On Tue, 30 Jun 2026 16:20:21 GMT, Marius Hanl <[email protected]> wrote:
>> It's a bit odd it was overridden in the first place to do something >> different from what most others `List`s do (although it does look a bit more >> like what you see in CSS that way, but still without the leading dots). >> >> The format is not documented (you'd have to document it on `ObservableList` >> or on the `getStyleClass` method) so we would be allowed to make this >> change. Nobody should be relying on `toString` output, of essentially an >> `ObservableList`, being in a specific format. > > I agree with @hjohn here hence the changes I made. > There is no change for the `CssParser`. > If applications rely on the unspecified and not documented `toString`, then > yes, they may break. But that would be bad decision from the start - why > would you do that instead of calling `contains`? Well, I basically agree with you both, but I still want to get an answer 1. what is the rationale for this change? was there something that the old format couldn't represent correctly? 2. I did not ask whether the `CssParser` was changed, but whether it relied in any way on the old format ------------- PR Review Comment: https://git.openjdk.org/jfx/pull/2191#discussion_r3500318014
