License question

[Jeremy Lavergne]

Subports make sense iff the portfiles greatly overlap.

Mark Brethen <mark.brethen at gmail.com> wrote:

>
>On Jan 2, 2012, at 5:08 PM, Lawrence Velázquez wrote:
>
>> On Jan 2, 2012, at 5:24 p.m., Mark Brethen wrote:
>> 
>>> Looking at the pure Portfile, In don't quite follow:
>>> 
>>> if {${name} == ${subport}} {
>>> 
>>> }
>>> 
>>> Is everything in-between the curly brackets read only if user issues
>'port install pure'?
>> 
>> Yes. Another common idiom is
>> 
>>> if {${name} != ${subport}} {
>>>    <lots of stuff>
>>> }
>> 
>> where the port itself is treated as a stub port that just depends on
>one of its subports, and the meat of the port only takes effect when a
>subport is selected. There are a couple of decent examples linked at
>the bottom of this MacPorts wiki page:
>https://trac.macports.org/wiki/Python
>> 
>> vq
>
>I have a couple of choices: 1) as mentioned above, split the reduce
>port into 2 subports: reduce-csl (which builds the csl lisp base
>reduce) and reduce-psl (which just fetches precompiled psl lisp
>binaries) or  2) reduce builds csl by default and a subport,
>reduce-psl, which fetches the psl binaries. CSL has many options of its
>own, such as building wx instead of fox. And both have the option of
>building 32-bit instead of 64-bit.
>
>Which makes the most sense? 
>
>Mark
>
>
>
>
>_______________________________________________
>macports-dev mailing list
>macports-dev at lists.macosforge.org
>http://lists.macosforge.org/mailman/listinfo.cgi/macports-dev