h$isObject is implemented incorrectly in jsbits/utils.js.

[crocket]

Let's have a look at

ghcjs-base/jsbits/utils.js

Lines 26 to 28 in 9d89e12

	function h$isObject(o) {
	return typeof(o) === 'object';
	}

function h$isObject(o) {
    return typeof(o) === 'object';
}

typeof(null) is 'object', thus you should replace that with something like

function h$isObject(o) {
    return (o instanceof Object);
}

I don't know if instanceof Object is totally accurate yet. I'll add comments on this.

[crocket]

According to http://stackoverflow.com/a/8511350/2104107 and http://stackoverflow.com/a/22482737/2104107,

function h$isObject(o) {
    return o !== null && typeof o === 'object';
}

seems correct. Since functions are considered objects in javascript specification, the following function could be considered correct, too.

function h$isObject(o) {
    return o !== null && (typeof o === 'object' || typeof o === 'function');
}

[hamishmack]

Is there an example use case where the current implementation is problematic? It seems like it would be best to mirror the behaviour of JavaScript in this case unless there is a reason not to.

[crocket]

I stopped programming for money. I'm transitioning into art. Nowadays, I rarely ever write code except when I write some haskell code in XMonad and small command line programs. Thus, I also stopped caring about user interface, javascript, and GHCJS.

That said, I don't know what you mean by mirroring the behavior of javascript.
Does javascript have isObject?
Anyway, I defer this issue entirely to you or anyone else.

[hamishmack]

I stopped programming for money.

Ah.

I'm transitioning into art.

Nice.

Nowadays, I rarely ever write code except when I write some haskell code in XMonad and small command line programs. Thus, I also stopped caring about user interface, javascript, and GHCJS.

Doh!

That said, I don't know what you mean by mirroring the behavior of javascript.
Does javascript have isObject?

Good point. I guess I just was following the pattern that the other isX options were all the typeof of the value. We already have a wrapper for typeof so I think you are right it would be nicer for isObject to be a bit more OO in there behaviour (exclude null and include functions).

Also reading through that post in more detail suggests o === Object(o) works the same as (typeof(o) === 'object' \|\| typeof(o) === 'function') && o !== null. I think that might be the way to go because it could even be inlined inexpensively.