Tom Eyckmans wrote:
...
Some questions:
- does anybody use per-test forking? use-cases?
- what have you been missing regarding test execution in your build? we
would like to know!
1. My biggest desire is to be able to run the tests in a single project without
having to re-run all the tests in all dependent projects. Jira GRADLE-220 covers
this. In a big build system (many projects) it is quite painful during normal
development to have to run all tests every time when I know that I just edited
files in one project and want to test that change. As a result I end up running
my normal development tests through the IDE instead of using gradle :(
2. Ability to run a single test (or package of tests) (Jira GRADLE-133). When
you are trying to get a new feature working it is nice to re-run a single test
without waiting for all tests in a build to execute.
3. Simple console output of test results when a test fails. When I have a few
tests failing, I would like to see the results in the console instead of having
to go look at the HTML report.
4. Automatic filtering to run only valid test cases (derived from appropriate
base class or has appropriate annotations and is not abstract). The system
should be able to find the appropriate tests without using a naming pattern.
5. Some type of summary results for a multi-project build. Currently the test
results are scattered in each project.
6. Some type of test group for forking would be nice (ex. fork every 100 tests).
See comments in Jira GRADLE-448. This would help with memory usage (permgen
space).
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
