[
https://issues.apache.org/jira/browse/MATH-744?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13206514#comment-13206514
]
Thundre commented on MATH-744:
------------------------------
I don't have the test classes working in Eclipse, so I can't do it there.
Here's a main method that demonstrates. (10^401+1)/(10^400) is just a smidgen
more than 10.
public static void main(String args[]) {
double x = new BigFraction(BigInteger.TEN.pow(401).add(BigInteger.ONE),
BigInteger.TEN.pow(400)).doubleValue();
System.out.println(x);
System.exit(0);
}
> BigFraction.doubleValue() returns Double.NaN for large numerators or
> denominators
> ---------------------------------------------------------------------------------
>
> Key: MATH-744
> URL: https://issues.apache.org/jira/browse/MATH-744
> Project: Commons Math
> Issue Type: Bug
> Affects Versions: 2.2
> Reporter: Thundre
> Attachments: site.patch
>
>
> The current implementation of doubleValue() divides numerator.doubleValue() /
> denominator.doubleValue(). BigInteger.doubleValue() fails for any number
> greater than Double.MAX_VALUE. So if the user has 308-digit numerator or
> denominator, the resulting quotient fails, even in cases where the result
> would be well inside Double's range.
> I have a patch to fix it, if I can figure out how to attach it here I will.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
