.html('<input value="<http://example.org/>" />') produces <input value="<http://example.org></http:>">

[holger@…]

Example:

    <div id="reloadable"></div>
    <script>
        $('#reloadable').html('<input value="<http://example.org/>" />');
    </script>

This corrupts the HTML snippet - it apparently tries to convert the "XML tag" attribute value to balanced pair of XML tags. However, this is an attribute value only, and jQuery shouldn't change that.

When I use

    $('#reloadable')[0].innerHTML = '<input value="<http://example.org/>" />';

then it works. The issue above is problematic because .html is also used by the .load() function and I see no way to work around this effect.

[Rick Waldron]

Use: '<input value="&lt;​http://example.org/&gt;" />'

[anonymous]

According to the W3C HTML validator, using <.../> unescaped inside of an attribute is valid. Also, Xalan produces this HTML when rendering a DOM tree to HTML. I therefore don't see why you simply close this as a wontfix.

If this is a built-in feature of jQuery to "clean up" HTML, then please provide an option to switch this off because my server already produces well-balanced HTML tags.

Also, could you give me a better work-around - not everyone has the luxury of changing the server side code when this is coming from the server via .load(). Thanks.

[holger@…]

I have narrowed it down to the following line in jQuery:

rxhtmlTag = /<(?!area|br|col|embed|hr|img|input|link|meta|param)(([\w:]+)[^>]*)\/>/ig,

This regular expression should skip anything within "..." and '...' quotes so that only real tags are handled and not attribute values. Please reopen this bug.