On Wed, 22 Sep 2004, Johnny C. Lam wrote:

> On Tue, Sep 21, 2004 at 07:50:32PM -0400, Dan Winship wrote:
>> (Yeah, yeah, I know. I kept meaning to test the wrapper framework, but
>> I was busy and didn't get around to it :-)
>>
>> The wrapper framework makes it impossible to build anything on Darwin,
>> because the gcc wrapper ends up being overwritten by the cpp wrapper
>> (because ${CPP} is "gcc -E") or something, and gcc turns into an
>> infinite loop. The attached patch can be used as a workaround until
>> it's fixed. (Or maybe it's even the right fix?)
>>
>> -- Dan
>>
>
>> Index: bsd.wrapper.mk
>> ===================================================================
>> RCS file: /cvsroot/pkgsrc/mk/wrapper/bsd.wrapper.mk,v
>> retrieving revision 1.1
>> diff -u -r1.1 bsd.wrapper.mk
>> --- bsd.wrapper.mk	21 Sep 2004 15:01:41 -0000	1.1
>> +++ bsd.wrapper.mk	21 Sep 2004 23:45:27 -0000
>> @@ -132,7 +132,9 @@
>>  #
>>  _WRAPPEES+=	AS
>>  _WRAPPEES+=	CC
>> +.if ${CPP:N-*} != ${CC}
>>  _WRAPPEES+=	CPP
>> +.endif
>>  _WRAPPEES+=	CXX
>>  _WRAPPEES+=	FC
>>  .if defined(USE_X11)
>
> This looks right to me.  I'll extend it to check for ${LD} == gcc as
> well, since that's also quite common.

This looks like a good candidate for a regression test. :)

-- 
Gillette - the best a man can forget