https://svn.dev.java.net/svn/hudson/branches/rc
........
r25878 | mindless | 2010-01-15 12:22:17 -0800 (Fri, 15 Jan 2010) | 4 lines
[HUDSON-5236] remove Dependency.getBuildActions API, as its computation may
overlap with shouldTriggerBuild. Instead, add List<Action> parameter in
shouldTriggerBuild which impl can add Actions to before returning boolean.
........
git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@25882 71c3de6d-444a-0410-be80-ed276b4c234a
(11:18:15 AM) rpetti: ok, it doesn't find it when doing mvn hpi:run
(11:18:59 AM) rpetti: It's probably a configuration issue on my end, but I have no idea where to look
(11:19:19 AM) kohsuke: mvn dependency:tree and stack trace of the problem?
(11:22:17 AM) rpetti: k, my maven repo must be screwed or something
(11:22:44 AM) rpetti: I'm trying to get commons-codec-1.4.jar, and dependency:tree says:
(11:23:00 AM) rpetti: commons-codec:commons-codec:jar:1.4:compile
(11:23:11 AM) rpetti: as opposed to 'provided' like the others
(11:23:51 AM) kohsuke: rpetti: maybe the dependency is colliding with that of Hudson core
(11:42:25 AM) rpetti: kohsuke: it must be loading 1.3 instead of 1.4, the particular method I'm using doesn't exist pre-1.4.
(11:43:09 AM) rpetti: kohsuke: is there any way to get around that, or am I limited to whatever core is using?
(11:48:28 AM) kohsuke: rpetti: unfortunately no, but we'd like to do something about that
(11:49:49 AM) kohsuke: I'll bump up the core to use 1.4 for now
(11:49:53 AM) kohsuke: that should fix this particular case
git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@25879 71c3de6d-444a-0410-be80-ed276b4c234a
overlap with shouldTriggerBuild. Instead, add List<Action> parameter in
shouldTriggerBuild which impl can add Actions to before returning boolean.
git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@25878 71c3de6d-444a-0410-be80-ed276b4c234a
https://svn.dev.java.net/svn/hudson/branches/rc
........
r25820 | mindless | 2010-01-14 08:00:16 -0800 (Thu, 14 Jan 2010) | 5 lines
[FIXED HUDSON-5236] one fix for r25751.. DependencyGraph as Comparator returns
zero for jobs not dependending on eachother.. in a TreeMap this means keys
overwrite eachother and not all builds are triggered.
Switch to sorting in a List.
........
git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@25829 71c3de6d-444a-0410-be80-ed276b4c234a
zero for jobs not dependending on eachother.. in a TreeMap this means keys
overwrite eachother and not all builds are triggered.
Switch to sorting in a List.
git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@25820 71c3de6d-444a-0410-be80-ed276b4c234a
whether builds are triggered and provide Actions for the triggered build.
(previously only tasks.BuildTrigger had hardcoded support for logic on whether or
not to trigger.. now it uses this mechanism)
git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@25751 71c3de6d-444a-0410-be80-ed276b4c234a
and apparently without having extension preloaded at the parent, the main artifact of the 'test' module
ends up installed with 'stapler-jar' extension (which normally is an indication that the ArtifactTypeHandler
defined in this extension is not getting picked up.)
To avoid this problem, I'm defining an extension here. Not sure if the nested <dependencies> is necessary.
It's also possible that I misdiagnosed the problem and the root cause is something entirely different.
To test if you can remove this work around, do a rebuild from main and see if the test harness
is installed with the right extension into the local repository.
The typical indication of this problem is:
[INFO] Installing /usr/home/jitu/src/hudson/main/test/target/hudson-test-harness-1.341-SNAPSHOT.jar to /usr/home/jitu/.m2/repository/org/jvnet/hudson/main/hudson-test-harness/1.341-SNAPSHOT/hudson-test-harness-1.341-SNAPSHOT.stapler-jar
git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@25727 71c3de6d-444a-0410-be80-ed276b4c234a