Last active
July 23, 2016 15:15
-
-
Save n13i/5421671 to your computer and use it in GitHub Desktop.
#moyasearch
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 warnings; | |
use strict; | |
use utf8; | |
use FindBin qw($Bin); | |
use Encode; | |
use YAML; | |
use Net::Twitter::Lite::WithAPIv1_1; | |
use DateTime; | |
use Schedule::At; | |
use Web::Scraper; | |
use LWP::UserAgent; | |
use URI::Escape; | |
my $MyName = $Bin . '/post.pl'; | |
binmode STDOUT, ':encoding(utf8)'; | |
#exit if($#ARGV < 0); | |
my $conffile = $Bin . '/conf/config.yaml'; | |
my $conf = YAML::LoadFile($conffile) or die('load config failed'); | |
my $twit_r = Net::Twitter::Lite::WithAPIv1_1->new( | |
consumer_key => $conf->{twitter}->{consumer_key}, | |
consumer_secret => $conf->{twitter}->{consumer_secret}, | |
legacy_lists_api => 0, | |
ssl => 1, | |
); | |
$twit_r->access_token($conf->{twitter}->{read}->{access_token}); | |
$twit_r->access_token_secret($conf->{twitter}->{read}->{access_token_secret}); | |
print "Requesting rate limit status\n"; | |
print Dump($twit_r->rate_limit_status()); | |
my $dt_now = DateTime->now(time_zone => $conf->{timezone}); | |
printf "Now: %s\n", $dt_now->strftime('%Y/%m/%d %H:%M:%S'); | |
my $last_ids = &load_last_postid; | |
print Dump($last_ids); | |
print "Requesting mentions\n"; | |
my $dms = $twit_r->direct_messages({ since_id => $last_ids->{dm} }); | |
#my $mentions = $twit_r->mentions({ since_id => $last_ids->{mentions} }); | |
my $mentions = []; | |
my $max_ids = $last_ids; | |
my @queries = (); | |
foreach my $item (@{$dms}) | |
{ | |
print Dump($item); | |
next if($item->{sender}->{id_str} ne $conf->{accept_sender_id}); | |
if($item->{id} > $max_ids->{dm}) | |
{ | |
$max_ids->{dm} = $item->{id}; | |
} | |
push(@queries, $item); | |
} | |
foreach my $item (@{$mentions}) | |
{ | |
print Dump($item); | |
next if($item->{user}->{id_str} ne $conf->{accept_sender_id}); | |
if($item->{id} > $max_ids->{mentions}) | |
{ | |
$max_ids->{mentions} = $item->{id}; | |
} | |
push(@queries, $item); | |
} | |
foreach my $query (@queries) | |
{ | |
my $text = $query->{text}; | |
$text =~ s/^\@n13i_g\s+//g; | |
printf "%s--> %s\n", $query->{id}, $text; | |
#next; | |
my $keyword = '"' . $text . '"'; | |
my $result = &google_it($keyword); | |
my $post = undef; | |
if(defined($result->{error2})) | |
{ | |
$post = $result->{error2}; | |
} | |
elsif(defined($result->{error})) | |
{ | |
$post = $result->{error}; | |
$post =~ s/検索ツールをリセット//g; | |
} | |
elsif(defined($result->{count})) | |
{ | |
$post = $keyword . ' ' . $result->{count}; | |
} | |
$post .= ' #moyasearch'; | |
&post_it($post); | |
} | |
&save_last_postid($max_ids); | |
exit; | |
sub post_it | |
{ | |
my $post = shift || die; | |
my $twit_w = Net::Twitter::Lite::WithAPIv1_1->new( | |
consumer_key => $conf->{twitter}->{consumer_key}, | |
consumer_secret => $conf->{twitter}->{consumer_secret}, | |
ssl => 1, | |
); | |
$twit_w->access_token($conf->{twitter}->{write}->{access_token}); | |
$twit_w->access_token_secret($conf->{twitter}->{write}->{access_token_secret}); | |
if(defined($post)) | |
{ | |
#$post .= ' [auto]'; | |
printf "post: %s\n", $post; | |
my $status = undef; | |
eval { | |
$status = $twit_w->update($post); | |
}; | |
if($@) | |
{ | |
warn "$@\n"; | |
} | |
} | |
} | |
sub google_it | |
{ | |
my $query = shift || die; | |
my $ua = LWP::UserAgent->new(agent => 'Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US) AppleWebKit/534.16 (KHTML, like Gecko) Chrome/10.0.648.151 Safari/534.16'); | |
#my $res = $ua->get('http://www.google.co.jp/search?q=' . uri_escape_utf8($query) . '&nfpr=1&tbs=li:1&safe=off'); | |
my $res = $ua->get('http://www.google.co.jp/search?q=' . uri_escape_utf8($query) . '&nfpr=1&safe=off'); | |
my $google_html = decode('utf8', $res->content); | |
#print $google_html; | |
my $scraper = scraper { | |
process 'div#resultStats', 'count' => 'TEXT'; | |
process 'div#topstuff div.med', 'error' => 'TEXT'; | |
#process 'div#topstuff div.med p:nth-child(2)', 'error2' => 'TEXT'; | |
process 'div#topstuff div.med p:nth-child(2)', 'error2' => 'TEXT'; | |
}; | |
my $r = $scraper->scrape($google_html); | |
#print Dump($r); | |
return $r; | |
} | |
sub save_last_postid | |
{ | |
my $postid = shift || return; | |
YAML::DumpFile($Bin . '/lastid', $postid) or die; | |
#open FH, '>', $Bin . '/lastid' or die; | |
#print FH $postid . "\n"; | |
#close FH; | |
} | |
sub load_last_postid | |
{ | |
my $lastid; | |
eval { | |
$lastid = YAML::LoadFile($Bin . '/lastid'); | |
}; | |
if($@) | |
{ | |
$lastid = { dm => 722471195827544063, mentions => 607471195827544063 }; | |
} | |
#open FH, '<', $Bin . '/lastid' or return 0; | |
#my $lastid = <FH>; | |
#close FH; | |
return $lastid; | |
} | |
# vim: noexpandtab |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment