#80 Fixed bugs related to exponential format and end-of-number processing and added support for CP_NUM type
nvalid exponents may be confused with text to skip. Examples: “1.2E” and “1.2E+”, which are both treated as valid numbers. Also “1.2egg” was considered to have the text “gg” following the number rather than the text “egg”. Cases similar to these may affect results depending on the “whole” argument and the ft_strictnumparse flag.
The action based on ft_strictnumparse was not as documented and was changed to comply.
If the string being converted is an integer and if that integer can be represented as an int without truncation, the return value is set to +1 to indicate that tne numbe can be converted to a CP_NUM.
So now the return codes for this function are -1 on conversion failure, +1 on conversion of a string that can be a CP_NUM and 0 for the case where the number was not in integer format (e.g., 1.) or not an integer.
This fix was included with the fix for patch #74.