RE: [security-services] the Time Issue

From
Hal Lockhart <>
Date
2002-01-31T18:23:56+00:00
ID
Thread
RE: [security-services] the Time Issue
Title: RE: [security-services] the Time Issue

Rich,

> The issue is, basically, this.  XML is a "printed" 

> representation, so if 

> you send schema-typed data, you have all the standard round-off and 

> truncation kinds of issues for floating-point numbers.  For example, 

> what happens when your SOAP layer converts the XML text node 

> "2.99999999999999999" into a local floating-point number?  The XML 

> Schema spec makes no requirements.  And the problem is 

> compounded when 

> you convert the local floating-point number back out to an 

> XML text node 

> for output.

I am with Phill here. Assuming we have no requirement for round tripping (we can debate that seperately if you wish) I don't think there is a problem.

No matter how we represent it in XML it will have to be represented internally somehow. As long as this conversion does not change a less than relationship into a greater than relationship  or vice versa, I don't see an issue. I don't see doing math on the a seconds field of a date/time. I realize that 2.99999999999999999 might end up equal to 2.99999999999999998, but this will also happen if we limit the precision. 

Hal