getAttribute('href') in IE

K

kangax

According to MSDN [1], if iFlag is set to the integer 2, the
getAttribute() method should return "... the value exactly as it was
set in script or in the source document."

Yet, when appending anchor element via innerHTML, calling
getAttribute('href', 2) on that element always returns string
representing absolute url (ignoring the actual value).

Does anyone know of a way to get value of href attribute exactly as it
was defined in the document?

[1] http://msdn2.microsoft.com/en-us/library/ms536429.aspx

Thank you.
 
V

VK

According to MSDN [1], if iFlag is set to the integer 2, the
getAttribute() method should return "... the value exactly as it was
set in script or in the source document."

Yet, when appending anchor element via innerHTML, calling
getAttribute('href', 2) on that element always returns string
representing absolute url (ignoring the actual value).

Does anyone know of a way to get value of href attribute exactly as it
was defined in the document?

[1]http://msdn2.microsoft.com/en-us/library/ms536429.aspx

Thank you.

You have actually set a URL and the method hapilly reports what:
either you have set www.ExAmPlE.CoM or www.example.com

If you have actually have to know either www.ExAmPlE.CoM or www.example.com
had been set then your solution is having much more serious problems
than stated in our OP.
 
T

tomtom.wozniak

According to MSDN [1], if iFlag is set to the integer 2, the
getAttribute() method should return "... the value exactly as it was
set in script or in the source document."

Yet, when appending anchor element via innerHTML, calling
getAttribute('href', 2) on that element always returns string
representing absolute url (ignoring the actual value).

Does anyone know of a way to get value of href attribute exactly as it
was defined in the document?

[1]http://msdn2.microsoft.com/en-us/library/ms536429.aspx

Thank you.

I've duplicated your problem here:

<!-- Problem: setting href=document.html alerts as "http://host/
document.html" and not "document.html" -->
<html><body><span id=mystuff></span></body>
<script>
document.getElementById("mystuff").innerHTML = "<a id=mycall
href=document.html>My Link</a>";
alert(document.getElementById("mycall").getAttribute("href",2));
</script>
</html>

Don't know how to get around the browser translating the href into an
absolute although I do have a workaround.

When writing out the elements to innerHTML, simply add another
attribute. I called my new attribute "xref" in the example below:

<!-- Workaround: use additional attribute "xref" to skip around
browser resolving relative-to-absolute URLs -->
<html><body><span id=mystuff></span></body>
<script>
document.getElementById("mystuff").innerHTML = "<a id=mycall
href=document.html xref=document.html>My Link</a>";
alert(document.getElementById("mycall").getAttribute("xref",2));
</script>
</html>

This skirts around the issue but could get you going for now.

Cheers,

-Tom Woz
 
K

kangax

According to MSDN [1], if iFlag is set to the integer 2, the
getAttribute() method should return "... the value exactly as it was
set in script or in the source document."
Yet, when appending anchor element via innerHTML, calling
getAttribute('href', 2) on that element always returns string
representing absolute url (ignoring the actual value).
Does anyone know of a way to get value of href attribute exactly as it
was defined in the document?

Thank you.

I've duplicated your problem here:

<!-- Problem: setting href=document.html alerts as "http://host/
document.html" and not "document.html" -->
<html><body><span id=mystuff></span></body>
<script>
document.getElementById("mystuff").innerHTML = "<a id=mycall
href=document.html>My Link</a>";
alert(document.getElementById("mycall").getAttribute("href",2));
</script>
</html>

Don't know how to get around the browser translating the href into an
absolute although I do have a workaround.

When writing out the elements to innerHTML, simply add another
attribute. I called my new attribute "xref" in the example below:

<!-- Workaround: use additional attribute "xref" to skip around
browser resolving relative-to-absolute URLs -->
<html><body><span id=mystuff></span></body>
<script>
document.getElementById("mystuff").innerHTML = "<a id=mycall
href=document.html xref=document.html>My Link</a>";
alert(document.getElementById("mycall").getAttribute("xref",2));
</script>
</html>

This skirts around the issue but could get you going for now.

Cheers,

-Tom Woz

Tom,

This is an interesting solution. One thing I'm concerned about is the
cost of maintenance when using such workaround. In order to overcome
bugs in IE's DOM implementation, I would have to duplicate all the
"href" attributes throughout all of my "templates" (relatively large
chunks of html which are then inserted into a document).

What's more disappointing is that element.outerHTML and
element.parentNode.innerHTML properties both contain absolute (!!!)
url representation as well. *sigh* I would be more than happy if
anyone else could shed some light on the issue.

Thanks.
 
R

RobG

According to MSDN [1], if iFlag is set to the integer 2, the
getAttribute() method should return "... the value exactly as it was
set in script or in the source document."

Yet, when appending anchor element via innerHTML, calling
getAttribute('href', 2) on that element always returns string
representing absolute url (ignoring the actual value).

Does anyone know of a way to get value of href attribute exactly as it
was defined in the document?

[1]http://msdn2.microsoft.com/en-us/library/ms536429.aspx

It seems to me that you are trying to use the href property for
something that it was not intended to be used for. It doesn't seem
like a good idea to create a dependency on an implementation detail of
the unspecified innerHTML property or the non-standard iFlag argument
for getAttribute.

If you want to distinguish between identical links where one is
relative and one absolute in the original source, you could use a
value in the class attribute or add a query string after the link,
e.g.:

<a href="index.html" class="var_relative ..." ...>

<a href="index.html?relative" ...>
 
K

kangax

According to MSDN [1], if iFlag is set to the integer 2, the
getAttribute() method should return "... the value exactly as it was
set in script or in the source document."
Yet, when appending anchor element via innerHTML, calling
getAttribute('href', 2) on that element always returns string
representing absolute url (ignoring the actual value).
Does anyone know of a way to get value of href attribute exactly as it
was defined in the document?

It seems to me that you are trying to use the href property for
something that it was not intended to be used for. It doesn't seem
like a good idea to create a dependency on an implementation detail of
the unspecified innerHTML property or the non-standard iFlag argument
for getAttribute.

If you want to distinguish between identical links where one is
relative and one absolute in the original source, you could use a
value in the class attribute or add a query string after the link,
e.g.:

<a href="index.html" class="var_relative ..." ...>

<a href="index.html?relative" ...>

Rob,

I understand your concerns but the idea is to have a consistent result
of getting an href attribute across browsers. The fact that part of
document was modified via innerHTML is something that could have
happened or could have not. We can't be sure if it will or will not be
used. We only care about "reading" href attribute in a consistent way.
innerHTML just hapenns to trigger this bug.

Am I missing something?

Thanks.
 
R

RobG

According to MSDN [1], if iFlag is set to the integer 2, the
getAttribute() method should return "... the value exactly as it was
set in script or in the source document."
Yet, when appending anchor element via innerHTML, calling
getAttribute('href', 2) on that element always returns string
representing absolute url (ignoring the actual value).
Does anyone know of a way to get value of href attribute exactly as it
was defined in the document?
[1]http://msdn2.microsoft.com/en-us/library/ms536429.aspx
It seems to me that you are trying to use the href property for
something that it was not intended to be used for. It doesn't seem
like a good idea to create a dependency on an implementation detail of
the unspecified innerHTML property or the non-standard iFlag argument
for getAttribute.
If you want to distinguish between identical links where one is
relative and one absolute in the original source, you could use a
value in the class attribute or add a query string after the link,
e.g.:
<a href="index.html" class="var_relative ..." ...>
<a href="index.html?relative" ...>

I understand your concerns but the idea is to have a consistent result
of getting an href attribute across browsers.

It seems to me that you may be looking for a general solution to
retrieving attribute values such as might be used in a library.
Features like getting attribute values consistently in a general way
are extremely difficult because of the variability in different user
agents (it's one of the arguments against general libraries, and a
reason why "general" libraries usually only work in a small subset of
browsers).

It is often possible to solve the difficult cases with ad hoc methods
(such as those proposed earlier, no doubt there are others such as
inspecting the string used for the innerHTML property or storing the
value somewhere), but that is counter to a general solution.

The fact that part of
document was modified via innerHTML is something that could have
happened or could have not. We can't be sure if it will or will not be
used. We only care about "reading" href attribute in a consistent way.
innerHTML just hapenns to trigger this bug.

I'm not sure I'd label it a bug - inconsistent perhaps, but since
there is no published specification for the innerHTML property (which
might, if it exists, specifiy that the value of an href attribute is
resolved from a relative URI to a full URI before adding to the DOM)
it's likely that inconsistencies will be found elsewhere too.

Am I missing something?

Maybe, but that's it from me! :)
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

Forum statistics

Threads
473,769
Messages
2,569,582
Members
45,066
Latest member
VytoKetoReviews

Latest Threads

Top