Re: Review request for 8016665: [macosx] JComponent behaviour doesn't comply API documentation (setComponentOrientation method), Aqua LAF

2016-01-14 Thread Alexander Scherbatiy 


  The fix looks good to me.

  Thanks,
  Alexandr.

On 1/14/2016 9:02 AM, Sergey Bylokhov wrote:

Looks fine.

On 14/01/16 07:37, Avik Niyogi wrote:

Hi All,

Please find code changes incorporating changes as suggested by inputs
provided:

http://cr.openjdk.java.net/~aniyogi/8016665/webrev.02/

With Regards,
Avik Niyogi


On 13-Jan-2016, at 11:38 pm, Sergey Bylokhov
> wrote:

On 13/01/16 09:27, Avik Niyogi wrote:

Hi Sergey,
If fFileList refers to a soft linked empty folder, we will will not
need to apply orientation to it’s sub-components.

I didn't understand what "soft linked empty folder" means here =((

Just to clarify. In the fix you have this code:
 95 if (fFileList != null) {
 96 fFileList.setComponentOrientation(o);
...
 98 }

Which means that fFileList can be null, but it is a final field in the
AquaFileSystemModel and it is initialized in the
AquaFileChooserUI.createList() to a non null value:

   fFileList = new JTableExtension();
.
   model = new AquaFileSystemModel(fc, fFileList, fColumnNames);

Or am I missing something?



With Regards,
Avik Niyogi


On 12-Jan-2016, at 7:19 pm, Sergey Bylokhov

> wrote:

Hi, Avik.
Is it possible that fFileList can be null? I see that in other places
we do not check it to null?

On 04/01/16 11:52, Avik Niyogi wrote:

Hi All,

Please review the webrev.01 :
http://cr.openjdk.java.net/~aniyogi/8016665/webrev.01/
incorporated with the inputs received.

With Regards,
Avik Niyogi


On 28-Dec-2015, at 10:23 am, Avik Niyogi 
> wrote:

Hi Alexandr,

Automated test may fail based on folder contents on individual 
systems

irrespective of the fix directly not depending on the same.
Also, to confirm this fix, it will need visual confirmation and 
hence,

no automated test was provided.

With Regards,
Avik Niyogi

On 23-Dec-2015, at 7:32 pm, Alexander Scherbatiy

> wrote:



The fix looks good to me.

Is it possible to write an automated test for the fix?

Thanks,
Alexandr.

On 12/21/2015 2:55 PM, Avik Niyogi wrote:

Hi All,

Kindly review the bug fix for JDK 9.

*Bug:*
https://bugs.openjdk.java.net/browse/JDK-8016665

*Webrev:*
http://cr.openjdk.java.net/~aniyogi/8016665/webrev.00/


*Issue:*
The manual test: Swing_AllComponents/Manual/I18nSwingTests
in testsuite fails.

*Cause:*
Due to not honouring of RIGHT_TO_LEFT parameter for 
setOrientation

method applied for a JFileChooser for the AquaLookAndFeel only,
the fileChooser does not get displayed in RIGHT_TO_LEFT 
orientation.

This issue was verified to exist only in AquaLookAndFeel for
JFileChooser only due to wrong implementation in
AquaFileSystemModel.
Also, as provided in comments: "The Aqua LAF must support the RTL
orientation of JFileChooser."

*Fix:*
Added implementation for the check of RIGHT_TO_LEFT
ComponentOrientation and verified with test suite.


With Regards,
Avik Niyogi









--
Best regards, Sergey.





--
Best regards, Sergey.

Re: Review request for 8016665: [macosx] JComponent behaviour doesn't comply API documentation (setComponentOrientation method), Aqua LAF

2016-01-13 Thread Sergey Bylokhov 

On 13/01/16 09:27, Avik Niyogi wrote:

Hi Sergey,
If fFileList refers to a soft linked empty folder, we will  will not
need to apply orientation to it’s sub-components.

I didn't understand what "soft linked empty folder" means here =((

Just to clarify. In the fix you have this code:
  95 if (fFileList != null) {
  96 fFileList.setComponentOrientation(o);
...
  98 }

Which means that fFileList can be null, but it is a final field in the 
AquaFileSystemModel and it is initialized in the 
AquaFileChooserUI.createList() to a non null value:


fFileList = new JTableExtension();
.
model = new AquaFileSystemModel(fc, fFileList, fColumnNames);

Or am I missing something?



With Regards,
Avik Niyogi


On 12-Jan-2016, at 7:19 pm, Sergey Bylokhov
> wrote:

Hi, Avik.
Is it possible that fFileList can be null? I see that in other places
we do not check it to null?

On 04/01/16 11:52, Avik Niyogi wrote:

Hi All,

Please review the webrev.01 :
http://cr.openjdk.java.net/~aniyogi/8016665/webrev.01/
incorporated with the inputs received.

With Regards,
Avik Niyogi


On 28-Dec-2015, at 10:23 am, Avik Niyogi 
> wrote:

Hi Alexandr,

Automated test may fail based on folder contents on individual systems
irrespective of the fix directly not depending on the same.
Also, to confirm this fix, it will need visual confirmation and hence,
no automated test was provided.

With Regards,
Avik Niyogi

On 23-Dec-2015, at 7:32 pm, Alexander Scherbatiy

> wrote:



 The fix looks good to me.

 Is it possible to write an automated test for the fix?

 Thanks,
 Alexandr.

On 12/21/2015 2:55 PM, Avik Niyogi wrote:

Hi All,

Kindly review the bug fix for JDK 9.

*Bug:*
https://bugs.openjdk.java.net/browse/JDK-8016665

*Webrev:*
http://cr.openjdk.java.net/~aniyogi/8016665/webrev.00/


*Issue:*
The manual test: Swing_AllComponents/Manual/I18nSwingTests
in testsuite fails.

*Cause:*
Due to not honouring of RIGHT_TO_LEFT parameter for setOrientation
method applied for a JFileChooser for the AquaLookAndFeel only,
the fileChooser does not get displayed in RIGHT_TO_LEFT orientation.
This issue was verified to exist only in AquaLookAndFeel for
JFileChooser only due to wrong implementation in AquaFileSystemModel.
Also, as provided in comments: "The Aqua LAF must support the RTL
orientation of JFileChooser."

*Fix:*
Added implementation for the check of RIGHT_TO_LEFT
ComponentOrientation and verified with test suite.


With Regards,
Avik Niyogi









--
Best regards, Sergey.





--
Best regards, Sergey.

Re: Review request for 8016665: [macosx] JComponent behaviour doesn't comply API documentation (setComponentOrientation method), Aqua LAF

2016-01-13 Thread Sergey Bylokhov 

Looks fine.

On 14/01/16 07:37, Avik Niyogi wrote:

Hi All,

Please find code changes incorporating changes as suggested by inputs
provided:

http://cr.openjdk.java.net/~aniyogi/8016665/webrev.02/

With Regards,
Avik Niyogi


On 13-Jan-2016, at 11:38 pm, Sergey Bylokhov
> wrote:

On 13/01/16 09:27, Avik Niyogi wrote:

Hi Sergey,
If fFileList refers to a soft linked empty folder, we will  will not
need to apply orientation to it’s sub-components.

I didn't understand what "soft linked empty folder" means here =((

Just to clarify. In the fix you have this code:
 95 if (fFileList != null) {
 96 fFileList.setComponentOrientation(o);
...
 98 }

Which means that fFileList can be null, but it is a final field in the
AquaFileSystemModel and it is initialized in the
AquaFileChooserUI.createList() to a non null value:

   fFileList = new JTableExtension();
.
   model = new AquaFileSystemModel(fc, fFileList, fColumnNames);

Or am I missing something?



With Regards,
Avik Niyogi


On 12-Jan-2016, at 7:19 pm, Sergey Bylokhov

> wrote:

Hi, Avik.
Is it possible that fFileList can be null? I see that in other places
we do not check it to null?

On 04/01/16 11:52, Avik Niyogi wrote:

Hi All,

Please review the webrev.01 :
http://cr.openjdk.java.net/~aniyogi/8016665/webrev.01/
incorporated with the inputs received.

With Regards,
Avik Niyogi


On 28-Dec-2015, at 10:23 am, Avik Niyogi 
> wrote:

Hi Alexandr,

Automated test may fail based on folder contents on individual systems
irrespective of the fix directly not depending on the same.
Also, to confirm this fix, it will need visual confirmation and hence,
no automated test was provided.

With Regards,
Avik Niyogi

On 23-Dec-2015, at 7:32 pm, Alexander Scherbatiy

> wrote:



The fix looks good to me.

Is it possible to write an automated test for the fix?

Thanks,
Alexandr.

On 12/21/2015 2:55 PM, Avik Niyogi wrote:

Hi All,

Kindly review the bug fix for JDK 9.

*Bug:*
https://bugs.openjdk.java.net/browse/JDK-8016665

*Webrev:*
http://cr.openjdk.java.net/~aniyogi/8016665/webrev.00/


*Issue:*
The manual test: Swing_AllComponents/Manual/I18nSwingTests
in testsuite fails.

*Cause:*
Due to not honouring of RIGHT_TO_LEFT parameter for setOrientation
method applied for a JFileChooser for the AquaLookAndFeel only,
the fileChooser does not get displayed in RIGHT_TO_LEFT orientation.
This issue was verified to exist only in AquaLookAndFeel for
JFileChooser only due to wrong implementation in
AquaFileSystemModel.
Also, as provided in comments: "The Aqua LAF must support the RTL
orientation of JFileChooser."

*Fix:*
Added implementation for the check of RIGHT_TO_LEFT
ComponentOrientation and verified with test suite.


With Regards,
Avik Niyogi









--
Best regards, Sergey.





--
Best regards, Sergey.





--
Best regards, Sergey.

Re: Review request for 8016665: [macosx] JComponent behaviour doesn't comply API documentation (setComponentOrientation method), Aqua LAF

2016-01-13 Thread Avik Niyogi 
Hi All,

Please find code changes incorporating changes as suggested by inputs provided:

http://cr.openjdk.java.net/~aniyogi/8016665/webrev.02/ 


With Regards,
Avik Niyogi

> On 13-Jan-2016, at 11:38 pm, Sergey Bylokhov  
> wrote:
> 
> On 13/01/16 09:27, Avik Niyogi wrote:
>> Hi Sergey,
>> If fFileList refers to a soft linked empty folder, we will  will not
>> need to apply orientation to it’s sub-components.
> I didn't understand what "soft linked empty folder" means here =((
> 
> Just to clarify. In the fix you have this code:
>  95 if (fFileList != null) {
>  96 fFileList.setComponentOrientation(o);
> ...
>  98 }
> 
> Which means that fFileList can be null, but it is a final field in the 
> AquaFileSystemModel and it is initialized in the 
> AquaFileChooserUI.createList() to a non null value:
> 
>fFileList = new JTableExtension();
> .
>model = new AquaFileSystemModel(fc, fFileList, fColumnNames);
> 
> Or am I missing something?
> 
>> 
>> With Regards,
>> Avik Niyogi
>> 
>>> On 12-Jan-2016, at 7:19 pm, Sergey Bylokhov
>>> > wrote:
>>> 
>>> Hi, Avik.
>>> Is it possible that fFileList can be null? I see that in other places
>>> we do not check it to null?
>>> 
>>> On 04/01/16 11:52, Avik Niyogi wrote:
 Hi All,
 
 Please review the webrev.01 :
 http://cr.openjdk.java.net/~aniyogi/8016665/webrev.01/
 incorporated with the inputs received.
 
 With Regards,
 Avik Niyogi
 
> On 28-Dec-2015, at 10:23 am, Avik Niyogi  
> > wrote:
> 
> Hi Alexandr,
> 
> Automated test may fail based on folder contents on individual systems
> irrespective of the fix directly not depending on the same.
> Also, to confirm this fix, it will need visual confirmation and hence,
> no automated test was provided.
> 
> With Regards,
> Avik Niyogi
>> On 23-Dec-2015, at 7:32 pm, Alexander Scherbatiy
>> 
>> > wrote:
>> 
>> 
>> 
>> The fix looks good to me.
>> 
>> Is it possible to write an automated test for the fix?
>> 
>> Thanks,
>> Alexandr.
>> 
>> On 12/21/2015 2:55 PM, Avik Niyogi wrote:
>>> Hi All,
>>> 
>>> Kindly review the bug fix for JDK 9.
>>> 
>>> *Bug:*
>>> https://bugs.openjdk.java.net/browse/JDK-8016665
>>> 
>>> *Webrev:*
>>> http://cr.openjdk.java.net/~aniyogi/8016665/webrev.00/
>>> 
>>> 
>>> *Issue:*
>>> The manual test: Swing_AllComponents/Manual/I18nSwingTests
>>> in testsuite fails.
>>> 
>>> *Cause:*
>>> Due to not honouring of RIGHT_TO_LEFT parameter for setOrientation
>>> method applied for a JFileChooser for the AquaLookAndFeel only,
>>> the fileChooser does not get displayed in RIGHT_TO_LEFT orientation.
>>> This issue was verified to exist only in AquaLookAndFeel for
>>> JFileChooser only due to wrong implementation in AquaFileSystemModel.
>>> Also, as provided in comments: "The Aqua LAF must support the RTL
>>> orientation of JFileChooser."
>>> 
>>> *Fix:*
>>> Added implementation for the check of RIGHT_TO_LEFT
>>> ComponentOrientation and verified with test suite.
>>> 
>>> 
>>> With Regards,
>>> Avik Niyogi
>> 
> 
 
>>> 
>>> 
>>> --
>>> Best regards, Sergey.
>> 
> 
> 
> -- 
> Best regards, Sergey.

Re: Review request for 8016665: [macosx] JComponent behaviour doesn't comply API documentation (setComponentOrientation method), Aqua LAF

2016-01-12 Thread Avik Niyogi 
Hi Sergey,
If fFileList refers to a soft linked empty folder, we will  will not need to 
apply orientation to it’s sub-components.

With Regards,
Avik Niyogi

> On 12-Jan-2016, at 7:19 pm, Sergey Bylokhov  
> wrote:
> 
> Hi, Avik.
> Is it possible that fFileList can be null? I see that in other places we do 
> not check it to null?
> 
> On 04/01/16 11:52, Avik Niyogi wrote:
>> Hi All,
>> 
>> Please review the webrev.01 :
>> http://cr.openjdk.java.net/~aniyogi/8016665/webrev.01/ 
>> 
>> incorporated with the inputs received.
>> 
>> With Regards,
>> Avik Niyogi
>> 
>>> On 28-Dec-2015, at 10:23 am, Avik Niyogi >> 
>>> >> wrote:
>>> 
>>> Hi Alexandr,
>>> 
>>> Automated test may fail based on folder contents on individual systems
>>> irrespective of the fix directly not depending on the same.
>>> Also, to confirm this fix, it will need visual confirmation and hence,
>>> no automated test was provided.
>>> 
>>> With Regards,
>>> Avik Niyogi
 On 23-Dec-2015, at 7:32 pm, Alexander Scherbatiy
 
 >> wrote:
 
 
 
  The fix looks good to me.
 
  Is it possible to write an automated test for the fix?
 
  Thanks,
  Alexandr.
 
 On 12/21/2015 2:55 PM, Avik Niyogi wrote:
> Hi All,
> 
> Kindly review the bug fix for JDK 9.
> 
> *Bug:*
> https://bugs.openjdk.java.net/browse/JDK-8016665 
> 
> 
> *Webrev:*
> http://cr.openjdk.java.net/~aniyogi/8016665/webrev.00/ 
> 
>  >
> 
> *Issue:*
> The manual test: Swing_AllComponents/Manual/I18nSwingTests
> in testsuite fails.
> 
> *Cause:*
> Due to not honouring of RIGHT_TO_LEFT parameter for setOrientation
> method applied for a JFileChooser for the AquaLookAndFeel only,
> the fileChooser does not get displayed in RIGHT_TO_LEFT orientation.
> This issue was verified to exist only in AquaLookAndFeel for
> JFileChooser only due to wrong implementation in AquaFileSystemModel.
> Also, as provided in comments: "The Aqua LAF must support the RTL
> orientation of JFileChooser."
> 
> *Fix:*
> Added implementation for the check of RIGHT_TO_LEFT
> ComponentOrientation and verified with test suite.
> 
> 
> With Regards,
> Avik Niyogi
 
>>> 
>> 
> 
> 
> -- 
> Best regards, Sergey.

Re: Review request for 8016665: [macosx] JComponent behaviour doesn't comply API documentation (setComponentOrientation method), Aqua LAF

2016-01-12 Thread Avik Niyogi 
Hi All,
The comment in the mail trail is regarding different bug 8015748. Since, No 
inputs remain for code review related to this fix, please provide +1 for the 
fixes.
Thank you in advance.
With Regards,
Avik Niyogi
> On 12-Jan-2016, at 2:52 pm, Avik Niyogi  wrote:
> 
> Hi Alexander,
> 
> Due to animation effects on Aqua look and feel, the color checks can not be 
> used as seen in the image below:
> 
> The Blue color is not homogeneously distributed and varies continuously. 
> Since this is a fix for Aqua look and feel mainly, we need to perform a 
> visual confirmation itself.
> Hence, manual test is essential in this case.
> 
> With Regards,
> Avik Niyogi
> 
>> On 11-Jan-2016, at 9:12 pm, Alexander Scherbatiy 
>> > 
>> wrote:
>> 
>> 
>> Is it possible to make an automated test which compare colors from the 5/6 
>> part of the progress bar
>> when its value is set to 0 and 30? If these colors are the same it means 
>> that the progress has been painted on the left side instead of right.
>> 
>> Thanks,
>> Alexandr.
>>  
>> 
>> On 04/01/16 12:52, Avik Niyogi wrote:
>>> Hi All,
>>> 
>>> Please review the webrev.01 :  
>>> http://cr.openjdk.java.net/~aniyogi/8016665/webrev.01/
>>>  
>>> incorporated with the inputs received.
>>> 
>>> With Regards,
>>> Avik Niyogi
>>> 
 On 28-Dec-2015, at 10:23 am, Avik Niyogi > wrote:
 
 Hi Alexandr,
 
 Automated test may fail based on folder contents on individual systems 
 irrespective of the fix directly not depending on the same. 
 Also, to confirm this fix, it will need visual confirmation and hence, no 
 automated test was provided.
 
 With Regards,
 Avik Niyogi
> On 23-Dec-2015, at 7:32 pm, Alexander Scherbatiy < 
> alexandr.scherba...@oracle.com 
> > wrote:
> 
> 
> 
>   The fix looks good to me.
> 
>   Is it possible to write an automated test for the fix?
> 
>   Thanks,
>   Alexandr.
> 
> On 12/21/2015 2:55 PM, Avik Niyogi wrote:
>> Hi All,
>> 
>> Kindly review the bug fix for JDK 9.
>> 
>> Bug:
>> https://bugs.openjdk.java.net/browse/JDK-8016665 
>> 
>> 
>> Webrev:
>> http://cr.openjdk.java.net/~aniyogi/8016665/webrev.00/ 
>> 
>> 
>> Issue:
>> The manual test: Swing_AllComponents/Manual/I18nSwingTests
>> in testsuite fails.
>> 
>> Cause:
>> Due to not honouring of RIGHT_TO_LEFT parameter for setOrientation 
>> method applied for a JFileChooser for the AquaLookAndFeel only,
>> the fileChooser does not get displayed in RIGHT_TO_LEFT orientation. 
>> This issue was verified to exist only in AquaLookAndFeel for 
>> JFileChooser only due to wrong implementation in AquaFileSystemModel.
>> Also, as provided in comments: "The Aqua LAF must support the RTL 
>> orientation of JFileChooser."
>> 
>> Fix:
>> Added implementation for the check of RIGHT_TO_LEFT ComponentOrientation 
>> and verified with test suite.
>> 
>> 
>> With Regards,
>> Avik Niyogi
> 
 
>>> 
>> 
>

Re: Review request for 8016665: [macosx] JComponent behaviour doesn't comply API documentation (setComponentOrientation method), Aqua LAF

2016-01-12 Thread Avik Niyogi 
Hi Alexander,

Due to animation effects on Aqua look and feel, the color checks can not be 
used as seen in the image below:

The Blue color is not homogeneously distributed and varies continuously. Since 
this is a fix for Aqua look and feel mainly, we need to perform a visual 
confirmation itself.
Hence, manual test is essential in this case.

With Regards,
Avik Niyogi

> On 11-Jan-2016, at 9:12 pm, Alexander Scherbatiy 
>  wrote:
> 
> 
> Is it possible to make an automated test which compare colors from the 5/6 
> part of the progress bar
> when its value is set to 0 and 30? If these colors are the same it means that 
> the progress has been painted on the left side instead of right.
> 
> Thanks,
> Alexandr.
>  
> 
> On 04/01/16 12:52, Avik Niyogi wrote:
>> Hi All,
>> 
>> Please review the webrev.01 :  
>> http://cr.openjdk.java.net/~aniyogi/8016665/webrev.01/
>>  
>> incorporated with the inputs received.
>> 
>> With Regards,
>> Avik Niyogi
>> 
>>> On 28-Dec-2015, at 10:23 am, Avik Niyogi >> > wrote:
>>> 
>>> Hi Alexandr,
>>> 
>>> Automated test may fail based on folder contents on individual systems 
>>> irrespective of the fix directly not depending on the same. 
>>> Also, to confirm this fix, it will need visual confirmation and hence, no 
>>> automated test was provided.
>>> 
>>> With Regards,
>>> Avik Niyogi
 On 23-Dec-2015, at 7:32 pm, Alexander Scherbatiy < 
 alexandr.scherba...@oracle.com 
 > wrote:
 
 
 
   The fix looks good to me.
 
   Is it possible to write an automated test for the fix?
 
   Thanks,
   Alexandr.
 
 On 12/21/2015 2:55 PM, Avik Niyogi wrote:
> Hi All,
> 
> Kindly review the bug fix for JDK 9.
> 
> Bug:
> https://bugs.openjdk.java.net/browse/JDK-8016665 
> 
> 
> Webrev:
> http://cr.openjdk.java.net/~aniyogi/8016665/webrev.00/ 
> 
> 
> Issue:
> The manual test: Swing_AllComponents/Manual/I18nSwingTests
> in testsuite fails.
> 
> Cause:
> Due to not honouring of RIGHT_TO_LEFT parameter for setOrientation method 
> applied for a JFileChooser for the AquaLookAndFeel only,
> the fileChooser does not get displayed in RIGHT_TO_LEFT orientation. 
> This issue was verified to exist only in AquaLookAndFeel for JFileChooser 
> only due to wrong implementation in AquaFileSystemModel.
> Also, as provided in comments: "The Aqua LAF must support the RTL 
> orientation of JFileChooser."
> 
> Fix:
> Added implementation for the check of RIGHT_TO_LEFT ComponentOrientation 
> and verified with test suite.
> 
> 
> With Regards,
> Avik Niyogi
 
>>> 
>> 
>

Re: Review request for 8016665: [macosx] JComponent behaviour doesn't comply API documentation (setComponentOrientation method), Aqua LAF

2016-01-12 Thread Sergey Bylokhov 

Hi, Avik.
Is it possible that fFileList can be null? I see that in other places we 
do not check it to null?


On 04/01/16 11:52, Avik Niyogi wrote:

Hi All,

Please review the webrev.01 :
http://cr.openjdk.java.net/~aniyogi/8016665/webrev.01/
incorporated with the inputs received.

With Regards,
Avik Niyogi


On 28-Dec-2015, at 10:23 am, Avik Niyogi > wrote:

Hi Alexandr,

Automated test may fail based on folder contents on individual systems
irrespective of the fix directly not depending on the same.
Also, to confirm this fix, it will need visual confirmation and hence,
no automated test was provided.

With Regards,
Avik Niyogi

On 23-Dec-2015, at 7:32 pm, Alexander Scherbatiy
> wrote:



  The fix looks good to me.

  Is it possible to write an automated test for the fix?

  Thanks,
  Alexandr.

On 12/21/2015 2:55 PM, Avik Niyogi wrote:

Hi All,

Kindly review the bug fix for JDK 9.

*Bug:*
https://bugs.openjdk.java.net/browse/JDK-8016665

*Webrev:*
http://cr.openjdk.java.net/~aniyogi/8016665/webrev.00/


*Issue:*
The manual test: Swing_AllComponents/Manual/I18nSwingTests
in testsuite fails.

*Cause:*
Due to not honouring of RIGHT_TO_LEFT parameter for setOrientation
method applied for a JFileChooser for the AquaLookAndFeel only,
the fileChooser does not get displayed in RIGHT_TO_LEFT orientation.
This issue was verified to exist only in AquaLookAndFeel for
JFileChooser only due to wrong implementation in AquaFileSystemModel.
Also, as provided in comments: "The Aqua LAF must support the RTL
orientation of JFileChooser."

*Fix:*
Added implementation for the check of RIGHT_TO_LEFT
ComponentOrientation and verified with test suite.


With Regards,
Avik Niyogi









--
Best regards, Sergey.

Re: Review request for 8016665: [macosx] JComponent behaviour doesn't comply API documentation (setComponentOrientation method), Aqua LAF

2016-01-12 Thread Alexandr Scherbatiy 


I am sorry. The comment really was about the issue 8015748.

The fix looks good for me.

Thanks,
Alexandr.

On 1/12/2016 1:44 AM, Avik Niyogi wrote:

Hi All,
The comment in the mail trail is regarding different bug 8015748. 
Since, No inputs remain for code review related to this fix, please 
provide +1 for the fixes.

Thank you in advance.
With Regards,
Avik Niyogi
On 12-Jan-2016, at 2:52 pm, Avik Niyogi > wrote:


Hi Alexander,

Due to animation effects on Aqua look and feel, the color checks can 
not be used as seen in the image below:


The Blue color is not homogeneously distributed and varies 
continuously. Since this is a fix for Aqua look and feel mainly, we 
need to perform a visual confirmation itself.

Hence, manual test is essential in this case.

With Regards,
Avik Niyogi

On 11-Jan-2016, at 9:12 pm, Alexander Scherbatiy 
> wrote:



Is it possible to make an automated test which compare colors from 
the 5/6 part of the progress bar
when its value is set to 0 and 30? If these colors are the same it 
means that the progress has been painted on the left side instead of 
right.


Thanks,
Alexandr.


On 04/01/16 12:52, Avik Niyogi wrote:

Hi All,

Please review the webrev.01 : 
http://cr.openjdk.java.net/~aniyogi/8016665/webrev.01/

incorporated with the inputs received.

With Regards,
Avik Niyogi

On 28-Dec-2015, at 10:23 am, Avik Niyogi > wrote:


Hi Alexandr,

Automated test may fail based on folder contents on individual 
systems irrespective of the fix directly not depending on the same.
Also, to confirm this fix, it will need visual confirmation and 
hence, no automated test was provided.


With Regards,
Avik Niyogi
On 23-Dec-2015, at 7:32 pm, Alexander Scherbatiy 
 wrote:




  The fix looks good to me.

  Is it possible to write an automated test for the fix?

  Thanks,
  Alexandr.

On 12/21/2015 2:55 PM, Avik Niyogi wrote:

Hi All,

Kindly review the bug fix for JDK 9.

*Bug:*
https://bugs.openjdk.java.net/browse/JDK-8016665

*Webrev:*
http://cr.openjdk.java.net/~aniyogi/8016665/webrev.00/ 



*Issue:*
The manual test: Swing_AllComponents/Manual/I18nSwingTests
in testsuite fails.

*Cause:*
Due to not honouring of RIGHT_TO_LEFT parameter for 
setOrientation method applied for a JFileChooser for the 
AquaLookAndFeel only,
the fileChooser does not get displayed in RIGHT_TO_LEFT 
orientation.
This issue was verified to exist only in AquaLookAndFeel for 
JFileChooser only due to wrong implementation 
in AquaFileSystemModel.
Also, as provided in comments: "The Aqua LAF must support the 
RTL orientation of JFileChooser."


*Fix:*
Added implementation for the check of RIGHT_TO_LEFT 
ComponentOrientation and verified with test suite.



With Regards,
Avik Niyogi

Re: Review request for 8016665: [macosx] JComponent behaviour doesn't comply API documentation (setComponentOrientation method), Aqua LAF

2016-01-11 Thread Alexander Scherbatiy 


Is it possible to make an automated test which compare colors from the 
5/6 part of the progress bar
when its value is set to 0 and 30? If these colors are the same it means 
that the progress has been painted on the left side instead of right.


Thanks,
Alexandr.


On 04/01/16 12:52, Avik Niyogi wrote:

Hi All,

Please review the webrev.01 : 
http://cr.openjdk.java.net/~aniyogi/8016665/webrev.01/ 


incorporated with the inputs received.

With Regards,
Avik Niyogi

On 28-Dec-2015, at 10:23 am, Avik Niyogi > wrote:


Hi Alexandr,

Automated test may fail based on folder contents on individual 
systems irrespective of the fix directly not depending on the same.
Also, to confirm this fix, it will need visual confirmation and 
hence, no automated test was provided.


With Regards,
Avik Niyogi
On 23-Dec-2015, at 7:32 pm, Alexander Scherbatiy 
> wrote:




  The fix looks good to me.

  Is it possible to write an automated test for the fix?

  Thanks,
  Alexandr.

On 12/21/2015 2:55 PM, Avik Niyogi wrote:

Hi All,

Kindly review the bug fix for JDK 9.

*Bug:*
https://bugs.openjdk.java.net/browse/JDK-8016665

*Webrev:*
http://cr.openjdk.java.net/~aniyogi/8016665/webrev.00/ 



*Issue:*
The manual test: Swing_AllComponents/Manual/I18nSwingTests
in testsuite fails.

*Cause:*
Due to not honouring of RIGHT_TO_LEFT parameter for setOrientation 
method applied for a JFileChooser for the AquaLookAndFeel only,

the fileChooser does not get displayed in RIGHT_TO_LEFT orientation.
This issue was verified to exist only in AquaLookAndFeel for 
JFileChooser only due to wrong implementation in AquaFileSystemModel.
Also, as provided in comments: "The Aqua LAF must support the RTL 
orientation of JFileChooser."


*Fix:*
Added implementation for the check of RIGHT_TO_LEFT 
ComponentOrientation and verified with test suite.



With Regards,
Avik Niyogi

Re: Review request for 8016665: [macosx] JComponent behaviour doesn't comply API documentation (setComponentOrientation method), Aqua LAF

2016-01-04 Thread Avik Niyogi 
Hi All,

Please review the webrev.01 : 
http://cr.openjdk.java.net/~aniyogi/8016665/webrev.01/ 

incorporated with the inputs received.

With Regards,
Avik Niyogi

> On 28-Dec-2015, at 10:23 am, Avik Niyogi  wrote:
> 
> Hi Alexandr,
> 
> Automated test may fail based on folder contents on individual systems 
> irrespective of the fix directly not depending on the same. 
> Also, to confirm this fix, it will need visual confirmation and hence, no 
> automated test was provided.
> 
> With Regards,
> Avik Niyogi
>> On 23-Dec-2015, at 7:32 pm, Alexander Scherbatiy 
>> > 
>> wrote:
>> 
>> 
>> 
>>   The fix looks good to me.
>> 
>>   Is it possible to write an automated test for the fix?
>> 
>>   Thanks,
>>   Alexandr.
>> 
>> On 12/21/2015 2:55 PM, Avik Niyogi wrote:
>>> Hi All,
>>> 
>>> Kindly review the bug fix for JDK 9.
>>> 
>>> Bug:
>>> https://bugs.openjdk.java.net/browse/JDK-8016665 
>>> 
>>> 
>>> Webrev:
>>> http://cr.openjdk.java.net/~aniyogi/8016665/webrev.00/ 
>>> 
>>> 
>>> Issue:
>>> The manual test: Swing_AllComponents/Manual/I18nSwingTests
>>> in testsuite fails.
>>> 
>>> Cause:
>>> Due to not honouring of RIGHT_TO_LEFT parameter for setOrientation method 
>>> applied for a JFileChooser for the AquaLookAndFeel only,
>>> the fileChooser does not get displayed in RIGHT_TO_LEFT orientation. 
>>> This issue was verified to exist only in AquaLookAndFeel for JFileChooser 
>>> only due to wrong implementation in AquaFileSystemModel.
>>> Also, as provided in comments: "The Aqua LAF must support the RTL 
>>> orientation of JFileChooser."
>>> 
>>> Fix:
>>> Added implementation for the check of RIGHT_TO_LEFT ComponentOrientation 
>>> and verified with test suite.
>>> 
>>> 
>>> With Regards,
>>> Avik Niyogi
>> 
>

Re: Review request for 8016665: [macosx] JComponent behaviour doesn't comply API documentation (setComponentOrientation method), Aqua LAF

2015-12-27 Thread Avik Niyogi 
Hi Alexandr,

Automated test may fail based on folder contents on individual systems 
irrespective of the fix directly not depending on the same. 
Also, to confirm this fix, it will need visual confirmation and hence, no 
automated test was provided.

With Regards,
Avik Niyogi
> On 23-Dec-2015, at 7:32 pm, Alexander Scherbatiy 
>  wrote:
> 
> 
> 
>   The fix looks good to me.
> 
>   Is it possible to write an automated test for the fix?
> 
>   Thanks,
>   Alexandr.
> 
> On 12/21/2015 2:55 PM, Avik Niyogi wrote:
>> Hi All,
>> 
>> Kindly review the bug fix for JDK 9.
>> 
>> Bug:
>> https://bugs.openjdk.java.net/browse/JDK-8016665 
>> 
>> 
>> Webrev:
>> http://cr.openjdk.java.net/~aniyogi/8016665/webrev.00/ 
>> 
>> 
>> Issue:
>> The manual test: Swing_AllComponents/Manual/I18nSwingTests
>> in testsuite 
>> http://sqe-hg.us.oracle.com/hg/index.cgi/testbase/javase/functional/7/swing 
>> 
>>  fails.
>> 
>> Cause:
>> Due to not honouring of RIGHT_TO_LEFT parameter for setOrientation method 
>> applied for a JFileChooser for the AquaLookAndFeel only,
>> the fileChooser does not get displayed in RIGHT_TO_LEFT orientation. 
>> This issue was verified to exist only in AquaLookAndFeel for JFileChooser 
>> only due to wrong implementation in AquaFileSystemModel.
>> Also, as provided in comments: "The Aqua LAF must support the RTL 
>> orientation of JFileChooser."
>> 
>> Fix:
>> Added implementation for the check of RIGHT_TO_LEFT ComponentOrientation and 
>> verified with test suite.
>> 
>> 
>> With Regards,
>> Avik Niyogi
>

Re: Review request for 8016665: [macosx] JComponent behaviour doesn't comply API documentation (setComponentOrientation method), Aqua LAF

2015-12-23 Thread Alexander Scherbatiy 

  
  

  
    The fix looks good to me.
  
    Is it possible to write an automated test for the fix?
  
    Thanks,
    Alexandr.
  
  On 12/21/2015 2:55 PM, Avik Niyogi wrote:


  
  Hi All,
  
  
  Kindly review the bug fix for JDK 9.
  
  
  Bug:
  https://bugs.openjdk.java.net/browse/JDK-8016665
  
  
  Webrev:
  http://cr.openjdk.java.net/~aniyogi/8016665/webrev.00/
  
  
  Issue:
  The manual test:
Swing_AllComponents/Manual/I18nSwingTests
  in testsuite http://sqe-hg.us.oracle.com/hg/index.cgi/testbase/javase/functional/7/swing fails.
  
  
  
Cause:
Due to not honouring of RIGHT_TO_LEFT parameter
  for setOrientation method applied for a JFileChooser for the
  AquaLookAndFeel only,
the fileChooser does not get displayed in
  RIGHT_TO_LEFT orientation. 
This issue was verified to exist only in
  AquaLookAndFeel for JFileChooser only due to wrong
  implementation in AquaFileSystemModel.
Also, as provided in comments: "The Aqua LAF must
  support the RTL orientation of JFileChooser."


Fix:
Added implementation for the check of
  RIGHT_TO_LEFT ComponentOrientation and verified with test
  suite.




With Regards,
Avik Niyogi

Review request for 8016665: [macosx] JComponent behaviour doesn't comply API documentation (setComponentOrientation method), Aqua LAF

2015-12-21 Thread Avik Niyogi 
Hi All,

Kindly review the bug fix for JDK 9.

Bug:
https://bugs.openjdk.java.net/browse/JDK-8016665 


Webrev:
http://cr.openjdk.java.net/~aniyogi/8016665/webrev.00/ 


Issue:
The manual test: Swing_AllComponents/Manual/I18nSwingTests
in testsuite 
http://sqe-hg.us.oracle.com/hg/index.cgi/testbase/javase/functional/7/swing 
 
fails.

Cause:
Due to not honouring of RIGHT_TO_LEFT parameter for setOrientation method 
applied for a JFileChooser for the AquaLookAndFeel only,
the fileChooser does not get displayed in RIGHT_TO_LEFT orientation. 
This issue was verified to exist only in AquaLookAndFeel for JFileChooser only 
due to wrong implementation in AquaFileSystemModel.
Also, as provided in comments: "The Aqua LAF must support the RTL orientation 
of JFileChooser."

Fix:
Added implementation for the check of RIGHT_TO_LEFT ComponentOrientation and 
verified with test suite.


With Regards,
Avik Niyogi