GNOME Bugzilla – Bug 667954
test_properties.TestProperties.testRange fails under Gentoo Linux
Last modified: 2013-01-08 16:28:47 UTC
====================================================================== ERROR: testRange (test_properties.TestProperties) ---------------------------------------------------------------------- Traceback (most recent call last):
+ Trace 229449
obj.set_property(key, max)
---------------------------------------------------------------------- Ran 90 tests in 0.853s FAILED (errors=1) My system is an almost stable Gentoo system.
Here are more info : https://bugs.gentoo.org/show_bug.cgi?id=396505
Please post the URL in the downstream report and paste useful information from downstream here in this report so it's searchable.
Tests fail because ebuild tries to run ipv6 tests though no ipv6 support is available. It passes tests when ipv6 support is there (in the kernel config).
This is a bug report against glib, not pygobject, reassigning.
Erk, sorry. I was looking in the referenced upstream bug (https://bugs.gentoo.org/show_bug.cgi?id=396505), which is against glib (testing IPv6 sockets). So this is not the bug we are looking for. Is this still an issue for you with current PyGObject 3.2.0 and GLib 2.32? If so, which architecture is that?
Created attachment 212628 [details] build log I re-tested it again here under an up-to-date stable x86 Gentoo and got : ====================================================================== FAIL: test_python_calls_sync (test_gdbus.TestGDBusClient) ---------------------------------------------------------------------- Traceback (most recent call last):
+ Trace 230123
self.assertTrue('Timeout' in str(e), str(e))
OK, that's an entirely different error now. It's actually supposed to time out immediately, but that might be different with newer D-BUS versions.
Should be fixed with http://git.gnome.org/browse/pygobject/commit/?id=b9f24b4fbc2ca9f9b94b86f029c59b2fc3e8590f Can you please re-test with this patch or with current upstream git master? Thanks!
(In reply to comment #8) > Should be fixed with > > http://git.gnome.org/browse/pygobject/commit/?id=b9f24b4fbc2ca9f9b94b86f029c59b2fc3e8590f > > Can you please re-test with this patch or with current upstream git master? > Thanks! Hhm, with that pacth on top of 3.0.2 I get : ====================================================================== ERROR: testRange (test_properties.TestProperties) ---------------------------------------------------------------------- Traceback (most recent call last):
+ Trace 230126
----------------------------------------------------------------------
This is a completely unrelated test to the gdbus one (and back to the original), so I assume you got that failure before already? Can you please check whether this test case still fails with the 3.2.0 release tarball or master?
Also, can you please add some print statements there which show the values of "key", "gtype", "min", and "max" in the loop, so that we can see where it's failing?
(In reply to comment #10) The stable Gentoo stills stays at dev-libs/glib-2.30.2, therefore the test cannot be run here at my machine. (In reply to comment #11) > Also, can you please add some print statements there which show the values of > "key", "gtype", "min", and "max" in the loop, so that we can see where it's > failing? I'm not too familiar with Python and don't want to introduce own bugs, do you have patches for that ?
Something like: --- a/tests/test_properties.py +++ b/tests/test_properties.py @@ -355,6 +355,7 @@ class TestProperties(unittest.TestCase): obj = RangeCheck() for key, (gtype, min, max) in types.items(): + print('checking: key %s, gtype %s, min %s, max %s' % (key, gtype, min, max)) self.assertEqual(obj.get_property(key), getattr(RangeCheck.props, key).default_value) and then copy&pasting the output from TEST_NAMES=test_properties.TestProperties.testRange make check Thanks!
Toralf, can you work with the patch provided?
(In reply to comment #14) > Toralf, can you work with the patch provided? I could, but first /me wonders what's with this comment https://bugs.gentoo.org/show_bug.cgi?id=396505#c9 ?
(In reply to comment #15) > I could, but first /me wonders what's with this comment > https://bugs.gentoo.org/show_bug.cgi?id=396505#c9 ? Wrong bug? This looks unrelated. Anyway, this bug has been lingering around without being reproducible for months. Is it still actually an issue in current Gentoo with current pygobject?
Created attachment 232980 [details] build log This is the current situation : ====================================================================== FAIL: test_python_calls_sync (test_gdbus.TestGDBusClient) ---------------------------------------------------------------------- Traceback (most recent call last):
+ Trace 231353
(In reply to comment #17) > This is the current situation : > > ====================================================================== > FAIL: test_python_calls_sync (test_gdbus.TestGDBusClient) This got fixed in http://git.gnome.org/browse/pygobject/commit/?id=94a6cc93 So it seems the rest works now, so we can close this. Thanks! Martin