@leptonite/parse-int-strict v1.0.0

@leptonite/parse-int-strict

@leptonite/parse-int-strict provides the function parseIntStrict which parses a string into an integer. It is much stricter than parseInt which means it rejects a lot of inputs that parseInt accepts.

Motivation

I used to use parseInt(someString) or parseInt(someString, 10) to parse strings containing integral numbers. And I check if parsing has been successful by checking if the result is NaN via Number.isNaN(result).

But some day I was suprised about the behavior of parseInt: It happily acceps inputs that do not look like integral numbers to me. parseInt(someString) seems to work like “if someString is not exactly an integral number let’s just parse some part of someString that is”, but I want “if someString is not exactly an integral number let’s just reject it by returning NaN”.

While Number(someString) is stricter than parseInt(someString, 10) in some cases it’s not an alternative because it parses fractional numbers.

Examples:

Although this behavior matches the spec it is just not what I want.

parseIntStrict

parseIntStrict(someString) is a stricter replacement for parseInt(someString, 10). It returns NaN in all examples above. There are in fact even more cases where it returns NaN.

To be more specific:

• parseIntStrict(someString) returns an integral number (let’s call it int) if and only if all of the following is true:
  • someString is exactly equal to int.toString().
  • int is within the range Number.MIN_SAFE_INTEGER .. Number.MAX_SAFE_INTEGER.
• In all other cases parseIntStrict(someString) returns NaN.
• If someString is not of type string, than the behavior of parseIntStrict(someString) is not defined.

It follows in particular:

• parseIntStrict(someString) always returns either an integral number or NaN, it never throws.
• If parseIntStrict(someString) returns an integral number, than it returns the same number as parseInt(someString, 10).