cpplib: Fix __cplusplus definition

Neil Booth neil@daikokuya.demon.co.uk
Mon Jan 29 10:49:00 GMT 2001


Phil Edwards wrote:-

> While cleaning out my mailbox, I found that old discussion that Neil
> discovered in Google.  There was also a followup from Zack at the time:
> 
> > Would extending STDC_0_IN_SYSTEM_HEADERS help here?  Presently on
> > Solaris cpp causes __STDC__ to be 1 in user code and 0 in system headers.
> > It would be easy to apply that to __cplusplus as well, so it would be
> > 199711L in user code and 1 or 0 in the headers.
> 
> Neil, did you see this in the archives?  Does it still apply?

No and yes.

> Is this the way we want to go?  I'm still wavering between "try to make
> use of their attempt at standards compliance" and "just subvert the whole
> mess to our indomitable will".

I don't know.  I'd rather we don't hack it if possible.

I don't understand what the STDC_0... macro works around either - I've
never used a Solaris box.

Neil.


More information about the Libstdc++ mailing list