scan html: sep = "<td>"

Hi
I try to import html text and I need to split the fields at each <td> or 
</td> entry

How can I succeed? sep = '<td>' doens't yield the right result

thanks for hints