[GiNaC-list] building GiNaC in mixed Cygwin/MinGW environment

[Alexei Sheplyakov]

On Wed, Jan 02, 2008 at 11:53:41AM +0100, Richard B. Kreckel wrote:
 
> You're right, the code is ELF-specific because of dlopen and friends. I 
> don't see what makes it glibc-specific, though.

I thought mktemp remove, and co are POSIXish functions, and GNU libstdc++
provides them, but other implementations might not do so. Apparently,
this is not true. These functions are required by C89, thus, they should
be declared in <cstdlib> and <cstdio>, respectively. So, excompiler.cpp
should include those headers. But we got a different problem.

> But then, the entire code is pretty much disabled to throwing stubs
> #if !defined(HAVE_LIBDL).

I guess the configure script got confused due to mixed Cygwin/MinGW
environment. I think we should tell woe32 users to build GiNaC either
in MinGW/msys environment, or Cygwin, but not a mixture of thereof
(unless they know *what* are they doing).

Best regards,
	Alexei
 
-- 
All science is either physics or stamp collecting.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 827 bytes
Desc: Digital signature
Url : http://www.cebix.net/pipermail/ginac-list/attachments/20080102/b2e411bc/attachment.pgp