openembedded-core.git/meta/recipes-support, branch pyro

libproxy: speed up upstream version check

2017-04-28T10:26:07+00:00

Something in the fetched webpage made the default regex matching really slow.

Signed-off-by: Alexander Kanavin 
Signed-off-by: Ross Burton

boost: fix upstream version check

2017-04-28T10:26:07+00:00

Signed-off-by: Alexander Kanavin 
Signed-off-by: Ross Burton

libcap: drop obsolete attr PACKAGECONFIG option and libattr dependency

2017-04-28T10:26:06+00:00

In the 2.25 release, libcap dropped its dependency on an external
libattr library:

  https://git.kernel.org/pub/scm/linux/kernel/git/morgan/libcap.git/commit/?id=85f38a573fc47472ab792e813b6f6b6f0b1df112

Signed-off-by: Andre McCurdy 
Signed-off-by: Ross Burton

acl: fix race issue when do_compile

2017-04-19T09:16:22+00:00

Fixed race issue:
 In file included from acl_copy_entry.c:22:0:
 libacl.h:19:21: fatal error: sys/acl.h: No such file or directory
  #include 

[snip]

 compilation terminated.
 acl_get_file.c:27:24: fatal error: acl/libacl.h: No such file or directory
  #include 
                         ^

The acl.h is in "include" directory, and include/Makefile creates
symlink "sys" and "acl" poinst to current dirctory:
$ ls include/ -l
acl -> .
sys -> .

So if "libacl" target runs before "include", the error would happen
since no "acl" or "sys" directory.

Let libacl depend on include can fix the problem.

[YOCTO #11349]

Signed-off-by: Robert Yang 
Signed-off-by: Ross Burton

gpgme: add a dependency on python-unixadmin to Python bindings package

2017-04-13T22:57:38+00:00

Signed-off-by: Alexander Kanavin 
Signed-off-by: Richard Purdie

gpgme: correctly avoid host contamination

2017-04-13T22:57:38+00:00

Existing patch was actually doing the wrong thing and
sometimes removing a linking flag (-lgpgme) that should be present.

Instead, gpgme-config actually has internal logic to remove /usr/lib from the output,
which works only in non-multilib setups, so it is adjusted to include all possible
/usr/lib* and /lib* directories.

Signed-off-by: Alexander Kanavin 
Signed-off-by: Richard Purdie

ptest-runner: Upgrade to minor version 2.0.2

2017-04-11T17:09:21+00:00

To fix a problem when print ERROR after a ptest timeout, this
causes the user confusion about if a test ends or not.

[YOCTO #10842]

Signed-off-by: Aníbal Limón 
Signed-off-by: Richard Purdie

gpgme: fix configure if 'import distutils' causes output on stderr

2017-04-05T08:44:10+00:00

There are a number of reasons that importing a module could cause output on
stderr that isn't a fatal error (compatibilty problems with inputrc, or encoding
warnings) so backport a patch from autoconf-archive to only check the exit code
instead of asserting that stderr is empty.

[ YOCTO #11231 ]

Signed-off-by: Ross Burton 
Signed-off-by: Richard Purdie

boost: various cleanups

2017-03-26T12:17:02+00:00

Re-order some assignments to be logically arranged.

Remove the set -ex statements as they serve no purpose.

Pass --debug-configuration to see what configuration steps boost is taking.

Signed-off-by: Ross Burton 
Signed-off-by: Richard Purdie

boost: port boost-python to Python 3

2017-03-26T12:17:02+00:00

As Python 3 is the default Python version, change Boost to build against Python
3 instead of Python 2 if enabled.  It's not simple to support both, so this
means that support for building boost-python against Python 2 has been removed.

This involves backporting a number of patches upstream to fix Python 3 support,
and telling Boost precisely where to find the Python headers and libraries so
that it doesn't try to invoke the host Python to determine these values.

[ YOCTO #11104 ]

Signed-off-by: Ross Burton 
Signed-off-by: Richard Purdie