On Tue, 17 Feb 2026 10:45:27 GMT, Daishi Tabata <[email protected]> wrote:
>> [JDK-8348760](https://bugs.openjdk.org/browse/JDK-8348760) fixed an issue in >> Windows L&F JMenuItem layout whereby radio bullet/checkmark was rendered in >> different columnspace than menuitem imageicon so radiobullet/checkmark is >> rendered in first column and imageicon is rendered in 2nd column but this >> rendering of imageicon in 2nd columnspace was done invariably for all >> JMenuItem irrespective of if it is JRadioButtonMenuItem or JCheckBoxMenuItem >> or JMenuItem, which is wrong. >> >> Normal JMenuItem (which are not JRadioButtonMenuItem or JCheckBoxMenuItem) >> imageicon rendering should be done in first columnspace as was done before >> JDK-8348760 fix because there is no radiobullet/checkmark to render for >> those menuitems so no need to reserve columnspace for those bullet/checkmark >> icon >> >> Before fix >> >> <img width="205" height="127" alt="image" >> src="https://github.com/user-attachments/assets/13a1e352-5e8d-4251-b7a7-032935eab74e"; >> /> >> >> >> After fix >> >> <img width="195" height="131" alt="image" >> src="https://github.com/user-attachments/assets/84ec3ee6-2823-4bf7-840d-b53f2e9d44c3"; >> /> > > I don't have review permissions, but I'd like to comment on something that > caught my eye. > > The following image is a screenshot of the test > `test/jdk/javax/swing/JMenuItem/TestRadioAndCheckMenuItemWithIcon.java` that > was added by [JDK-8348760](https://bugs.openjdk.org/browse/JDK-8348760). > > Before > <img width="250" height="248" alt="before" > src="https://github.com/user-attachments/assets/51c44938-bf61-44f0-a6fe-f5cf6b05a2c3"; > /> > > After > <img width="248" height="247" alt="after" > src="https://github.com/user-attachments/assets/b5d01603-3bc7-4f4b-a538-20fcb13e74d9"; > /> > > Before this change, the left edges of all the red squares are aligned, and > the text of each menu item is also left-aligned. > However, after this change, for menu items without radio buttons or > checkboxes, the red squares and the text are misaligned compared to the > others. > Is this an expected behavioral change introduced by this fix? @tabata-d If you have an OpenJDK id, you can still review, and your review will be recorded in `Reviewed-by` section. Yet your review won't count towards “Change must be properly reviewed”. ------------- PR Comment: https://git.openjdk.org/jdk/pull/29730#issuecomment-3914202063
