https://svn.dev.java.net/svn/hudson/branches/rc
........
r27441 | kohsuke | 2010-02-12 16:47:05 -0800 (Fri, 12 Feb 2010) | 1 line
[maven-release-plugin] prepare release hudson-1_346
........
r27443 | kohsuke | 2010-02-12 16:47:17 -0800 (Fri, 12 Feb 2010) | 1 line
[maven-release-plugin] prepare for next development iteration
........
r27447 | kohsuke | 2010-02-12 17:28:50 -0800 (Fri, 12 Feb 2010) | 1 line
updated changelog as a part of the release
........
git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@27449 71c3de6d-444a-0410-be80-ed276b4c234a
also set text/plain content type on ajax summary response so browser doesn't try
to parse response (firefox 3.6 showed error in console with text/html).
git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@27444 71c3de6d-444a-0410-be80-ed276b4c234a
https://svn.dev.java.net/svn/hudson/branches/rc
........
r27147 | kohsuke | 2010-02-08 11:10:47 -0800 (Mon, 08 Feb 2010) | 1 line
[FIXED HUDSON-5536] in 1.345 and added a test case.
........
r27149 | kohsuke | 2010-02-08 12:01:18 -0800 (Mon, 08 Feb 2010) | 1 line
added a backward compatibility measure to expose variables in both forms
........
r27150 | kohsuke | 2010-02-08 12:02:19 -0800 (Mon, 08 Feb 2010) | 1 line
forgot to commit this(?)
........
r27151 | kohsuke | 2010-02-08 12:24:51 -0800 (Mon, 08 Feb 2010) | 1 line
[maven-release-plugin] prepare release hudson-1_345
........
r27153 | kohsuke | 2010-02-08 12:25:04 -0800 (Mon, 08 Feb 2010) | 1 line
[maven-release-plugin] prepare for next development iteration
........
git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@27154 71c3de6d-444a-0410-be80-ed276b4c234a
try to parse response (resulting in "no element found" in firefox error console).
Not the fix for 5536 itself, but nice to avoid js error after successful update.
git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@27145 71c3de6d-444a-0410-be80-ed276b4c234a
Link to top level /people/ for AllView, otherwise to people for this view.
Hide link only for non-default, non-All view that would list no people.
Add heading on People page and a link to top level /people/ when appropriate.
git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@27129 71c3de6d-444a-0410-be80-ed276b4c234a
See:
- SCM.compareRemoteRevisionWith
- SCM.calcRevisionsFromBuild
- SCMRevisionState
for details. This feature was originally developed in a branch,
whose merge records are as follows:
------------------
Merged revisions 24271,27060,27065,27083-27084 via svnmerge from
https://www.dev.java.net/svn/hudson/branches/HUDSON-2180/main
........
r24271 | kohsuke | 2009-11-30 19:41:30 -0800 (Mon, 30 Nov 2009) | 1 line
committing my work in progress
........
r27060 | kohsuke | 2010-02-05 18:11:05 -0800 (Fri, 05 Feb 2010) | 7 lines
Made SVNRevisionState non-comparable, since doing so and distinguishing
significant/insignificant changes is rather involving work.
So that's why I originally opted for the compareRemoteRevisionWith method approach
of telling SCM upfront about what it's comparing the state with.
........
r27065 | kohsuke | 2010-02-06 08:49:50 -0800 (Sat, 06 Feb 2010) | 1 line
crucial bug fix
........
r27083 | kohsuke | 2010-02-06 10:37:46 -0800 (Sat, 06 Feb 2010) | 1 line
doc improvement
........
r27084 | kohsuke | 2010-02-06 10:39:50 -0800 (Sat, 06 Feb 2010) | 1 line
doc improvement
........
git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@27103 71c3de6d-444a-0410-be80-ed276b4c234a
https://svn.dev.java.net/svn/hudson/branches/rc
........
r27047 | kohsuke | 2010-02-05 14:07:35 -0800 (Fri, 05 Feb 2010) | 1 line
[maven-release-plugin] prepare release hudson-1_344
........
r27049 | kohsuke | 2010-02-05 14:07:55 -0800 (Fri, 05 Feb 2010) | 1 line
[maven-release-plugin] prepare for next development iteration
........
r27053 | kohsuke | 2010-02-05 17:05:11 -0800 (Fri, 05 Feb 2010) | 1 line
updated changelog as a part of the release
........
git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@27059 71c3de6d-444a-0410-be80-ed276b4c234a
While investigating the sudden memory usage spike, I discovered that XMPP/MXParser doesn't have a capability of returning a long PCDATA as a series of events. Instead, it tries to buffer the whole thing and return it as a single string.
This obviously results in a horrible performance problem if the submission is large, as was the case.
Switching to StAX to prevent this problem. StAX API isn't a part of Java5, but we already bundle woodstox.
git-svn-id: https://hudson.dev.java.net/svn/hudson/trunk/hudson/main@27042 71c3de6d-444a-0410-be80-ed276b4c234a
abayer
godin
wyukawa
kohsuke
mindless
sogabe