RFR: 8233177: Remove testcase for JDK-8001470 as fix has been reverted
Prasanta Sadhukhan
psadhukhan at openjdk.org
Wed Feb 14 09:12:06 UTC 2024
On Tue, 13 Feb 2024 15:35:13 GMT, Alexey Ivanov <aivanov at openjdk.org> wrote:
> For me, it fails on Windows because no suitable font is found;
It is not because of fonts availability...It's because of the order of `pack `and `setVisible `being called which is causing some issue with `i18nFieldView `layouting. If the `pack` is called **before** `setVisible` then it works in windows and other platforms, so one can see it's NOT because of font availability
Now, in one of the test-sprint, it was mentioned that _setLocationRelativeTo followed by pack cause some issues_ so it is recommended to call `pack `before `setLocationRelativeTo`
https://github.com/openjdk/jdk/pull/13633#discussion_r1178208629
so in similar vein, I guess this can also argued that setVisible followed by pack can cause some issue and be recommended that `pack` should be called before `setVisible`
> And this test demonstrates it. The test is not to be removed.
The fix and the number of regressions it caused (as mentioned in my previous comment) demonstrated that the understanding was not correct so why should we impose the test on the fixer of JDK-8001470..Let him/her devise a new testcase for the fix he/she proposes (if we still think pack can be called after setvisible but it can be argued why different thought process for order of setLocationRelativeTo/pack and setVisible/pack)
-------------
PR Comment: https://git.openjdk.org/jdk/pull/17528#issuecomment-1943345035
More information about the client-libs-dev
mailing list