You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@harmony.apache.org by "Catherine Hope (JIRA)" <ji...@apache.org> on 2009/07/13 19:21:15 UTC

[jira] Created: (HARMONY-6271) [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales

[classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales
------------------------------------------------------------------------------------------------------------------

                 Key: HARMONY-6271
                 URL: https://issues.apache.org/jira/browse/HARMONY-6271
             Project: Harmony
          Issue Type: Bug
          Components: Classlib
         Environment: all
            Reporter: Catherine Hope
            Priority: Minor


BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory when called with a scale that can't be satisfied with the available memory.  The exception defined in the specification is:
"throws ArithmeticException - if divisor is zero, roundingMode==RoundingMode.UNNECESSARY and the specified scale is insufficient to represent the result of the division exactly."
Which doesn't say what to do in this precise occurrence (though it doesn't say it may throw an OutOfMemoryError), but both Sun and IBM Java implementations have this behaviour.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Updated: (HARMONY-6271) [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales

Posted by "Catherine Hope (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HARMONY-6271?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Catherine Hope updated HARMONY-6271:
------------------------------------

    Attachment: 6271-patch.txt

Fix testcase error

> [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales
> ------------------------------------------------------------------------------------------------------------------
>
>                 Key: HARMONY-6271
>                 URL: https://issues.apache.org/jira/browse/HARMONY-6271
>             Project: Harmony
>          Issue Type: Bug
>          Components: Classlib
>         Environment: all
>            Reporter: Catherine Hope
>            Assignee: Mark Hindess
>            Priority: Minor
>         Attachments: 6271-patch.txt, bigdecimal-patch.txt, bigdecimal-patch.txt
>
>
> BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory when called with a scale that can't be satisfied with the available memory.  The exception defined in the specification is:
> "throws ArithmeticException - if divisor is zero, roundingMode==RoundingMode.UNNECESSARY and the specified scale is insufficient to represent the result of the division exactly."
> Which doesn't say what to do in this precise occurrence (though it doesn't say it may throw an OutOfMemoryError), but both Sun and IBM Java implementations have this behaviour.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Updated: (HARMONY-6271) [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales

Posted by "Catherine Hope (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HARMONY-6271?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Catherine Hope updated HARMONY-6271:
------------------------------------

    Attachment: bigdecimal-patch.txt

Fix simply to change the OutOfMemoryError to an ArithmeticException (the error message remains the same as this shouldn't be matched).
Test added to BigIntegerDivideTest.java (testCase25)

> [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales
> ------------------------------------------------------------------------------------------------------------------
>
>                 Key: HARMONY-6271
>                 URL: https://issues.apache.org/jira/browse/HARMONY-6271
>             Project: Harmony
>          Issue Type: Bug
>          Components: Classlib
>         Environment: all
>            Reporter: Catherine Hope
>            Priority: Minor
>         Attachments: bigdecimal-patch.txt
>
>
> BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory when called with a scale that can't be satisfied with the available memory.  The exception defined in the specification is:
> "throws ArithmeticException - if divisor is zero, roundingMode==RoundingMode.UNNECESSARY and the specified scale is insufficient to represent the result of the division exactly."
> Which doesn't say what to do in this precise occurrence (though it doesn't say it may throw an OutOfMemoryError), but both Sun and IBM Java implementations have this behaviour.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Commented: (HARMONY-6271) [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales

Posted by "Mark Hindess (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/HARMONY-6271?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12736035#action_12736035 ] 

Mark Hindess commented on HARMONY-6271:
---------------------------------------

Applied patch in r798486.  Please confirm by closing this JIRA that the patch was applied as expected.


> [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales
> ------------------------------------------------------------------------------------------------------------------
>
>                 Key: HARMONY-6271
>                 URL: https://issues.apache.org/jira/browse/HARMONY-6271
>             Project: Harmony
>          Issue Type: Bug
>          Components: Classlib
>         Environment: all
>            Reporter: Catherine Hope
>            Assignee: Mark Hindess
>            Priority: Minor
>         Attachments: 6271-patch.txt, bigdecimal-patch.txt, bigdecimal-patch.txt
>
>
> BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory when called with a scale that can't be satisfied with the available memory.  The exception defined in the specification is:
> "throws ArithmeticException - if divisor is zero, roundingMode==RoundingMode.UNNECESSARY and the specified scale is insufficient to represent the result of the division exactly."
> Which doesn't say what to do in this precise occurrence (though it doesn't say it may throw an OutOfMemoryError), but both Sun and IBM Java implementations have this behaviour.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Updated: (HARMONY-6271) [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales

Posted by "Catherine Hope (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HARMONY-6271?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Catherine Hope updated HARMONY-6271:
------------------------------------

    Attachment: bigdecimal-patch.txt

Please ignore previous patch - I added the testcase to the wrong junit class.  Now moved in org.apache.harmony.tests.java.math.BigDecimalArithmeticTest

> [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales
> ------------------------------------------------------------------------------------------------------------------
>
>                 Key: HARMONY-6271
>                 URL: https://issues.apache.org/jira/browse/HARMONY-6271
>             Project: Harmony
>          Issue Type: Bug
>          Components: Classlib
>         Environment: all
>            Reporter: Catherine Hope
>            Priority: Minor
>         Attachments: bigdecimal-patch.txt, bigdecimal-patch.txt
>
>
> BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory when called with a scale that can't be satisfied with the available memory.  The exception defined in the specification is:
> "throws ArithmeticException - if divisor is zero, roundingMode==RoundingMode.UNNECESSARY and the specified scale is insufficient to represent the result of the division exactly."
> Which doesn't say what to do in this precise occurrence (though it doesn't say it may throw an OutOfMemoryError), but both Sun and IBM Java implementations have this behaviour.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Closed: (HARMONY-6271) [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales

Posted by "Catherine Hope (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HARMONY-6271?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Catherine Hope closed HARMONY-6271.
-----------------------------------


Verified

> [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales
> ------------------------------------------------------------------------------------------------------------------
>
>                 Key: HARMONY-6271
>                 URL: https://issues.apache.org/jira/browse/HARMONY-6271
>             Project: Harmony
>          Issue Type: Bug
>          Components: Classlib
>         Environment: all
>            Reporter: Catherine Hope
>            Assignee: Mark Hindess
>            Priority: Minor
>             Fix For: 5.0M11
>
>         Attachments: 6271-patch.txt, bigdecimal-patch.txt, bigdecimal-patch.txt
>
>
> BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory when called with a scale that can't be satisfied with the available memory.  The exception defined in the specification is:
> "throws ArithmeticException - if divisor is zero, roundingMode==RoundingMode.UNNECESSARY and the specified scale is insufficient to represent the result of the division exactly."
> Which doesn't say what to do in this precise occurrence (though it doesn't say it may throw an OutOfMemoryError), but both Sun and IBM Java implementations have this behaviour.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Assigned: (HARMONY-6271) [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales

Posted by "Mark Hindess (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HARMONY-6271?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Mark Hindess reassigned HARMONY-6271:
-------------------------------------

    Assignee: Mark Hindess

> [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales
> ------------------------------------------------------------------------------------------------------------------
>
>                 Key: HARMONY-6271
>                 URL: https://issues.apache.org/jira/browse/HARMONY-6271
>             Project: Harmony
>          Issue Type: Bug
>          Components: Classlib
>         Environment: all
>            Reporter: Catherine Hope
>            Assignee: Mark Hindess
>            Priority: Minor
>         Attachments: bigdecimal-patch.txt, bigdecimal-patch.txt
>
>
> BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory when called with a scale that can't be satisfied with the available memory.  The exception defined in the specification is:
> "throws ArithmeticException - if divisor is zero, roundingMode==RoundingMode.UNNECESSARY and the specified scale is insufficient to represent the result of the division exactly."
> Which doesn't say what to do in this precise occurrence (though it doesn't say it may throw an OutOfMemoryError), but both Sun and IBM Java implementations have this behaviour.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Commented: (HARMONY-6271) [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales

Posted by "Mark Hindess (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/HARMONY-6271?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12735613#action_12735613 ] 

Mark Hindess commented on HARMONY-6271:
---------------------------------------

When I run this test on Sun 1.5 and Sun 6.0 (linux/x86_64) I get a NegativeArraySizeException rather than ArithmeticException.  Perhaps we should follow RI behaviour?  Having said that ArithmeticException does seem more appropriate.



> [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales
> ------------------------------------------------------------------------------------------------------------------
>
>                 Key: HARMONY-6271
>                 URL: https://issues.apache.org/jira/browse/HARMONY-6271
>             Project: Harmony
>          Issue Type: Bug
>          Components: Classlib
>         Environment: all
>            Reporter: Catherine Hope
>            Assignee: Mark Hindess
>            Priority: Minor
>         Attachments: bigdecimal-patch.txt, bigdecimal-patch.txt
>
>
> BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory when called with a scale that can't be satisfied with the available memory.  The exception defined in the specification is:
> "throws ArithmeticException - if divisor is zero, roundingMode==RoundingMode.UNNECESSARY and the specified scale is insufficient to represent the result of the division exactly."
> Which doesn't say what to do in this precise occurrence (though it doesn't say it may throw an OutOfMemoryError), but both Sun and IBM Java implementations have this behaviour.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Resolved: (HARMONY-6271) [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales

Posted by "Mark Hindess (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HARMONY-6271?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Mark Hindess resolved HARMONY-6271.
-----------------------------------

       Resolution: Fixed
    Fix Version/s: 5.0M11

> [classlib][math] BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory for too large scales
> ------------------------------------------------------------------------------------------------------------------
>
>                 Key: HARMONY-6271
>                 URL: https://issues.apache.org/jira/browse/HARMONY-6271
>             Project: Harmony
>          Issue Type: Bug
>          Components: Classlib
>         Environment: all
>            Reporter: Catherine Hope
>            Assignee: Mark Hindess
>            Priority: Minor
>             Fix For: 5.0M11
>
>         Attachments: 6271-patch.txt, bigdecimal-patch.txt, bigdecimal-patch.txt
>
>
> BigDecimal.divide should throw an ArithmeticException instead of OutOfMemory when called with a scale that can't be satisfied with the available memory.  The exception defined in the specification is:
> "throws ArithmeticException - if divisor is zero, roundingMode==RoundingMode.UNNECESSARY and the specified scale is insufficient to represent the result of the division exactly."
> Which doesn't say what to do in this precise occurrence (though it doesn't say it may throw an OutOfMemoryError), but both Sun and IBM Java implementations have this behaviour.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.