harmony-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sian January (JIRA)" <j...@apache.org>
Subject [jira] Updated: (HARMONY-319) BigDecimal and Biginteger test failures should be fixed
Date Tue, 11 Apr 2006 12:56:19 GMT
     [ http://issues.apache.org/jira/browse/HARMONY-319?page=all ]

Sian January updated HARMONY-319:

    Attachment: mathpatch.txt

I have opened a separate bug to track the endless loop problem, which is actually related
to the Double class (see HARMONY-329).  This patch fixes all the other problems seen for BigDecimal
and BigInteger.  The patch includes some test changes as well as some changes to BigDecimal
and BigInteger because some of the tests were testing for the wrong thing, and were also failing
on both IBM's and Sun's 1.5.0 implementations.  

I also found another problem with BigDecimal, where calling 'new BigDecimal("")' should throw
a NumberFormatException and was instead throwing a StringIndexOutOfBoundsException.  I added
test test_constructor_String_empty for this.

In the patch I have commented out the call that caused the endless loop (in BigDecimalTest.test_doubleValue())
so that the tests could be added back in to the suite if desired.

The patch should be applied to the modules directory.

> BigDecimal and Biginteger test failures should be fixed
> -------------------------------------------------------
>          Key: HARMONY-319
>          URL: http://issues.apache.org/jira/browse/HARMONY-319
>      Project: Harmony
>         Type: Bug

>   Components: Classlib
>     Reporter: Mark Hindess
>     Priority: Minor
>  Attachments: mathpatch.txt
> The original API tests from BigInteger and BigDecimal from the IBM implementation fail
on the new Intel implementation.  See the too exclusions in modules/math/make/common/build.xml
>     <!-- This test seems to loop -->
>     <exclude name="**/api/java/math/BigDecimalTest.java" />
>     <!-- This test fails -->
>     <exclude name="**/api/java/math/BigIntegerTest.java" />
> I commented on this when I attached the integration scripts/diffs but no one seems to
have picked up on it so I thought I'd better raise a JIRA to make sure it wasn't overlooked.

This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
For more information on JIRA, see:

View raw message