...ied by threadCount
This is optimization of thread counts limited to the real number of parallel JUnit Runners.
SUREFIRE  Surefire does not use the total amount of threads spe…
…cified by threadCount
Yet incomplete. I will finalize in two days.
unit tests for algorithm
PC optimization refactoring (Only unit tests. Missing IT tests refact…
fixed tests for SUREFIRE
[SUREFIRE-1055] Incorrect reported test result for parallel run mode
Test case supplied by Chris Hansen
[SUREFIRE-1055] Integration Tests
IT submitted. Last thing which remains is to check old documentation, I know there are some typos.
Cool. Will you rebase and push to master on your own? Let me know if I can help you with it, or if I shall do the push again.
fixed typos and simplification of documentation
It looks like there is some kind of problem with my credentials, so pls trigger the build. Feel free to push.
BTW, I was thinking of opening an account on CloudBees for us. This means that we may get build result after we triggered the build on origin/master and our local GitHub repositories/branches.
Java Hamcrest has quite nice solution with Travis CI build process, so that every commit on a pr will trigger new build.
I've rebased the commits to the master branch and pushed them upstream. However, 18 tests in JUnit47ParallelIT fail with an error on my Win7 box. Plus, the test takes over three minutes to complete. Here's my test-result xml: https://gist.github.com/agudian/9453630
Let's see what Jenkins has to say.
Jenkins is also done now and tells me the same (plus one additional error)
My bad in pom.xml.
Now the JUnit47ParallelIT passed.
I triggered local build, but I guess should be ok.
IT speedup not possible - reverted back
Could you fetch the latest master from asf and apply your pom change there? It would be a good occasion for you to test your write-access again.
I would then start rolling the release tonight.
[SUREFIRE-1041] Do not crash with NPE in case a JUnit runner fails
Submitted by: Dan Fabulich
o Applied without changes
Merged with HEAD
I tried to "Push to..." asf master but did not succeed. :(
Before pushing, could you please rebase your Tibor17@73cb18c to master? You should then only have that one commit left to push. That's always nicer than merge-commits.
Are these commands correct?
git remote add apache https://git-wip-us.apache.org/repos/asf/maven-surefire.git
git rebase --interactive apache/master
In the interactive mode an editor is opened and I will remove the line with the latest two commitment e9252d0 and eaa4b04.
git push apache
that should do it. I just tried it myself with your branch and I did not have to change anything in the popped-up rebase file.
Watch out: if you are on your s2 branch, you should call git push apache master so that the changes land on the master branch instead of creating a new branch "s2" in the apache repository.
I always do a fast-forward merge to my local master, check the log with SmartGit (or gitk if you like) to see if it is really what I want to push upstream. Then I push from my local master to the apache master.
git push apache master
The last command git push apache master fails with error message. I am getting the same with SmartGit/Hg. I tried to use the same command SmartGit displays in the logs and used the switch "--force" but still getting the same error.
Pushing to https://git-wip-us.apache.org/repos/asf/maven-surefire.git
error: no DAV locking support on https://git-wip-us.apache.org/repos/asf/maven-surefire.git/
fatal: git-http-push failed
Are you behind a proxy or something? Very strange...
I pushed it myself now, as I wanted to get the release going.
But I'm sure there's plenty of other stuff where you can test out the commit-problems. We should continue that discussion together with Kristian via mail.
Could you close this PR for now?