Rails spinoffs - Jun 2007 - preventing Prototype from extending HTMLElement.prototype

Debugging IE is a serious pain in the neck, and I am running into bugs
because IE doesnt allow extension of HTMLElement.prototype.  I want to
temporarily prevent Prototype from extending HTMLElement.prototype in
Firefox so I can use firebug to burn all those nasty little buggers.
Does anyone have any ideas on how to do this?

Thanks
RR


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

RighteousRaven wrote:
> Debugging IE is a serious pain in the neck, and I am running into bugs
> because IE doesnt allow extension of HTMLElement.prototype.  I want to
> temporarily prevent Prototype from extending HTMLElement.prototype in
> Firefox so I can use firebug to burn all those nasty little buggers.
> Does anyone have any ideas on how to do this?
>
> Thanks
> RR
>
>   
Never tried, but you could try messing with Prototype.BrowserFeatures.  
Let us know what you find; sounds interesting.

--Ken


In prototype.js 1.5.1 line 21
--    ElementExtensions: !!window.HTMLElement,
++   ElementExtensions: false,

and comment out lines 1819-1824
/*if (!Prototype.BrowserFeatures.ElementExtensions &&
 document.createElement(''div'').__proto__) {
  window.HTMLElement = {};
  window.HTMLElement.prototype =
document.createElement(''div'').__proto__;
  Prototype.BrowserFeatures.ElementExtensions = true;
}*/

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

I got it working.  The only change from Ken''s suggestion is to also
set SpecificElementExtensions to false.  I had to upgrade to 1.5.1
cause I couldn''t figure it out on 1.5.0.

Thanks for your help Ken,

RR

On Jun 13, 4:38 pm, Ken Snyder
<kendsny...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:
> RighteousRaven wrote:
> > Debugging IE is a serious pain in the neck, and I am running into bugs
> > because IE doesnt allow extension of HTMLElement.prototype.  I want to
> > temporarily prevent Prototype from extending HTMLElement.prototype in
> > Firefox so I can use firebug to burn all those nasty little buggers.
> > Does anyone have any ideas on how to do this?
>
> > Thanks
> > RR
>
> Never tried, but you could try messing with Prototype.BrowserFeatures.
> Let us know what you find; sounds interesting.
>
> --Ken
>
> In prototype.js 1.5.1 line 21
> --    ElementExtensions: !!window.HTMLElement,
> ++   ElementExtensions: false,
>
> and comment out lines 1819-1824
> /*if (!Prototype.BrowserFeatures.ElementExtensions &&
>  document.createElement(''div'').__proto__) {
>   window.HTMLElement = {};
>   window.HTMLElement.prototype =
document.createElement(''div'').__proto__;
>   Prototype.BrowserFeatures.ElementExtensions = true;
>
> }*/

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

You know, Gentlemen,
I became aware of SpecificElementExtensions only because my FF 2.0.xxx
started blurting out errors with Prototype v 1.5.1.1 that certain
functions on extended DOM elements ( $(..) ) don''t actually exist.
Then I found SpecificElementExtensions, and then - this pretty
discussion.

Hasn''t it been tested, that apparently Prototype v1.5.1.1
doesn''t work
well with FF 2.0? How come no one gets warned from
http://prototypejs.org''s
blog? Aah.

Anyhow, thanks for the suggestions on this point. It worked with me.

	BrowserFeatures: {
		XPath: !!document.evaluate,
		ElementExtensions: false, //!!window.HTMLElement,
http://groups.google.com/group/rubyonrails-spinoffs/browse_thread/thread/b2c29ae55f5e9739/76664ecad4546be4?lnk=gst&q=SpecificElementExtensions&rnum=1#76664ecad4546be4
		SpecificElementExtensions: false
		/*
			(document.createElement(''div'').__proto__ !=			
document.createElement(''form'').__proto__)
			*/
	}

and..

#1828, after "Object.extend(Element, Element.Methods);":

/*
http://groups.google.com/group/rubyonrails-spinoffs/browse_thread/thread/b2c29ae55f5e9739/76664ecad4546be4?lnk=gst&q=SpecificElementExtensions&rnum=1#76664ecad4546be4
if (!Prototype.BrowserFeatures.ElementExtensions &&
 document.createElement(''div'').__proto__) {
	window.HTMLElement = {};
	window.HTMLElement.prototype
document.createElement(''div'').__proto__;
	Prototype.BrowserFeatures.ElementExtensions = true;
}*/

Works well. Thanks once again. Waiting for the working update.


Best regards,

Andrew Revinsky


On Jun 14, 1:27 am, RighteousRaven
<righteousra...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:
> I got it working.  The only change from Ken''s suggestion is to
also
> setSpecificElementExtensionsto false.  I had to upgrade to 1.5.1
> cause I couldn''t figure it out on 1.5.0.
>
> Thanks for your help Ken,
>
> RR
>
> On Jun 13, 4:38 pm, Ken Snyder
<kendsny...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>
>
>
> > RighteousRaven wrote:
> > > Debugging IE is a serious pain in the neck, and I am running into
bugs
> > > because IE doesnt allow extension of HTMLElement.prototype.  I
want to
> > > temporarily prevent Prototype from extending
HTMLElement.prototype in
> > > Firefox so I can use firebug to burn all those nasty little
buggers.
> > > Does anyone have any ideas on how to do this?
>
> > > Thanks
> > > RR
>
> > Never tried, but you could try messing with Prototype.BrowserFeatures.
> > Let us know what you find; sounds interesting.
>
> > --Ken
>
> > In prototype.js 1.5.1 line 21
> > --    ElementExtensions: !!window.HTMLElement,
> > ++   ElementExtensions: false,
>
> > and comment out lines 1819-1824
> > /*if (!Prototype.BrowserFeatures.ElementExtensions &&
> >  document.createElement(''div'').__proto__) {
> >   window.HTMLElement = {};
> >   window.HTMLElement.prototype =
document.createElement(''div'').__proto__;
> >   Prototype.BrowserFeatures.ElementExtensions = true;
>
> > }*/- Hide quoted text -
>
> - Show quoted text -

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Hi Andrew,

Could you be a it more specific and / or open a ticket for the missing
methods, as this issue is unheard of as far as I know.

If this is true it is definitely a bug and needs to be dealt with.

Thanks and best regards,

Tobie
> I became aware of SpecificElementExtensions only because my FF 2.0.xxx
> started blurting out errors with Prototype v 1.5.1.1 that certain
> functions on extended DOM elements ( $(..) ) don''t actually exist.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Tobie, hello!

I use FF 2.0.x.x with Firebug on Win XP RU. I''m not sure if the latter
part matters, anyhow.

Just after I''ve upgraded to 1.5.1.1, all calls to $() function were as
if I just did document.getEl..tById(), not extended. And it was so
only in FireFox. However, in IE elements were extended fully.

So I noticed FireBug was giving me errors like ".up() method is null
or undefined" and ".addClassName method is null or undefined"
(and
such).

Tracing back the callstack I figured, the Element.extend method was to
blame. There''s an "if" statement (the second line):

	if (!element || !element.tagName || element.nodeType == 3 ||
	 element._extended || F.SpecificElementExtensions || element =window) ...

And in my FF, F.SpecificElementExtensions always resolved to true,
while in IE - false. So, in FF, the element was returned back, having
no extended methods whatsoever. Really upset me.

I first checked your blog, then went here. Just to find that mild
solution to my problem by eliminating those niceties. I believe, this
describes my problems.


Oh, by the way, is the $$() function supposed to return unextended
elements?

Like, I happen to use $() so that this snippet would work (3rd line):

	$$(''.menuBar a'').each(function(el) {
		if (!window.location.href.endsWith(el.href)) return;
	
$(el).addClassName(''selected'').up(0).addClassName(''selected'');
		throw $break;
	});

Thanks!


Best regards,

Andrew Revinsky



On Jul 18, 7:17 pm, Tobie Langel
<tobie.lan...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:
> Hi Andrew,
>
> Could you be a it more specific and / or open a ticket for the missing
> methods, as this issue is unheard of as far as I know.
>
> If this is true it is definitely a bug and needs to be dealt with.
>
> Thanks and best regards,
>
> Tobie
>
>
>
> > I became aware of SpecificElementExtensions only because my FF 2.0.xxx
> > started blurting out errors with Prototype v 1.5.1.1 that certain
> > functions on extended DOM elements ( $(..) ) don''t actually
exist.- Hide quoted text -
>
> - Show quoted text -

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Hi Andrew,

You''re misunderstanding a lot of different things here.

I suggest you read the article "How Prototype Extends the DOM":
http://prototypejs.org/learn/extensions

Hope that helps.

Regards,

Tobie

On Jul 24, 8:39 am, Andrew Red
<arevin...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:
> Tobie, hello!
>
> I use FF 2.0.x.x with Firebug on Win XP RU. I''m not sure if the
latter
> part matters, anyhow.
>
> Just after I''ve upgraded to 1.5.1.1, all calls to $() function
were as
> if I just did document.getEl..tById(), not extended. And it was so
> only in FireFox. However, in IE elements were extended fully.
>
> So I noticed FireBug was giving me errors like ".up() method is null
> or undefined" and ".addClassName method is null or
undefined" (and
> such).
>
> Tracing back the callstack I figured, the Element.extend method was to
> blame. There''s an "if" statement (the second line):
>
>         if (!element || !element.tagName || element.nodeType == 3 ||
>          element._extended || F.SpecificElementExtensions || element =>
window) ...
>
> And in my FF, F.SpecificElementExtensions always resolved to true,
> while in IE - false. So, in FF, the element was returned back, having
> no extended methods whatsoever. Really upset me.
>
> I first checked your blog, then went here. Just to find that mild
> solution to my problem by eliminating those niceties. I believe, this
> describes my problems.
>
> Oh, by the way, is the $$() function supposed to return unextended
> elements?
>
> Like, I happen to use $() so that this snippet would work (3rd line):
>
>         $$(''.menuBar a'').each(function(el) {
>                 if (!window.location.href.endsWith(el.href)) return;
>                
$(el).addClassName(''selected'').up(0).addClassName(''selected'');
>                 throw $break;
>         });
>
> Thanks!
>
> Best regards,
>
> Andrew Revinsky
>
> On Jul 18, 7:17 pm, Tobie Langel
<tobie.lan...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>
> > Hi Andrew,
>
> > Could you be a it more specific and / or open a ticket for the missing
> > methods, as this issue is unheard of as far as I know.
>
> > If this is true it is definitely a bug and needs to be dealt with.
>
> > Thanks and best regards,
>
> > Tobie
>
> > > I became aware of SpecificElementExtensions only because my FF
2.0.xxx
> > > started blurting out errors with Prototype v 1.5.1.1 that certain
> > > functions on extended DOM elements ( $(..) ) don''t
actually exist.- Hide quoted text -
>
> > - Show quoted text -

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

On Jul 24, 12:22 pm, Tobie Langel
<tobie.lan...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:
> Hi Andrew,
>
> You''re misunderstanding a lot of different things here.
>
> I suggest you read the article "How Prototype Extends the
DOM":http://prototypejs.org/learn/extensions
The problem appears to be that the test in BrowserFeatures to know
whether to set SpecificElementExtensions to true or false is returning
"true" in Firefox 2.0.x.x.  Therefore, in Element.extend, all of the
code to actually perform the extensions are not called.

"So?" you may ask.  "Surely the point of the test is that the
extensions don''t need to be done; that''s the whole point of
that
flag."  Generally that''s true - but what about when code in one
document is called from code in another document, one which doesn''t
even include prototype.js?

It is in exactly this case that I''m being affected by the behavior. My
application involves a lot of iframes, and a *lot* of pages. We''ve
only begun using prototype lately, so many legacy pages do not yet
include prototype.js.  When a call is made via a reference to "top" or
"parent", things can go wrong.


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

That is a very strong statement, Tobie. Overly. Thank you very much. )

I''d rather you break it down to parts. Like, "that particular
snippet
of yours.." or somethin''. ''Cos "I''ve eaten
a god on Prototype" (how
they say it in our lands) by now. I understand you''re in the Prototype
team and all. And the rest of the world knows much less about this
child than you do.. ) But.. I, too, regularly study all changes even
to minor versions and have created a Prototype reference cheatsheet
for colleagues way before it was available on the web.

Now, what I understand about "F.SpecificElementExtensions", it
doesn''t
necessarily return the right value when it should. That''s the fact.
I''ve given you an example. Perhaps, some FF plugins were preventing it
from working the right way.

Then, second question, what I understand about $$(), it returns a
plain array with unextended elements (if IE is used, for instance). Is
it by design?

Thanks.

Best regards,
Andrew Revinsky

On Jul 24, 9:22 pm, Tobie Langel
<tobie.lan...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:
> Hi Andrew,
>
> You''re misunderstanding a lot of different things here.
>
> I suggest you read the article "How Prototype Extends the
DOM":http://prototypejs.org/learn/extensions
>
> Hope that helps.
>
> Regards,
>
> Tobie
>
> On Jul 24, 8:39 am, Andrew Red
<arevin...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>
>
>
> > Tobie, hello!
>
> > I use FF 2.0.x.x with Firebug on Win XP RU. I''m not sure if
the latter
> > part matters, anyhow.
>
> > Just after I''ve upgraded to 1.5.1.1, all calls to $()
function were as
> > if I just did document.getEl..tById(), not extended. And it was so
> > only in FireFox. However, in IE elements were extended fully.
>
> > So I noticed FireBug was giving me errors like ".up() method is
null
> > or undefined" and ".addClassName method is null or
undefined" (and
> > such).
>
> > Tracing back the callstack I figured, the Element.extend method was to
> > blame. There''s an "if" statement (the second line):
>
> >         if (!element || !element.tagName || element.nodeType == 3 ||
> >          element._extended || F.SpecificElementExtensions || element
=> > window) ...
>
> > And in my FF, F.SpecificElementExtensions always resolved to true,
> > while in IE - false. So, in FF, the element was returned back, having
> > no extended methods whatsoever. Really upset me.
>
> > I first checked your blog, then went here. Just to find that mild
> > solution to my problem by eliminating those niceties. I believe, this
> > describes my problems.
>
> > Oh, by the way, is the $$() function supposed to return unextended
> > elements?
>
> > Like, I happen to use $() so that this snippet would work (3rd line):
>
> >         $$(''.menuBar a'').each(function(el) {
> >                 if (!window.location.href.endsWith(el.href)) return;
> >                
$(el).addClassName(''selected'').up(0).addClassName(''selected'');
> >                 throw $break;
> >         });
>
> > Thanks!
>
> > Best regards,
>
> > Andrew Revinsky
>
> > On Jul 18, 7:17 pm, Tobie Langel
<tobie.lan...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>
> > > Hi Andrew,
>
> > > Could you be a it more specific and / or open a ticket for the
missing
> > > methods, as this issue is unheard of as far as I know.
>
> > > If this is true it is definitely a bug and needs to be dealt
with.
>
> > > Thanks and best regards,
>
> > > Tobie
>
> > > > I became aware of SpecificElementExtensions only because my
FF 2.0.xxx
> > > > started blurting out errors with Prototype v 1.5.1.1 that
certain
> > > > functions on extended DOM elements ( $(..) ) don''t
actually exist.- Hide quoted text -
>
> > > - Show quoted text -- Hide quoted text -
>
> - Show quoted text -

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Hey,

I can''t believe this thread is so long...  There''s just no way
we won''t
extend prototypes on elements, I think.  It''s at the core of what we
do.

Andrew Red a écrit :
> Now, what I understand about "F.SpecificElementExtensions", it
doesn''t
> necessarily return the right value when it should. That''s the
fact.
Sounds weird, and you''re welcome to submit a reproduceable case and the
matching tested patch to the Trac.
> Then, second question, what I understand about $$(), it returns a
> plain array with unextended elements (if IE is used, for instance). Is
Actually this is incorrect.  $$ branches out to either the browser''s 
native DOM Level 3 XPath support or its internal matcher.  Both do 
return actual JS arrays.

If in the former case, it doesn''t bother extending the results, because
it doesn''t need to: DOM 3 XPath is supported only on browsers (Opera, 
FF, WebKit) that happen to feature HTMLElement.prototype, so we''re 
pretty much in the clear: the only sore point you might hit is 
tag-specific extensions on Opera, which would require an extra $ call.

In the latter case (custom JS matcher), elements are extended as their 
internal duplicates are filtered out, in Selector.handlers.unique.  So 
all extensions, be they generic or tag-specific, are guaranteed.  This 
applies in MSIE and Safari, and only causes delay in MSIE (if your $$ 
resultset is very large).

''HTH

-- 
Christophe Porteneuve aka TDD
tdd-x+CfDp/qHev2eFz/2MeuCQ@public.gmane.org

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Hmm, thanks.

On Aug 2, 5:07 pm, Christophe Porteneuve
<t...-x+CfDp/qHev2eFz/2MeuCQ@public.gmane.org>
wrote:
> Hey,
>
> I can''t believe this thread is so long...  There''s just
no way we won''t
> extend prototypes on elements, I think.  It''s at the core of what
we do.
>
> Andrew Red a écrit :
>
> > Now, what I understand about "F.SpecificElementExtensions",
it doesn''t
> > necessarily return the right value when it should. That''s the
fact.
>
> Sounds weird, and you''re welcome to submit a reproduceable case
and the
> matching tested patch to the Trac.
>
> > Then, second question, what I understand about $$(), it returns a
> > plain array with unextended elements (if IE is used, for instance). Is
>
> Actually this is incorrect.  $$ branches out to either the
browser''s
> native DOM Level 3 XPath support or its internal matcher.  Both do
> return actual JS arrays.
>
> If in the former case, it doesn''t bother extending the results,
because
> it doesn''t need to: DOM 3 XPath is supported only on browsers
(Opera,
> FF, WebKit) that happen to feature HTMLElement.prototype, so we''re
> pretty much in the clear: the only sore point you might hit is
> tag-specific extensions on Opera, which would require an extra $ call.
>
> In the latter case (custom JS matcher), elements are extended as their
> internal duplicates are filtered out, in Selector.handlers.unique.  So
> all extensions, be they generic or tag-specific, are guaranteed.  This
> applies in MSIE and Safari, and only causes delay in MSIE (if your $$
> resultset is very large).
>
> ''HTH
>
> --
> Christophe Porteneuve aka TDD
> t...-x+CfDp/qHev2eFz/2MeuCQ@public.gmane.org

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

On Aug 2, 8:07 am, Christophe Porteneuve
<t...-x+CfDp/qHev2eFz/2MeuCQ@public.gmane.org>
wrote:
> Hey,
>
> I can''t believe this thread is so long...  There''s just
no way we won''t
> extend prototypes on elements, I think.  It''s at the core of what
we do.
Yes, there definitely is a way, in Firefox:

1) Create a page with an iframe. The page should import prototype.js.
2) Create a page to be loaded in the iframe. That page should *not*
import prototype.js.
3) Have the iframe page call
"parent.something(document.getElementById(''x''))",
such that the parent
"something" function calls $(arg) on its argument and then expects the
object to have all the Prototype element extensions.
4) Boom.

"That''s silly", you may say, "just import prototype in
your iframe
page."  That is indeed the best approach to take, but it''s a non-
trivial issue in an extensive application with hundreds of pages that
all use services on the "top" page.  The "top" code cannot
expect
using $() to extend correctly incoming (as parameters) elements until
all pages have been changed to import Prototype.



--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Pointy5,

I''ve run across the issue similar to it. I ruled it out with the
notion of the following:
When you call functions accross windows, the call chain tracks you
back to the original window (windows have separate JS engines). So
even if you''re searching for element in window "A" having
started from
window "B", the code of $() function will do the seeking in the
"B"
window. In order to break the call chain, you (I did it) have to set a
timeout on window "A" (having started from "B" window) with
minimal
time (0 or 1) to call a function local to "A" to do the rest. This
breaks the chain, ''window'' object hereafter will refer to
window "A",
not "B". To make the call safe.

With best regards,

Andrew



On Aug 5, 10:03 pm, pointym5 <m...-Xhj3G7Rj6JI@public.gmane.org>
wrote:
> On Aug 2, 8:07 am, Christophe Porteneuve
<t...-x+CfDp/qHev2eFz/2MeuCQ@public.gmane.org> wrote:
>
> > Hey,
>
> > I can''t believe this thread is so long...  There''s
just no way we won''t
> > extend prototypes on elements, I think.  It''s at the core of
what we do.
>
> Yes, there definitely is a way, in Firefox:
>
> 1) Create a page with an iframe. The page should import prototype.js.
> 2) Create a page to be loaded in the iframe. That page should *not*
> import prototype.js.
> 3) Have the iframe page call
>
"parent.something(document.getElementById(''x''))",
such that the parent
> "something" function calls $(arg) on its argument and then
expects the
> object to have all the Prototype element extensions.
> 4) Boom.
>
> "That''s silly", you may say, "just import prototype
in your iframe
> page."  That is indeed the best approach to take, but it''s a
non-
> trivial issue in an extensive application with hundreds of pages that
> all use services on the "top" page.  The "top" code
cannot expect
> using $() to extend correctly incoming (as parameters) elements until
> all pages have been changed to import Prototype.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---