parseInt() scientific notation

Question

Was playing around with JS and noticed this.

Why does parseInt(1e+21) return 1 and parseInt(1e+20) return 100000000000000000000 ? parseInt(10e+20) also returns 1?

Why does parseInt(1.7976931348623157E+10308); return NaN while parseFloat(1.7976931348623157E+10308) returns Infinity?

---

Even made a fibonacci sequence and any value that has +21 exponents only returns the first digit:

parseFloat() will return the correct number till Infinity.

Fibonacci demo (int : parseInt() : parseFloat())

( tested in Chrome )

Answer 1

parseInt and parseFloat are supposed to parse strings and convert them to Numbers. So, when you pass them a Number, it gets coerced to a string by the interpreter. So, at least in Firefox, "" + 1e+20 outputs "100000000000000000000", and "" + 1e+21 outputs "1e+21".

Let's read the documentation of parseInt:

If parseInt encounters a character that is not a numeral in the specified radix, it ignores it and all succeeding characters and returns the integer value parsed up to that point

so parseInt("1.whatever") returns 1 and parseInt("32.231e+something") returns 32.

A similar thing must be happening with parseFloat.

It's the weak typing's fault. IMO the correct behaviour when doing what you do would be to throw an exception.