On Mon, 9 Nov 2026 13:42:40 GMT, Marius Hanl <[email protected]> wrote:

> Question: Should we still keep the failign scenarios in a new disabled test? 
> We can write a small javadoc comment that those might be behavior we may aim 
> to have at one point -- or not. No formal specification as you wrote.
> 
> But if someone checks the code in 5 years, this may will help to document, 
> that we were indeed aware of this.

I was thinking about leaving it in, but ultimately decided to remove them to 
avoid people from being misled into thinking the behavior shown in the test can 
be achieved without
a) breaking backwards compatibility
b) breaking internal consistency with word-wise navigation (Ctrl+Left/Right)

If I were "king of JavaFX" :-) I would certainly change (at least) these two 
things to better suit _my_ needs, removing the use of BreakIterators and 
instead going for the "character class" solution I sketched.
But as I understand it, JavaFX has a much stronger focus on keeping thinks 
working the way they are now and a change in the way word-navigation works will 
certainly break someones software at some point.

-------------

PR Comment: https://git.openjdk.org/jfx/pull/2363#issuecomment-6744153760

Reply via email to