From Wikipedia, the free encyclopedia

This page is for code snippets related to improving and maintaining Wikipedia. These are not bots but standalone problems. All of these are released under the GFDL.

Rollback checker

This tool scans a user's last 500(5000 if a username/password of a bot or admin is given) contributions and returns diff urls for every usage of rollback seen.

If you have a bot or admin account you can set your username in the "login" command and create a file called "password" and put just your password in it, do not press enter after the password. This will let it scan 5000 revisions instead of 500.

Usage
./rollbackChecker.pl some user name
#!/usr/bin/perl
use strict;
use LWP::UserAgent;;
use HTTP::Request;
use HTTP::Request::Common;
use XML::Simple;

# Create ua object
my $ua = LWP::UserAgent->new('agent' => 'Rollback Checker 0.02 - http://en.wikipedia.org/wiki/User:Chillum/source#Rollback_checker');
$ua->default_header('Accept-Encoding' => 'gzip, deflate');
$ua->cookie_jar({ });
my $bite = 500;
$bite = login('SomeUser'); # Only useful if your account is a bot or admin. This will do nothing unless there is a "password" file containing only your password
print "Using bite size of: $bite\n";

my $resp = api(action=>'query',list=>'usercontribs',ucprop=>'ids|comment',uclimit=>$bite,ucuser=>join(' ', @ARGV)) || die "API call failed\n";
foreach my $rh_contrib (@{${$resp}{query}{usercontribs}{item}}) {
  next unless (${$rh_contrib}{comment} =~ m/\[\[Help:Reverting\|Reverted\]\] edits by .*? to last version by .*$/);
  print "http://en.wikipedia.org/?title=-&diff=prev&oldid=${$rh_contrib}{revid}\n";
}

sub login {
  open(PASS,'password') || return 500; sysread(PASS, my $password, -s(PASS)); close(PASS);
  my $xml = api(action=>'login',lgname=>shift,lgpassword=>$password);
  print "The log in result was '${$xml}{login}{result}'. ";
  return ((${$xml}{login}{result} eq 'Success') ? (5000) : (500));
}

sub api {return XMLin($ua->request(POST 'http://en.wikipedia.org/w/api.php', @_,format=>'xml'])->decoded_content());}

CGI version:

See: http://wiki.highinbc.com/cgi-bin/rollback

#!/usr/bin/perl
use strict;
use LWP::UserAgent;
use HTTP::Request;
use HTTP::Request::Common;
use XML::Simple;
use Cache::FileCache;
use CGI qw/:standard/;
use CGI::Carp 'fatalsToBrowser';
$CGI::POST_MAX=1024; # max 1K posts
$CGI::DISABLE_UPLOADS = 1; # no uploads

my $username = param('wpusername');

my $cache = new Cache::FileCache( );

print header('text/html'),
      start_html('Rollback scanner'),
      h1('Rollback scanner'),
      p('Enter the username of a Wikipedian and I will scan the last 500 edits for rollback usage and give you links.'),
      start_form,
      textfield('wpusername'),
      submit,
      end_form,
      hr(),
      $username 
        ? ('Most recent rollbacks by user '.a({href=>'http://en.wikipedia.org/wiki/User:'.$username},$username).' '.sup(small(a({href => self_url()},'(Link to this query)')))) 
        : '',
      br(),
      br();

my $ua;
if($username) {
  # Create ua object
  $ua = LWP::UserAgent->new('agent' => 'Rollback Checker 0.02 - http://en.wikipedia.org/wiki/User:Chillum/source#Rollback_checker');
  $ua->default_header('Accept-Encoding' => 'gzip, deflate');
  my $resp = api(action=>'query',list=>'usercontribs',ucprop=>'ids|comment|title|timestamp',uclimit=>500,ucuser=>$username);
  foreach my $rh_contrib (@{${$resp}{query}{usercontribs}{item}}) {
    next unless (${$rh_contrib}{comment} =~ m/\[\[Help:Reverting\|Reverted\]\] edits by .*? to last version by .*$/);
    my $link    = "http://en.wikipedia.org/?diff=prev&oldid=${$rh_contrib}{revid}";
    my $label   = "${$rh_contrib}{timestamp} - ${$rh_contrib}{title}";
    ${$rh_contrib}{comment} =~ s/\[\[.*?\|(.*?)\]\]/$1/g;
    print a({href=>$link},$label),
          b(": ${$rh_contrib}{comment}"),
          br();
  }
}
print end_html();
 
sub api {
  my @params = @_;
  my $apiCacheKey = "$0-lastAPICall";
  my $cacheKey = join(':', @params);
  my $xml = $cache->get( $cacheKey );
  if ( not defined $xml ) {
    my $throttle = 5; # One request every 5 seconds
    my $lastRun = $cache->get( $apiCacheKey ) || 0;
    if ((time() - $lastRun) < $throttle) {
      my $waitFor = $throttle - (time() - $lastRun);
      print h1('Too much traffic'),
            p('Please '.a({href => self_url()},'try again').' in '.$waitFor.' seconds'),
            end_html();
      exit;
    }
    $cache->set( $apiCacheKey, time(), "never" );
    $xml = $ua->request(POST 'http://en.wikipedia.org/w/api.php', @_,format=>'xml'])->decoded_content();
    $cache->set( $cacheKey, $xml, "5 minutes" );
  }  
  return XMLin($xml);
}
From Wikipedia, the free encyclopedia

This page is for code snippets related to improving and maintaining Wikipedia. These are not bots but standalone problems. All of these are released under the GFDL.

Rollback checker

This tool scans a user's last 500(5000 if a username/password of a bot or admin is given) contributions and returns diff urls for every usage of rollback seen.

If you have a bot or admin account you can set your username in the "login" command and create a file called "password" and put just your password in it, do not press enter after the password. This will let it scan 5000 revisions instead of 500.

Usage
./rollbackChecker.pl some user name
#!/usr/bin/perl
use strict;
use LWP::UserAgent;;
use HTTP::Request;
use HTTP::Request::Common;
use XML::Simple;

# Create ua object
my $ua = LWP::UserAgent->new('agent' => 'Rollback Checker 0.02 - http://en.wikipedia.org/wiki/User:Chillum/source#Rollback_checker');
$ua->default_header('Accept-Encoding' => 'gzip, deflate');
$ua->cookie_jar({ });
my $bite = 500;
$bite = login('SomeUser'); # Only useful if your account is a bot or admin. This will do nothing unless there is a "password" file containing only your password
print "Using bite size of: $bite\n";

my $resp = api(action=>'query',list=>'usercontribs',ucprop=>'ids|comment',uclimit=>$bite,ucuser=>join(' ', @ARGV)) || die "API call failed\n";
foreach my $rh_contrib (@{${$resp}{query}{usercontribs}{item}}) {
  next unless (${$rh_contrib}{comment} =~ m/\[\[Help:Reverting\|Reverted\]\] edits by .*? to last version by .*$/);
  print "http://en.wikipedia.org/?title=-&diff=prev&oldid=${$rh_contrib}{revid}\n";
}

sub login {
  open(PASS,'password') || return 500; sysread(PASS, my $password, -s(PASS)); close(PASS);
  my $xml = api(action=>'login',lgname=>shift,lgpassword=>$password);
  print "The log in result was '${$xml}{login}{result}'. ";
  return ((${$xml}{login}{result} eq 'Success') ? (5000) : (500));
}

sub api {return XMLin($ua->request(POST 'http://en.wikipedia.org/w/api.php', @_,format=>'xml'])->decoded_content());}

CGI version:

See: http://wiki.highinbc.com/cgi-bin/rollback

#!/usr/bin/perl
use strict;
use LWP::UserAgent;
use HTTP::Request;
use HTTP::Request::Common;
use XML::Simple;
use Cache::FileCache;
use CGI qw/:standard/;
use CGI::Carp 'fatalsToBrowser';
$CGI::POST_MAX=1024; # max 1K posts
$CGI::DISABLE_UPLOADS = 1; # no uploads

my $username = param('wpusername');

my $cache = new Cache::FileCache( );

print header('text/html'),
      start_html('Rollback scanner'),
      h1('Rollback scanner'),
      p('Enter the username of a Wikipedian and I will scan the last 500 edits for rollback usage and give you links.'),
      start_form,
      textfield('wpusername'),
      submit,
      end_form,
      hr(),
      $username 
        ? ('Most recent rollbacks by user '.a({href=>'http://en.wikipedia.org/wiki/User:'.$username},$username).' '.sup(small(a({href => self_url()},'(Link to this query)')))) 
        : '',
      br(),
      br();

my $ua;
if($username) {
  # Create ua object
  $ua = LWP::UserAgent->new('agent' => 'Rollback Checker 0.02 - http://en.wikipedia.org/wiki/User:Chillum/source#Rollback_checker');
  $ua->default_header('Accept-Encoding' => 'gzip, deflate');
  my $resp = api(action=>'query',list=>'usercontribs',ucprop=>'ids|comment|title|timestamp',uclimit=>500,ucuser=>$username);
  foreach my $rh_contrib (@{${$resp}{query}{usercontribs}{item}}) {
    next unless (${$rh_contrib}{comment} =~ m/\[\[Help:Reverting\|Reverted\]\] edits by .*? to last version by .*$/);
    my $link    = "http://en.wikipedia.org/?diff=prev&oldid=${$rh_contrib}{revid}";
    my $label   = "${$rh_contrib}{timestamp} - ${$rh_contrib}{title}";
    ${$rh_contrib}{comment} =~ s/\[\[.*?\|(.*?)\]\]/$1/g;
    print a({href=>$link},$label),
          b(": ${$rh_contrib}{comment}"),
          br();
  }
}
print end_html();
 
sub api {
  my @params = @_;
  my $apiCacheKey = "$0-lastAPICall";
  my $cacheKey = join(':', @params);
  my $xml = $cache->get( $cacheKey );
  if ( not defined $xml ) {
    my $throttle = 5; # One request every 5 seconds
    my $lastRun = $cache->get( $apiCacheKey ) || 0;
    if ((time() - $lastRun) < $throttle) {
      my $waitFor = $throttle - (time() - $lastRun);
      print h1('Too much traffic'),
            p('Please '.a({href => self_url()},'try again').' in '.$waitFor.' seconds'),
            end_html();
      exit;
    }
    $cache->set( $apiCacheKey, time(), "never" );
    $xml = $ua->request(POST 'http://en.wikipedia.org/w/api.php', @_,format=>'xml'])->decoded_content();
    $cache->set( $cacheKey, $xml, "5 minutes" );
  }  
  return XMLin($xml);
}

Videos

Youtube | Vimeo | Bing

Websites

Google | Yahoo | Bing

Encyclopedia

Google | Yahoo | Bing

Facebook