On Thu, 23 Sep 2021 18:41:38 GMT, Alexey Ivanov <[email protected]> wrote:

> The test is being moved from the closed to open. It's nearly a complete 
> re-write of the original test.
> 
> I made the test headless by painting into image.
> 
> To detect failure, the test uses the position of the caret at the end of 
> first line of the paragraph where the wrapping occurs.
> 
> I made sure that the updated test reproduces the original problem in 
> [JDK-6318524](https://bugs.openjdk.java.net/browse/JDK-6318524). (There are 
> some tweaks required to compile it with JDK 6.) The updated test fails when 
> run with 6-b52. In fact, it also fails with 6-b53 up to 6-b76 where 
> [JDK-6364882](https://bugs.openjdk.java.net/browse/JDK-6364882) is fixed.
> 
> The behaviour in 6-b76 coincides with that in the latest JDK.
> 
> There are three parameters provided for debugging purposes:
> • `-show`: Show frame for visual inspection
> • `-save`: Save *the start image* after the first paragraph is justified, and 
> *the last image* before it's checked that the first paragraph remains 
> justified
> • `-saveAll`: Save images for all the intermediate steps
> 
> [1] [JDK-6318524](https://bugs.openjdk.java.net/browse/JDK-6318524): 
> Regression: views in paragraph are not children of Paragraph.Row is some cases
> [2] [JDK-6364882](https://bugs.openjdk.java.net/browse/JDK-6364882): Last 
> line of a paragraph should not be justified

Probably it will be better to use descriptive name for the test rather than 
bugid. Probably it was done to maintain consistency with closed test but since 
it is supposedly a rewrite and we have done away with bugid in testname, a 
descriptive name will be better. Also @bug tag will let us know for which it 
was written..

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

PR: https://git.openjdk.java.net/jdk/pull/5660

Reply via email to