ICEfaces
  1. ICEfaces
  2. ICE-6143

SettableLocaleViewRoot optimization

    Details

    • Type: Bug Bug
    • Status: Open
    • Priority: Major Major
    • Resolution: Unresolved
    • Affects Version/s: 1.8.2-EE-GA_P01
    • Fix Version/s: None
    • Component/s: Framework
    • Labels:
      None
    • Environment:
      ICEfaces

      Description

      SettableLocaleViewRoot is implemented by inspecting the stack to accept the locale only from certain callers:

          public void setLocale(Locale locale) {
              //ignore locale set by RestoreViewPhase since it is using the first locale in the Accept-Language list,
              //instead it should calculate the locale
              StackTraceElement[] ste = (new RuntimeException()).getStackTrace();

      This approach is performance intensive and should be revised.

        Activity

        Hide
        Deryk Sinotte added a comment -

        Perhaps there is a way to use a PhaseListener here?

        Show
        Deryk Sinotte added a comment - Perhaps there is a way to use a PhaseListener here?
        Hide
        Ted Goddard added a comment -

        It's also very possible that this code is no longer necessary at all – removing the stack trace check allowed component showcase to change locales on tomcat with no problems.

        Show
        Ted Goddard added a comment - It's also very possible that this code is no longer necessary at all – removing the stack trace check allowed component showcase to change locales on tomcat with no problems.
        Hide
        Ken Fyten added a comment -

        Since we don't have a firm understanding of the regression risk associated with removing this logic, and that it would require some substantial effort to create reasonable tests to ascertain the actual risk, and also that there is no obvious way to implement the existing logic using a more performant technique, I am moving this out of P02 for time/resource constraint reasons.

        Show
        Ken Fyten added a comment - Since we don't have a firm understanding of the regression risk associated with removing this logic, and that it would require some substantial effort to create reasonable tests to ascertain the actual risk, and also that there is no obvious way to implement the existing logic using a more performant technique, I am moving this out of P02 for time/resource constraint reasons.

          People

          • Assignee:
            Unassigned
            Reporter:
            Ted Goddard
          • Votes:
            0 Vote for this issue
            Watchers:
            0 Start watching this issue

            Dates

            • Created:
              Updated: