Re: Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-21 Thread Alexander Scherbatiy 


  The fix looks good to me.

 Thanks,
 Alexandr.

On 21/03/16 18:02, Sergey Bylokhov wrote:

Looks fine then.

On 18.03.16 8:46, Avik Niyogi wrote:

Hi Sergey,
I have tested the test cases as mentioned.
1) They are not related to the current fix
2) ScreenMenuBarInputTwice.java issue is not reproducible.
3) There is a test bug in ActionListenerCalledTwiceTest.java, hence a
new test case called ScreenMenuBarInputTwice.java
was created to encompass actual expectations from Java on OS X as per
native behaviour.

Please review my current fix as available in the mail trail. Thank you
in advance.

With Regards,
Avik Niyogi



On 14-Mar-2016, at 6:48 pm, Sergey Bylokhov
> wrote:

Hi, Avik.
Can you please take a look to these two tests before fixing this bug:

TEST: javax/swing/JMenuItem/8139169/ScreenMenuBarInputTwice.java
--
TEST:
javax/swing/JMenuItem/ActionListenerCalledTwice/ActionListenerCalledTwiceTest.java 



I remember they passed on jdk8, but it seems we have a regression in
jdk9 and both of them fail.

On 14.03.16 8:05, Avik Niyogi wrote:

Hi All,
A gentle reminder, please review my code changes.

With Regards,
Avik Niyogi

On 08-Mar-2016, at 9:39 pm, Avik Niyogi 
> wrote:

Hi All,

Kindly review the bug fix for JDK 9.

*Bug:*

_https://bugs.openjdk.java.net/browse/JDK-8148555_
_
_
*Webrev:*

_http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/_

*Issue:*
Emoji selection in Character Viewer was causing exception in JNI

*Cause:*
Emojis are considered to be of different class type (namely,
NSConcreteMutableAttributedString) from NSString which other
characters are because of a surrogate pair for them.

*Fix:*
Major changes done for condition of emojis in JNI. Albeit 
rendering is

not yet supported, they will appear as blank “Missing font” notation.
Also, added debug point in case of issue with glyph arrises.

With Regards,
Avik Niyogi





--
Best regards, Sergey.

Re: Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-21 Thread Sergey Bylokhov 

Looks fine then.

On 18.03.16 8:46, Avik Niyogi wrote:

Hi Sergey,
I have tested the test cases as mentioned.
1) They are not related to the current fix
2) ScreenMenuBarInputTwice.java issue is not reproducible.
3) There is a test bug in ActionListenerCalledTwiceTest.java, hence a
new test case called ScreenMenuBarInputTwice.java
was created to encompass actual expectations from Java on OS X as per
native behaviour.

Please review my current fix as available in the mail trail. Thank you
in advance.

With Regards,
Avik Niyogi



On 14-Mar-2016, at 6:48 pm, Sergey Bylokhov
> wrote:

Hi, Avik.
Can you please take a look to these two tests before fixing this bug:

TEST: javax/swing/JMenuItem/8139169/ScreenMenuBarInputTwice.java
--
TEST:
javax/swing/JMenuItem/ActionListenerCalledTwice/ActionListenerCalledTwiceTest.java

I remember they passed on jdk8, but it seems we have a regression in
jdk9 and both of them fail.

On 14.03.16 8:05, Avik Niyogi wrote:

Hi All,
A gentle reminder, please review my code changes.

With Regards,
Avik Niyogi

On 08-Mar-2016, at 9:39 pm, Avik Niyogi 
> wrote:

Hi All,

Kindly review the bug fix for JDK 9.

*Bug:*

_https://bugs.openjdk.java.net/browse/JDK-8148555_
_
_
*Webrev:*

_http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/_

*Issue:*
Emoji selection in Character Viewer was causing exception in JNI

*Cause:*
Emojis are considered to be of different class type (namely,
NSConcreteMutableAttributedString) from NSString which other
characters are because of a surrogate pair for them.

*Fix:*
Major changes done for condition of emojis in JNI. Albeit rendering is
not yet supported, they will appear as blank “Missing font” notation.
Also, added debug point in case of issue with glyph arrises.

With Regards,
Avik Niyogi





--
Best regards, Sergey.





--
Best regards, Sergey.

Re: Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-21 Thread Sergey Bylokhov 

Please take a look to my comment below before pushing this fix.

On 14.03.16 16:18, Sergey Bylokhov wrote:

Hi, Avik.
Can you please take a look to these two tests before fixing this bug:

TEST: javax/swing/JMenuItem/8139169/ScreenMenuBarInputTwice.java
--
TEST:
javax/swing/JMenuItem/ActionListenerCalledTwice/ActionListenerCalledTwiceTest.java


I remember they passed on jdk8, but it seems we have a regression in
jdk9 and both of them fail.

On 14.03.16 8:05, Avik Niyogi wrote:

Hi All,
A gentle reminder, please review my code changes.

With Regards,
Avik Niyogi

On 08-Mar-2016, at 9:39 pm, Avik Niyogi > wrote:

Hi All,

Kindly review the bug fix for JDK 9.

*Bug:*

_https://bugs.openjdk.java.net/browse/JDK-8148555_
_
_
*Webrev:*

_http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/_

*Issue:*
Emoji selection in Character Viewer was causing exception in JNI

*Cause:*
Emojis are considered to be of different class type (namely,
NSConcreteMutableAttributedString) from NSString which other
characters are because of a surrogate pair for them.

*Fix:*
Major changes done for condition of emojis in JNI. Albeit rendering is
not yet supported, they will appear as blank “Missing font” notation.
Also, added debug point in case of issue with glyph arrises.

With Regards,
Avik Niyogi








--
Best regards, Sergey.

Re: Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-21 Thread Manajit Halder 
Hi Avik,

Changes looks good to me.

Regards,
Manajit

> On 21-Mar-2016, at 4:16 pm, Rajeev Chamyal  wrote:
> 
> Test code looks good to me.
>  
> Regards,
> Rajeev Chamyal
>  
> From: Avik Niyogi 
> Sent: 21 March 2016 14:02
> To: Manajit Halder; Alexander Scherbatiy
> Cc: swing-dev@openjdk.java.net; Rajeev Chamyal
> Subject: Re:  Review Request of 8148555: [macosx] An uncaught 
> exception was raised entering Emoji into JTextArea
>  
> Hi All,
> Please review the below code changes as per the inputs received.
> http://cr.openjdk.java.net/~aniyogi/8148555/webrev.01/ 
> 
>  
>  
> With Regards,
> Avik Niyogi
> On 21-Mar-2016, at 12:45 pm, Rajeev Chamyal  > wrote:
>  
> Hello Avik,
>  
> I can’t comment on objective C code.
> As far as test is concerned below are my comments.
>  
> 1)  UI should be created in Swing thread.
> 2)  Switch case in  actionPerformed should be refactored.
>  
> Regards,
> Rajeev Chamyal
>  
> From: Avik Niyogi 
> Sent: 21 March 2016 12:20
> To: Sergey Bylokhov
> Cc: swing-dev@openjdk.java.net ; Alexander 
> Scherbatiy; Rajeev Chamyal
> Subject: Re:  Review Request of 8148555: [macosx] An uncaught 
> exception was raised entering Emoji into JTextArea
>  
> Hi Rajeev,
> Please review the following code changes.
>  
> With Regards,
> Avik Niyogi
> On 21-Mar-2016, at 12:17 pm, Avik Niyogi  > wrote:
>  
> Hi Sergey,
>  
> Please review the following code changes.
> With Regards,
> Avik Niyogi
> On 17-Mar-2016, at 7:03 pm, Alexander Scherbatiy 
> > 
> wrote:
>  
> 
> The fix looks good to me.
> 
> Just a small note: it is better to remove comment "527 //" since it 
> does not have a description.
> 
> Thanks,
> Alexandr.
> 
> On 17/03/16 17:21, Avik Niyogi wrote:
> It can be made into a class method, but herein this case it is needed for 
> that instance only and hence the need for instance method and referred with 
> “self”. 
>  
> With Regards,
> Avik Niyogi
> On 16-Mar-2016, at 11:55 pm, Alexander Scherbatiy 
> > 
> wrote:
>  
> 
> Could the -(NSMutableString *) parseString: method be declared as class 
> method instead of instance?
> 
> Thanks,
> Alexandr.
> 
> On 14/03/16 17:18, Sergey Bylokhov wrote:
> Hi, Avik. 
> Can you please take a look to these two tests before fixing this bug: 
> 
> TEST: javax/swing/JMenuItem/8139169/ScreenMenuBarInputTwice.java 
> -- 
> TEST: 
> javax/swing/JMenuItem/ActionListenerCalledTwice/ActionListenerCalledTwiceTest.java
> 
> I remember they passed on jdk8, but it seems we have a regression in jdk9 and 
> both of them fail. 
> 
> On 14.03.16 8:05, Avik Niyogi wrote: 
> 
> 
> Hi All, 
> A gentle reminder, please review my code changes. 
> 
> With Regards, 
> Avik Niyogi 
> 
> 
> On 08-Mar-2016, at 9:39 pm, Avik Niyogi   
>  > wrote: 
> 
> Hi All, 
> 
> Kindly review the bug fix for JDK 9. 
> 
> *Bug:* 
> 
> _https://bugs.openjdk.java.net/browse/JDK-8148555_ 
>  
> _ 
> _ 
> *Webrev:* 
> 
> _http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/_ 
>  
> 
> *Issue:* 
> Emoji selection in Character Viewer was causing exception in JNI 
> 
> *Cause:* 
> Emojis are considered to be of different class type (namely, 
> NSConcreteMutableAttributedString) from NSString which other 
> characters are because of a surrogate pair for them. 
> 
> *Fix:* 
> Major changes done for condition of emojis in JNI. Albeit rendering is 
> not yet supported, they will appear as blank “Missing font” notation. 
> Also, added debug point in case of issue with glyph arrises. 
> 
> With Regards, 
> Avik Niyogi

Re: Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-21 Thread Rajeev Chamyal 
Test code looks good to me.

 

Regards,

Rajeev Chamyal

 

From: Avik Niyogi 
Sent: 21 March 2016 14:02
To: Manajit Halder; Alexander Scherbatiy
Cc: swing-dev@openjdk.java.net; Rajeev Chamyal
Subject: Re:  Review Request of 8148555: [macosx] An uncaught 
exception was raised entering Emoji into JTextArea

 

Hi All,

Please review the below code changes as per the inputs received.

http://cr.openjdk.java.net/~aniyogi/8148555/webrev.01/

 

 

With Regards,

Avik Niyogi

On 21-Mar-2016, at 12:45 pm, Rajeev Chamyal mailto:rajeev.cham...@oracle.com"rajeev.cham...@oracle.com> wrote:

 

Hello Avik,

 

I can’t comment on objective C code.

As far as test is concerned below are my comments.

 

1)  UI should be created in Swing thread.

2)  Switch case in  actionPerformed should be refactored.

 

Regards,

Rajeev Chamyal

 

From: Avik Niyogi 
Sent: 21 March 2016 12:20
To: Sergey Bylokhov
Cc: HYPERLINK "mailto:swing-dev@openjdk.java.net"swing-dev@openjdk.java.net; 
Alexander Scherbatiy; Rajeev Chamyal
Subject: Re:  Review Request of 8148555: [macosx] An uncaught 
exception was raised entering Emoji into JTextArea

 

Hi Rajeev,

Please review the following code changes.

 

With Regards,

Avik Niyogi

On 21-Mar-2016, at 12:17 pm, Avik Niyogi mailto:avik.niy...@oracle.com"avik.niy...@oracle.com> wrote:

 

Hi Sergey,

 

Please review the following code changes.

With Regards,

Avik Niyogi

On 17-Mar-2016, at 7:03 pm, Alexander Scherbatiy mailto:alexandr.scherba...@oracle.com"alexandr.scherba...@oracle.com> wrote:

 


The fix looks good to me.

Just a small note: it is better to remove comment "527 //" since it 
does not have a description.

Thanks,
Alexandr.

On 17/03/16 17:21, Avik Niyogi wrote:

It can be made into a class method, but herein this case it is needed for that 
instance only and hence the need for instance method and referred with “self”. 

 

With Regards,

Avik Niyogi

On 16-Mar-2016, at 11:55 pm, Alexander Scherbatiy mailto:alexandr.scherba...@oracle.com"alexandr.scherba...@oracle.com> wrote:

 


Could the -(NSMutableString *) parseString: method be declared as class method 
instead of instance?

Thanks,
Alexandr.

On 14/03/16 17:18, Sergey Bylokhov wrote:

Hi, Avik. 
Can you please take a look to these two tests before fixing this bug: 

TEST: javax/swing/JMenuItem/8139169/ScreenMenuBarInputTwice.java 
-- 
TEST: 
javax/swing/JMenuItem/ActionListenerCalledTwice/ActionListenerCalledTwiceTest.java

I remember they passed on jdk8, but it seems we have a regression in jdk9 and 
both of them fail. 

On 14.03.16 8:05, Avik Niyogi wrote: 




Hi All, 
A gentle reminder, please review my code changes. 

With Regards, 
Avik Niyogi 




On 08-Mar-2016, at 9:39 pm, Avik Niyogi mailto:avik.niy...@oracle.com"avik.niy...@oracle.com 
HYPERLINK "mailto:avik.niy...@oracle.com;> 
wrote: 

Hi All, 

Kindly review the bug fix for JDK 9. 

*Bug:* 

_https://bugs.openjdk.java.net/browse/JDK-8148555_ 
_ 
_ 
*Webrev:* 

_HYPERLINK 
"http://cr.openjdk.java.net/%7Eaniyogi/8148555/webrev.00/_"http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/_
 

*Issue:* 
Emoji selection in Character Viewer was causing exception in JNI 

*Cause:* 
Emojis are considered to be of different class type (namely, 
NSConcreteMutableAttributedString) from NSString which other 
characters are because of a surrogate pair for them. 

*Fix:* 
Major changes done for condition of emojis in JNI. Albeit rendering is 
not yet supported, they will appear as blank “Missing font” notation. 
Also, added debug point in case of issue with glyph arrises. 

With Regards, 
Avik Niyogi 

 

Re: Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-21 Thread Avik Niyogi 
Hi All,
Please review the below code changes as per the inputs received.
http://cr.openjdk.java.net/~aniyogi/8148555/webrev.01/ 



With Regards,
Avik Niyogi
> On 21-Mar-2016, at 12:45 pm, Rajeev Chamyal  wrote:
> 
> Hello Avik,
>  
> I can’t comment on objective C code.
> As far as test is concerned below are my comments.
>  
> 1)  UI should be created in Swing thread.
> 2)  Switch case in  actionPerformed should be refactored.
>  
> Regards,
> Rajeev Chamyal
>  
> From: Avik Niyogi 
> Sent: 21 March 2016 12:20
> To: Sergey Bylokhov
> Cc: swing-dev@openjdk.java.net; Alexander Scherbatiy; Rajeev Chamyal
> Subject: Re:  Review Request of 8148555: [macosx] An uncaught 
> exception was raised entering Emoji into JTextArea
>  
> Hi Rajeev,
> Please review the following code changes.
>  
> With Regards,
> Avik Niyogi
> On 21-Mar-2016, at 12:17 pm, Avik Niyogi  > wrote:
>  
> Hi Sergey,
>  
> Please review the following code changes.
> With Regards,
> Avik Niyogi
> On 17-Mar-2016, at 7:03 pm, Alexander Scherbatiy 
> > 
> wrote:
>  
> 
> The fix looks good to me.
> 
> Just a small note: it is better to remove comment "527 //" since it 
> does not have a description.
> 
> Thanks,
> Alexandr.
> 
> On 17/03/16 17:21, Avik Niyogi wrote:
> It can be made into a class method, but herein this case it is needed for 
> that instance only and hence the need for instance method and referred with 
> “self”. 
>  
> With Regards,
> Avik Niyogi
> On 16-Mar-2016, at 11:55 pm, Alexander Scherbatiy 
> > 
> wrote:
>  
> 
> Could the -(NSMutableString *) parseString: method be declared as class 
> method instead of instance?
> 
> Thanks,
> Alexandr.
> 
> On 14/03/16 17:18, Sergey Bylokhov wrote:
> Hi, Avik. 
> Can you please take a look to these two tests before fixing this bug: 
> 
> TEST: javax/swing/JMenuItem/8139169/ScreenMenuBarInputTwice.java 
> -- 
> TEST: 
> javax/swing/JMenuItem/ActionListenerCalledTwice/ActionListenerCalledTwiceTest.java
> 
> I remember they passed on jdk8, but it seems we have a regression in jdk9 and 
> both of them fail. 
> 
> On 14.03.16 8:05, Avik Niyogi wrote: 
> 
> Hi All, 
> A gentle reminder, please review my code changes. 
> 
> With Regards, 
> Avik Niyogi 
> 
> On 08-Mar-2016, at 9:39 pm, Avik Niyogi   
>  > wrote: 
> 
> Hi All, 
> 
> Kindly review the bug fix for JDK 9. 
> 
> *Bug:* 
> 
> _https://bugs.openjdk.java.net/browse/JDK-8148555_ 
>  
> _ 
> _ 
> *Webrev:* 
> 
> _http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/_ 
>  
> 
> *Issue:* 
> Emoji selection in Character Viewer was causing exception in JNI 
> 
> *Cause:* 
> Emojis are considered to be of different class type (namely, 
> NSConcreteMutableAttributedString) from NSString which other 
> characters are because of a surrogate pair for them. 
> 
> *Fix:* 
> Major changes done for condition of emojis in JNI. Albeit rendering is 
> not yet supported, they will appear as blank “Missing font” notation. 
> Also, added debug point in case of issue with glyph arrises. 
> 
> With Regards, 
> Avik Niyogi

Re: Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-21 Thread Rajeev Chamyal 
Hello Avik,

 

I can’t comment on objective C code.

As far as test is concerned below are my comments.

 

1)  UI should be created in Swing thread.

2)  Switch case in  actionPerformed should be refactored.

 

Regards,

Rajeev Chamyal

 

From: Avik Niyogi 
Sent: 21 March 2016 12:20
To: Sergey Bylokhov
Cc: swing-dev@openjdk.java.net; Alexander Scherbatiy; Rajeev Chamyal
Subject: Re:  Review Request of 8148555: [macosx] An uncaught 
exception was raised entering Emoji into JTextArea

 

Hi Rajeev,

Please review the following code changes.

 

With Regards,

Avik Niyogi

On 21-Mar-2016, at 12:17 pm, Avik Niyogi mailto:avik.niy...@oracle.com"avik.niy...@oracle.com> wrote:

 

Hi Sergey,

 

Please review the following code changes.

With Regards,

Avik Niyogi

On 17-Mar-2016, at 7:03 pm, Alexander Scherbatiy mailto:alexandr.scherba...@oracle.com"alexandr.scherba...@oracle.com> wrote:

 


The fix looks good to me.

Just a small note: it is better to remove comment "527 //" since it 
does not have a description.

Thanks,
Alexandr.

On 17/03/16 17:21, Avik Niyogi wrote:

It can be made into a class method, but herein this case it is needed for that 
instance only and hence the need for instance method and referred with “self”. 

 

With Regards,

Avik Niyogi

On 16-Mar-2016, at 11:55 pm, Alexander Scherbatiy mailto:alexandr.scherba...@oracle.com"alexandr.scherba...@oracle.com> wrote:

 


Could the -(NSMutableString *) parseString: method be declared as class method 
instead of instance?

Thanks,
Alexandr.

On 14/03/16 17:18, Sergey Bylokhov wrote:

Hi, Avik. 
Can you please take a look to these two tests before fixing this bug: 

TEST: javax/swing/JMenuItem/8139169/ScreenMenuBarInputTwice.java 
-- 
TEST: 
javax/swing/JMenuItem/ActionListenerCalledTwice/ActionListenerCalledTwiceTest.java

I remember they passed on jdk8, but it seems we have a regression in jdk9 and 
both of them fail. 

On 14.03.16 8:05, Avik Niyogi wrote: 



Hi All, 
A gentle reminder, please review my code changes. 

With Regards, 
Avik Niyogi 



On 08-Mar-2016, at 9:39 pm, Avik Niyogi mailto:avik.niy...@oracle.com"avik.niy...@oracle.com 
HYPERLINK "mailto:avik.niy...@oracle.com;> 
wrote: 

Hi All, 

Kindly review the bug fix for JDK 9. 

*Bug:* 

_https://bugs.openjdk.java.net/browse/JDK-8148555_ 
_ 
_ 
*Webrev:* 

_HYPERLINK 
"http://cr.openjdk.java.net/%7Eaniyogi/8148555/webrev.00/_"http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/_
 

*Issue:* 
Emoji selection in Character Viewer was causing exception in JNI 

*Cause:* 
Emojis are considered to be of different class type (namely, 
NSConcreteMutableAttributedString) from NSString which other 
characters are because of a surrogate pair for them. 

*Fix:* 
Major changes done for condition of emojis in JNI. Albeit rendering is 
not yet supported, they will appear as blank “Missing font” notation. 
Also, added debug point in case of issue with glyph arrises. 

With Regards, 
Avik Niyogi 

 

 

 

 

 

 

 

Re: Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-21 Thread Avik Niyogi 
Hi Manajit,
Please review the following code changes.

With Regards,
Avik Niyogi
> On 21-Mar-2016, at 12:20 pm, Avik Niyogi  wrote:
> 
> Hi Rajeev,
> Please review the following code changes.
> 
> With Regards,
> Avik Niyogi
>> On 21-Mar-2016, at 12:17 pm, Avik Niyogi > > wrote:
>> 
>> Hi Sergey,
>> 
>> Please review the following code changes.
>> With Regards,
>> Avik Niyogi
>>> On 17-Mar-2016, at 7:03 pm, Alexander Scherbatiy 
>>> > 
>>> wrote:
>>> 
>>> 
>>> The fix looks good to me.
>>> 
>>> Just a small note: it is better to remove comment "527 //" since it 
>>> does not have a description.
>>> 
>>> Thanks,
>>> Alexandr.
>>> 
>>> On 17/03/16 17:21, Avik Niyogi wrote:
 It can be made into a class method, but herein this case it is needed for 
 that instance only and hence the need for instance method and referred 
 with “self”.
 
 With Regards,
 Avik Niyogi
> On 16-Mar-2016, at 11:55 pm, Alexander Scherbatiy 
> > 
> wrote:
> 
> 
> Could the -(NSMutableString *) parseString: method be declared as class 
> method instead of instance?
> 
> Thanks,
> Alexandr.
> 
> On 14/03/16 17:18, Sergey Bylokhov wrote:
>> Hi, Avik. 
>> Can you please take a look to these two tests before fixing this bug: 
>> 
>> TEST: javax/swing/JMenuItem/8139169/ScreenMenuBarInputTwice.java 
>> -- 
>> TEST: 
>> javax/swing/JMenuItem/ActionListenerCalledTwice/ActionListenerCalledTwiceTest.java
>> 
>> I remember they passed on jdk8, but it seems we have a regression in 
>> jdk9 and both of them fail. 
>> 
>> On 14.03.16 8:05, Avik Niyogi wrote: 
>>> Hi All, 
>>> A gentle reminder, please review my code changes. 
>>> 
>>> With Regards, 
>>> Avik Niyogi 
 On 08-Mar-2016, at 9:39 pm, Avik Niyogi  
  > 
 wrote: 
 
 Hi All, 
 
 Kindly review the bug fix for JDK 9. 
 
 *Bug:* 
 
 _https://bugs.openjdk.java.net/browse/JDK-8148555_ 
  
 _ 
 _ 
 *Webrev:* 
 
 _http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/_ 
  
 
 *Issue:* 
 Emoji selection in Character Viewer was causing exception in JNI 
 
 *Cause:* 
 Emojis are considered to be of different class type (namely, 
 NSConcreteMutableAttributedString) from NSString which other 
 characters are because of a surrogate pair for them. 
 
 *Fix:* 
 Major changes done for condition of emojis in JNI. Albeit rendering is 
 not yet supported, they will appear as blank “Missing font” notation. 
 Also, added debug point in case of issue with glyph arrises. 
 
 With Regards, 
 Avik Niyogi 
>>> 
>> 
>> 
> 
 
>>> 
>> 
>

Re: Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-21 Thread Avik Niyogi 
Hi Rajeev,
Please review the following code changes.

With Regards,
Avik Niyogi
> On 21-Mar-2016, at 12:17 pm, Avik Niyogi  wrote:
> 
> Hi Sergey,
> 
> Please review the following code changes.
> With Regards,
> Avik Niyogi
>> On 17-Mar-2016, at 7:03 pm, Alexander Scherbatiy 
>> > 
>> wrote:
>> 
>> 
>> The fix looks good to me.
>> 
>> Just a small note: it is better to remove comment "527 //" since it 
>> does not have a description.
>> 
>> Thanks,
>> Alexandr.
>> 
>> On 17/03/16 17:21, Avik Niyogi wrote:
>>> It can be made into a class method, but herein this case it is needed for 
>>> that instance only and hence the need for instance method and referred with 
>>> “self”.
>>> 
>>> With Regards,
>>> Avik Niyogi
 On 16-Mar-2016, at 11:55 pm, Alexander Scherbatiy 
 > 
 wrote:
 
 
 Could the -(NSMutableString *) parseString: method be declared as class 
 method instead of instance?
 
 Thanks,
 Alexandr.
 
 On 14/03/16 17:18, Sergey Bylokhov wrote:
> Hi, Avik. 
> Can you please take a look to these two tests before fixing this bug: 
> 
> TEST: javax/swing/JMenuItem/8139169/ScreenMenuBarInputTwice.java 
> -- 
> TEST: 
> javax/swing/JMenuItem/ActionListenerCalledTwice/ActionListenerCalledTwiceTest.java
> 
> I remember they passed on jdk8, but it seems we have a regression in jdk9 
> and both of them fail. 
> 
> On 14.03.16 8:05, Avik Niyogi wrote: 
>> Hi All, 
>> A gentle reminder, please review my code changes. 
>> 
>> With Regards, 
>> Avik Niyogi 
>>> On 08-Mar-2016, at 9:39 pm, Avik Niyogi >>  
>>>  > wrote: 
>>> 
>>> Hi All, 
>>> 
>>> Kindly review the bug fix for JDK 9. 
>>> 
>>> *Bug:* 
>>> 
>>> _https://bugs.openjdk.java.net/browse/JDK-8148555_ 
>>>  
>>> _ 
>>> _ 
>>> *Webrev:* 
>>> 
>>> _http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/_ 
>>>  
>>> 
>>> *Issue:* 
>>> Emoji selection in Character Viewer was causing exception in JNI 
>>> 
>>> *Cause:* 
>>> Emojis are considered to be of different class type (namely, 
>>> NSConcreteMutableAttributedString) from NSString which other 
>>> characters are because of a surrogate pair for them. 
>>> 
>>> *Fix:* 
>>> Major changes done for condition of emojis in JNI. Albeit rendering is 
>>> not yet supported, they will appear as blank “Missing font” notation. 
>>> Also, added debug point in case of issue with glyph arrises. 
>>> 
>>> With Regards, 
>>> Avik Niyogi 
>> 
> 
> 
 
>>> 
>> 
>

Re: Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-21 Thread Avik Niyogi 
Hi Sergey,

Please review the following code changes.
With Regards,
Avik Niyogi
> On 17-Mar-2016, at 7:03 pm, Alexander Scherbatiy 
>  wrote:
> 
> 
> The fix looks good to me.
> 
> Just a small note: it is better to remove comment "527 //" since it 
> does not have a description.
> 
> Thanks,
> Alexandr.
> 
> On 17/03/16 17:21, Avik Niyogi wrote:
>> It can be made into a class method, but herein this case it is needed for 
>> that instance only and hence the need for instance method and referred with 
>> “self”.
>> 
>> With Regards,
>> Avik Niyogi
>>> On 16-Mar-2016, at 11:55 pm, Alexander Scherbatiy 
>>> > 
>>> wrote:
>>> 
>>> 
>>> Could the -(NSMutableString *) parseString: method be declared as class 
>>> method instead of instance?
>>> 
>>> Thanks,
>>> Alexandr.
>>> 
>>> On 14/03/16 17:18, Sergey Bylokhov wrote:
 Hi, Avik. 
 Can you please take a look to these two tests before fixing this bug: 
 
 TEST: javax/swing/JMenuItem/8139169/ScreenMenuBarInputTwice.java 
 -- 
 TEST: 
 javax/swing/JMenuItem/ActionListenerCalledTwice/ActionListenerCalledTwiceTest.java
 
 I remember they passed on jdk8, but it seems we have a regression in jdk9 
 and both of them fail. 
 
 On 14.03.16 8:05, Avik Niyogi wrote: 
> Hi All, 
> A gentle reminder, please review my code changes. 
> 
> With Regards, 
> Avik Niyogi 
>> On 08-Mar-2016, at 9:39 pm, Avik Niyogi >  
>>  > wrote: 
>> 
>> Hi All, 
>> 
>> Kindly review the bug fix for JDK 9. 
>> 
>> *Bug:* 
>> 
>> _https://bugs.openjdk.java.net/browse/JDK-8148555_ 
>>  
>> _ 
>> _ 
>> *Webrev:* 
>> 
>> _http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/_ 
>>  
>> 
>> *Issue:* 
>> Emoji selection in Character Viewer was causing exception in JNI 
>> 
>> *Cause:* 
>> Emojis are considered to be of different class type (namely, 
>> NSConcreteMutableAttributedString) from NSString which other 
>> characters are because of a surrogate pair for them. 
>> 
>> *Fix:* 
>> Major changes done for condition of emojis in JNI. Albeit rendering is 
>> not yet supported, they will appear as blank “Missing font” notation. 
>> Also, added debug point in case of issue with glyph arrises. 
>> 
>> With Regards, 
>> Avik Niyogi 
> 
 
 
>>> 
>> 
>

Re: Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-19 Thread Avik Niyogi 
Hi Sergey,
I have tested the test cases as mentioned.
1) They are not related to the current fix
2) ScreenMenuBarInputTwice.java issue is not reproducible.
3) There is a test bug in ActionListenerCalledTwiceTest.java, hence a new test 
case called ScreenMenuBarInputTwice.java 
was created to encompass actual expectations from Java on OS X as per native 
behaviour.

Please review my current fix as available in the mail trail. Thank you in 
advance.

With Regards,
Avik Niyogi


> On 14-Mar-2016, at 6:48 pm, Sergey Bylokhov  
> wrote:
> 
> Hi, Avik.
> Can you please take a look to these two tests before fixing this bug:
> 
> TEST: javax/swing/JMenuItem/8139169/ScreenMenuBarInputTwice.java
> --
> TEST: 
> javax/swing/JMenuItem/ActionListenerCalledTwice/ActionListenerCalledTwiceTest.java
> 
> I remember they passed on jdk8, but it seems we have a regression in jdk9 and 
> both of them fail.
> 
> On 14.03.16 8:05, Avik Niyogi wrote:
>> Hi All,
>> A gentle reminder, please review my code changes.
>> 
>> With Regards,
>> Avik Niyogi
>>> On 08-Mar-2016, at 9:39 pm, Avik Niyogi >> 
>>> >> wrote:
>>> 
>>> Hi All,
>>> 
>>> Kindly review the bug fix for JDK 9.
>>> 
>>> *Bug:*
>>> 
>>> _https://bugs.openjdk.java.net/browse/JDK-8148555_ 
>>> 
>>> _
>>> _
>>> *Webrev:*
>>> 
>>> _http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/_
>>> 
>>> *Issue:*
>>> Emoji selection in Character Viewer was causing exception in JNI
>>> 
>>> *Cause:*
>>> Emojis are considered to be of different class type (namely,
>>> NSConcreteMutableAttributedString) from NSString which other
>>> characters are because of a surrogate pair for them.
>>> 
>>> *Fix:*
>>> Major changes done for condition of emojis in JNI. Albeit rendering is
>>> not yet supported, they will appear as blank “Missing font” notation.
>>> Also, added debug point in case of issue with glyph arrises.
>>> 
>>> With Regards,
>>> Avik Niyogi
>> 
> 
> 
> -- 
> Best regards, Sergey.

Re: Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-19 Thread Avik Niyogi 
It can be made into a class method, but herein this case it is needed for that 
instance only and hence the need for instance method and referred with “self”.

With Regards,
Avik Niyogi
> On 16-Mar-2016, at 11:55 pm, Alexander Scherbatiy 
>  wrote:
> 
> 
> Could the -(NSMutableString *) parseString: method be declared as class 
> method instead of instance?
> 
> Thanks,
> Alexandr.
> 
> On 14/03/16 17:18, Sergey Bylokhov wrote:
>> Hi, Avik. 
>> Can you please take a look to these two tests before fixing this bug: 
>> 
>> TEST: javax/swing/JMenuItem/8139169/ScreenMenuBarInputTwice.java 
>> -- 
>> TEST: 
>> javax/swing/JMenuItem/ActionListenerCalledTwice/ActionListenerCalledTwiceTest.java
>> 
>> I remember they passed on jdk8, but it seems we have a regression in jdk9 
>> and both of them fail. 
>> 
>> On 14.03.16 8:05, Avik Niyogi wrote: 
>>> Hi All, 
>>> A gentle reminder, please review my code changes. 
>>> 
>>> With Regards, 
>>> Avik Niyogi 
 On 08-Mar-2016, at 9:39 pm, Avik Niyogi  
  > wrote: 
 
 Hi All, 
 
 Kindly review the bug fix for JDK 9. 
 
 *Bug:* 
 
 _https://bugs.openjdk.java.net/browse/JDK-8148555_ 
  
 _ 
 _ 
 *Webrev:* 
 
 _http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/_ 
  
 
 *Issue:* 
 Emoji selection in Character Viewer was causing exception in JNI 
 
 *Cause:* 
 Emojis are considered to be of different class type (namely, 
 NSConcreteMutableAttributedString) from NSString which other 
 characters are because of a surrogate pair for them. 
 
 *Fix:* 
 Major changes done for condition of emojis in JNI. Albeit rendering is 
 not yet supported, they will appear as blank “Missing font” notation. 
 Also, added debug point in case of issue with glyph arrises. 
 
 With Regards, 
 Avik Niyogi 
>>> 
>> 
>> 
>

Re: Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-19 Thread Alexander Scherbatiy 


Could the -(NSMutableString *) parseString: method be declared as class 
method instead of instance?


Thanks,
Alexandr.

On 14/03/16 17:18, Sergey Bylokhov wrote:

Hi, Avik.
Can you please take a look to these two tests before fixing this bug:

TEST: javax/swing/JMenuItem/8139169/ScreenMenuBarInputTwice.java
--
TEST: 
javax/swing/JMenuItem/ActionListenerCalledTwice/ActionListenerCalledTwiceTest.java


I remember they passed on jdk8, but it seems we have a regression in 
jdk9 and both of them fail.


On 14.03.16 8:05, Avik Niyogi wrote:

Hi All,
A gentle reminder, please review my code changes.

With Regards,
Avik Niyogi

On 08-Mar-2016, at 9:39 pm, Avik Niyogi > wrote:

Hi All,

Kindly review the bug fix for JDK 9.

*Bug:*

_https://bugs.openjdk.java.net/browse/JDK-8148555_
_
_
*Webrev:*

_http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/_

*Issue:*
Emoji selection in Character Viewer was causing exception in JNI

*Cause:*
Emojis are considered to be of different class type (namely,
NSConcreteMutableAttributedString) from NSString which other
characters are because of a surrogate pair for them.

*Fix:*
Major changes done for condition of emojis in JNI. Albeit rendering is
not yet supported, they will appear as blank “Missing font” notation.
Also, added debug point in case of issue with glyph arrises.

With Regards,
Avik Niyogi

Re: Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-19 Thread Alexander Scherbatiy 


The fix looks good to me.

Just a small note: it is better to remove comment "527 //" since 
it does not have a description.


Thanks,
Alexandr.

On 17/03/16 17:21, Avik Niyogi wrote:
It can be made into a class method, but herein this case it is needed 
for that instance only and hence the need for instance method and 
referred with “self”.


With Regards,
Avik Niyogi
On 16-Mar-2016, at 11:55 pm, Alexander Scherbatiy 
> wrote:



Could the -(NSMutableString *) parseString: method be declared as 
class method instead of instance?


Thanks,
Alexandr.

On 14/03/16 17:18, Sergey Bylokhov wrote:

Hi, Avik.
Can you please take a look to these two tests before fixing this bug:

TEST: javax/swing/JMenuItem/8139169/ScreenMenuBarInputTwice.java
--
TEST: 
javax/swing/JMenuItem/ActionListenerCalledTwice/ActionListenerCalledTwiceTest.java


I remember they passed on jdk8, but it seems we have a regression in 
jdk9 and both of them fail.


On 14.03.16 8:05, Avik Niyogi wrote:

Hi All,
A gentle reminder, please review my code changes.

With Regards,
Avik Niyogi

On 08-Mar-2016, at 9:39 pm, Avik Niyogi > wrote:

Hi All,

Kindly review the bug fix for JDK 9.

*Bug:*

_https://bugs.openjdk.java.net/browse/JDK-8148555_
_
_
*Webrev:*

_http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/_

*Issue:*
Emoji selection in Character Viewer was causing exception in JNI

*Cause:*
Emojis are considered to be of different class type (namely,
NSConcreteMutableAttributedString) from NSString which other
characters are because of a surrogate pair for them.

*Fix:*
Major changes done for condition of emojis in JNI. Albeit 
rendering is

not yet supported, they will appear as blank “Missing font” notation.
Also, added debug point in case of issue with glyph arrises.

With Regards,
Avik Niyogi

Re: Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-14 Thread Sergey Bylokhov 

Hi, Avik.
Can you please take a look to these two tests before fixing this bug:

TEST: javax/swing/JMenuItem/8139169/ScreenMenuBarInputTwice.java
--
TEST: 
javax/swing/JMenuItem/ActionListenerCalledTwice/ActionListenerCalledTwiceTest.java


I remember they passed on jdk8, but it seems we have a regression in 
jdk9 and both of them fail.


On 14.03.16 8:05, Avik Niyogi wrote:

Hi All,
A gentle reminder, please review my code changes.

With Regards,
Avik Niyogi

On 08-Mar-2016, at 9:39 pm, Avik Niyogi > wrote:

Hi All,

Kindly review the bug fix for JDK 9.

*Bug:*

_https://bugs.openjdk.java.net/browse/JDK-8148555_
_
_
*Webrev:*

_http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/_

*Issue:*
Emoji selection in Character Viewer was causing exception in JNI

*Cause:*
Emojis are considered to be of different class type (namely,
NSConcreteMutableAttributedString) from NSString which other
characters are because of a surrogate pair for them.

*Fix:*
Major changes done for condition of emojis in JNI. Albeit rendering is
not yet supported, they will appear as blank “Missing font” notation.
Also, added debug point in case of issue with glyph arrises.

With Regards,
Avik Niyogi





--
Best regards, Sergey.

Review Request of 8148555: [macosx] An uncaught exception was raised entering Emoji into JTextArea

2016-03-08 Thread Avik Niyogi 
Hi All,

Kindly review the bug fix for JDK 9.

Bug:

https://bugs.openjdk.java.net/browse/JDK-8148555 


Webrev:

http://cr.openjdk.java.net/~aniyogi/8148555/webrev.00/ 


Issue:
Emoji selection in Character Viewer was causing exception in JNI

Cause:
Emojis are considered to be of different class type (namely, 
NSConcreteMutableAttributedString) from NSString which other characters are 
because of a surrogate pair for them.

Fix:
Major changes done for condition of emojis in JNI. Albeit rendering is not yet 
supported, they will appear as blank “Missing font” notation.
Also, added debug point in case of issue with glyph arrises.

With Regards,
Avik Niyogi