This seems like a web page parsing question. Another approach can be as follows if you know the limiting token strings: a.split(' <TD WIDTH=175><FONT SIZE=2>')[1].split('</FONT></TD>\r\n')[0]