[25413] trunk/dports/print/ghostscript/Portfile

N_Ox n.oxyde at gmail.com
Tue May 22 10:43:16 PDT 2007


I think it breaks it as it breaks ICU port, -I${prefix}/include get  
pulled in before the internal -I.. stuff, and headers from installed  
version got included instead of the sources ones.

Le 22 mai 07 à 03:52, Ryan Schmidt a écrit :

> On May 21, 2007, at 11:36, source_changes at macosforge.org wrote:
>
>> Revision: 25413
>>           http://trac.macosforge.org/projects/macports/changeset/ 
>> 25413
>> Author:   digdog at macports.org
>> Date:     2007-05-21 09:36:54 -0700 (Mon, 21 May 2007)
>>
>> Log Message:
>> -----------
>> patch for #11901
>>
>> Modified Paths:
>> --------------
>>     trunk/dports/print/ghostscript/Portfile
>>
>> Modified: trunk/dports/print/ghostscript/Portfile
>> ===================================================================
>> --- trunk/dports/print/ghostscript/Portfile	2007-05-21 16:28:48  
>> UTC (rev 25412)
>> +++ trunk/dports/print/ghostscript/Portfile	2007-05-21 16:36:54  
>> UTC (rev 25413)
>> @@ -26,6 +26,12 @@
>>          ghostscript-fonts-std-8.11.tar.gz md5  
>> 6865682b095f8c4500c54b285ff05ef6 \
>>          ghostscript-fonts-other-6.0.tar.gz md5  
>> 1a643ae62ef166562e4d422b1a601272
>>
>> +configure {
>> +        # MacPorts 1.4.1 or higher adds -I${prefix}/include and -L 
>> ${prefix}/lib flags to the configure by default.
>> +        # This change may break building ghostscript. Here is a  
>> workaround to avoid this issue.
>> +        system "cd $worksrcpath && ./configure --prefix=${prefix}"
>> +}
>> +
>>  build.target
>>
>>  destroot.destdir    prefix=${destroot}/${prefix}
>
>
> How does the MacPorts change break ghostscript? And are you sure  
> this is the best solution? Wouldn't it be better to just remove the  
> -I${prefix}/include and -L${prefix}/lib flags from their respective  
> variables? I think doing it the way you're doing it may break other  
> things, since now you're not having any of the env variables that  
> MacPorts uses. I'm guessing this would also break the +universal  
> variant (if in fact it ever worked with ghostscript).
>
>
> _______________________________________________
> macports-dev mailing list
> macports-dev at lists.macosforge.org
> http://lists.macosforge.org/mailman/listinfo/macports-dev

Anthony Ramine, a lazy french student.
nox at macports.org







More information about the macports-dev mailing list