Build.joomla.com discrepency

42 views
Skip to first unread message

garyamort

unread,
Mar 25, 2012, 4:50:43 PM3/25/12
to joomla-de...@googlegroups.com
According to:  http://build.joomla.org/job/platform/670/testReport/?
There are 2295 tests, 0 failures, and 0 skipped 

There are  3233 tests 0 failures, 659 incomplete, and 279 skipped

So which number is currently correct?

brian teeman

unread,
Mar 25, 2012, 5:13:44 PM3/25/12
to joomla-de...@googlegroups.com

 

3233 - 659 - 279 = 2295

garyamort

unread,
Mar 25, 2012, 5:59:39 PM3/25/12
to joomla-de...@googlegroups.com


On Sunday, March 25, 2012 5:13:44 PM UTC-4, brian teeman wrote:

 

3233 - 659 - 279 = 2295



So shouldn't the results be:
2576 tests, 0 failures, and 279 skipped ?

Or are there different definitions for skipped? 

Amy Stephen

unread,
Mar 25, 2012, 9:13:45 PM3/25/12
to joomla-de...@googlegroups.com
Gary -

I get the 2nd set of 3233 tests when I run unittests locally on current framework.

I think they are working on getting jools running again - the two sets of values might be related to that. (?)

Time: 20 seconds, Memory: 153.00Mb

OK, but incomplete or skipped tests!
Tests: 3233, Assertions: 4894, Incomplete: 650, Skipped: 307.
Amy-Stephen:joomla-platform amystephen$

Ian

unread,
Mar 27, 2012, 9:37:13 AM3/27/12
to joomla-de...@googlegroups.com
Yeah, it is the difference between the tally report in PHPUnit and the way that the xUnit parser in Jenkins parses the report.  So basically two different organizations (Sebastian Bergmann and PHPUnit vs whoever wrote the xUnit parser from the Jenkins crew).

Ian
Reply all
Reply to author
Forward
0 new messages