PRB: VAL() Does Not Return Error When Number Is Out of RangeLast reviewed: August 21, 1995Article ID: Q135115 |
The information in this article applies to:
SYMPTOMSAssigning a variable a number that is out of the allowed range for double precision numbers doesn't generate an error.
STATUSMicrosoft is researching this behavior and will post new information here in the Microsoft Knowledge Base as it becomes available.
MORE INFORMATION
Code to Reproduce BehaviorCreate and run this program: CREATE TABLE test (numeric N(10,5), float F(10,5), ; double D, currency M, integer I)APPEND BLANK nBig=VAL('1E+700') && largest double precision number has an exp of 308
? TYPE ('nBig') && prints 'N' ? nBig && prints ******** ? STR(nBig,203,7) && prints 203 *'s REPLACE numeric WITH nBig && Numeric overflow. Data was lost REPLACE float WITH nBig && Numeric overflow. Data was lost REPLACE double WITH nBig && data type mismatchREPLACE currency WITH nBig && data type mismatch REPLACE integer WITH nBig && Numeric overflow. Data was lost ? LOG(nBig) && prints 20 asterisks |
Additional reference words: 3.00 VFoxWin
© 1998 Microsoft Corporation. All rights reserved. Terms of Use. |