FS#50428 - {namcap} building *namcap-git 3.2.7.9.g* errors, "FAILED (failures=1, errors=2, expected failures=2)
Attached to Project:
Arch Linux
Opened by James Harvey (jamespharvey20) - Thursday, 18 August 2016, 02:19 GMT
Last edited by Kyle Keen (keenerd) - Thursday, 22 February 2018, 23:18 GMT
Opened by James Harvey (jamespharvey20) - Thursday, 18 August 2016, 02:19 GMT
Last edited by Kyle Keen (keenerd) - Thursday, 22 February 2018, 23:18 GMT
|
Details
On 6/8/2016, I successfully built
namcap-git-3.2.7.2.g4424a23-1.
Today, 8/17/2016, I cannot. Nor can I build master, commit 99bf6961. Leads me to believe a dependency's new version is at fault. Using devtools testing-x86_64-build. Also fails if I disable the test repos, to rule out an un-released package causing this. Can you replicate? libelf-0.166-1 elfutils-0.166-1 licenses-20140629-1 pyalpm-git-0.8.r0.g1b5abcf-1 python-3.5.2-1 python-pyelftools-0.24-1 perl-error-0.17024-1 python-packaging-16.7-1 python-pyparsing-2.1.8-1 python-six-1.10.0-2 git-2.9.3-1 python-setuptools-1:25.2.0-1 Entire build log here: http://pastebin.ca/3695741 |
This task depends upon
======================================================================
FAIL: test_satisfied2 (Namcap.tests.test_depends.DependsTests)
----------------------------------------------------------------------
Traceback (most recent call last):
File "/build/namcap-git/src/namcap-git/Namcap/tests/test_depends.py", line 68, in test_satisfied2
self.assertEqual(w, expected_w)
AssertionError: Lists differ: [] != [('dependency-already-satisfied %s', 'python')]
Second list contains 1 additional elements.
First extra element 0:
('dependency-already-satisfied %s', 'python')
- []
+ [('dependency-already-satisfied %s', 'python')]
----------------------------------------------------------------------
Searching for others getting that, I see that I ran into that about a year ago. I must have disabled the test back then to get around it. I build AUR packages using devtools, using a core-x86_64-build script based off testing-x86_64-build, which uses core. Perhaps that's why I'm running into it and apparently not others - don't think many use devtools with AUR. I've disabled that test again, so I'm happy.