<Swing Dev> [9] Review Request: 8015085 [macosx] Label shortening via " ... " broken when String contains combining diaeresis

[Sergey Bylokhov]

Hi, Alexander.
You are right!
http://cr.openjdk.java.net/~serb/8015085/webrev.02

On 02.02.2015 17:10, Alexander Zvegintsev wrote:
> Hi Sergey,
>
> The fix looks good to me too, but the test using a valid third party 
> domain[1].
> I suggest to replace it with something else.
>
> [1] https://whoisology.com/archive_8/aaaaaaaaaaaaaaaaaaaaaa.com
>
> Thanks,
>
> Alexander.
>
> On 01/29/2015 04:08 PM, Alexander Scherbatiy wrote:
>>
>>   The fix looks good to me.
>>
>>   Thanks,
>>   Alexandr.
>>
>> On 1/20/2015 8:16 PM, Sergey Bylokhov wrote:
>>> Still looking for volunteers =(
>>>
>>> On 12.01.2015 18:46, Sergey Bylokhov wrote:
>>>> Hello.
>>>> Please review a fix for jdk 9.
>>>> Description:
>>>> The problem is that we consider the string which contains a 
>>>> combining diaeresis as string which needs TextLayout. In such case 
>>>> SwingUtilities2.clipString will use LineBreakMeasurer to calculate 
>>>> the characters inside some advance. But by default 
>>>> LineBreakMeasurer will use LineSeparator instead of CharSeparator, 
>>>> and this causes a difference against a normal case where we use per 
>>>> character calculation(SwingUtilities2 lines 453 - 461).
>>>> Note that in the test I paint two strings one with combining 
>>>> diaeresis/one without, and expect that they will be rendered in the 
>>>> same way.
>>>>
>>>> Bug: https://bugs.openjdk.java.net/browse/JDK-8015085
>>>> Webrev can be found at: 
>>>> http://cr.openjdk.java.net/~serb/8015085/webrev.01
>>>>
>>>
>>>
>>
>


-- 
Best regards, Sergey.