On Mon, 22 Feb 2021 19:17:29 GMT, Sergey Bylokhov <s...@openjdk.org> wrote:
> > > > The test has to ensure the string gets clipped so that ‘…’ replaces the > > part that does not fit in the size provided, and the result of clipping > > must be the same. This should not depend on the uiScale. > > The test tries to set the bounds of the window to the "special" size which > trigger the clipping of only the last characters, if it fails to do so then > it just check that the non-clipped text is the same for both strings. Since I cannot see the clipping occurs with uiScale=1, the test has to adjust its special size, probably by measuring the text before setting the size. > If 200% caused the test to fail because the wrong part of the text is clipped > mean that this is not a test bug. And the change in the test should include > using more scale/text variants to increase coverage, not to set it to the > uiScale="1". That's exactly my point. The current behaviour looks like a product bug. So there's more to it than just setting uiScale=1, which makes the test pass but then the test does not clip the string at all as we discussed above. ------------- PR: https://git.openjdk.java.net/jdk/pull/2502
