_map.h>, and so on.
| Maybe they figured out a way to solve (or work around) those problems,
| preferable to removing just a subset of the headers which we don't
| ""like"" :-)

Certainly on my side, it is not a "I don't like" thingy.  If you want
to remove files because you don't like them, then that is certainly
not a good enough reason. 
My rationale is not based on affection.

   #include <stdio.h>
   #include <complex.h>

   int main { }

which header do you want to pick, given the C99 compatibilty? 

-- Gaby