Distinguish float and integer types from string

[Adam D. Ruppe]

On Saturday, 9 March 2019 at 18:11:09 UTC, Jacob Shtokolov wrote:w
> One of the task was to take a string from STDIN and detect its 
> type.

The way I'd do this is a very simple loop:

enum Type { String, Float, Int }
if(str.length && str[0] == '-') {
    str = str[1 .. $];
}
Type type = str.length ? Type.Int : Type.String;
foreach(ch; str) {
    if(ch == '.' && type = Type.Int)
         type = Type.Float;
    else if(ch < '0' || ch > '9') {
        type = Type.String;
        break;
    }
}

And if you need to support other details, add them on top of 
that. For example, exponents on floats may be a second clause 
like how I put negative ahead.

You may also choose to use a regular expression though I think 
that is overkill for this.

>     if (data.isNumeric) {

There are several caveats on that isNumeric function: it sees if 
something looks like a D numeric literal, which might not be what 
you want. For example, isNumeric("1UL") passes because the U and 
L suffixes are allowed in D literals...


> But I think that's ugly. The thing is that in PHP, for example, 
> I would do that like this:
>
> ```
> if (is_integer($data)) {

Simiarly, this also will not od what you want. is_integer("1") 
will return false. "1" is of type string. Those functions check 
the dynamic type tag, not the contents of a string.

(actually, you arguably can just always return "string" cuz stdin 
is basically just a string or a binary stream anyway :P )

PHP's is_numeric returns true for both integer and floating point 
things, similarly to D's...