Re: svn commit: r558075 - in /lenya/trunk/src/modules: editors/resources/javascript/ editors/resourc

Richard Frovarp wrote:
> Jörn Nettingsmeier wrote:
>> rfrovarpapache.org wrote:
>>> Author: rfrovarp
>>> Date: Fri Jul 20 10:53:59 2007
>>> New Revision: 558075
>>>
>>> URL: 
http://svn.apache.org/viewvc?view=rev&rev=558075
>>> Log:
>>> Prevents backspace in browsers from making the
browser to go back a 
>>> page.
>>
>> very useful. i had never encountered this, but
after checking it out 
>> on windows, it's really a major issue.
>>
>> i have a minor nitpick, though:
>>
>>> Modified:
lenya/trunk/src/modules/fckeditor/usecases/fckeditor.jx
>>> URL: 
>>> http://svn.apache.org/viewvc/lenya/
trunk/src/modules/fckeditor/usecases/fckeditor.jx?view=diff&
amp;rev=558075&r1=558074&r2=558075 
>>>
>>>
============================================================
================== 
>>>
>> <...>
>>> +  <script type="text/javascript"

>>>
src="/modules/editors/javascript/disablebackspace.js&qu
ot;>&#160;</script>
>> <...>
>>> -        <page:body>
>>> +            <page:body
onkeydown="return checkBackspace(event)" 
>>> onkeypress="return
checkBackspace(event)" >
>>
>> why should we have to add two snippets to each
editor?
>> in the o'reilly book on javascript, flanagan
mentions the concept of 
>> "unobtrusive javascript", i.e. avoiding
javascript snippets in the 
>> html as much as possible. i like that idea very
much.
>>
>> what he suggests is to register such handlers via a
window.onload call 
>> that is part of the included script file, like so:
>>
>> function LenyaDisableBackspace() {
>>   var body = document.getElementByTagName('body');
>>   body.onkeydown = new Function(event) {
>>     // checkBackspace function body here
>>   };
>>   body.onkeypress = body.onkeydown;
>> }
>> window.onload = LenyaDisableBackspace;
>>
>> this has the additional benefit of using anonymous
nested functions, 
>> which keep the surrounding object namespace clean
(you never know if 
>> some editor might have a global checkBackspace
function...)
>>
>> note: this is pseudocode, i haven't tested it
yet... just as food for 
>> thought.
>>
>> best,
>>
>> jörn
>>
> 
> Good idea. I'm not that familiar with javascript. I am
unable to get it 
> to work on body. I can get it to work on document
however. Not sure 
> there is much of a difference in this instance between
the two values.

i've done some experiments with this as well, and i've hit a
snag:

there can only be one such handler per element, and only one

window.onload function. which means that it's not as modular
as it 
looks: if two included scripts use the window.onload trick,
one will 
have its function overwritten...
there is a better event handling mechanism available that
avoids this 
problem (addEventListener()), but the joke that passes for a
browser in 
redmond still does not support it :(
you could add global code to the included js like so:

lenyaDisableBackspace();

but it won't work either, because at the time it is
executed, a <body/> 
node does not exist yet. hrrmpf.

morale: your approach is more robust (but mine looks better
:-D)

regards,

jörn


-- 
jörn nettingsmeier

home://germany/45128 essen/lortzingstr. 11/
http://spunk.dnsalias.org
phone://+49/201/491621

Kurt is up in Heaven now.

------------------------------------------------------------
---------
To unsubscribe, e-mail: dev-unsubscribelenya.apache.org
For additional commands, e-mail: dev-helplenya.apache.org

Joern Nettingsmeier again spoke too soon when he wrote:
>> Good idea. I'm not that familiar with javascript. I
am unable to get 
>> it to work on body. I can get it to work on
document however. Not sure 
>> there is much of a difference in this instance
between the two values.
> 
> i've done some experiments with this as well, and i've
hit a snag:
> 
> there can only be one such handler per element, and
only one 
> window.onload function. which means that it's not as
modular as it 
> looks: if two included scripts use the window.onload
trick, one will 
> have its function overwritten...
> there is a better event handling mechanism available
that avoids this 
> problem (addEventListener()), but the joke that passes
for a browser in 
> redmond still does not support it :(
> you could add global code to the included js like so:
> 
> lenyaDisableBackspace();
> 
> but it won't work either, because at the time it is
executed, a <body/> 
> node does not exist yet. hrrmpf.
> 
> morale: your approach is more robust (but mine looks
better :-D)

given that functions are objects in js, the following might
work:

var somebodyElsesOnloadHandler = window.onload;
window.onload = function() {
   somebodyElsesOnloadHandler();
   ourOwnStuff();
}

need to test that some time...



-- 
jörn nettingsmeier

home://germany/45128 essen/lortzingstr. 11/
http://spunk.dnsalias.org
phone://+49/201/491621

Kurt is up in Heaven now.

------------------------------------------------------------
---------
To unsubscribe, e-mail: dev-unsubscribelenya.apache.org
For additional commands, e-mail: dev-helplenya.apache.org