HTML5 script and encoding

[eek]

If I had that much control of the output, I wouldn't write the CDATA in the first place.

if you can't add CDATA's your screwed mate. Clientco have the same issue with our document management system and we ended up with js files running custom view functions based on whether a particular div is on the page.

[VectraMan]

Can't you do something like:

//<![CDATA[
stuff
//]]>


so that the CDATA bit is commented out as far as the javascript parser is concerned?

If I had that much control of the output, I wouldn't write the CDATA in the first place.

[Bunk]

Can't you do something like:

//<![CDATA[
stuff
//]]>


so that the CDATA bit is commented out as far as the javascript parser is concerned?

[VectraMan]

I'm using MSXML to write HTML5 output, which mostly works except in a few subtle ways where some genius has decided the new standard shouldn't follow the rules of XML.

One of which is to do with script tags. The problem I currently have is anything I write inside a <script> is encoded as markup, so this turns < into &lt; and && into &amp;&amp;, but the browser interprets the contents of <script> as raw javascript rather than following the rules and decoding it.

The way to put raw unencoded text into XML is to use CDATA, which writes the script correctly, but then the browser tries to interpret the [CDATA tag as part of the script.

Does anyone know if there's a good answer to this problem? At the moment all I can think to do is post-process the produced HTML to either decode the contents of a script tag, or probably easier, remove the CDATA tag. I could write the script as an external file, but I'm trying to avoid that.