Bug 212730 – TypedArray.prototype.set is incorrect with primitives

• Status: RESOLVED
• Resolution: FIXED
• Component: JavaScriptCore

[NWU_NISL]

#version: dbae081
https://github.com/WebKit/webkit/tree/dbae081ad7e22d9ab61edf2f337f6c2bb593c7f8

#Testcase:
var NISLFuzzingFunc = function(){
	var e = '123';
	A = new Uint8Array(5);
	A.set(e);
	print(A);
};
NISLFuzzingFunc();

#Command:
./webkit/WebKitBuild/Release/bin/jsc testcase.js

#Output:
Exception: TypeError: First argument should be an object

#Expected output:
1,2,3,0,0

#Description:

According to ES2019 standard, the steps of `%TypedArray%.prototype.set` are as follows.

> 22.2.3.23.1 %TypedArray%.prototype.set ( array [ , offset ] )
>
> 1. Assert: array is any ECMAScript language value other than an Object with a [[TypedArrayName]] internal slot. If it is such an Object, the definition in 22.2.3.23.2 applies.
>
>    ...
>
> 15. Let src be ? ToObject(array).

In this testcase, `ToObject(e)` should not throw a TypeError. So I suspect it is a bug of javascriptcore.

#Reference:
http://ecma-international.org/ecma-262/10.0/#sec-%typedarray%.prototype.set-array-offset

#Additional info:
Contributor: QuXing

[Radar WebKit Bug Importer]

<rdar://problem/64095236>

[Alexey Shvayka]

In genericTypedArrayViewProtoFuncSet(), JSC casts an argument to JSObject* instead of performing toObject().
SpiderMonkey gets this right, while V8 handles all primitives per spec except for numbers.
I will submit a patch as soon as test262 coverage (https://github.com/tc39/test262/pull/2651) is merged & synced.

[Alexey Shvayka]

*** Bug 188877 has been marked as a duplicate of this bug. ***

[Alexey Shvayka]

Created attachment 402161 [details]
Patch

[Yusuke Suzuki]

Comment on attachment 402161 [details]
Patch

r=me

[Alexey Shvayka]

Created attachment 402198 [details]
Patch

Adjust WebGL tests.

[EWS]

Committed r263216: <https://trac.webkit.org/changeset/263216>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 402198 [details].