Extracting info from html page using perl

[fox]

I need to extract patterns from a line in a web page and these patterns sometimes show up twice in the same line so using grep with the pattern only grabs one.

Exaple is I need
<td width="30%" class="navtext" >Sample1</td><width="7%"> 20</td>
<td width="30%" class="navtext" >Sample2</td><td width="7%">18</td>

extracted from the bottom code line.

My final result wants to be
Sample1 20
Sample2 18
Outputed to a seperate file.
Is there a way of using grep or sed to do this from as perl script

<td width="7%"></td><td width="30%" class="navtext" >Sample1</td><td width="7%">20</td><td width="7%"></td><td width="30%" class="navtext" >Sample2</td><td width="7%">18</td></font>

Thanks

[KevinADC]

grep returns a list, lets see the code you have been using.

[fox]

This the code I use which finds the line its on but then prints everything not just the
Sample1:18
Sample2:20

my $grepline=`grep '<td width=\"30%\" class=\"navtext \">' tmp52.html`;
if ($grepline=~/<td width=\"30%\" class=\"navtext \">(.+)<\/td><td width=\"7%\">(. +)<\/td>/){
open (SCORE, ">tmp55.htm l");
print SCORE "$1:$2\n";
print"$1:$2\n";
}

[KevinADC]

this might work, might not:

open (FH, 'tmp52.html') or die "$!";
open (SCORE, ">", 'tmp55.html') or die "$!";
while(<FH>){
   if (m|<td width="30%" class="navtext">(.+?)</td><td width="7%">(.+?)</td>|) {
      print SCORE "$1:$2\n";
   }
}
close SCORE;
close FH;