Re: Fwd: Re: RFR: 8229337: java.lang.Math class doc should be adjusted regarding -Exact methods

2019-08-13 Thread Julia Boes 

Hi Roger,

Changed - Thanks for reviewing!

Cheers,

Julia

On 13/08/2019 14:48, Roger Riggs wrote:

Hi Julia,

Looks good.

In the compatibility description, I would lead with the "No behavioral 
change".
Since this is part of the specification, the code (implementation) is 
not relevant to the change.


Thanks, Roger


On 8/13/19 9:33 AM, Julia Boes wrote:

Forwarding this thread to the right list!



 Forwarded Message 
Subject: Re: RFR: 8229337: java.lang.Math class doc should be 
adjusted regarding -Exact methods

Date: Tue, 13 Aug 2019 13:51:02 +0100
From: Julia Boes 
To: Lance Andersen 
CC: jdk-...@openjdk.java.net



Hi Lance,

Thanks for your suggestions. I incorporated them, except for the 
change of the compatibility risk because that Jira doesn't accept the 
option 'None' for this field :(


Best,

Julia

On 13/08/2019 12:55, Lance Andersen wrote:

Hi Julia,

I added myself as a reviewer.

A couple of suggestions to consider:

Update:

——
A minor fix of the javadoc in java.lang.Math. The methods 
decrementExact(), incrementExact(), and negateExact()  declare they 
throw a ArithmeticException, and as such should be included in the 
list of operations that can detect overflow in the class-level javadoc.



to

———
Clarify the javadoc class summary for java.lang.Math class  to 
indicate that the methods s decrementExact(), incrementExact() and 
negateExact()   throw a ArithmeticException when the results overflow



For compatibility risk it  would be none (there is no behavioral 
change);


For the compatibility description  I would also clarify that thereis 
no behavioral change


Compatibility kind is probably best as behavioral given you are 
clarifying what actually can happen for the methods.



You are good to go and you can finalize as this will be an easy 
(famous last words) review/approval :-)


HTH

Best
Lance

On Aug 13, 2019, at 7:32 AM, Julia Boes > wrote:


Hi Lance,

Thanks for the info.

CSR: https://bugs.openjdk.java.net/browse/JDK-8229473

Best,

Julia

On 13/08/2019 11:25, Lance Andersen wrote:

Hi Julia,

I think this looks OK.  I would create a CSR as the changes while 
minor should get tracked and a final approval via a CSR.


Best
Lance
On Aug 13, 2019, at 6:16 AM, Julia Boes > wrote:


Hi,

This is a minor fix of the javadoc in java.lang.Math. The methods 
decrementExact(), incrementExact() and negateExact()  declare 
they throw a ArithmeticException, and as such should be included 
in the list of operations that can detect overflow in the 
class-level javadoc.


Bug: https://bugs.openjdk.java.net/browse/JDK-8229337
Webrev: http://cr.openjdk.java.net/~dfuchs/jboes/8229337/webrev.00/

Is a CSR request needed?

Thanks,
Julia




 

 

Lance 
Andersen| Principal Member of Technical Staff | +1.781.442.2037

Oracle Java Engineering
1 Network Drive
Burlington, MA 01803
lance.ander...@oracle.com 






 

Lance 
Andersen| Principal Member of Technical Staff | +1.781.442.2037

Oracle Java Engineering
1 Network Drive
Burlington, MA 01803
lance.ander...@oracle.com

Re: Fwd: Re: RFR: 8229337: java.lang.Math class doc should be adjusted regarding -Exact methods

2019-08-13 Thread Roger Riggs 

Hi Julia,

Looks good.

In the compatibility description, I would lead with the "No behavioral 
change".
Since this is part of the specification, the code (implementation) is 
not relevant to the change.


Thanks, Roger


On 8/13/19 9:33 AM, Julia Boes wrote:

Forwarding this thread to the right list!



 Forwarded Message 
Subject: Re: RFR: 8229337: java.lang.Math class doc should be 
adjusted regarding -Exact methods

Date: Tue, 13 Aug 2019 13:51:02 +0100
From: Julia Boes 
To: Lance Andersen 
CC: jdk-...@openjdk.java.net



Hi Lance,

Thanks for your suggestions. I incorporated them, except for the 
change of the compatibility risk because that Jira doesn't accept the 
option 'None' for this field :(


Best,

Julia

On 13/08/2019 12:55, Lance Andersen wrote:

Hi Julia,

I added myself as a reviewer.

A couple of suggestions to consider:

Update:

——
A minor fix of the javadoc in java.lang.Math. The methods 
decrementExact(), incrementExact(), and negateExact()  declare they 
throw a ArithmeticException, and as such should be included in the 
list of operations that can detect overflow in the class-level javadoc.



to

———
Clarify the javadoc class summary for java.lang.Math class  to 
indicate that the methods s decrementExact(), incrementExact() and 
negateExact()   throw a ArithmeticException when the results overflow



For compatibility risk it  would be none (there is no behavioral 
change);


For the compatibility description  I would also clarify that thereis 
no behavioral change


Compatibility kind is probably best as behavioral given you are 
clarifying what actually can happen for the methods.



You are good to go and you can finalize as this will be an easy 
(famous last words) review/approval :-)


HTH

Best
Lance

On Aug 13, 2019, at 7:32 AM, Julia Boes > wrote:


Hi Lance,

Thanks for the info.

CSR: https://bugs.openjdk.java.net/browse/JDK-8229473

Best,

Julia

On 13/08/2019 11:25, Lance Andersen wrote:

Hi Julia,

I think this looks OK.  I would create a CSR as the changes while 
minor should get tracked and a final approval via a CSR.


Best
Lance
On Aug 13, 2019, at 6:16 AM, Julia Boes > wrote:


Hi,

This is a minor fix of the javadoc in java.lang.Math. The methods 
decrementExact(), incrementExact() and negateExact()  declare they 
throw a ArithmeticException, and as such should be included in the 
list of operations that can detect overflow in the class-level 
javadoc.


Bug: https://bugs.openjdk.java.net/browse/JDK-8229337
Webrev: http://cr.openjdk.java.net/~dfuchs/jboes/8229337/webrev.00/

Is a CSR request needed?

Thanks,
Julia




 

 

Lance 
Andersen| Principal Member of Technical Staff | +1.781.442.2037

Oracle Java Engineering
1 Network Drive
Burlington, MA 01803
lance.ander...@oracle.com 






 

Lance 
Andersen| Principal Member of Technical Staff | +1.781.442.2037

Oracle Java Engineering
1 Network Drive
Burlington, MA 01803
lance.ander...@oracle.com