Created
January 25, 2012 19:52
-
-
Save brianjmiller/1678212 to your computer and use it in GitHub Desktop.
data extraction using HTML::DOM
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/perl | |
use strict; | |
use warnings; | |
use Data::Dumper; | |
use DBI; | |
use HTML::DOM; | |
my $dbh = DBI->connect( | |
'dbi:mysql:dbname=...;host=localhost;sslmode=disable', | |
'...', | |
'...', | |
{ | |
RaiseError => 1, | |
}, | |
); | |
my $st = q{ | |
SELECT | |
* | |
FROM | |
table | |
WHERE | |
some_field = ? | |
}; | |
my @bind = qw( some_value ); | |
my $dom = HTML::DOM->new; | |
my $struct = { | |
products => [], | |
}; | |
my $products = $dbh->selectall_arrayref($st, { Slice => {} }, @bind); | |
for my $product (@$products) { | |
eval { | |
my $ref = { | |
product_id => $product->{id}, | |
}; | |
push @{ $struct->{products} }, $ref; | |
$dom->open; | |
$dom->write($product->{html_field}); | |
my $h1_list = $dom->getElementsByTagName('h1'); | |
if ($h1_list->length == 1) { | |
my $h1 = $h1_list->item(0); | |
$ref->{product_name} = $h1->as_text; | |
} | |
elsif ($h1_list->length == 0) { | |
die "No product name h1 detected\n"; | |
} | |
else { | |
die "Cannot handle multiple h1s\n"; | |
} | |
my $display_order = 0; | |
for my $tab_link ($dom->getElementsByClassName('tab')) { | |
my $tab_ref = { | |
display_order => $display_order, | |
}; | |
$display_order = $display_order + 1; | |
push @{ $ref->{tabs} }, $tab_ref; | |
my ($span) = $tab_link->getElementsByTagName('span'); | |
$tab_ref->{tab_title} = $span->as_text; | |
my $href = $tab_link->href; | |
if ($href =~ /\A#(.+)\z/) { | |
my $tab = $dom->getElementById($1); | |
my $tab_html = $tab->innerHTML; | |
if (defined $tab_html and $tab_html ne '') { | |
$tab_html =~ s/{include file="((?!").*)"}/{include file="$1"}/sg; | |
} | |
$tab_ref->{tab_html} = $tab_html; | |
# | |
# check to see if this tab has store_options (actions) | |
# | |
for my $action_link ($tab->getElementsByClassName('green_link')) { | |
my $href = $action_link->href; | |
if ($href =~ /javascript:store_options\(\\'(.+)\\'\)/) { | |
$ref->{actions}->{$1} = 1; | |
} | |
} | |
} | |
else { | |
die "Unrecognized href as tab link: $href\n"; | |
} | |
} | |
$dom->close; | |
}; | |
if ($@) { | |
warn "Failed to build product info ($product->{id}): $@\n"; | |
} | |
} | |
print Dumper($struct); | |
__END__ |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment