#!/usr/bin/perl -w use strict; use HTML::TreeBuilder::XPath; use Encode qw( _utf8_on ); my $data = join "", <>; _utf8_on( $data ); my $tree= HTML::TreeBuilder::XPath->new; $tree->parse( $data ); my( @content ) = $tree->findvalues("//li"); for my $line ( @content ) { $line =~ s/[#\x{c2}\x{a0}\s]+/ /g; next if $line !~ /^\d/; my( $date, $val ) = split " ", $line; next if !defined $val; print "$date,$val\n"; }