>>its simple to do, simply don't select it to begin with
then in that case, we need to list ALL those tags we're interested in. won't this endup in a long list?
HTML::Parser has a method ignore_tags() which could be used to ignore tags. I used it as below & tried to get the text, but it returned many nested arrays. I could not figure out how to access to final extracted text from this "@array"
my @array;
my $p = HTML::Parser->new(api_version => 3,
handlers => { text => [\@array, "text"]});
$p->ignore_tags(qw(table img));
$p->parse($page);
print "Size of array=$#array\n";
foreach my $aline (@array)
{
print $aline;
}
print "\n";
Meanwhile, I found an alternative, but seems it is quite slower than what we could have achieved with HTML::Parser.
my $link = 'somelinek';
my $page = get($link) or die $!;
my $stream = HTML::TokeParser->new(\$page);
my $doparse = 1; ## 0 means don't parse
while (my $token = $stream->get_token)
{
if ($token->[0] eq 'S')
{
if ($token->[1] eq 'table')
{
$doparse = 0;
}
elsif ($token->[1] eq 'img')
{
;;
}
}
elsif ($token->[0] eq 'E' and $token->[1] eq 'table')
{
$doparse = 1;
}
elsif ($token->[0] eq 'C')
{
;;
}
elsif ($token->[0] eq 'T' and $doparse eq 1)
{ # text process the text in $token->[1]
# skip: empty lines, " "
if (defined ($token->[1]))
{
$token->[1] =~ s/ / /ig;
$token->[1] =~ s/’/'/ig;
$token->[1] =~ s/[7-8];/"/ig;
$token->[1] =~ s/—//ig;
$token->[1] =~ s/&/&/ig;
$token->[1] =~ s/-{2,}//ig;
print "$token->[1]";
}
}
}
This above use of TokeParser gives lot of broken text.
Which could be better way?
Thanks |