[thelist] xml in javascript ie does show it

Jeff Howden jeff at jeffhowden.com
Mon Oct 24 00:43:22 CDT 2005


Hi,

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> From: gruppenfreizeit
> 
> > Just because it allows you to do it does *not* mean
> > it's correct.  If you want to refer to the cells in a
> > table, you *must* go through either thead, tbody, or
> > tfoot and then that element's tr.  The fact that 
> > Mozilla doesn't require this is an egregious flaw.
> 
> That's excatly what you called "ie forgives"!
> Mozilla doesn't need a reference to the cell.
> When you want to change the node inside a td
> it enough to change the child node element of the td
> like happily IE supports, too.
> 
> print 'var td=x.getElementById("td'.$o.'");';
> print 'td.lastChild.nodeValue="another text";';
> 
> Now I've to create and delete a useless TBODY only to
> keep IE away from creating one automatically, which
> shuffles my whole dom tree.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

The TBODY element is *not* a useless element.  It has semantic significance.
Whether or not you like the extra step isn't important.  What matters is
that IE gets this correct and Mozilla does not.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > Sorry, without actually comparative code samples I'm
> > having difficulty seeing the parallel.
> 
> el.setAttribute("style","color:black;paddingRight:5px;...")
> 
> instead of
> 
> el.style.color="black"
> el.style.paddingRight="5px"
> .
> .
> .
> you see?
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Yes, I see.  However, I personally prefer the second method as it's much
more clear what's going on.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> Last example was the dependency of windows.
> Specially for IE I had to write a loop which is
> searching and closing each single dependend subwindow,
> cause IE ignores this argument.  Mozilla closes all
> dependend subwindows automatically, when you close
> the upper window.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

As you didn't provide a code sample, it's difficult to say how exactly
you're creating dependant child windows.  If so, you're using a browser
specific feature that's *not* a part of ECMAScript.  If so, then it's no
wonder it's not supported by IE.  However, that's not to say there isn't an
IE-specific way to accomplish something like that, like a modeless dialog.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > The semi-colon is *optional*.  There isn't a single
> > browser out there that is ECMAScript compliant that
> > will balk at the omission of a semi-colon (except
> > when required to separate statements, as defined by 
> > the language).
> 
> When you wright js script code in one line, the semicolon
> is absolutly necessary!!  Any browser could read it
> without.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Agreed, which is why I included the exception to my claim -- "except when
required to separate statements, as defined by the language".

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> The information is lost inside the inserted document.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

Perhaps what you're inserting into the other document is overwriting the
function you're calling.

 [>] Jeff Howden
     jeff at jeffhowden.com
     http://jeffhowden.com/




More information about the thelist mailing list