Created
February 2, 2015 14:51
-
-
Save marcioferreira/4d893a82be285f5b3066 to your computer and use it in GitHub Desktop.
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/env perl | |
use use qw( common::sense YADA Web::Scraper DDP ); | |
my $dom_sitemap = scraper { process 'loc', 'url[]' => 'text'; }; | |
my $sitemap = 'http://www.tricae.com.br/sitemapProducts.xml'; | |
#my ($loc) = $dom_sitemap->scrape( URI->new($sitemap) ); | |
my $dom = scraper { | |
process 'h1', 'name' => 'TEXT'; | |
process 'div.txt-brand', 'brand' => 'TEXT'; | |
process '.breadcrumb li:not(:first-child) a', 'categories[]' => 'TEXT'; | |
process '//div[@property="gr:description"]', 'description' => 'TEXT'; | |
process 'div.subtit-product', 'resume' => 'TEXT'; | |
process '.header-product .box-rating', 'rating[]' => '@class';#FIXME | |
process '#productMoreImagesList li', 'images[]' => scraper{ | |
process './/', 'normal' => '@data-image-product'; | |
process './/', 'zoom' => '@data-image-big'; | |
process 'img', 'thumb' => '@alt'; | |
}; | |
process 'div.box-price', 'price' => scraper{ | |
process 'span.price-old', 'from' => ['TEXT', sub{ s/R\$//g; } ]; | |
process '.richsnippets.price b', 'price' => ['TEXT', sub{ s/,/./g; } ]; | |
process '.plots span', 'split' => ['TEXT', sub{ s/,/./g; s/[^x0-9\.]//gi }]; | |
}; | |
process '#prd-attributes-box table tr', '_esp[]' => scraper{ | |
process '//td[1]', 'fdl' => 'text'; | |
process '//td[2]', 'val' => 'text'; | |
}; | |
# process '.item-opinion', 'comments[]' => scraper{ | |
# process 'span:first-child', 'comment' => 'text'; | |
# process '.box-rating', 'rating' => '@class'; | |
# }; | |
# process '', 'recommendation' => 'TEXT'; | |
}; | |
YADA->new( http_response => 1, max => 10, common_opts => { FOLLOWLOCATION => 1 } )->append( | |
#[ @{ $loc->{url} } ] => sub { | |
[ qw( | |
http://www.tricae.com.br/carrinho-de-bebe-de-bebe-bye-bye-cinza-e-azul-burigotto-11084.html | |
http://www.tricae.com.br/guittara-sonora-carros-2-elka--6753.html | |
) ] => sub { | |
my ($me) = @_; | |
return if not $me->response->is_success; | |
my $res = $dom->scrape( $me->response->decoded_content ); | |
$res->{url} = $me->final_url .''; | |
$res->{especification}{ $_->{fdl} } = $_->{val} for @{$res->{_esp}}; | |
delete $res->{$_} for qw/ _esp /; | |
p$res; | |
} | |
)->wait; |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment