JavaScript is made fun of a lot in the world of memes. While some things actually are unique to JS, more often than not, if you would try similar things in other languages, you probably would also get unintuitive results.
I want to look at some of the weird, unintuitive results and demystify how they happen.
[].every()
always returns true
Array.prototype.every()
is a great function to check if every element of an array fulfills the given condition. It would seem logical that an empty array will return false
since there is no element that could fulfill the condition.
But if we take a look at the actual code, we can find the culprit:
The function checks in a while loop if the current item does not honor the condition. If not, it returns false
. This makes sense, if the first of 100 elements don’t meet the condition, there is no reason to check the rest. If the while loop is finished, we know that the condition is true for every element and the function returns now true
.
So, if there is no element in the array, the while loop is skipped and the function gets to the return true
immediately.
A similar behavior can be found if you try to call [].some()
. It will always return false
. The check is pretty much the same, but with the booleans swapped. It checks for the first element that returns true and skips the rest. If none meet the condition, it returns false.
true + true + true === 3
I don’t understand why I see this all the time. The numeric value of true
is 1, the value of false
is 0. It’s just binary.
This taps into the concept of truthiness
and falsiness
, where only a few values are considered false
:
null
undefined
0
""
(meaning empty string)NaN
- and of course
false
1 + "1" === "11"
and 1 - "1" === 0
Something that may have happened to you was to add a string to a number and then getting something completely different than what you expected. But if you tried to subtract a string from a number, you actually get the result you expected. Why?
Well, the reason behind this behavior is simply that +
is doing much more than just add two numbers. What the +
operator will do is defined by the operands. If one is a string, it’s going for string concatenation.
The -
operator on the other hand only works for subtracting. The string is then coerced to a number and will be subtracted from the first operand.
[] == ""
, but [] != []
How can an empty array be the same as an empty string, but not the same as another array? Type coercion is to blame again, since I did not use ===
or strict equality here.
When comparing the empty array against a string, the array is cast as a string. So string is the same as string. But if you compare two objects with each other (and as we know, arrays are objects in Javascript), they are compared by reference - you can think of memory address. Since they are separate objects they are stored in different locations in the heap.
true != "true"
Another case of type coercion that feels weird, but is actually pretty logical. Trying this comparison will change true
as 1
, the number. Therefore, it will try parseInt("true")
which will of course return NaN
. And 1
is not NaN
.
Trailing commas
Let’s take a look at the following:
Trailing commas are great, it removes some pain when coding. But this results in the last comma being ignored in an array, hence the result above!
Arrow functions
If we create an arrow function called foo like this, everything will work as expected:
But if we change "bar"
to {}
, something unexpected happens:
Why? The curly braces are normal arrow function syntax and since it does not return anything, the result of foo is undefined
. To get foo to actually return the curlies, you have to wrap them in parentheses:
Conclusion
Yes, JavaScript can be weird sometimes. But let’s be honest: we don’t encounter the issues I described here in our daily work.
Most of the JavaScript weirdness are more fun challenges to figure out. Or at least for me!