This Bugzilla instance is deprecated, and most Eclipse projects now use GitHub or Eclipse GitLab. Please see the deprecation plan for details.
Bug 213100 - automate publishing of failed tests
Summary: automate publishing of failed tests
Status: RESOLVED FIXED
Alias: None
Product: z_Archived
Classification: Eclipse Foundation
Component: Mylyn (show other bugs)
Version: dev   Edit
Hardware: PC Windows XP
: P3 enhancement (vote)
Target Milestone: ---   Edit
Assignee: Steffen Pingel CLA
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks: 116487
  Show dependency tree
 
Reported: 2007-12-15 22:27 EST by Steffen Pingel CLA
Modified: 2007-12-20 02:35 EST (History)
0 users

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Steffen Pingel CLA 2007-12-15 22:27:35 EST
Implement infrastructure to send out notifications about failed JUnit tests.
Comment 1 Steffen Pingel CLA 2007-12-16 04:22:33 EST
I have added a headless application for publishing test results to the sandbox in the org.eclipse.mylyn.tests.report project. A little bit of documentation is available here:

http://wiki.eclipse.org/Mylyn/Build
Comment 2 Eugene Kuleshov CLA 2007-12-16 14:00:58 EST
Few random ideas and observations:

-- assign these tickets to the person who made commit that broke that test (unless ticket is already open)
-- consider using fqn name for the test class in the subject
-- consider assigning issues to the same components as used in project's Bugzilla (maybe even use project's Bugzilla instead of separate Trac instance)
-- issue type is not set
Comment 3 Eugene Kuleshov CLA 2007-12-19 23:31:27 EST
When opening or closing these reports it does make sense to leave comment which build id did that, and if build been triggered by some commit provide association with the corresponding change set.
Comment 4 Steffen Pingel CLA 2007-12-19 23:40:20 EST
Currently a comment is added denoting the build id in which a test was fixed.
Comment 5 Steffen Pingel CLA 2007-12-20 02:35:23 EST
Additional ideas:

- create a master task for each test class and make test cases child tasks
- run tests on each commit and reopen/comment on the bug report that is related to the commit
- publish test results for Eclipse 3.3 as well as for 3.4 by putting a tag in the summary, e.g. [e3.3]