Thanks for the archaeology.
It means that on Windows this probably was caused by freetype being used
and in b17 there was a specific fix for rotation
https://bugs.openjdk.java.net/browse/JDK-8203485
-phil.
On 4/26/20, 9:23 PM, Prasanta Sadhukhan wrote:
It seems in my windows 10 system, the test started failing on
jdk11-b05 and started passing from jdk11-b17.
+1
Regards
Prasanta
On 27-Apr-20 5:29 AM, Philip Race wrote:
Bug: https://bugs.openjdk.java.net/browse/JDK-8197797
webrev : http://cr.openjdk.java.net/~prr/8197797/
I can't reproduce this anywhere despite trying hard.
The closest is I saw 8u221 on mac 10.13.6 show *slight* differences in
alignment but not nearly enough for the test to fail.
More details of testing in the bug report. No need to repeat them here.
So I'd like to de-problem list it.
If someone then sees it fail we can get exact details of their config.
Or if anyone can reproduce it now the same applies.
Otherwise, right now there's nothing to be done.
-phil.
