[OE-core] [PATCH 1/1] pkgdata: fix test_find_path to consider multilib

Richard Purdie richard.purdie at linuxfoundation.org
Mon Jun 17 15:51:42 UTC 2019


On Mon, 2019-06-17 at 16:32 +0800, Chen Qi wrote:
> Fix test_find_path test case to take into consideration of multilib
> being enabled. After this change, no matter multilib is enabled or
> not, the test case should pass.
> 
> Signed-off-by: Chen Qi <Qi.Chen at windriver.com>
> ---
>  meta/lib/oeqa/selftest/cases/pkgdata.py | 7 ++++---
>  1 file changed, 4 insertions(+), 3 deletions(-)
> 
> diff --git a/meta/lib/oeqa/selftest/cases/pkgdata.py b/meta/lib/oeqa/selftest/cases/pkgdata.py
> index 833a180..2bd15eb 100644
> --- a/meta/lib/oeqa/selftest/cases/pkgdata.py
> +++ b/meta/lib/oeqa/selftest/cases/pkgdata.py
> @@ -47,10 +47,11 @@ class OePkgdataUtilTests(OESelftestTestCase):
>          self.assertGreater(pkgsize, 1, "Size should be greater than 1. %s" % result.output)
>  
>      def test_find_path(self):
> -        result = runCmd('oe-pkgdata-util find-path /lib/libz.so.1')
> -        self.assertEqual(result.output, 'zlib: /lib/libz.so.1')
> +        base_libdir = get_bb_var('base_libdir')
> +        result = runCmd('oe-pkgdata-util find-path %s/libz.so.1' % base_libdir)
> +        self.assertEqual(result.output, 'zlib: %s/libz.so.1' % base_libdir)
>          result = runCmd('oe-pkgdata-util find-path /usr/bin/m4')
> -        self.assertEqual(result.output, 'm4: /usr/bin/m4')
> +        self.assertTrue('m4: /usr/bin/m4' in result.output)
>          result = runCmd('oe-pkgdata-util find-path /not/exist', ignore_status=True)
>          self.assertEqual(result.status, 1, "Status different than 1. output: %s" % result.output)
>          self.assertEqual(result.output, 'ERROR: Unable to find any package producing path /not/exist')

For selftest we really want a "known" configuration so I'm not sure we
want to go down this route as there are a lot of configurations which
could confuse the test results. 

It would probably be better to have selftest only build a clean config
and accept a limited number of variables. The parallel execution code
did go a long way to getting us to the point where it would do that.

Cheers,

Richard




More information about the Openembedded-core mailing list