matching a decimal digit

When you write a tokenizer, for example for a math expression evaluator, then it is unavoidable to find the category of a single character, whether it’s a number, an operator, etc. While the most logical way is to use regular expression, what if you would like to do it without one?

A typical JavaScript function to carry out this monumental job is as follows:

function isDigit(ch)
{
    var c = ch.charCodeAt();
    return (c >= 48) && (c < = 57);
}

If you want to use some slightly complicated branch statement, then do something like this (or again, a version which uses the character code instead):

function isDigit(ch)
{
    switch (ch) {
    case '0':
    case '1':
    case '2':
    case '3':
    case '4':
    case '5':
    case '6':
    case '7':
    case '8':
    case '9':
        return true;
    default:
        return false;
    }
}

Should you know the internals of modern JavaScript engines, the above two forms translate to a nicely optimized form, ready to be executed at a blazing speed by the engine.

However, if you don’t really care about microsecond-level of performance (your bottleneck is probably somewhere else, not in the above innocent function), an alternative would be (warning: careful JavaScript experts might be provoked into another variant using the property lookup of a constant object):

function isDigit(ch)
{
    return '0123456789'.indexOf(ch) >= ;
}

String searching can’t beat the speed of simple integer comparison yet from time to time I find the construct depicted above is just beautiful and easy on eyes.

Your take?