Details
-
Bug
-
Status: Resolved
-
Minor
-
Resolution: Fixed
-
2.10.0
-
None
Description
Multiple issues addressed:
1) Total digits for precision decimal is t in "values are expressible in scientific notation using at most t digits for the coefficient." So for "0.003", total digits is 1, not 3 or 4.
2) Precision decimal comparison. a) In compareTo(), if the other value is NaN, should return INDETERMINATE. b) If one value is +0 and the other -0, should return equal. c) In equals(), if both are NaN, treat as equal.
3) min/maxScale of precisionDecimal are always satisfied by the special values that don't have precision: NaN, INF, and -INF. The TypeValidator needs to answer whether a value has a precision.
These issues were fixed by Sandy in SVN rev 1031774, 1031800, 1031811.
Just opening the JIRA issue at this point for tracking purposes.