Javascript / Client Side Development :: new Foo(); as a statement? (ECMAScript spec question)


	new Foo(); as a statement? (ECMAScript spec question) Hi, I am reading the ECMAScript specs trying to figure out if the next line is a legal statement or not new Foo(); I think the above code may only be legal as an expression and not as a stand alone statement. Would this make the above a bug? Douglas Crockford's JSLint will choke on the above line of code and stop parsing. All the browsers seem to accept it as ok and work as I expect: the returned object just doesn't get assigned to anything. The time I have used a line like the above is when the constructor has side effects and the "class" keeps track of all its instances. Any ideas what is right or wrong in this case? Thanks, Peter On May 13, 6:05 am, Peter Michaux <petermich@gmail.com> wrote: > Hi, > I am reading the ECMAScript specs trying to figure out if the next > line is a legal statement or not > new Foo(); > I think the above code may only be legal as an expression and not as a > stand alone statement. Would this make the above a bug? I can't see the point in agonising over the difference between an expression and a statement - it's moot. e.g. who cares whether the following is an exression or a statement? foo && foo(); > Douglas Crockford's JSLint will choke on the above line of code and > stop parsing. All the browsers seem to accept it as ok and work as I > expect: the returned object just doesn't get assigned to anything. JSLint is a verifier, not just a validator, so it looks for errors beyond straight language validity to where it thinks you might have made an error. > The time I have used a line like the above is when the constructor has > side effects and the "class" keeps track of all its instances. > Any ideas what is right or wrong in this case? Seems OK to me, but I'd question why you are using a constructor this way - there are usually many ways to achieve the same result. "Right" or "wrong" can only really be judged in context, "better" or "best" could be more appropriate terms. -- Rob. On May 12, 1:05 pm, Peter Michaux <petermich@gmail.com> wrote: > Hi, > I am reading the ECMAScript specs trying to figure out if the next > line is a legal statement or not > new Foo(); > I think the above code may only be legal as an expression and not as a > stand alone statement. Would this make the above a bug? <...> > Any ideas what is right or wrong in this case? ECMA-262/3 12.4 should help with the answer. -- ../rh