<div class="snippet-clipboard-content notranslate position-relative overflow-auto" data-snippet-clip

Original comment by mikesamuel@gmail.com on 15 Aug 20

CDATA doesn't get treated as javascript about google-code-prettify HOT 6 CLOSED

axtutuu commented on July 23, 2024

CDATA doesn't get treated as javascript

from google-code-prettify.

Comments (6)

GoogleCodeExporter commented on July 23, 2024

Original comment by [email protected] on 15 Aug 2007 at 6:22

Changed state: Accepted

from google-code-prettify.

GoogleCodeExporter commented on July 23, 2024

Original comment by [email protected] on 15 Aug 2007 at 8:43

from google-code-prettify.

GoogleCodeExporter commented on July 23, 2024

Thanks for the testcase.

The lexer treats the chunk as markup if the first non-whitespace character is 
'<'. 
Since "<![CDATA[" starts with '<' it's getting confused.

I will change the splitSourceNodes pass to force non markup lexing.

Original comment by [email protected] on 30 Aug 2007 at 9:29

from google-code-prettify.

GoogleCodeExporter commented on July 23, 2024

Actually, I'll just do it in the chunkifier.

Original comment by [email protected] on 30 Aug 2007 at 9:32

from google-code-prettify.

GoogleCodeExporter commented on July 23, 2024

I tested, and
  <pre class="prettyprint">
  <script><![CDATA[
    mySourceDoe();
  ]]></script>
works properly.

In the testPrettify.html example though:

    <pre class="prettyprint"><![CDATA[
    /**
     * @method isPrettyGood
     * This is a test.
     * @param foo {String} is it a "bar" string?
     * function returns a new this. document is open for each class.
     */
    isPrettyGood = function isPrettyGood(/* string*/foo ) {
        var x = new String( "poop" );
        // first char is "p".
        var p = x.charAt( 0 );
        var p = new Poop( p );
        var t = new this( s );
    };
    ]]>

    <!-- foo html comment function -->
    <?xml version="1.0" encoding="ISO-8859-1"?>
    <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
    "http://www.w3.org/TR/html4/strict.dtd">

    </pre>

What do you expect to happen?
It looks like the content is markup, and should be parsed as such.  The CDATA 
section
is a text node, not something that a browser would treat as javascript, so what
indicator is there that it should be treated as javascript?

Original comment by [email protected] on 30 Aug 2007 at 9:46

from google-code-prettify.

GoogleCodeExporter commented on July 23, 2024