Negative Numbers and NumberFormat

[hh]

>>> Bob S. wrote:
>>> You should rather say the number format determines the num of DIGITS. 
>> hh wrote: 
>> That's what you wish to have (good idea). But it's not like that, its 
>> the length of the number as string. 
> Bob S. wrote
> Really? Then why does :
> set the numberformat to 0.0; put -.1 + 0 result in -0.1 ?  

You are mixing input methods and output methods.

"-.x" (or ".x") is allowed as input, but then converted to
"-0.x" (or "0.x"). Just the same for "00.x" or "-00.x"

*After* that the numberformat is applied (and doesn't cut if leading is
longer). That's why

  set the numberformat to "0.0"
  put 0-".1", 0+"-0.1", 0+"-00.1" after s -- you may leave out quotes
  set the numberformat to "00.0"
  put 0-".1", 0+"-0.1", 0+"-00.1" after s -- you may leave out quotes

both yield -0.1,-0.1,-0.1 (in LC 6/7/8/9).

To see more clear the string length is applied to the number:
You can use _ANY_ other one-byte-char other than the separator (".").

set the numberformat to "ab.c"
is equivalent to
set the numberformat to "00.0"

This is comfortable if you wish to have the numbers length
dependent on any string varibale's length:

  put "Bob.S" into str
  set the numberformat to str
  put 0-12 & " _and_ " & 0+13

yields -12.0 _and_ 013.0, as I said (in LC 6/7/8/9).

*** Once again, I like and support your idea. ***
But I suggest to name it a feature request, not a bug:

Perhaps the option to apply the numberformat to the absolute value
only and choose/apply a prefix: either +, -, space or empty?
The empty is for people who distinguish negative and positive nums
by labels only (debit and credit) or colors only (red and black).