Legacy ( eZ Publish ) do not perform much validation of html attributes ( ie. xml attribuites with the xhtml namespace ). As a consequence, user may experience validation errors when converting ezxmltext to richtext and displays content in eZ Platform.

A example of ezxmltext which will be rendered into invalid xhtml is here (xhtml:colspan contains invalid value):

<?xml version="1.0" encoding="utf-8"?>
<section
        xmlns:image="http://ez.no/namespaces/ezpublish3/image/"
        xmlns:xhtml="http://ez.no/namespaces/ezpublish3/xhtml/"
        xmlns:custom="http://ez.no/namespaces/ezpublish3/custom/">
    <paragraph
            xmlns:tmp="http://ez.no/namespaces/ezpublish3/temporary/">
        <table width="100%">
            <tr>
                <td xhtml:colspan="{sub($span1,1)}">
                    <paragraph>aaa</paragraph>
                </td>
            </tr>
        </table>
    </paragraph>
</section>

Even though the rendered output produced by eZ Publish will be invalid, browsers will cope with it just fine. eZ Platform will on the other hand validate the output and show error messages. Also, the conversion script will complain about such errors, but it will be vague in the hinting of what the validation problem is.