On Jul 4, 2013, at 12:05, [email protected] wrote: > Revision: 107723 > https://trac.macports.org/changeset/107723 > Author: [email protected] > Date: 2013-07-04 10:05:50 -0700 (Thu, 04 Jul 2013) > Log Message: > ----------- > Added universal test (includes universal-2 test). > > Added Paths: > ----------- > branches/gsoc13-tests/tests/test/universal/DESCRIPTION > branches/gsoc13-tests/tests/test/universal/Portfile.in > branches/gsoc13-tests/tests/test/universal/test.tcl > > Removed Paths: > ------------- > branches/gsoc13-tests/tests/test/universal/Makefile > branches/gsoc13-tests/tests/test/universal/Portfile > branches/gsoc13-tests/tests/test/universal/master > > Added: branches/gsoc13-tests/tests/test/universal/DESCRIPTION > =================================================================== > --- branches/gsoc13-tests/tests/test/universal/DESCRIPTION > (rev 0) > +++ branches/gsoc13-tests/tests/test/universal/DESCRIPTION 2013-07-04 > 17:05:50 UTC (rev 107723) > @@ -0,0 +1,7 @@ > +This test checks the handling of the universal variant in MacPorts base. > +Every port should have an universal variant unless 'universal_variant no' is > +specified.
There are actually many situations that result in the default universal variant not being available: * universal_variant no * use_configure no * supported_archs noarch In addition, ports can include any number of portgroups, and any of those portgroups could change the behavior. For example, they could disable the universal variant if it is known that all ports of that class of ports cannot be built universal. Portgroups and individual ports can also add a custom universal variant, if the default one won't work. > +There are 2 test cases. One for universal variant set and another for no > +universal variant. For each test a specific Portfile is generated from the > +Port.in file. _______________________________________________ macports-dev mailing list [email protected] https://lists.macosforge.org/mailman/listinfo/macports-dev
