You and your "best practices" and "good coding habits".
On Jan 20, 11:59 pm, Aaron Newton <aa...@iminta.com> wrote:
> var x = 'x';
>
> var foo = function(){
> var x = 'foo';};
>
> foo();
> //x is still 'x'
> var bar = function(){
> x = 'bar';};
>
> bar();
> //x is now 'bar'
>
> window.addEvent('domready', bar); //changes x to 'bar' on domready
> window.addEvent('domready', function(){
> x = 'bar';
>
> }); //the *exact* same behavior
>
> Best practice:
>
> var x;
> window.addEvent('domready', function(){
> x = 'whatever';
>
> });
>
> If you want a global variable, declare one. Don't *imply* one by not
> declaring it and not using var inside your domready statement or elsewhere.
>
>
>
> On Wed, Jan 20, 2010 at 6:05 PM, hairbo <bmuel...@gmail.com> wrote:
> > Yes, sorry. I assumed this was so obvious to the rest of you that I
> > didn't bother explaining. It would appear that declaring "var STUFF"
> > keeps the "STUFF" object scoped (and here's where I get a little
> > fuzzy) to the function that is fired on 'domready'. Removing "var"
> > makes "STUFF" a global var accessible to anything on the page.
>
> > I learned about the importance of declaring "var" more recently than I
> > should have (like, only 4 years ago or so), and so I've become more of
> > a nazi about it. In retrospect, I suppose it should have been obvious
> > that declaring "var STUFF" would keep that object's scope narrow.
> > However, I suppose I assumed that the "domready" event was special in
> > some way...like, why would you *ever* want to keep variables scoped
> > just to that function? Probably very rarely. Anyway, it was just me
> > being careless.
>
> > Though...thinking back, this probably explains some of the other
> > headaches I've endured in the past, since I normally do say "var
> > STUFF" in the domready function.
>
> > On Jan 20, 6:32 pm, Daniel Lohse <annismcken...@googlemail.com> wrote:
> > > Care to share your revelation with me/us?
>
> > > Is it because when defined with "var" no global variable is defined and
> > so the class instance (STUFF) is not accessible in the HTML?
>
> > > Cheers, Daniel
>
> > > On 21.01.2010, at 00:21, hairbo wrote:
>
> > > > Hey Aaron,
>
> > > > Thanks for the reply. Let me just say...OMFG.
>
> > > > I did (and usually have done) this:
>
> > > > window.addEvent('domready',function(){
> > > > var STUFF = new MyClass();
> > > > });
>
> > > > ...instead of this:
>
> > > > window.addEvent('domready',function(){
> > > > STUFF = new MyClass();
> > > > });
>
> > > > sigh. Thanks for the help
>
> > > > On Jan 20, 4:39 pm, Aaron Newton <aa...@iminta.com> wrote:
> > > >> On Wed, Jan 20, 2010 at 2:36 PM, hairbo <bmuel...@gmail.com> wrote:
> > > >>> Okay, there may be things here that I don't understand, so
> > > >>> apologies...
>
> > > >>> 1) I thought this function was supposed to be able to handle a mix of
> > > >>> HTML and JS, and would be smart enough to evaluate any JS it finds in
> > > >>> the returned data. If this is the case (as it does appear to be),
> > > >>> then removing the script tag would surely break.
>
> > > >> This is the case; it should script the scripts from the response
> > (which is
> > > >> why you don't see it if you log the response html) and then evaluates
> > it
> > > >> after updating your element.
>
> > > >>> 2) It turns out that my problem was the JS I had returned wasn't
> > being
> > > >>> evaluated because it didn't recognize a Mootools class I had
> > > >>> instantiated. This I don't think I understand. I created a class,
> > > >>> like this:
>
> > > >>> var MyClass = new Class({
>
> > > >>> ...bla bla bla...
>
> > > >>> });
>
> > > >>> MyClass.implement (new Options, new Events);
>
> > > >>> ...then, I did this in the page:
>
> > > >>> window.addEvent('domready',function(){
> > > >>> var STUFF = new MyClass();
> > > >>> });
>
> > > >>> I expected that the JS being returned by my Ajax page would be able
> > to
> > > >>> do this:
>
> > > >>> STUFF.subMethod();
>
> > > >>> ...but that failed. I don't get why it failed, though, since I'm
> > able
> > > >>> to just create a regular JS function on the page, and reference
> > > >>> *that*, like this:
>
> > > >>> function myFunction () {}
>
> > > >>> ...and then my JS in the AJAX response is this:
> > > >>> myFunction()
>
> > > >>> Why does that second example work, and the first not work? What
> > basic
> > > >>> JS principle am I missing?
>
> > > >> This is where we can't help you so much. You need to use console
> > (firebug)
> > > >> and start logging things you are expecting to be present to get to the
> > > >> bottom of it. I can say that Form.Request is well tested and works (I
> > use it
> > > >> constantly)...
