On Mon, 21 Oct 2024 18:16:01 GMT, Kevin Rushforth <k...@openjdk.org> wrote:
>> Removed "not yet fully implemented" USE_MULTIPLE_NODES and related code. >> >> I would like to remove the early unfinished idea of using multiple Text >> nodes in TextAreaSkin to clean up the code, to make it easier to do fixes >> for [JDK-8342233](https://bugs.openjdk.org/browse/JDK-8342233) and >> [JDK-8296266](https://bugs.openjdk.org/browse/JDK-8296266). >> >> Also some minor cleanup. >> >> ## Summary of Changes >> >> - removed USE_MULTIPLE_NODES and code paths that correspond to its `true` >> value >> - permanently adding one Text node to `paragraphNodes` Group, keeping the >> latter for compatibility purposes >> - removed any code that scans `paragraphNodes` children >> - using `getTextNode()` in place of >> `((Text)paragraphNodes.getChildren().get(0))` > > modules/javafx.controls/src/main/java/javafx/scene/control/skin/TextAreaSkin.java > line 729: > >> 727: @Override >> 728: protected PathElement[] getUnderlineShape(int start, int end) { >> 729: return getTextNode().underlineShape(start, end); > > The former code would return `null` if `start` > > `textNode.textProperty().getValueSafe().length()`. I think this is OK, given > that the implementation of `Text::underlineShape` checks -- what do you > think? This applies to `getRangeShape`, too. I think it's ok. The javadoc for `TextInputControlSkin::underlineShape()` does not specify a null return value. In fact, the javadoc for `Text::rangeShape()` and `::underlineShape()` say `@return an array of {@code PathElement} which can be used to create a {@code Shape}` implying that it should never be `null` as the return value is typically used in `Shape::getElements().setAll()` or `.addAll()` which will throw an NPE if the thing is null. In fact, `Text::rangeShape()` (used for underline and range shape) always returns a non-null value. ------------- PR Review Comment: https://git.openjdk.org/jfx/pull/1601#discussion_r1809385959
