the file triggers an attempt to rebuild autoconf files, which breaks the build
if automake-1.11 is installed. Worked out by Nigel Taylor.
Fix typo in DESCR while there. The patch was there because these flags
broke the build in an earlier version of libnice, but are no longer needed
(checked gcc 3 and 4; port didn't build on gcc 2 arch anyway). ok jasper@
move it from a BUILD+RUN_DEPENDS to WANTLIB+LIB_DEPENDS so that the dependency
is tracked and opal gets updated if the FFmpeg API changes. Comment to explain
why check-lib-depends says it's extra. Pointed out by Brad, ajacoutot@ hates
it but is ok with it.
no package change as the python module is only included if the python
flavour is selected (and unless that's included, MODPY_WANTLIB is blank),
but it's clearer this way.
side-note, python bindings should probably be split into a subpackage
sometime, there seems to be no need to have a separate flavour for this.
library dependency to be tracked properly. even though it's
dlopen(3)'d not linked directly, we still need to make sure the
package is compatible with the library.
it will show as Extra in lib-depends-check, that's not important
(but add a comment to make sure people don't think it can be removed).
pointed out by jakemsr and discussed with brad and espie.
script to use "gtar" not "tar". This is not added as a run dep to avoid
pulling in extra dependencies, especially xz which requires gcc3+ so is
not available on some arch and is more difficult to build on others.
(There's a clear error message at runtime if gtar is unavailable.)
ok dcoppa@ ckuethe@
from base causing a build failure in apr-util (as there's been nothing
changed since then to trigger an update). Set cache value to autoconf
to make sure nawk isn't picked up if for some reason it's hanging
around, though the REVISION bump alone is probably enough to take
care of most problems.
Asm6 is another 6502 assembler. It was built to do NES development, but
you can probably use it for just about anything. It was written because
the author thought most other assemblers either were too finicky, had
weird syntax, took too much work to set up, or were too bug-ridden to be
useful.
so it isn't a WANTLIB. Therefore the LIB_DEPENDS entry for multimedia/xvidcore
gets stripped out of the packing list at package build time. This means that
A) the dependency isn't pulled in directly (though as it's provided by other
deps the package does work OK) and B) print-package-signature doesn't
match pkg_info -S, triggering unnecessary rebuilds with dpb -R.
Move xvidcore from LIB_DEPENDS to BUILD+RUN_DEPENDS to fix this.
From Nigel Taylor.
* cut it into separate targets for readability and better testing
* filter libraries more efficiently
* use internal variable for resolve-lib
create two targets for solving wantlib: one (_fake-wantlib-args) which uses
the information under the fake directory and installed packages, and another
(_port-wantlib-args) which walks packing-lists. The second one can be used
to collect meta-info even when nothing is installed, and thus provide better
package-signature accuracy, or help with lib-depends-check.
The first one uses what's actually in place when a package is built.
For now, we're paranoid and use both when building a package, erroring out
if they don't match, even though _port-wantlib-args is somewhat slower.
there's a safeguard against self-depends anyways.
This prevents ports from getting moved from "built" to "installable"
prematurely.
For instance, devel/libsoup,-gnome doesn't directly depend on
net/glib2-networking, but it depends on devel/libsoup,-main which has
the net/glib2-networking.
Without this patch, dependencies on devel/libsoup,-gnome would often build
prematurely and build net/glib2-networking with them.
