Created
January 10, 2011 05:31
-
-
Save en45masao/772414 to your computer and use it in GitHub Desktop.
Yahoo! Japanの英検過去問題をダウンロードしてCSVファイル化するスクリプト
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 utf8; | |
use WWW::Mechanize; | |
use Text::CSV_XS; | |
use Web::Scraper; | |
my ($grade, $year, $sq) = @ARGV; | |
die "$0 <1|P1|2|P2|3|4|5> [[yyyy] 1|2|3]" unless defined $grade; | |
$year = '' unless defined $year; | |
$sq = '' unless defined $sq; | |
my $scraper = scraper { | |
process '//div[@id="answer"]/h2', 'year' => ['TEXT', sub {m/([0-9]+)年度/; return $1;}]; | |
process '//div[@id="answer"]/h2', 'sq' => ['TEXT', sub {m/第([0-9]+)回/; return $1;}]; | |
process '//div[@id="answer"]/h2', 'no' => ['TEXT', sub {m/\(([0-9]+)\)/; return $1;}]; | |
process '//div[@id="a-question"]/dl[1]/dt', 'question' => 'TEXT'; | |
process '//div[@id="a-question"]/dl[1]/dd/ul/li', 'choices[]' => 'TEXT'; | |
process '//dl[@id="a-match"]/dd[2]', 'answer' => 'TEXT'; | |
}; | |
my $mech = new WWW::Mechanize(); | |
$mech->get('http://stepup.yahoo.co.jp/english/eiken/minitest.html'); | |
my $csv = Text::CSV_XS->new({binary => 1}); | |
foreach my $top ($mech->find_all_links(text_regex => qr/${year}年度第${sq}/, url_abs_regex => qr(^http://stepup\.yahoo\.co\.jp/.*gr=${grade}))) { | |
$mech->get($top->url()); | |
$mech->follow_link(text => '1', url_regex => qr/.*daily_a\.html.*/); | |
my $tmp = $scraper->scrape($mech->content, $mech->uri); | |
my $filename = "eiken_${grade}-$tmp->{year}-$tmp->{sq}.csv"; | |
open my $file, '>:encoding(cp932)', $filename or die $!; # force output in CP932 | |
print "$filename: Now scraping...\n"; | |
for (;;) { | |
my $scrap = $scraper->scrape($mech->content, $mech->uri); | |
$csv->combine($scrap->{year}, $scrap->{sq}, $scrap->{no}, $scrap->{question}, @{$scrap->{choices}}, $scrap->{answer}); | |
print $file $csv->string() . "\n"; | |
sleep 1; | |
my $link = $mech->find_link(text => '次の問題', url_regex => qr/.*daily_q\.html.*/); | |
last unless defined $link; | |
$mech->get($link); | |
sleep 1; | |
$mech->follow_link(text => '1', url_regex => qr/.*daily_a\.html.*/); | |
} | |
close $file; | |
} | |
=head1 NAME | |
makecsv_yahooeiken - A script to download "Yahoo! Japan EIKEN Daily Test" as CSV files. | |
=head1 SYNOPSIS | |
makecsv_yahooeiken <1|P1|2|P2|3|4|5> [[yyyy] 1|2|3] | |
=head1 EXAMPLES | |
perl makecsv_yahooeiken P1 | |
perl makecsv_yahooeiken P1 2009 | |
perl makecsv_yahooeiken P1 2009 3 | |
=head1 DESCRIPTION | |
このスクリプトを使って、Yahoo! Japanの英検デイリーミニテスト<L<http://stepup.yahoo.co.jp/english/eiken/>>の問題をCSVファイルに書き出すことができます。 | |
=head1 AUTHOR | |
en45masao |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment