How can I get a specific attribute from a html dom node using libtidy

Question

I am using C++ and libtidy to parse the html page. but the document for libtidy is so incomplete that I can hardly understand what do the API functions do. What I want is to get the specific attribute value from a html dom node, for example:

<table class="xxxx" ...

I want to get the "xxxx" out. How can I do this? Do I have to traverse all the attributes under the table node like this?

if (tidyNodeIsTABLE(tdNode))
{
    TidyAttr attr;
    for (attr=tidyAttrFirst(tdNode); attr; attr=tidyAttrNext(attr))
    {
        if (tidyAttrGetId(attr) == TidyAttr_CLASS)
        {
            std::string value = tidyAttrValue(attr);
            break;
        }
    }
}

or is there any more convenient way? Thanks.