util: isDeepStrictEqual always returns false to compare WeakMap/WeakSet

[yosuke-furukawa]

solve #18227

Checklist

• make -j4 test (UNIX), or vcbuild test (Windows) passes
• tests and/or benchmarks are included
• documentation is changed or added
• commit message follows commit guidelines

Affected core subsystem(s)

util

[BridgeAR]

I thought about this before and I feel like the current way is the more correct one.

The comparison always relies on what is possible to detect. Since Weap(Map|Set) entries do not provide a way to compare them, they should just be ignored.

This PR will also make it impossible to distinguish differences like these:

consta=newWeakMap();constb=newWeakMap();constc=newWeakMap();c.isClearlyDifferent=true;util.isDeepStrictEqual(a,b);// => false even though they are indeed equalutil.isDeepStrictEqual(a,c);// => false without distinguishing this from the former check

Because of those reasons I am -1 on landing this.

[jasnell]

I wonder if it wouldn't be better to throw when one of the arguments is a WeakMap/WeakSet specifically because of the limitations here.

[BridgeAR]

@jasnell hm... I am not certain if it is a good default behavior. But there is again no definite right or wrong.

[yosuke-furukawa]

@jasnell my ideal is so, they are not comparable then throw an Error. First I feel JavaScript engineers do not like Error so much. so I implemented this behavior, but some core members thinks throwing Error is better, I will change this PR.

[TimothyGu]

As stated in #18248 (comment), I don't see why WeakMap and WeakSet should be treated differently from regular objects. Thus, -1 on this PR.