Hiding script in certain browsers

Christopher Benson-Manica · Sep 14, 2004

Obviously, compiliant browsers that encounter

<script type="text/javascript">

</script>

will either execute the script or ignore it completely. However, we are
encountering some browsers running on mobile phones that can't handle
the <script> tag, and will fail to render any page containing the tag.
Can anyone suggest some other mechanism of hiding script from such
browsers in such a way that they will render the page normally aside
from the script? Might there be some correctable mistake causing the
problem?

Michael Winter · Sep 14, 2004

Obviously, compiliant browsers that encounter

<script type="text/javascript">
<!--

Don't place script content in SGML. That practice is outdated and
unnecessary. If you want to prevent a script from being rendered, place it
in an external file. It's advisable to do this for production code, anyway.

// your script here
// -->
</script>

will either execute the script or ignore it completely. However, we are
encountering some browsers running on mobile phones that can't handle
the <script> tag, and will fail to render any page containing the tag.

I find it difficult to believe that the page won't render if a user agent
can't handle an element. The HTML Specification explicitly instructs user
agents to ignore the element and render the contents if it's unknown. As
you're scripts should have no content, there will be no effect.

Try placing an empty SCRIPT element in a minimal page (that is, the
smallest, valid page). If it doesn't render the page, inform the vendors:
that's a serious bug.

This should do the trick:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
"http://www.w3.org/TR/html4/strict.dtd">

<html>
<head>
<meta http-equiv="Content-Type"
content="text/html; charset=iso-8859-1">

<title>Test</title>

<script type="text/javascript"></script>
</head>

<body>
<p>This is a test page.</p>
</body>
</html>

If your server sends the content type and character set, you can remove
the META element.

Can anyone suggest some other mechanism of hiding script from such
browsers in such a way that they will render the page normally aside
from the script?

If they're choking on the SCRIPT element, rather than the script itself,
it'll be difficult. The only thing I can think of at the moment is to
sniff the browser on the server and send a different page with no SCRIPT
elements. Hardly a nice solution, though.

[snip]

Good luck,
Mike

Ben Measures · Sep 14, 2004

Christopher said:
Obviously, compiliant browsers that encounter

<script type="text/javascript">

</script>

will either execute the script or ignore it completely. However, we are
encountering some browsers running on mobile phones that can't handle
the <script> tag, and will fail to render any page containing the tag.

Can anyone suggest some other mechanism of hiding script from such
browsers in such a way that they will render the page normally aside
from the script?

Put the JS in an external file and link it with
<script type="text/javascript" src="uri_to.js"></script>

Richard Cornford · Sep 14, 2004

Michael said:
Don't place script content in SGML. That practice is outdated ...

<snip> ^

SGML Comments?

I find it difficult to believe that the page won't render
if a user agent can't handle an element.

I also find it difficult to believe that "some browsers" have been
written for mobile devices (and so presumably relatively recently) such
that they do not conform to the HTML specs (while accepting that content
type). However, we might finally have examples of browsers that take the
rules about un-escaped '</' sequences within javascript string literals
seriously.

... . The only thing I can think of at the moment
is to sniff the browser on the server and send a different
page with no SCRIPT elements. Hardly a nice solution, though.

The only information the server has is the UA header, so that won't be
of much use.

Richard.

Michael Winter · Sep 14, 2004

Michael Winter wrote:
[snip]

Don't place script content in SGML. That practice is outdated ...

Click to expand...

<snip> ^

SGML Comments?

Umm, yeah. Thanks.

[snip]

The only information the server has is the UA header, so that won't be
of much use.

It may be of use, but how much depends on reliability. Assuming that the
UA in question doesn't flawlessly spoof another browser, what's the
likelihood of someone spoofing a mobile browser? There isn't much benefit
in pretending to be an embedded Opera browser, for example.

But first things first: is this a browser bug that needs to be
circumvented, or an error on the OP's part? We'll have to wait and see.

Mike

Just for the record, I don't like entertaining the idea of browser
sniffing, but if this is a browser bug, there aren't many alternatives,
are there?

Sencha Touch--Support 2 browsers in just 228K!	64	Jul 16, 2010
Image upload not working in browser	4	Sep 9, 2022
HTML5 in Firefox and other browsers	1	Aug 24, 2012
need help looking for a cross-browsers vertical scrolling script	2	Mar 12, 2006
Example of HTML5 to be viewed on 4 commom browsers	2	Sep 19, 2010
Position absolute different in different browsers	9	May 9, 2010
HTML5 <video> and <audio> support for several browsers	15	Oct 4, 2010
How to include "<script>" in a string ?	3	Feb 24, 2008

Hiding script in certain browsers

Christopher Benson-Manica

Michael Winter

Ben Measures

Richard Cornford

Michael Winter

Ask a Question

Similar Threads

Members online

Forum statistics

Latest Threads