On Wed, 3 Aug 2022 15:11:31 GMT, Kevin Rushforth <k...@openjdk.org> wrote:
>> Thanks for all the feedback. I've updated as per your suggestions >> @jperedadnr and @kevinrushforth. Let me know how this looks to you - I'm >> learning a lot here :-) >> >> I ran the tests as per your example and got `802 tests completes, 28 failed, >> 108 skipped`, which is exactly the same as when I run them on the `master` >> branch. Is there something I should be doing on a Mac to get zero failures? >> (Although, it's a lot better than my Linux box which gets 200 failed!) > > @gargoyle To clarify my inline comments, the minimal change you need to make > is: > > 1. Change the tests of the four `jField` variables to remove the `> 0` (so it > becomes a boolean test for non-null). > 2. Change the tests of `width` and `height` to be `> 0` (rather than `> 1`) > > My other suggestions are optional. Thanks @kevinrushforth. Thinking about it, is there merit in changing the `scalex` and `scaley` checks to be `!= 0`, as The main reason for checking at all is to prevent a division by zero. Although unlikely, could fractional values < 1 reach this code? ------------- PR: https://git.openjdk.org/jfx/pull/743
