In general, a regex approach is not the best for parsing HTML, but here's a start anyway:

>perl -wMstrict -le
"my $s = '<P_contrib-author>Mariel Miller</P_contrib-author>,'
       . '<P_contrib-author>Allyson Fiona Hadwin</P_contrib-author>,'
       . '<P_contrib-author>  Jenna Cambria  </P_contrib-author>'
       ;
 print qq{[[$s]]};
 ;;
 my $tag = qr{ P_contrib-author }xms;
 my $open_tag  = qr{ <   $tag > }xms;
 my $close_tag = qr{ < / $tag > }xms;
 ;;
 while ($s =~ m{ $open_tag \s* (.*?) \s* $close_tag }xmsg) {
   print qq{name: '$1'};
   }
"
[[<P_contrib-author>Mariel Miller</P_contrib-author>,<P_contrib-author
+>Allyson Fiona Hadwin</P_contrib-author>,<P_contrib-author>  Jenna Ca
+mbria  </P_contrib-author>]]
name: 'Mariel Miller'
name: 'Allyson Fiona Hadwin'
name: 'Jenna Cambria'
[download]

Perhaps some monks better versed in XML parsing than I can suggest a more robust approach.