Patch which fixes bug #6703377 (font outline transformation)

[Pavel Tisnovsky]

Omair Majid wrote:
> On 05/03/2011 05:10 AM, Pavel Tisnovsky wrote:
>> Omair Majid wrote:
>>> On 10/15/2010 11:07 AM, Pavel Tisnovsky wrote:
>>>> Hi all,
>>>>
>>>> I prepared yet another patch for IcedTea6 HEAD. This patch refine
>>>> computation of x and y coordinates during font outline transformation
>>>> when negative values are used
>>>> (see http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6703377)
>>>>
>>>> Attached files:
>>>>
>>>> 6703377-freetypescaler.patch - the patch itself
>>>> hg_export_icedtea6.diff - changes I would like to make in IcedTea
>>>>
>>>> Correct behaviour of outline transformation is (partially) checked by
>>>> regression test:
>>>> java/awt/font/Rotate/TranslatedOutlineTest.java
>>>>
>>>> Can anyone please review this patch?
>>>
>>> Upstream has resolved this bug using another method:
>>> http://hg.openjdk.java.net/jdk7/2d/jdk/rev/7fba83f5f5e0
>>>
>>> Is your patch fixing the exact same problem? Do you think it might be
>>> worth it to switch to upstream's fix?
>>>
>>> Thanks,
>>> Omair
>>
>> Hi Omair,
>>
>> AFAIK the upstream patch you mentioned is already included in IcedTea6,
>> but it is itself does not sufficient to resolve this bug when negative
>> coordinates are used.
>>
> 
> Ok. I was mislead because the icedtea patch was (re)using the same bug
> number. I have updated the jdk 6 patch web page [1] to reflect this. Any
> chance you could push this upstream? If this is a problem in IcedTea6,
> it will likely affect IcedTea7 (and IcedTea8 soon).

Yeah, I'm going to look at it.

Cheers,
Pavel
> 
> Thanks,
> Omair
> 
> http://icedtea.classpath.org/wiki/IcedTea_JDK6_Patches