Well one instance is when I have a build editor function that builds an RTE and returns the container div. This editor can be built from other things (Eg. a another function), or directly via a 3 different links. So I can't make it return false, because some other functions need it to return its container div. So I either put it inside an anonymous function with return false when adding the event to the link (which is messy IMO), make another function in the class which performs buildEditor and returns false (which is a bit better), or I make a generic function to do that (which I chose and asked about here).
Lewis. On Apr 23, 10:42 am, Sebastian Markbåge <[email protected]> wrote: > How are you using this pattern in a real world app? Are you looping > through a list of links in a domready event or what is the use case? > > Seems that you wouldn't need this pattern if you had a Class based > approach to this. > > On Apr 23, 10:32 am, Lewis <[email protected]> wrote: > > > Wow thanks for the interesting conversation. Whilst I can see the > > reasoning behind both sides of the argument, to me it's cleaner and > > more convenient to tell the bind function to return false, than it is > > to create anonymous functions for it. Don't forget, a lot of the time > > people will have to bind the anonymous function, which looks even more > > messy. Eg: > > > function(){ > > this.doSomething(); > > return false; > > > }.bind(this); > > > Compared to: > > this.doSomething.returnFalse(this); > > > Although perhaps a better design would be to allow the 3rd argument of > > bind (and similar functions) to be an explicit return value. Or even > > just allowing an explicit return value in function.create so people > > can create their own wrapper functions (and it would only add a few > > bytes of extra code). > > > Lewis. > > > On Apr 22, 11:16 pm, Sanford Whiteman <[email protected]> > > wrote: > > > > > addEvent('click',function(){return f() && false;}) > > > > amounts to the same thing. > > > > I know it's the same as > > > > f(); return false; > > > > but it might be more easily interpreted as wrapping the function and > > > its forced-false return in one, and that's closer to what the OP was > > > wondering about (the two-at-once/wrapper concept). > > > > > I still don't see the need here. "return false" is just not that big > > > > of a burden. > > > > Me neither, just playing with other ways.... > > > > --Sandy
