Uploaded image for project: 'Jenkins'
  1. Jenkins
  2. JENKINS-33567

Job configuration scrolls "persist" between page loads incorrectly

      I'm not sure why this "feature" exists, but it appears the freestyle job configuration page is persisting the scroll state in a cookie. That by itself isn't the problem, but if you navigate away from the page after having scrolled to a tab, and then come back, the configuration page scrolls to the incorrect location.

          [JENKINS-33567] Job configuration scrolls "persist" between page loads incorrectly

          R. Tyler Croy added a comment -

          I should note that the persistence of scroll position is applied across every job configuration page. In effect if you scroll to "Post-build Actions" on JobA and then navigate to the configuration of JobB, the page will scroll incorrectly

          R. Tyler Croy added a comment - I should note that the persistence of scroll position is applied across every job configuration page. In effect if you scroll to "Post-build Actions" on JobA and then navigate to the configuration of JobB, the page will scroll incorrectly

          Daniel Beck added a comment -

          This made sense when it was real tabs (Jenkins remembers where you were!), but needs to be removed now.

          Daniel Beck added a comment - This made sense when it was real tabs (Jenkins remembers where you were!), but needs to be removed now.

          Gus is planning to remove this PR-2117

          Spike Washburn added a comment - Gus is planning to remove this PR-2117

          gus reiber added a comment -

          I kind of like the tab memory, even in the scroll situation. The fact that the tabs didn't show up is a real bug, though. I am submitting a fix.

          gus reiber added a comment - I kind of like the tab memory, even in the scroll situation. The fact that the tabs didn't show up is a real bug, though. I am submitting a fix.

          Daniel Beck added a comment -

          gusreiber The problem is that memory does not seem to behave in any way predictable. With shorter sections, you can never really know which section you're in. And the scrolling in the case of a short config form is also weird, as it doesn't even reach your previous section (too many sections on the screen). I mean, I knew of this feature in tabs, it was obvious it's still in there in scrolling, but I had no idea WTF it's doing. Imagine not knowing about it.

          Daniel Beck added a comment - gusreiber The problem is that memory does not seem to behave in any way predictable. With shorter sections, you can never really know which section you're in. And the scrolling in the case of a short config form is also weird, as it doesn't even reach your previous section (too many sections on the screen). I mean, I knew of this feature in tabs, it was obvious it's still in there in scrolling, but I had no idea WTF it's doing. Imagine not knowing about it.

          gus reiber added a comment -

          I thought I had fixed this, but will double check to see if it has been re-introduced.

          gus reiber added a comment - I thought I had fixed this, but will double check to see if it has been re-introduced.

          Daniel Beck added a comment -

          gusreiber Haven't seen this is a while, so if the scroll is gone, that would be great and you can just resolve this.

          Daniel Beck added a comment - gusreiber Haven't seen this is a while, so if the scroll is gone, that would be great and you can just resolve this.

          gus reiber added a comment -

          danielbeck In checking this guy out I was able to catch the tab bar not pegged at the top of the screen doing the sort of back-n-forth testing you are describing. The issue is a similar root cause, scroll position faked out by browser's own state memory. So I have offered this PR: https://github.com/jenkinsci/jenkins/pull/2219

          gus reiber added a comment - danielbeck In checking this guy out I was able to catch the tab bar not pegged at the top of the screen doing the sort of back-n-forth testing you are describing. The issue is a similar root cause, scroll position faked out by browser's own state memory. So I have offered this PR: https://github.com/jenkinsci/jenkins/pull/2219

          gus reiber added a comment -

          closing the issue, since Daniel isn't seeing it. I think I only saw a similar issue because Chrome was tripping over the debugger. Will check again when we open back up for 2.1

          gus reiber added a comment - closing the issue, since Daniel isn't seeing it. I think I only saw a similar issue because Chrome was tripping over the debugger. Will check again when we open back up for 2.1

            gusreiber gus reiber
            rtyler R. Tyler Croy
            Votes:
            0 Vote for this issue
            Watchers:
            3 Start watching this issue

              Created:
              Updated:
              Resolved: