Inconsistency in `string-to-number'

[Davis Herring]

(string-to-number "1:") => 1
(string-to-number "1.2:") => 1
(string-to-number "1.2") => 1.2

Because of the call to isfloat_string in Fstring_to_number, the lenience
shown integers is not extended to floats, with the bizarre result that the
function succeeds but ignores part of its input.

Should there be a lenient form of isfloat_string that suppresses the check
for whitespace or NUL after the float, or is this not part of
string-to-number's "documented behavior"?  (It isn't actually documented
either to allow garbage after ints or to forbid it after floats.)

Davis

-- 
This product is sold by volume, not by mass.  If it appears too dense or
too sparse, it is because mass-energy conversion has occurred during
shipping.