RFR: 8326661: sun/java2d/cmm/ColorConvertOp/ColConvTest.java assumes profiles were generated by LCMS
Sergey Bylokhov
serb at openjdk.org
Tue Mar 5 03:51:54 UTC 2024
On Tue, 5 Mar 2024 00:58:42 GMT, Sergey Bylokhov <serb at openjdk.org> wrote:
>> Updated several tests to avoid potential failure with recent LCMS update and non-LCMS generated profile.
>
> test/jdk/sun/java2d/cmm/ColorConvertOp/ColConvCCMTest.java line 63:
>
>> 61: 10.5, // GRAY
>> 62: 215.0, // PYCC
>> 63: 56.0 // CIEXYZ
>
> All this change will delete all information about "old and good" values and start to use lcms thresholds for all profiles? I think it should be the opposite - use the good data for all except lcms(or any other not that good as jdk8 profiles), like we do it now.
Actually the purpose of these particular tests is validation of the quality of the built-in color profiles.
-------------
PR Review Comment: https://git.openjdk.org/jdk/pull/18097#discussion_r1512099541
More information about the client-libs-dev
mailing list