Temporary variable declared too late

[peblpebl]

Code

var _b;
class C {
    static properties: { [key: string]: string } = {
        ["foo"] : "bar1",        
    };
    static otherproperties = (_b = {}, _b["foo"] = "bar2", _b);
}

// Use C
console.log(C.properties["foo"] + " " + C.otherproperties["foo"])
// End of original src

Expected behavior:

I would have expected that the sugar/javascript for constructing 'properties' would have declared the tmp variable before the construction. In the compile js the _a is declared after the last expression (but before that last comment??).

Google closure compiler warns about this.

Actual behavior:

var _b;
var C = (function () {
    function C() {
    }
    return C;
}());
C.properties = (_a = {},
    _a["foo"] = "bar1",
    _a);
C.otherproperties = (_b = {}, _b["foo"] = "bar2", _b);
// Use C
console.log(C.properties["foo"] + " " + C.otherproperties["foo"]);
var _a;
// End of original src

[ahejlsberg]

See #4138 and #7297.

[peblpebl]

This issue is better explained in #4138.

[peblpebl]

Hmm, this is not quite the same.
In #4138 and #7397 the declaration is in a function and if I understand ecma script correct - then valid everywhere within that function.

_a is not declared in a function here?

[RyanCavanaugh]

It doesn't matter. All var declarations are hoisted to the top of their enclosing lexical scope

[mhegazy]

This issue seems to have been already addressed or is unactionable at the moment. I am auto-closing it for now.