[Scons-dev] Problematic conflicts between C/C++ and D

[Russel Winder]

On Tue, 2017-08-29 at 07:39 -0700, Bill Deegan wrote:
> env.get is just like a python dictionary.

I can live with that. :-)

> the order will depend on the order the tools are loaded into the
> Environment()..

This is where the problem comes. I think there has to be a single,
deterministic loading order and that, to avoid breaking changes, that
the cc and c++ tools win for the definition of SHOBJSUFFIX. I am
surprised though that there have not previously been more hassles over
this as not all C and C++ compilers use .os extension.

I think we just have to accept that the default system is broken, and
that the GCC toolchain is the winner. We just have to warn dmd and ldc
toolchain users that this is the case and that using an explicit rather
than implicit/default toolchain, the problem goes away. (gdc is part of
GCC Suite and so doesn't have the problem.)

The current set of people using SCons for D builds haven't actively
complained so there is no real driver for "a fix" other than me wanting
it to work.
 
-- 
Russel.
=============================================================================
Dr Russel Winder      t: +44 20 7585 2200   voip: sip:russel.winder at ekiga.net
41 Buckmaster Road    m: +44 7770 465 077   xmpp: russel at winder.org.uk
London SW11 1EN, UK   w: www.russel.org.uk  skype: russel_winder
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: This is a digitally signed message part
URL: <https://pairlist2.pair.net/pipermail/scons-dev/attachments/20170830/fa7b30a8/attachment.pgp>