Created
May 6, 2012 08:12
-
-
Save kkobayashi/2621015 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
use strict; | |
use warnings; | |
use Web::Scraper; | |
use URI; | |
use YAML::Syck; | |
use utf8; | |
$YAML::Syck::ImplicitUnicode = 1; | |
# 一覧: アニメ(終了/再放送) - しょぼいカレンダー | |
# http://cal.syoboi.jp/list?cat=10 | |
# 一覧: アニメ - しょぼいカレンダー | |
# http://cal.syoboi.jp/list?cat=1 | |
my @uri_list = qw(http://cal.syoboi.jp/list?cat=10 http://cal.syoboi.jp/list?cat=1); | |
my $yaml_file = 'syoboi.yaml'; | |
my $yaml = (-f $yaml_file) ? LoadFile($yaml_file) : {}; | |
my $result = {}; | |
my @list = (); | |
### $yaml | |
foreach my $u (@uri_list){ | |
### $u | |
print STDERR "scraping $u ...\n"; | |
my $info = scraper { | |
process 'table#TitleList tbody tr', 'h[]' => scraper { | |
process '//td[1]/a', 'title' => 'TEXT'; | |
process '//td[1]/a', 'url' => '@href'; | |
process '//td[2]', 'begin' => 'TEXT'; | |
process '//td[3]', 'end' => 'TEXT'; | |
process '//td[4]', 'tid' => 'TEXT'; | |
}; | |
result 'h'; | |
}->scrape(URI->new($u)); | |
### $info | |
push(@list, @$info); | |
sleep 1; | |
} | |
foreach my $l(@list){ | |
### $l | |
my $tid = $l->{tid}; | |
if(exists $yaml->{$tid}){ | |
print STDERR " tid $tid found\n"; | |
$result->{$tid} = { %{$yaml->{$tid}}, %$l }; | |
} | |
else{ | |
print STDERR " tid $tid not found. scraping $l->{url} ...\n"; | |
$result->{$tid} = $l; | |
my $v = scraper { | |
process 'table.cast table.data td a.keyword', 'v[]' => 'TEXT'; | |
result 'v'; | |
}->scrape($l->{url}); | |
### $v | |
sleep 1; | |
$result->{$tid}->{v} = $v; | |
} | |
} | |
DumpFile($yaml_file, $result); |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment