Hello gang,
even though, we live in hard pre-F16 times, I have an AutoQA related update. There is a new branch in git "resultsdb_integration" http://git.fedorahosted.org/git/?p=autoqa.git;a=shortlog;h=refs/heads/resultsdb_integration, which contains the changes required for basic autoqa-resulsdb cooperation. All the tests should now be storing test results, and relevant/reasonable data (need to check up on this later with the test devs though ;-))
autoqa.conf#102: Let's not put our private URL into the config file.
all tests: What about
self.detail.resultsdb['keyval_pairs'] =
convert to
self.detail.keyvals =
?
depcheck.py: why do you need to add the following line?
from copy import deepcopy
test.py:
* AutoQATest.resultsdb attribute could be private, tests doesn't have to operate on it directly. * Does it make sense to move AutoQATest.is_multiresult to AutoQATest.detail.is_multiresult? * AutoQATest.job_id doesn't seem to be needed by test authors. Can we move it e.g. inside resultsdb attribute and hide it from the test author? * Does it make sense to replace:
if test_detail.resultsdb['testrun_id'] in [0, None]:
to
if not test_detail.resultsdb['testrun_id']:
? * I haven't seen the ResultsDBApi class. What happens when there is an exception/timeout/error during starting or ending testruns/jobs? Will it crash our test?