Javascript, Firefox and document.links

[andreister]

He there!
I've discovered that the

=============== =============== =============== ====
document.links( 'link_id_here') .href = "something" ;
=============== =============== =============== ====

is not applicable for the Firefox (even for the 1.0.6), and I'm falling
into passing through the array and checking each link id.

Could anybody confirm it's true and/or direct to a more elegant
solution if any?
Thanx,
Andrew

[Martin Honnen]

an********@gmai l.com wrote:

I've discovered that the document.links( 'link_id_here') .href = "something" ; is not applicable for the Firefox (even for the 1.0.6)

That will give you an error "document.l inks is not a function" indeed.
Why do you think that document.links is a function?
And if you really have an id why don't you simply use
document.getEle mentById('link_ id_here')
?

--

Martin Honnen
http://JavaScript.FAQTs.com/

[Martin Kurz]

Martin Honnen schrieb:

an********@gmai l.com wrote:

I've discovered that the

document.links( 'link_id_here') .href = "something" ;

is not applicable for the Firefox (even for the 1.0.6)

That will give you an error "document.l inks is not a function" indeed.
Why do you think that document.links is a function?
And if you really have an id why don't you simply use
document.getEle mentById('link_ id_here')
?

Correct, document.links is an array, you can get a single element by ie
document.links[0] for the first link in document, the links-count is in
document.links. length and the href isn't called hraf but target:
document.links[0].target.

And the functionality you're looking for has been posted by Martin Honnen, then
you can set the href by

document.getEle mentById('link_ id_here').setAt tribute("href", "something" );

[Randy Webb]

Martin Kurz said the following on 8/31/2005 2:48 PM:

Martin Honnen schrieb:

an********@gm ail.com wrote:

I've discovered that the

document.lin ks('link_id_her e').href = "something" ;

is not applicable for the Firefox (even for the 1.0.6)

That will give you an error "document.l inks is not a function" indeed.
Why do you think that document.links is a function?
And if you really have an id why don't you simply use
document.getEle mentById('link_ id_here')
?

Correct, document.links is an array, you can get a single element by ie
document.links[0] for the first link in document, the links-count is in
document.links. length and the href isn't called hraf but target:
document.links[0].target.

And the functionality you're looking for has been posted by Martin Honnen, then
you can set the href by

document.getEle mentById('link_ id_here').setAt tribute("href", "something" );

And wonder why it doesn't work properly in the majority of browsers
since IE has major problems with setAttribute? Set it dierectly:

document.links['linkNameNotID'].href="new HREF for the link";

--
Randy
comp.lang.javas cript FAQ - http://jibbering.com/faq & newsgroup weekly
Answer:It destroys the order of the conversation
Question: Why?
Answer: Top-Posting.
Question: Whats the most annoying thing on Usenet?

[Martin Kurz]

>> document.getEle mentById('link_ id_here').setAt tribute("href",

"something" );

And wonder why it doesn't work properly in the majority of browsers
since IE has major problems with setAttribute? Set it dierectly:

document.links['linkNameNotID'].href="new HREF for the link";

Really? Didn't noticed yet (but all Clients I'm scripting for are on Intranet
and so they're all IE6 (if IE at all) and there I didn't got any problems yet)

But document.links['linkNameNotID'] isn't correct, links is (as posted earlier
an arry, not a hash) - some browser may support this but at least IE 6 does not
(document.links ... is null or not an object)

Martin

[Martin Kurz]

Martin Kurz schrieb:

document.get ElementById('li nk_id_here').se tAttribute("hre f",
"something") ;

And wonder why it doesn't work properly in the majority of browsers
since IE has major problems with setAttribute? Set it dierectly:

document.link s['linkNameNotID'].href="new HREF for the link";

Really? Didn't noticed yet (but all Clients I'm scripting for are on Intranet
and so they're all IE6 (if IE at all) and there I didn't got any problems yet)

But document.links['linkNameNotID'] isn't correct, links is (as posted earlier
an arry, not a hash) - some browser may support this but at least IE 6 does not
(document.links ... is null or not an object)

Martin

Well, setAttribute on IE seems to be a little bit intolerant about speeling of
the attributes (in special about the case):

object.setAttri bute(sName, vValue [, iFlags])

If the specified attribute is not already present, the setAttribute method
adds the attribute to the object and sets the value.

Be careful when spelling attribute names. If you set iFlags to 1 and the
sName parameter does not have the same uppercase and lowercase letters as the
attribute, a new attribute is created for the object.

If two or more attributes have the same name, differing only in case, and
iFlags is set to 0, this method assigns values only to the first attribute
created with this name. All other attributes of the same name are ignored.

When setting the CLASS attribute using this method, set the sName to be
"className" , which is the corresponding Dynamic HTML (DHTML) property.

See
http://msdn.microsoft.com/workshop/a...tattribute.asp

So setAttribute("h ref", "URL") shouldn't bring up any problems in IE (as every
other attribute too when regarding the spelling and case) and is working in
Firefox like requested.

Martin

[ASM]

Martin Kurz wrote:

[attributes]

When setting the CLASS attribute using this method, set the sName to be
"className" , which is the corresponding Dynamic HTML (DHTML) property.

See
http://msdn.microsoft.com/workshop/a...tattribute.asp

So setAttribute("h ref", "URL") shouldn't bring up any problems in IE (as every
other attribute too when regarding the spelling and case) and is working in
Firefox like requested.

Well, I have a question :

What is, for IE, 'type' in an input ?
is that an attribute ?
It seems it is but "readonly" -> link given by M$ ->
http://www.w3.org/TR/2000/WD-DOM-Lev...tml#ID-6043025

My IE (5.2 Mac)
aborts, breaks, functions when 'type' is trying to set
(see example below)

In addition :
My IE needs attribute 'className' as specified in link you gave
but ...
this attribute is not understood by FF ... !
example :

<html>
<script type="text/javascript">
var ind = 0;
function newTextField(id ) {
ind++;
var ntf = document.create Element('INPUT' );
ntf.setAttribut e('className',' inpText'); // IE
ntf.setAttribut e('class','inpT ext'); // others
ntf.id = id+ind; // don't know if ID is an attribute ?

ntf.setAttribut e('type','text' ); // IE breaks function here
ntf.type = 'text'; // all as with this way

var val = 'new field id = '+id+ind;
ntf.setAttribut e('value',val);
var place = document.create Element('P');
place.appendChi ld(ntf);
document.body.a ppendChild(plac e);
}
</script>
<style type="text/css">
..inpText { background:yell ow; border:2px solid red; color:blue }
</style>
<button onclick="newTex tField('nouv'); ">New Field</button>
</html>
--
Stephane Moriaux et son [moins] vieux Mac

[RobG]

Martin Kurz wrote:
[...]

But document.links['linkNameNotID'] isn't correct, links is (as posted earlier
an arry, not a hash)

It's not an array, it's an HTML collection. It is a bit like an array -
it has a length property and you can access the items by index or
name/id - but it does not have array methods like sort, join, etc.

<URL:http://www.w3.org/TR/2000/PR-DOM-Level-2-HTML-20000927/html.html#ID-75708506>
--
Rob

[Martin Kurz]

Well, I have a question :

What is, for IE, 'type' in an input ?
is that an attribute ?
It seems it is but "readonly" -> link given by M$ ->
http://www.w3.org/TR/2000/WD-DOM-Lev...tml#ID-6043025
Yes, type on input-element seems to be readonly.
My IE (5.2 Mac)
aborts, breaks, functions when 'type' is trying to set
(see example below)
Well, seems to be a not so good errorhandling on that IE-Version. But what do
you expect in setting a readonly-Attribute? :o)
In addition :
My IE needs attribute 'className' as specified in link you gave
but ...
this attribute is not understood by FF ... !

And that seems to be an error in the FF-Implemantation. className isn't just
specified in the link I gave, it's specified in the
/WD-DOM-Level-1-20000929-Link you gave too. Easiest way seems to me to test an
element of which you know it has a class-attrib in the html-source. So you could
check the correct class-Attribute-name:

if (document.getEl ementById(eleme ntid).getAttrib ute("className" )) {
// use setAttribute("c lassName", "")
} else if (document.getEl ementById(eleme ntid).getAttrib ute("class")) {
// use setAttribute("c lass", "")
} else {
// something else
}

Not the easiest way for setting class-attribs - but the correct one.

What's really interesting on the ff-behavior: Setting the Class the following way

document.getEle mentById(elemen tid).className = "green";

works in both browsers, IE and FF, while

document.getEle mentById(elemen tid).class = "green";

works in none, so FF seems to be a little bit inconsequent.

I prefer this method because of it's the "oficial" way and working - so why
should I use another method that's working on some actual browsers but maybe
won't work in the next generation. If using it this way, if it isn't working in
any browser, at least I can say, that it's the browsers fault, not mine, and
trying to get around the bug in these special cases ;o)

greetings,

Martin

[ASM]

Martin Kurz wrote:

thanks for answer

Well, I have a question :

What is, for IE, 'type' in an input ?
is that an attribute ?
But, all my question is based on a createElement
and setting its (new) attributes on the fly

where (setting attributes to a virtual element) :
- my IE fall down, errores, breaks, if type it's tried to set
whatever I can test
- my IE wants 'className' as attribute for new class
while my FF wants 'class' for same attribute

All was in my comemted example I give here back :

example :

<html>
<script type="text/javascript">
var ind = 0;
function newTextField(id ) {
ind++;
var ntf = document.create Element('INPUT' );
ntf.setAttribut e('className',' inpText'); // IE way
ntf.setAttribut e('class','inpT ext'); // FF way
ntf.id = id+ind; // don't know if ID is an attribute ?

ntf.setAttribut e('type','text' ); // IE breaks function here
ntf.type = 'text'; // all as with this way

var val = 'new field id = '+id+ind;
ntf.setAttribut e('value',val);
var place = document.create Element('P');
place.appendChi ld(ntf);
document.body.a ppendChild(plac e);
}
</script>
<style type="text/css">
..inpText { background:yell ow; border:2px solid red; color:blue }
</style>
<button onclick="newTex tField('nouv'); ">New Field</button>
</html>
It seems it is but "readonly" -> link given by M$ ->
http://www.w3.org/TR/2000/WD-DOM-Lev...tml#ID-6043025

Yes, type on input-element seems to be readonly.

My IE (5.2 Mac)
aborts, breaks, functions when 'type' is trying to set
(see example below)

Well, seems to be a not so good errorhandling on that IE-Version. But what do
you expect in setting a readonly-Attribute? :o)

I try to tell to that new input element (createElement)
it's attribute would be one of types :
- text
- checkbox
- radio
- button
- submit
- reset
If I (and you, and they, we) can't set this type,
how to do to get some more radio-buttons, or check-boxes, or ... ?

[className]

this attribute is not understood by FF ... !

And that seems to be an error in the FF-Implemantation.

....
if (document.getEl ementById(eleme ntid).getAttrib ute("className" )) {
// use setAttribute("c lassName", "")
No, I am still in createElement
I can't get an attribute not yet set ...
document.getEle mentById(elemen tid).className = "green";

works in both browsers, IE and FF,
Yes, this way all work fine
document.getEle mentById(elemen tid).class = "green";

works in none, so FF seems to be a little bit inconsequent.

It seems to :-/

--
Stephane Moriaux et son [moins] vieux Mac