A new data type for Batch, that can represent large and floating-point numbers and enables calculation with those.

The basic syntax of Number.cmd looks like this:

Number <variable-name> = <operand1> <operator> <operand2> [precision:<digits>] [format:<pattern>]

• With <variable-name> being the plain string of the batch variable that should be set.

• With <operand1> and <operand2> being the two operands of the calculation. They can be either normal integers or numbers in the internally used notation with mantissa and exponent, seperated by an E.
  For example the number 1,23 would be 123E-2 in this notation. If you want, you can write normal integers in this notation as well: 1200 for example would be 12E2.
  The operands can also be constants like Number.pi or Number.e.

• And <operator> should be one of the following mathematical operators: +, -, *, /.

• If you want, you can also specify a custom precision by adding an additional parameter, starting with precision: or the short form p:. Then you can specify the amount of significant digits.
  For example 1 / 3 p:4 will give you an output with 4 digit precision: +3333E-4.
  (This works for all operations. Because division sometimes only terminates when specifying an maximum amount of digits, the precision has a default value of 8 that is used for division only.)

• To get the output in a specific format, you can provide a format pattern as optional argument.
  The argument needs to start with format: or, to keep it short, f:. Then you can specify the format-pattern for the output number: First the amount of digits before the floating point (leaving this unspecified will cause it to be dynamically set), then the symbol that should be used as floating point (has to be any other than [0-9]) and finally the amount of digits after the floating point (again you can leave this out for dynamic calcualtion).
  If you do not specify both amounts of digts and only provide a floating point symbol, the output will get adjusted completely dynamic and will never contain any exponents (exponent is adjusted to be zero and then omited).
  You can find more information about the format options here: #23 (comment).

Number x = 4 + 48
Number x = %x% * -E5
Number y = 50E5 + 2E5
Number result = %x% - %y% 
Number result = %result% / 3000 
Number result = %result% + 4 "format:."

echo result: %result%

Note: In batch files you have to call Number to continue execution.

Other examples for real use cases can be found in examples

The Number.cmd file does not need any dependencies and can be used as is without other files from this repository.

To use Number.cmd in your projects, you can of course manually download or clone the repository or just copy or clone the Number.cmd file. But the latest version of Number.cmd (and the whole repository) can also be found on git.tim-greller.de/number/Number.cmd. This allows you to import the latest version in your project using curl:

curl "https://tim-greller.de/git/number/Number.cmd" > Number.cmd

This way you won't have to add Number.cmd to your git-repository and you won't have to worry about manually checking for updates in order to recieve bugfixes or new features.

 0: Success
 1: First operand is not a number (NaN)
 2: Second operand is not a number (NaN)
 3: Unknown operator
 4: Missing parameter(s)

Number.cmd does not write anything to standard output (stdout) by default. If you want the result to be printed, for example when using Number.cmd directly in cmd, you can use # instead of a variable-name with an equal-sign.
The following command for example will output "+8E-1" to the standard output stream:

Number # 4 / 5