On Mon, 5 Sep 2022 07:08:13 GMT, Prasanta Sadhukhan <psadhuk...@openjdk.org> wrote:
>> For me, test is not failing without the fix. >> Also, if you test entire width and fail if it is not red, it might encounter >> black pixel of text font too and might fail, so I think it's better to check >> width/2 to width -1 and not entire width. > > Also, it's might be better to write the image in case of failure so that it > gives an idea why it failed. > For me, test is not failing without the fix. Also, if you test entire width > and fail if it is not red, it might encounter black pixel of text font too > and might fail, so I think it's better to check width/2 to width -1 and not > entire width. Test was not failing without the fix because the second argument of getRGB method is height/2, and for that pixel value the returned color is red. Changing the argument from height/2 to height/4 will fail the test without the fix otherwise it will pass the test. Also writing the image in case of failure. ------------- PR: https://git.openjdk.org/jdk/pull/10112
