<Swing Dev> RFR 8157339: Further stabilization for the SwingSet client sanity tests.

[Alexander Kouznetsov]

Shura,

Great changes.

I have several comments on them:

  * Seems like @key intermittent should be added to
    ButtonDemoScreenshotTest.
  * OptionPaneDemoTest.java:
      o Looks like the following line should be removed:

        88 final String labelText = COMPONENT_R3;

      o And index 1 is not needed in the variable name here:

        87 JDialogOperator jdo1 = new JDialogOperator(MESSAGE);


      o Shouldn't waitText be used here?

        97 if(textToType != null) {
        98 new JTextFieldOperator(jdo).typeText(textToType);
        99 }

        as it is used below here:

          191             jto.typeText(TEXT_TO_TYPE);
        192 jto.waitText(TEXT_TO_TYPE);

  * TreeDemoTest
      o Description is not correct here:

        + private void waitRowCount(JTreeOperator tree, int count) {
        + tree.waitState(new ComponentChooser() {
        + public boolean checkComponent(Component comp) {
        + return tree.getRowCount() == count;
        + }
        + public String getDescription() {
        + return "All nodes to be expanded in the tree";
        + }
        + });
        + }

  * I'd propose to introduce a lambda-compliant method to easily wait
    for any state without writing many lines of code. Something like:

          waitState("All nodes to be expanded in the tree", () ->
    tree.getRowCount() == NODES_TOTAL);

    That would allow to eliminate several methods like waitRowCount() above.

Best regards,
Alexander Kouznetsov
(408) 276-0387

On 5/19/2016 8:59 AM, Alexandre (Shura) Iline wrote:
> Hi.
>
> Please take a look on the changes in: http://cr.openjdk.java.net/~shurailine/8157339/webrev.00/
>
> I have gone through the code to discover potential instabilities. I was not changing the test logic much and mostly changed the code where I had to change it for stabilization reason. In some cases it was easier to move logic into methods to avoid copy-pasting it around.
>
> The other bits of work which I was able to identify (TODOs), do not affect test stability, as far as I can see.
>
> I am CCing Praveen Mohan and Alexander Kouznetsov who had worked on that code before.
>
> Shura.
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openjdk.java.net/pipermail/swing-dev/attachments/20160520/de91f1a5/attachment.html>