My hitch() method does this kind of:

http://higginsforpresident.net/js/jq.hitch.js

It would look like:

this._input.bind('change', $.hitch(this, "_onInputChange"));

Regards,
Peter Higgins

On Wed, Apr 8, 2009 at 12:03 PM, gregory <gregory.tomlin...@gmail.com> wrote:
>
> the only difficulty I am having with
> Balazs Endresz's approach (which I have also
> implemented in my environment) is if another developer passes a
> function as 'data' param, the results become unpredictable. Though I
> don't *think* anybody should be passing a function to access as
> event.data, it currently does work to do so.
>
> though changing the pattern to no longer have the handler as the last
> param may cause minor confusion, it should not cause any backward
> compatibility issues.
>
> I have never bench marked the performance of 'return toString.call
> (obj) === "[object Function]";' Is this faster than running typeof obj
> === "function" ?
>
> very, very interested in seeing the core of jquery improved to include
> a capability to apply correct scope to the handler function
>
> thanks!
> -gregory
>
> On Mar 29, 3:26 am, Azat Razetdinov <razetdi...@gmail.com> wrote:
>> From the updated jQuery 1.4 Roadmap:
>>
>> > If you need a different object for the scope, why not use the data 
>> > argument to transport it?
>>
>> In OOP-style applications the handler is often not an anonymous
>> function but a link to the current objects's prototype method:
>>
>> this._input.bind('change', this._onInputChange, this);
>>
>> And all prototype methods expect that 'this' points to the current
>> object. If one needs the jQuery object, he could happily use
>> event.currentTarget to reach it.
>>
>> One would recommend binding all handlers with anonymous functions,
>> e.g.:
>>
>> var that = this;
>> this._input.bind('change', function (event) { that._onInputChange
>> (event) });
>>
>> 1. It's more verbose. 2. There's no way to unbind this handler.
>>
>> On Feb 23, 11:56 pm, Azat Razetdinov <razetdi...@gmail.com> wrote:
>>
>> > Passing handler after scope is not suitable for two reasons:
>>
>> > 1. There's no way to determine whether data or scope is passed in a
>> > three-argument method call.
>> > 2. Passing scope after handler is common pattern in JavaScript 1.6
>> > methods like forEach.
>>
>> > On Dec 25 2008, 11:08 pm, "Eduardo Lundgren"
>>
>> > <eduardolundg...@gmail.com> wrote:
>> > > The isFunction is faster now but still has more coast that when you don't
>> > > need to call it.
>>
>> > > We should keep the handler as the last parameter to fit with the jQuery 
>> > > API,
>> > > the change is compatible with it.
>>
>> > >   $('div').bind('click', {data: true}, scope, *scope.internalHandler*);
>>
>> > > Scoping events is a good addition to jQuery.
>>
>> > > Ariel, Joern, John? Let me know if it make sense for you.
>>
>> > > Thanks,
>> > > Eduardo Lundgren
>>
>> > > On Thu, Dec 25, 2008 at 11:57 AM, Balazs Endresz
>> > > <balazs.endr...@gmail.com>wrote:
>>
>> > > > True, but the new isFunction is a couple of times faster than the old
>> > > > one, though it's still many times faster to directly call
>> > > > Object.prototype.toString, which is far below 0.001ms. But as the
>> > > > callback function is the last parameter everywhere in jQuery it might
>> > > > be confusing to change this pattern, it just looked more like binding
>> > > > the function with a native method for me.
>>
>> > > > On Dec 25, 7:06 pm, "Eduardo Lundgren" <eduardolundg...@gmail.com>
>> > > > wrote:
>> > > > > Hi Balazs,
>>
>> > > > > Thanks for give us your opinion.
>>
>> > > > > When you use $.isFunction(data) on the bind method it is very 
>> > > > > expensive
>> > > > when
>> > > > > you have a lot of iterations.
>>
>> > > > > Diff the file I attached with the original file (rev. 5996) I made 
>> > > > > only a
>> > > > > small change on the bind() method, and it's compatible with data and 
>> > > > > with
>> > > > > out API.
>>
>> > > > > On Thu, Dec 25, 2008 at 3:05 AM, Balazs Endresz <
>> > > > balazs.endr...@gmail.com>wrote:
>>
>> > > > > > Hi, I think this would be really useful! I've also modified jQuery 
>> > > > > > to
>> > > > > > do this a while ago (1.2.6) but with the new scope being the last
>> > > > > > argument, so it works without the data object as well:
>>
>> > > > > > jQuery.fn.bind=function( type, data, fn, bind ) {
>> > > > > >                return type == "unload" ? this.one(type, data, fn) :
>> > > > > > this.each
>> > > > > > (function(){
>> > > > > >                        if( $.isFunction(data) )
>> > > > > >                                jQuery.event.add( this, type, data,
>> > > > bind, fn
>> > > > > > );
>> > > > > >                        else
>> > > > > >                                jQuery.event.add( this, type, fn, 
>> > > > > > data,
>> > > > bind
>> > > > > > );
>> > > > > >                });
>> > > > > >        }
>>
>> > > > > > jQuery.event = {
>> > > > > >        add: function(elem, types, handler, data, bind) {
>> > > > > >                if ( elem.nodeType == 3 || elem.nodeType == 8 )
>> > > > > >                        return;
>>
>> > > > > >                if( bind != undefined )
>> > > > > >                        handler = jQuery.bind(handler, bind); 
>> > > > > > //change
>> > > > scope
>> > > > > > ...
>>
>> > > > > > jQuery.each( 
>> > > > > > ("blur,focus,load,resize,scroll,unload,click,dblclick," +
>>
>> > > >  "mousedown,mouseup,mousemove,mouseover,mouseout,mouseenter,mouseleave,"
>> > > > > > +
>> > > > > >        
>> > > > > > "change,select,submit,keydown,keypress,keyup,error").split(","),
>> > > > > > function(i, name){
>>
>> > > > > >        // Handle event binding
>> > > > > >        jQuery.fn[name] = function(fn, bind){ //second argument for 
>> > > > > > the
>> > > > > > scope
>> > > > > >                return fn ? this.bind(name, fn, bind) :
>> > > > this.trigger(name);
>> > > > > >        };
>> > > > > > });
>>
>> > > > > > where jQuery.bind is:
>>
>> > > > > > jQuery.bind=function( fn, bind ){
>> > > > > >        var args = $.makeArray( arguments ).slice(2);
>> > > > > >        if( args.length == 1 && $.isArray( args[0] ) )
>> > > > > >                args = args[0];
>> > > > > >        return function(){
>> > > > > >                return fn.apply( bind, args );
>> > > > > >        }
>> > > > > > }
>>
>> > > > > > On Dec 25, 10:38 am, "Eduardo Lundgren" <eduardolundg...@gmail.com>
>> > > > > > wrote:
>> > > > > > > Hi guys,
>>
>> > > > > > > The .bind() method consider the scope of the handler the element
>> > > > whose
>> > > > > > the
>> > > > > > > event is assigned - that is the correct as default.
>>
>> > > > > > > I've been playing with the event.js and implemented an 
>> > > > > > > alternative to
>> > > > > > call
>> > > > > > > .bind(), specifying another scope, that looks useful for our api.
>>
>> > > > > > > I've attached the event.js modified from the rev. 5996 from the
>> > > > trunk.
>> > > > > > > The changes are compatible with the current API.
>>
>> > > > > > > Here goes one example:
>>
>> > > > > > > var scopeTest = function() {
>> > > > > > >                 this.name = "iamanotherscope";
>>
>> > > > > > >                 this.internalHandler = function(event) {
>> > > > > > >                     console.log("I am another scope method.",
>> > > > this.name,
>> > > > > > > event, event.data);
>> > > > > > >                 };
>> > > > > > >             };
>>
>> > > > > > >             var scope = new scopeTest();
>>
>> > > > > > >  $('div').bind('click', {data: true}, globalHandler);
>> > > > > > >             $('div').bind('click', {data: true},
>> > > > scope.internalHandler);
>> > > > > > //
>> > > > > > > handler, data, default scope
>> > > > > > >             $('div').bind('click', {data: true}, scope,
>> > > > > > > scope.internalHandler); // handler, data, pre-defined scope
>>
>> > > > > > > Let me know if make sense for you.
>>
>> > > > > > > --
>> > > > > > > Eduardo Lundgren
>> > > > > > > Software Engineer
>> > > > > > > Liferay, Inc.
>> > > > > > > Enterprise. Open Source. For Life.
>>
>> > > > > > >  event.js
>> > > > > > > 26KViewDownload
>>
>> > > > > --
>> > > > > Eduardo Lundgren
>> > > > > Software Engineer
>> > > > > Liferay, Inc.
>> > > > > Enterprise. Open Source. For Life.
>>
>> > > --
>> > > Eduardo Lundgren
>> > > Software Engineer
>> > > Liferay, Inc.
>> > > Enterprise. Open Source. For Life.
>

Reply via email to