On Chrome (and possibly others), an inline element like span doesn't enlarge
its size to cover its content, so the 1px place holder wasn't taking effect.
To ensure the minimum height of 16px, make it an inline-block element. I'm
not entirely sure how an inline-block element with no fixed width would behave,
but with a bit of casual testing, the text inside seems to wrap OK.
For example, a new job might have a trigger setting to begin with,
or it might resolve a reference that was previously failing (for
example, 'up' job referring to 'down' job and the 'down' job is newly
created.)
Observations:
- When marked inside, moving the mouse to <A> and clicking the context menu becomes easier,
because you can move the mouse toward the hot spot area and it'll activate.
Otherwise you have to move the mouse over to <A>, then move the mouse on the side, click it,
then click the menu, which adds cognitive stress.
- Marking model link as inside creates an empty space to the right of <A>, which looks
odd if hyperlinks appear inline or if some other text follows it.
Given that, my guiding principle for this change is:
- any model link that is the last child of a block element gets marked as inside
- I've adjust a few text to get rid of the single-letter training text after a model link,
like "(#3)" to "- #3"
- If another inline element follows a model link, leave it as is.
This does create an ugly overlap.
I'm not really sure what this code is doing or what ID is supposed to look like. So this is just a minimum-intrusion change to avoid an exception.
java.lang.StringIndexOutOfBoundsException: String index out of range: -1
at java.lang.String.substring(String.java:1875)
at hudson.tasks.junit.PackageResult.findCorrespondingResult(PackageResult.java:86)
at hudson.tasks.junit.TestResult.findCorrespondingResult(TestResult.java:295)
at hudson.tasks.test.AbstractTestResultAction.findCorrespondingResult(AbstractTestResultAction.java:183)
at hudson.tasks.test.TestResult.getResultInBuild(TestResult.java:162)
at hudson.tasks.junit.History.getList(History.java:82)
at hudson.tasks.junit.History$1.createDataSet(History.java:105)
at hudson.tasks.junit.History$GraphImpl.createGraph(History.java:169)
at hudson.util.Graph.render(Graph.java:87)
at hudson.util.Graph.doPng(Graph.java:98)
Added logging more liberally, and fixed the problem.
The root cause was that when the mouse enters a hot spot, I wasn't cancelling the previous delayed hide timer.
This fix eliminates the need for the hack in the hide method where I was testing if the mouse was still over the hotspot.
but perhaps artifacts for modulebuilds should be deleted just as logs
are deleted. see https://issues.jenkins-ci.org/browse/JENKINS-17508
I did not fully test it yet, all unit tests ran fine, but didn't succeed
to run jobs inside Jenkins, due to classloader issues while running
Jenkins in Jetty, to verify deletion