-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi Daniel,
Daniel Stenberg wrote:
> On Sat, 6 May 2006, Bram Matthys (Syzop) wrote:
>
>> backported from libtool 1.6 by Paolo Bonzini
>
> Thanks for your mail. This silly dependency on a C++ compiler certainly
> is... silly.
*nod*
> So this is code originating from libtool? libtool is GPL licensed and we
> cannot easily just merge in code from them into c-ares.
I think you'll have to ask Paolo Bonzini to be sure. Or instead use your own
judgement.
Let me know if it cannot be used. Because TRE (another lib) used it, and TRE
recently went from GPL to LGPL, so if he cannot use it anymore, he probably
wants to know as well (even though MIT!=LGPL I suppose).
> But reading this patch, and if this is then how libtool works, can't we
> achieve an ever better result by simply setting all those tools'
> variables (like "CXX" and "F77" etc) to "no" before this function runs
> in configure?
>
> And if not, why does this patch include code for doing the C++ and F77
> checks etc when its sole purpose is to remove the dependencies on those
> tools?
If I'm not mistaken, this AC_LIBTOOL_TAGS is a general thing, which can be
called in a certain way so to enable for example both C and C++, or only
fortran, only xyz, etc. It allows you to choose...
So it's not designed to specificly remove everything except C. I wouldn't be
surprised if for (y)our purpose it can be made shorter.
I'm simply using this one, because I know from past experience it works ok.
Bram.
- --
Bram Matthys
Software developer/IT consultant syzop@vulnscan.org
PGP key: www.vulnscan.org/pubkey.asc
PGP fp: 8DD4 437E 9BA8 09AA 0A8D 1811 E1C3 D65F E6ED 2AA2
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (MingW32)
iD8DBQFEXkk/4cPWX+btKqIRAuxqAKDX3OuLND6bznXfSrcqD1GXcqJSCACgvuBd
j1K7VDmiLLpCtI+w2QU/5gw=
=s34A
-----END PGP SIGNATURE-----
Received on Sun May 7 21:23:47 2006