8388277: [REDO] Looked-up color fails for -fx-background-color in JavaFX CSS file by Maran23 · Pull Request #2225 · openjdk/jfx

Maran23 · 2026-07-27T20:57:49Z

Another much better try to fix the issue.
I recommend to read: #2201 first. All tests from there are included.
I added some new ones that succeed before and after, a first step for more CSS tests as discussed in: #2218 (comment)

My new idea is now the following constraint, which I think is also a much better approach:

A CssStyleHelper always has a correct firstStyleableAncestor. We can at any time trust and rely on it.
We will also reuse the existing loop for the isUserSetFont check to improve the performance a bit

Implementation:

A new private CssHelperState is introduced on Node. We need to know whether we can trust the styleHelper.
Not every Node has a styleHelper - it is only created when needed, so we can not attach the flag in there

This fixes the issue while a deep (optionally unstyled) scene graph has no performance penality.
The approach is similar than my previous PR, but more smart. And with the set constraint mentioned above.

I do think we can improve the CssStyleHelper more. But for another day.
Maybe at one point, with more tests and when all requirements are clear, we can find a way without CssHelperState and without creating an empty CssStyleHelper just to hold trigger states (because of that, we need to check styleHelper.cacheContainer != null a lot of times).

/issue add JDK-8187955

I confirm that I make this contribution in accordance with the OpenJDK Interim AI Policy.

Progress

Change must not contain extraneous whitespace
Commit message must refer to an issue
Change must be properly reviewed (2 reviews required, with at least 1 Reviewer, 1 Author)

Issues

JDK-8388277: [REDO] Looked-up color fails for -fx-background-color in JavaFX CSS file (Bug - P4)
JDK-8187955: CssStyleHelper : recursion in isUserSetFont() can be avoided (Bug - P4)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jfx.git pull/2225/head:pull/2225
$ git checkout pull/2225

Update a local copy of the PR:
$ git checkout pull/2225
$ git pull https://git.openjdk.org/jfx.git pull/2225/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 2225

View PR using the GUI difftool:
$ git pr show -t 2225

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jfx/pull/2225.diff

Using Webrev

Link to Webrev Comment

…aFX CSS file

bridgekeeper · 2026-07-27T20:58:40Z

👋 Welcome back mhanl! A progress list of the required criteria for merging this PR into master will be added to the body of your pull request. There are additional pull request commands available for use with this pull request.

openjdk · 2026-07-27T20:59:34Z

❗ This change is not yet ready to be integrated.
See the Progress checklist in the description for automated requirements.

openjdk · 2026-07-27T21:00:06Z

@Maran23
Adding additional issue to issue list: 8187955: CssStyleHelper : recursion in isUserSetFont() can be avoided.

openjdk · 2026-07-27T21:00:08Z

The total number of required reviews for this PR has been set to 2 based on the presence of this label: rfr. This can be overridden with the /reviewers command.

mlbridge · 2026-07-27T21:03:39Z

Webrevs

01: Full (b69bae08)
00: Full (1f36abaa)

andy-goryachev-oracle

please resolve the merge conflict.

…REDO]-Looked-up-color-fails-for--fx-background-color-in-JavaFX-CSS-file # Conflicts: # modules/javafx.graphics/src/test/java/test/javafx/scene/CssStyleHelperTest.java

subhramit · 2026-08-02T11:57:25Z

Our project JabRef is currently dependent on this fix for a critical performance bug report.
Would really appreciate if it could be expedited.

kevinrushforth · 2026-08-04T15:35:22Z

Our project JabRef is currently dependent on this fix for a critical performance bug report. Would really appreciate if it could be expedited.

GIven that the earlier version caused a regression, and that this touches an area of the code that is particularly tricky, it will need sufficient time to review and test it. It will then go into jfx28 and is unlikely to be backported, given the risk, at least not right away.

andy-goryachev-oracle · 2026-08-04T15:03:04Z

+     */
+    private static boolean isStyleableAncestor(Node parentNode) {
+        if (parentNode.cssHelperState == Node.CssHelperState.STALE) {
+            parentNode.cssHelperState = Node.CssHelperState.RESOLVED_EARLY;


there might be a problem here: this code sets RESOLVED_EARLY before creating the styleHelper. createStyleHelper() can invoke application listeners that might, for example, add new nodes and force applyCss() immediately. The re-entrance descendants see RESOLVED_EARLY and trust the old, stale, styleHelper.

would it make sense to add another state, or perhaps move the state into styleHelper ?

Moving to styleHelper was my very first idea but many nodes will have not have one (mostly layout containers). So it is not possible to do that. I can change the order of the statements though.

A layout container may have no style helper but because a parent style changed and now affects it, it needs one (and we need to set the state accordingly).

In the future, I would really like to revisit this. I hope this can be simplified, but there are more changes needed. Especially the creation of an empty CssStyleHelper just to hold parent states is weird. Maybe, on the way we will find an even better way, when other things are refactored/improved first.

andy-goryachev-oracle · 2026-08-04T15:26:56Z

        Styleable parent = node;
        int depth = 0;
-        while(parent != null) {
+        while (parent != null) {


there might be potential quadratic execution time here.

imagine a long chain of nodes

root + node1 + node2 ... + nodeN

start by adding a child to nodeN, this makes the entire chain dirty.

change style class in all the nodes in the chain

add a second child to nodeN

adding the second child recursively rebuilds every stale ancestor, walking all the way to the root.

this will only rebuild the style helper that needs it. And only the first ancestor. I can't see how this could be a problem - do you have a unit test in mind?
I tested several scenarios and could not spot any problem. Note that this is a very rare case that usually only happens for the scenarios I implemented as tests

ok, so here is the test that passes in master and fails spectacularly with this PR:

@Test void checkQuadraticPerformace() { scene.getStylesheets().add(toDataURL( """ .old { -fx-padding: 1.0; } .new { -fx-padding: 99.0; } """)); AtomicInteger counter = new AtomicInteger(); class TPane extends Pane { public TPane(String style) { getStyleClass().add("style"); } @Override public Styleable getStyleableParent() { counter.incrementAndGet(); return super.getStyleableParent(); } } int number = 16; ArrayList<TPane> chain = new ArrayList<>(); TPane top = new TPane("old"); chain.add(top); TPane p = top; for (int i = 1; i < number; i++) { TPane ch = new TPane("old"); p.getChildren().add(ch); chain.add(ch); p = ch; } scene.setRoot(top); top.applyCss(); counter.set(0); // mark the chain dirty p.getChildren().add(new Pane()); int baseline = counter.get(); for (TPane pane : chain) { pane.getStyleClass().setAll("new"); } counter.set(0); // should not result in quadratic performance p.getChildren().add(new Pane()); int newCount = counter.get(); assertTrue(newCount <= baseline * 4, () -> { return MessageFormat.format("Baseline={0}, observed={1}", baseline, newCount); }); }

andy-goryachev-oracle · 2026-08-04T15:56:24Z

+        }
+
+        if (node.styleHelper != null) {
+            setFirstStyleableAncestor(node.styleHelper, styleableAncestor);


this changes the existing helper's ancestor unconditionally. the helper might contain stale styles (especially during style or hierarchy updates from within the listeners)

not sure I follow: This just sets the ancestor, the same way we did before

andy-goryachev-oracle · 2026-08-10T19:24:41Z

+        }
+        return chain;
+    }
+


toDataUrl: similar to #2236 (comment)

perhaps we could create a Utils class with this method. no need to replicate utility code all over the place.

andy-goryachev-oracle · 2026-08-10T20:01:24Z

        Styleable parent = node;
        int depth = 0;
-        while(parent != null) {
+        while (parent != null) {


ok, so here is the test that passes in master and fails spectacularly with this PR:

@Test void checkQuadraticPerformace() { scene.getStylesheets().add(toDataURL( """ .old { -fx-padding: 1.0; } .new { -fx-padding: 99.0; } """)); AtomicInteger counter = new AtomicInteger(); class TPane extends Pane { public TPane(String style) { getStyleClass().add("style"); } @Override public Styleable getStyleableParent() { counter.incrementAndGet(); return super.getStyleableParent(); } } int number = 16; ArrayList<TPane> chain = new ArrayList<>(); TPane top = new TPane("old"); chain.add(top); TPane p = top; for (int i = 1; i < number; i++) { TPane ch = new TPane("old"); p.getChildren().add(ch); chain.add(ch); p = ch; } scene.setRoot(top); top.applyCss(); counter.set(0); // mark the chain dirty p.getChildren().add(new Pane()); int baseline = counter.get(); for (TPane pane : chain) { pane.getStyleClass().setAll("new"); } counter.set(0); // should not result in quadratic performance p.getChildren().add(new Pane()); int newCount = counter.get(); assertTrue(newCount <= baseline * 4, () -> { return MessageFormat.format("Baseline={0}, observed={1}", baseline, newCount); }); }

8388277: [REDO] Looked-up color fails for -fx-background-color in Jav…

1f36aba

…aFX CSS file

openjdk Bot added the rfr Ready for review label Jul 27, 2026

Maran23 mentioned this pull request Jul 28, 2026

8388378: [Regression] FX CSS: styled property is reset to its initial value on a pseudo-class-only transition after tree items change (TreeTableRow -fx-indent collapses) #2218

Closed

4 tasks

andy-goryachev-oracle reviewed Jul 30, 2026

View reviewed changes

Merge branch 'master' of https://github.com/openjdk/jfx into 8388277-…

b69bae0

…REDO]-Looked-up-color-fails-for--fx-background-color-in-JavaFX-CSS-file # Conflicts: # modules/javafx.graphics/src/test/java/test/javafx/scene/CssStyleHelperTest.java

subhramit mentioned this pull request Aug 2, 2026

[Experimental] Defer ControlsFX validation decoration to avoid entry editor freeze JabRef/jabref#16466

Closed

3 tasks

calixtus mentioned this pull request Aug 3, 2026

Improve entry editor first-open performance JabRef/jabref#16467

Merged

14 tasks

andy-goryachev-oracle reviewed Aug 4, 2026

View reviewed changes

andy-goryachev-oracle reviewed Aug 10, 2026

View reviewed changes

Uh oh!

Conversation

Maran23 commented Jul 27, 2026 • edited by openjdk Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Progress

Issues

Reviewing

Uh oh!

bridgekeeper Bot commented Jul 27, 2026

Uh oh!

openjdk Bot commented Jul 27, 2026

Uh oh!

openjdk Bot commented Jul 27, 2026

Uh oh!

openjdk Bot commented Jul 27, 2026

Uh oh!

mlbridge Bot commented Jul 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Webrevs

Uh oh!

andy-goryachev-oracle left a comment

Choose a reason for hiding this comment

Uh oh!

subhramit commented Aug 2, 2026 • edited by bridgekeeper Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

kevinrushforth commented Aug 4, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Milestone

Development

Uh oh!

4 participants

Maran23 commented Jul 27, 2026 •

edited by openjdk Bot

Loading

mlbridge Bot commented Jul 27, 2026 •

edited

Loading

subhramit commented Aug 2, 2026 •

edited by bridgekeeper Bot

Loading