On Tue, 22 Apr 2025 01:06:11 GMT, Damon Nguyen <dngu...@openjdk.org> wrote:

>> Open-sourcing and standardizing several swing tests (JScrollPane & JRootPane)
>
> Damon Nguyen has updated the pull request incrementally with one additional 
> commit since the last revision:
> 
>   Move width and height to EDT

test/jdk/javax/swing/JRootPane/bug4403624.java line 92:

> 90: 
> 91:             r.mouseMove(p.x + OFFSET, p.y + OFFSET);
> 92:             Color c = r.getPixelColor(p.x + OFFSET, p.y + OFFSET);

That might fail intermittently on the macOS because you take a pixel color 
exactly at the place where mouse cursor is and it's a known issue on m1 based 
macs that cursor sometimes messes up the screenshots and pixel color grabbing - 
either directly (cursor is visible on screenshot) or indirectly (macOS 
calculates cursor effects such as shadow leaving traces of it in the screen 
buffer that we analyze). Whenever it is logically possible i'm trying to move 
cursor at least couple of pixels away from the area i'm going to analyze for 
color comparison.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/24758#discussion_r2054489527

Reply via email to