in reply to Re: Extracting span and meta content with HTML::TreeBuilder
in thread Extracting span and meta content with HTML::TreeBuilder

Thanks poj, I first have to extract the "review-content" elements, and pull the span out of those. So I don't have that HTML snippet to work on directly. A nested look_down fails: 
for my $page (@$review_pages) {
    my $html = get $page->[1];
    $html =~ s/([^[:ascii:]]+)/unidecode($1)/ge;
    my $tree = HTML::TreeBuilder->new;    # empty tree
    $tree->parse($html);
    print "Review for $page->[0]\n";
    my @items = $tree->look_down( 'class', 'review-content' )
      or die("no items: $!\n");
    for my $item (@items) {
        my @meta = $item->look_down( '_tag', 'meta' ) 
          or die("no meta: $!\n");    # dies here
        for my $meta_item (@meta) {
            print $meta_item->attr('itemprop');
            print ' = ';
            print $meta_item->attr('content') . "\n";
        }
    }
}

[download]

Replies are listed 'Best First'.
Re^3: Extracting span and meta content with HTML::TreeBuilder
by poj (Abbot) on Jul 16, 2014 at 21:35 UTC
    I guessed that might be the case, how about using Xpath ? 
    #!perl
use strict;
use HTML::TreeBuilder::XPath; 
my $tree = HTML::TreeBuilder::XPath->new;  
$tree->parse_file(\*DATA);

my @items = $tree->findnodes( '//div[@class="review-content"]' )
   or die("no items: $!\n");
for my $item (@items) {
  for ( $item->findnodes( '//meta') ){
    print $_->attr('itemprop');
    print ' = ';
    print $_->attr('content')."\n";
  }
}

    [download]
    poj
[reply]
[d/l]
      That's getting the meta data, but also way too much of what I don't want.
[reply]
        poj has shown you how to get the meta properties - to get the date just add a test: 
        next unless $_->attr('itemprop') eq 'datePublished';

        [download]
[reply]
[d/l]
        Ok, try another approach 
        #!perl
use strict;
use HTML::TreeBuilder::XPath; 
my $tree = HTML::TreeBuilder::XPath->new;  
$tree->parse_file('test.htm'); # change to ->parse($html)

my $xpath = '//meta[@itemprop=~/author|datePublished|ratingValue/]';
my @items = $tree->findnodes( $xpath )
   or die("no items: $!\n");

my $rec={};
my $count = 0;
for my $item (@items) {
  my $prop = $item->attr('itemprop');
  $rec->{$prop} = $item->attr('content');
  if ($prop eq 'datePublished'){
    print ++$count." ";
    print $rec->{'author'}." ; ";
    print $rec->{'ratingValue'}." ; ";
    print $rec->{'datePublished'}."\n";
  };
}

        [download]
        poj
[reply]
[d/l]

In Section Seekers of Perl Wisdom