--- /dev/null
+* text=auto
-#! /usr/bin/perl -W\r
-#\r
-# feedlog.pl\r
-#\r
-# This script will log headers and other data to a database\r
-# for further analysis by parsing a feed from INN.\r
-# \r
-# It is part of the NewsStats package.\r
-#\r
-# Copyright (c) 2010 Thomas Hochstein <thh@inter.net>\r
-#\r
-# It can be redistributed and/or modified under the same terms under \r
-# which Perl itself is published.\r
-\r
-BEGIN {\r
- our $VERSION = "0.01";\r
- use File::Basename;\r
- push(@INC, dirname($0));\r
-}\r
-use strict;\r
-\r
-use NewsStats;\r
-\r
-use Sys::Syslog qw(:standard :macros);\r
-\r
-use Date::Format;\r
-use DBI;\r
-\r
-################################# Main program #################################\r
-\r
-### read commandline options\r
-my %Options = &ReadOptions('qd');\r
-\r
-### read configuration\r
-my %Conf = %{ReadConfig('newsstats.conf')};\r
-\r
-### init syslog\r
-openlog($MySelf, 'nofatal,pid', LOG_NEWS);\r
-syslog(LOG_NOTICE, "$MyVersion starting up.") if !$Options{'q'};\r
-\r
-### init database\r
-my $DBHandle = InitDB(\%Conf,0);\r
-if (!$DBHandle) {\r
- syslog(LOG_CRIT, 'Database connection failed: %s', $DBI::errstr);\r
- while (1) {}; # go into endless loop to suppress further errors and respawning\r
-};\r
-my $DBQuery = $DBHandle->prepare(sprintf("INSERT INTO %s.%s (day,date,mid,timestamp,token,size,peer,path,newsgroups,headers) VALUES (?,?,?,?,?,?,?,?,?,?)",$Conf{'DBDatabase'},$Conf{'DBTableRaw'}));\r
-\r
-### main loop\r
-while (<>) {\r
- chomp;\r
- # catch empty lines trailing or leading\r
- if ($_ eq '') {\r
- next;\r
- }\r
- # first line contains: mid, timestamp, token, size, peer, Path, Newsgroups\r
- my ($Mid, $Timestamp, $Token, $Size, $Peer, $Path, $Newsgroups) = split;\r
- # remaining lines contain headers\r
- my $Headers = "";\r
- while (<>) {\r
- chomp;\r
- # empty line terminates this article\r
- if ($_ eq '') {\r
- last;\r
- }\r
- # collect headers\r
- $Headers .= $_."\n" ;\r
- }\r
-\r
- # parse timestamp to day (YYYY-MM-DD) and to MySQL timestamp\r
- my $Day = time2str("%Y-%m-%d", $Timestamp);\r
- my $Date = time2str("%Y-%m-%d %H:%M:%S", $Timestamp);\r
-\r
- # write to database\r
- if (!$DBQuery->execute($Day, $Date, $Mid, $Timestamp, $Token, $Size, $Peer, $Path, $Newsgroups, $Headers)) {\r
- syslog(LOG_ERR, 'Database error: %s', $DBI::errstr);\r
- };\r
- $DBQuery->finish;\r
- \r
- warn sprintf("-----\nDay: %s\nDate: %s\nMID: %s\nTS: %s\nToken: %s\nSize: %s\nPeer: %s\nPath: %s\nNewsgroups: %s\nHeaders: %s\n",$Day, $Date, $Mid, $Timestamp, $Token, $Size, $Peer, $Path, $Newsgroups, $Headers) if !$Options{'d'};\r
-}\r
-\r
-### close handles\r
-$DBHandle->disconnect;\r
-syslog(LOG_NOTICE, "$MySelf closing down.") if !$Options{'q'};\r
-closelog();\r
-\r
-__END__\r
-\r
-################################ Documentation #################################\r
-\r
-=head1 NAME\r
-\r
-feedlog - log data from an INN feed to a database\r
-\r
-=head1 SYNOPSIS\r
-\r
-B<feedlog> [B<-Vhdq>]\r
-\r
-=head1 REQUIREMENTS\r
-\r
-See doc/README: Perl 5.8.x itself and the following modules from CPAN:\r
-\r
-=over 2\r
-\r
-=item -\r
-\r
-Config::Auto\r
-\r
-=item -\r
-\r
-Date::Format\r
-\r
-=item -\r
-\r
-DBI\r
-\r
-=back\r
-\r
-=head1 DESCRIPTION\r
-\r
-This script will log overview data and complete headers to a database\r
-table for further examination by parsing a feed from INN. It will\r
-parse that information and write it to a mysql database table in real\r
-time.\r
-\r
-All reporting is done to I<syslog> via I<news> facility. If B<feedlog>\r
-fails to initiate a database connection at startup, it will log to\r
-I<syslog> with I<CRIT> priority and go in an endless loop, as\r
-terminating would only result in a rapid respawn.\r
-\r
-=head2 Configuration\r
-\r
-F<feedlog.pl> will read its configuration from F<newsstats.conf> which\r
-should be present in the same directory via Config::Auto.\r
-\r
-See doc/INSTALL for an overview of possible configuration options.\r
-\r
-=head1 OPTIONS\r
-\r
-=over 3\r
-\r
-=item B<-V> (version)\r
-\r
-Print out version and copyright information on B<yapfaq> and exit.\r
-\r
-=item B<-h> (help)\r
-\r
-Print this man page and exit.\r
-\r
-=item B<-d> (debug)\r
-\r
-Output debugging information to STDERR while parsing STDIN. You'll\r
-find that information most probably in your B<INN> F<errlog> file.\r
-\r
-=item B<-q> (quiet)\r
-\r
-Suppress logging to syslog.\r
-\r
-=back\r
-\r
-=head1 INSTALLATION\r
-\r
-See doc/INSTALL.\r
-\r
-=head1 EXAMPLES\r
-\r
-Set up a feed like that in your B<INN> F<newsfeeds> file:\r
-\r
- ## gather statistics for NewsStats\r
- newsstats!\r
- :!*,de.*\r
- :Tc,WmtfbsPNH,Ac:/path/to/feedlog.pl\r
-\r
-See doc/INSTALL for further information.\r
-\r
-=head1 FILES\r
-\r
-=over 4\r
-\r
-=item F<feedlog.pl>\r
-\r
-The script itself.\r
-\r
-=item F<NewsStats.pm>\r
-\r
-Library functions for the NewsStats package.\r
-\r
-=item F<newsstats.conf>\r
-\r
-Runtime configuration file for B<yapfaq>.\r
-\r
-=back\r
-\r
-=head1 BUGS\r
-\r
-Please report any bugs or feature requests to the author or use the\r
-bug tracker at L<http://bugs.th-h.de/>!\r
-\r
-=head1 SEE ALSO\r
-\r
-=over 2\r
-\r
-=item -\r
-\r
-doc/README\r
-\r
-=item -\r
-\r
-doc/INSTALL\r
-\r
-=back\r
-\r
-This script is part of the B<NewsStats> package.\r
-\r
-=head1 AUTHOR\r
-\r
-Thomas Hochstein <thh@inter.net>\r
-\r
-=head1 COPYRIGHT AND LICENSE\r
-\r
-Copyright (c) 2010 Thomas Hochstein <thh@inter.net>\r
-\r
-This program is free software; you may redistribute it and/or modify it\r
-under the same terms as Perl itself.\r
-\r
-=cut\r
+#! /usr/bin/perl -W
+#
+# feedlog.pl
+#
+# This script will log headers and other data to a database
+# for further analysis by parsing a feed from INN.
+#
+# It is part of the NewsStats package.
+#
+# Copyright (c) 2010 Thomas Hochstein <thh@inter.net>
+#
+# It can be redistributed and/or modified under the same terms under
+# which Perl itself is published.
+
+BEGIN {
+ our $VERSION = "0.01";
+ use File::Basename;
+ push(@INC, dirname($0));
+}
+use strict;
+
+use NewsStats;
+
+use Sys::Syslog qw(:standard :macros);
+
+use Date::Format;
+use DBI;
+
+################################# Main program #################################
+
+### read commandline options
+my %Options = &ReadOptions('qd');
+
+### read configuration
+my %Conf = %{ReadConfig('newsstats.conf')};
+
+### init syslog
+openlog($MySelf, 'nofatal,pid', LOG_NEWS);
+syslog(LOG_NOTICE, "$MyVersion starting up.") if !$Options{'q'};
+
+### init database
+my $DBHandle = InitDB(\%Conf,0);
+if (!$DBHandle) {
+ syslog(LOG_CRIT, 'Database connection failed: %s', $DBI::errstr);
+ while (1) {}; # go into endless loop to suppress further errors and respawning
+};
+my $DBQuery = $DBHandle->prepare(sprintf("INSERT INTO %s.%s (day,date,mid,timestamp,token,size,peer,path,newsgroups,headers) VALUES (?,?,?,?,?,?,?,?,?,?)",$Conf{'DBDatabase'},$Conf{'DBTableRaw'}));
+
+### main loop
+while (<>) {
+ chomp;
+ # catch empty lines trailing or leading
+ if ($_ eq '') {
+ next;
+ }
+ # first line contains: mid, timestamp, token, size, peer, Path, Newsgroups
+ my ($Mid, $Timestamp, $Token, $Size, $Peer, $Path, $Newsgroups) = split;
+ # remaining lines contain headers
+ my $Headers = "";
+ while (<>) {
+ chomp;
+ # empty line terminates this article
+ if ($_ eq '') {
+ last;
+ }
+ # collect headers
+ $Headers .= $_."\n" ;
+ }
+
+ # parse timestamp to day (YYYY-MM-DD) and to MySQL timestamp
+ my $Day = time2str("%Y-%m-%d", $Timestamp);
+ my $Date = time2str("%Y-%m-%d %H:%M:%S", $Timestamp);
+
+ # write to database
+ if (!$DBQuery->execute($Day, $Date, $Mid, $Timestamp, $Token, $Size, $Peer, $Path, $Newsgroups, $Headers)) {
+ syslog(LOG_ERR, 'Database error: %s', $DBI::errstr);
+ };
+ $DBQuery->finish;
+
+ warn sprintf("-----\nDay: %s\nDate: %s\nMID: %s\nTS: %s\nToken: %s\nSize: %s\nPeer: %s\nPath: %s\nNewsgroups: %s\nHeaders: %s\n",$Day, $Date, $Mid, $Timestamp, $Token, $Size, $Peer, $Path, $Newsgroups, $Headers) if !$Options{'d'};
+}
+
+### close handles
+$DBHandle->disconnect;
+syslog(LOG_NOTICE, "$MySelf closing down.") if !$Options{'q'};
+closelog();
+
+__END__
+
+################################ Documentation #################################
+
+=head1 NAME
+
+feedlog - log data from an INN feed to a database
+
+=head1 SYNOPSIS
+
+B<feedlog> [B<-Vhdq>]
+
+=head1 REQUIREMENTS
+
+See doc/README: Perl 5.8.x itself and the following modules from CPAN:
+
+=over 2
+
+=item -
+
+Config::Auto
+
+=item -
+
+Date::Format
+
+=item -
+
+DBI
+
+=back
+
+=head1 DESCRIPTION
+
+This script will log overview data and complete headers to a database
+table for further examination by parsing a feed from INN. It will
+parse that information and write it to a mysql database table in real
+time.
+
+All reporting is done to I<syslog> via I<news> facility. If B<feedlog>
+fails to initiate a database connection at startup, it will log to
+I<syslog> with I<CRIT> priority and go in an endless loop, as
+terminating would only result in a rapid respawn.
+
+=head2 Configuration
+
+F<feedlog.pl> will read its configuration from F<newsstats.conf> which
+should be present in the same directory via Config::Auto.
+
+See doc/INSTALL for an overview of possible configuration options.
+
+=head1 OPTIONS
+
+=over 3
+
+=item B<-V> (version)
+
+Print out version and copyright information on B<yapfaq> and exit.
+
+=item B<-h> (help)
+
+Print this man page and exit.
+
+=item B<-d> (debug)
+
+Output debugging information to STDERR while parsing STDIN. You'll
+find that information most probably in your B<INN> F<errlog> file.
+
+=item B<-q> (quiet)
+
+Suppress logging to syslog.
+
+=back
+
+=head1 INSTALLATION
+
+See doc/INSTALL.
+
+=head1 EXAMPLES
+
+Set up a feed like that in your B<INN> F<newsfeeds> file:
+
+ ## gather statistics for NewsStats
+ newsstats!
+ :!*,de.*
+ :Tc,WmtfbsPNH,Ac:/path/to/feedlog.pl
+
+See doc/INSTALL for further information.
+
+=head1 FILES
+
+=over 4
+
+=item F<feedlog.pl>
+
+The script itself.
+
+=item F<NewsStats.pm>
+
+Library functions for the NewsStats package.
+
+=item F<newsstats.conf>
+
+Runtime configuration file for B<yapfaq>.
+
+=back
+
+=head1 BUGS
+
+Please report any bugs or feature requests to the author or use the
+bug tracker at L<http://bugs.th-h.de/>!
+
+=head1 SEE ALSO
+
+=over 2
+
+=item -
+
+doc/README
+
+=item -
+
+doc/INSTALL
+
+=back
+
+This script is part of the B<NewsStats> package.
+
+=head1 AUTHOR
+
+Thomas Hochstein <thh@inter.net>
+
+=head1 COPYRIGHT AND LICENSE
+
+Copyright (c) 2010 Thomas Hochstein <thh@inter.net>
+
+This program is free software; you may redistribute it and/or modify it
+under the same terms as Perl itself.
+
+=cut
-#! /usr/bin/perl -W\r
-#\r
-# gatherstats.pl\r
-#\r
-# This script will gather statistical information from a database\r
-# containing headers and other information from a INN feed.\r
-# \r
-# It is part of the NewsStats package.\r
-#\r
-# Copyright (c) 2010 Thomas Hochstein <thh@inter.net>\r
-#\r
-# It can be redistributed and/or modified under the same terms under \r
-# which Perl itself is published.\r
-\r
-BEGIN {\r
- our $VERSION = "0.01";\r
- use File::Basename;\r
- push(@INC, dirname($0));\r
-}\r
-use strict;\r
-\r
-use NewsStats qw(:DEFAULT :TimePeriods ListNewsgroups);\r
-\r
-use DBI;\r
-\r
-################################# Definitions ##################################\r
-\r
-# define types of information that can be gathered\r
-# all / groups (/ clients / hosts)\r
-my %LegalTypes;\r
-@LegalTypes{('all','groups')} = ();\r
-\r
-################################# Main program #################################\r
-\r
-### read commandline options\r
-my %Options = &ReadOptions('dom:p:t:n:r:g:c:s:');\r
-\r
-### read configuration\r
-my %Conf = %{ReadConfig('newsstats.conf')};\r
-\r
-### override configuration via commandline options\r
-my %ConfOverride;\r
-$ConfOverride{'DBTableRaw'} = $Options{'r'} if $Options{'r'};\r
-$ConfOverride{'DBTableGrps'} = $Options{'g'} if $Options{'g'};\r
-$ConfOverride{'DBTableClnts'} = $Options{'c'} if $Options{'c'};\r
-$ConfOverride{'DBTableHosts'} = $Options{'s'} if $Options{'s'};\r
-$ConfOverride{'TLH'} = $Options{'n'} if $Options{'n'};\r
-&OverrideConfig(\%Conf,\%ConfOverride);\r
-\r
-### get type of information to gather, defaulting to 'all'\r
-$Options{'t'} = 'all' if !$Options{'t'};\r
-die "$MySelf: E: Unknown type '-t $Options{'t'}'!\n" if !exists($LegalTypes{$Options{'t'}});\r
-\r
-### get time period (-m or -p)\r
-my ($StartMonth,$EndMonth) = &GetTimePeriod($Options{'m'},$Options{'p'});\r
-\r
-### init database\r
-my $DBHandle = InitDB(\%Conf,1);\r
-\r
-### get data for each month\r
-warn "$MySelf: W: Output only mode. Database is not updated.\n" if $Options{'o'};\r
-foreach my $Month (&ListMonth($StartMonth,$EndMonth)) {\r
-\r
- print "---------- $Month ----------\n" if $Options{'d'};\r
-\r
- if ($Options{'t'} eq 'all' or $Options{'t'} eq 'groups') {\r
- ### ----------------------------------------------\r
- ### get groups data (number of postings per group)\r
- # get groups data from raw table for given month\r
- my $DBQuery = $DBHandle->prepare(sprintf("SELECT newsgroups FROM %s.%s WHERE day LIKE ? AND NOT disregard",$Conf{'DBDatabase'},$Conf{'DBTableRaw'}));\r
- $DBQuery->execute($Month.'-%') or die sprintf("$MySelf: E: Can't get groups data for %s from %s.%s: $DBI::errstr\n",$Month,$Conf{'DBDatabase'},$Conf{'DBTableRaw'});\r
-\r
- # count postings per group\r
- my %Postings;\r
-\r
- while (($_) = $DBQuery->fetchrow_array) {\r
- # get list oft newsgroups and hierarchies from Newsgroups:\r
- my %Newsgroups = ListNewsgroups($_);\r
- # count each newsgroup and hierarchy once\r
- foreach (sort keys %Newsgroups) {\r
- # don't count newsgroup/hierarchy in wrong TLH\r
- next if(defined($Conf{'TLH'}) and !/^$Conf{'TLH'}/);\r
- $Postings{$_}++;\r
- };\r
- };\r
-\r
- print "----- GroupStats -----\n" if $Options{'d'};\r
- foreach my $Newsgroup (sort keys %Postings) {\r
- print "$Newsgroup => $Postings{$Newsgroup}\n" if $Options{'d'};\r
- if (!$Options{'o'}) {\r
- # write to database\r
- $DBQuery = $DBHandle->prepare(sprintf("REPLACE INTO %s.%s (month,newsgroup,postings) VALUES (?, ?, ?)",$Conf{'DBDatabase'},$Conf{'DBTableGrps'}));\r
- $DBQuery->execute($Month, $Newsgroup, $Postings{$Newsgroup}) or die sprintf("$MySelf: E: Can't write groups data for %s/%s to %s.%s: $DBI::errstr\n",$Month,$Newsgroup,$Conf{'DBDatabase'},$Conf{'DBTableGrps'});\r
- $DBQuery->finish;\r
- };\r
- };\r
- } else {\r
- # other types of information go here - later on\r
- };\r
-};\r
-\r
-### close handles\r
-$DBHandle->disconnect;\r
-\r
-__END__\r
-\r
-################################ Documentation #################################\r
-\r
-=head1 NAME\r
-\r
-gatherstats - process statistical data from a raw source\r
-\r
-=head1 SYNOPSIS\r
-\r
-B<gatherstats> [B<-Vhdo>] [B<-m> I<YYYY-MM>] [B<-p> I<YYYY-MM:YYYY-MM>] [B<-t> I<type>] [B<-n> I<TLH>] [B<-r> I<database table>] [B<-g> I<database table>] [B<-c> I<database table>] [B<-s> I<database table>]\r
-\r
-=head1 REQUIREMENTS\r
-\r
-See doc/README: Perl 5.8.x itself and the following modules from CPAN:\r
-\r
-=over 2\r
-\r
-=item -\r
-\r
-Config::Auto\r
-\r
-=item -\r
-\r
-DBI\r
-\r
-=back\r
-\r
-=head1 DESCRIPTION\r
-\r
-This script will extract and process statistical information from a\r
-database table which is fed from F<feedlog.pl> for a given time period\r
-and write its results to (an)other database table(s).\r
-\r
-The time period to act on defaults to last month; you can assign\r
-another month via the B<-m> switch or a time period via the B<-p>\r
-switch; the latter takes preference.\r
-\r
-By default B<gatherstats> will process all types of information; you\r
-can change that using the B<-t> switch and assigning the type of\r
-information to process. Currently only processing of the number of\r
-postings per group per month is implemented anyway, so that doesn't\r
-matter yet.\r
-\r
-Possible information types include:\r
-\r
-=over 3\r
-\r
-=item B<groups> (postings per group per month)\r
-\r
-B<gatherstats> will examine Newsgroups: headers. Crosspostings will be\r
-counted for each single group they appear in. Groups not in I<TLH>\r
-will be ignored.\r
-\r
-B<gatherstats> will also add up the number of postings for each\r
-hierarchy level, but only count each posting once. A posting to\r
-de.alt.test will be counted for de.alt.test, de.alt.ALL and de.ALL,\r
-respectively. A crossposting to de.alt.test and de.alt.admin, on the\r
-other hand, will be counted for de.alt.test and de.alt.admin each, but\r
-only once for de.alt.ALL and de.ALL.\r
-\r
-Data is written to I<DBTableGrps> (see doc/INSTALL).\r
-\r
-=back\r
-\r
-=head2 Configuration\r
-\r
-F<gatherstats.pl> will read its configuration from F<newsstats.conf>\r
-which should be present in the same directory via Config::Auto.\r
-\r
-See doc/INSTALL for an overview of possible configuration options.\r
-\r
-You can override configuration options via the B<-n>, B<-r>, B<-g>,\r
-B<-c> and B<-s> switches, respectively.\r
-\r
-=head1 OPTIONS\r
-\r
-=over 3\r
-\r
-=item B<-V> (version)\r
-\r
-Print out version and copyright information on B<yapfaq> and exit.\r
-\r
-=item B<-h> (help)\r
-\r
-Print this man page and exit.\r
-\r
-=item B<-d> (debug)\r
-\r
-Output debugging information to STDOUT while processing (number of\r
-postings per group).\r
-\r
-=item B<-o> (output only)\r
-\r
-Do not write results to database. You should use B<-d> in conjunction\r
-with B<-o> ... everything else seems a bit pointless.\r
-\r
-=item B<-m> I<YYYY-MM> (month)\r
-\r
-Set processing period to a month in YYYY-MM format. Ignored if B<-p>\r
-is set.\r
-\r
-=item B<-p> I<YYYY-MM:YYYY-MM> (period)\r
-\r
-Set processing period to a time period between two month, each in\r
-YYYY-MM format, separated by a colon. Overrides B<-m>.\r
-\r
-=item B<-t> I<type> (type)\r
-\r
-Set processing type to one of I<all> and I<groups>. Defaults to all\r
-(and is currently rather pointless as only I<groups> has been\r
-implemented).\r
-\r
-=item B<-n> I<TLH> (newsgroup hierarchy)\r
-\r
-Override I<TLH> from F<newsstats.conf>.\r
-\r
-=item B<-r> I<table> (raw data table)\r
-\r
-Override I<DBTableRaw> from F<newsstats.conf>.\r
-\r
-=item B<-g> I<table> (postings per group table)\r
-\r
-Override I<DBTableGrps> from F<newsstats.conf>.\r
-\r
-=item B<-c> I<table> (client data table)\r
-\r
-Override I<DBTableClnts> from F<newsstats.conf>.\r
-\r
-=item B<-s> I<table> (server/host data table)\r
-\r
-Override I<DBTableHosts> from F<newsstats.conf>.\r
-\r
-=back\r
-\r
-=head1 INSTALLATION\r
-\r
-See doc/INSTALL.\r
-\r
-=head1 EXAMPLES\r
-\r
-Process all types of information for lasth month:\r
-\r
- gatherstats\r
-\r
-Do a dry run, showing results of processing:\r
-\r
- gatherstats -do\r
-\r
-Process all types of information for January of 2010:\r
-\r
- gatherstats -m 2010-01\r
-\r
-Process only number of postings for the year of 2010:\r
-\r
- gatherstats -p 2010-01:2010-12 -t groups\r
-\r
-=head1 FILES\r
-\r
-=over 4\r
-\r
-=item F<gatherstats.pl>\r
-\r
-The script itself.\r
-\r
-=item F<NewsStats.pm>\r
-\r
-Library functions for the NewsStats package.\r
-\r
-=item F<newsstats.conf>\r
-\r
-Runtime configuration file for B<yapfaq>.\r
-\r
-=back\r
-\r
-=head1 BUGS\r
-\r
-Please report any bugs or feature requests to the author or use the\r
-bug tracker at L<http://bugs.th-h.de/>!\r
-\r
-=head1 SEE ALSO\r
-\r
-=over 2\r
-\r
-=item -\r
-\r
-doc/README\r
-\r
-=item -\r
-\r
-doc/INSTALL\r
-\r
-=back\r
-\r
-This script is part of the B<NewsStats> package.\r
-\r
-=head1 AUTHOR\r
-\r
-Thomas Hochstein <thh@inter.net>\r
-\r
-=head1 COPYRIGHT AND LICENSE\r
-\r
-Copyright (c) 2010 Thomas Hochstein <thh@inter.net>\r
-\r
-This program is free software; you may redistribute it and/or modify it\r
-under the same terms as Perl itself.\r
-\r
-=cut\r
+#! /usr/bin/perl -W
+#
+# gatherstats.pl
+#
+# This script will gather statistical information from a database
+# containing headers and other information from a INN feed.
+#
+# It is part of the NewsStats package.
+#
+# Copyright (c) 2010 Thomas Hochstein <thh@inter.net>
+#
+# It can be redistributed and/or modified under the same terms under
+# which Perl itself is published.
+
+BEGIN {
+ our $VERSION = "0.01";
+ use File::Basename;
+ push(@INC, dirname($0));
+}
+use strict;
+
+use NewsStats qw(:DEFAULT :TimePeriods ListNewsgroups);
+
+use DBI;
+
+################################# Definitions ##################################
+
+# define types of information that can be gathered
+# all / groups (/ clients / hosts)
+my %LegalTypes;
+@LegalTypes{('all','groups')} = ();
+
+################################# Main program #################################
+
+### read commandline options
+my %Options = &ReadOptions('dom:p:t:n:r:g:c:s:');
+
+### read configuration
+my %Conf = %{ReadConfig('newsstats.conf')};
+
+### override configuration via commandline options
+my %ConfOverride;
+$ConfOverride{'DBTableRaw'} = $Options{'r'} if $Options{'r'};
+$ConfOverride{'DBTableGrps'} = $Options{'g'} if $Options{'g'};
+$ConfOverride{'DBTableClnts'} = $Options{'c'} if $Options{'c'};
+$ConfOverride{'DBTableHosts'} = $Options{'s'} if $Options{'s'};
+$ConfOverride{'TLH'} = $Options{'n'} if $Options{'n'};
+&OverrideConfig(\%Conf,\%ConfOverride);
+
+### get type of information to gather, defaulting to 'all'
+$Options{'t'} = 'all' if !$Options{'t'};
+die "$MySelf: E: Unknown type '-t $Options{'t'}'!\n" if !exists($LegalTypes{$Options{'t'}});
+
+### get time period (-m or -p)
+my ($StartMonth,$EndMonth) = &GetTimePeriod($Options{'m'},$Options{'p'});
+
+### init database
+my $DBHandle = InitDB(\%Conf,1);
+
+### get data for each month
+warn "$MySelf: W: Output only mode. Database is not updated.\n" if $Options{'o'};
+foreach my $Month (&ListMonth($StartMonth,$EndMonth)) {
+
+ print "---------- $Month ----------\n" if $Options{'d'};
+
+ if ($Options{'t'} eq 'all' or $Options{'t'} eq 'groups') {
+ ### ----------------------------------------------
+ ### get groups data (number of postings per group)
+ # get groups data from raw table for given month
+ my $DBQuery = $DBHandle->prepare(sprintf("SELECT newsgroups FROM %s.%s WHERE day LIKE ? AND NOT disregard",$Conf{'DBDatabase'},$Conf{'DBTableRaw'}));
+ $DBQuery->execute($Month.'-%') or die sprintf("$MySelf: E: Can't get groups data for %s from %s.%s: $DBI::errstr\n",$Month,$Conf{'DBDatabase'},$Conf{'DBTableRaw'});
+
+ # count postings per group
+ my %Postings;
+
+ while (($_) = $DBQuery->fetchrow_array) {
+ # get list oft newsgroups and hierarchies from Newsgroups:
+ my %Newsgroups = ListNewsgroups($_);
+ # count each newsgroup and hierarchy once
+ foreach (sort keys %Newsgroups) {
+ # don't count newsgroup/hierarchy in wrong TLH
+ next if(defined($Conf{'TLH'}) and !/^$Conf{'TLH'}/);
+ $Postings{$_}++;
+ };
+ };
+
+ print "----- GroupStats -----\n" if $Options{'d'};
+ foreach my $Newsgroup (sort keys %Postings) {
+ print "$Newsgroup => $Postings{$Newsgroup}\n" if $Options{'d'};
+ if (!$Options{'o'}) {
+ # write to database
+ $DBQuery = $DBHandle->prepare(sprintf("REPLACE INTO %s.%s (month,newsgroup,postings) VALUES (?, ?, ?)",$Conf{'DBDatabase'},$Conf{'DBTableGrps'}));
+ $DBQuery->execute($Month, $Newsgroup, $Postings{$Newsgroup}) or die sprintf("$MySelf: E: Can't write groups data for %s/%s to %s.%s: $DBI::errstr\n",$Month,$Newsgroup,$Conf{'DBDatabase'},$Conf{'DBTableGrps'});
+ $DBQuery->finish;
+ };
+ };
+ } else {
+ # other types of information go here - later on
+ };
+};
+
+### close handles
+$DBHandle->disconnect;
+
+__END__
+
+################################ Documentation #################################
+
+=head1 NAME
+
+gatherstats - process statistical data from a raw source
+
+=head1 SYNOPSIS
+
+B<gatherstats> [B<-Vhdo>] [B<-m> I<YYYY-MM>] [B<-p> I<YYYY-MM:YYYY-MM>] [B<-t> I<type>] [B<-n> I<TLH>] [B<-r> I<database table>] [B<-g> I<database table>] [B<-c> I<database table>] [B<-s> I<database table>]
+
+=head1 REQUIREMENTS
+
+See doc/README: Perl 5.8.x itself and the following modules from CPAN:
+
+=over 2
+
+=item -
+
+Config::Auto
+
+=item -
+
+DBI
+
+=back
+
+=head1 DESCRIPTION
+
+This script will extract and process statistical information from a
+database table which is fed from F<feedlog.pl> for a given time period
+and write its results to (an)other database table(s).
+
+The time period to act on defaults to last month; you can assign
+another month via the B<-m> switch or a time period via the B<-p>
+switch; the latter takes preference.
+
+By default B<gatherstats> will process all types of information; you
+can change that using the B<-t> switch and assigning the type of
+information to process. Currently only processing of the number of
+postings per group per month is implemented anyway, so that doesn't
+matter yet.
+
+Possible information types include:
+
+=over 3
+
+=item B<groups> (postings per group per month)
+
+B<gatherstats> will examine Newsgroups: headers. Crosspostings will be
+counted for each single group they appear in. Groups not in I<TLH>
+will be ignored.
+
+B<gatherstats> will also add up the number of postings for each
+hierarchy level, but only count each posting once. A posting to
+de.alt.test will be counted for de.alt.test, de.alt.ALL and de.ALL,
+respectively. A crossposting to de.alt.test and de.alt.admin, on the
+other hand, will be counted for de.alt.test and de.alt.admin each, but
+only once for de.alt.ALL and de.ALL.
+
+Data is written to I<DBTableGrps> (see doc/INSTALL).
+
+=back
+
+=head2 Configuration
+
+F<gatherstats.pl> will read its configuration from F<newsstats.conf>
+which should be present in the same directory via Config::Auto.
+
+See doc/INSTALL for an overview of possible configuration options.
+
+You can override configuration options via the B<-n>, B<-r>, B<-g>,
+B<-c> and B<-s> switches, respectively.
+
+=head1 OPTIONS
+
+=over 3
+
+=item B<-V> (version)
+
+Print out version and copyright information on B<yapfaq> and exit.
+
+=item B<-h> (help)
+
+Print this man page and exit.
+
+=item B<-d> (debug)
+
+Output debugging information to STDOUT while processing (number of
+postings per group).
+
+=item B<-o> (output only)
+
+Do not write results to database. You should use B<-d> in conjunction
+with B<-o> ... everything else seems a bit pointless.
+
+=item B<-m> I<YYYY-MM> (month)
+
+Set processing period to a month in YYYY-MM format. Ignored if B<-p>
+is set.
+
+=item B<-p> I<YYYY-MM:YYYY-MM> (period)
+
+Set processing period to a time period between two month, each in
+YYYY-MM format, separated by a colon. Overrides B<-m>.
+
+=item B<-t> I<type> (type)
+
+Set processing type to one of I<all> and I<groups>. Defaults to all
+(and is currently rather pointless as only I<groups> has been
+implemented).
+
+=item B<-n> I<TLH> (newsgroup hierarchy)
+
+Override I<TLH> from F<newsstats.conf>.
+
+=item B<-r> I<table> (raw data table)
+
+Override I<DBTableRaw> from F<newsstats.conf>.
+
+=item B<-g> I<table> (postings per group table)
+
+Override I<DBTableGrps> from F<newsstats.conf>.
+
+=item B<-c> I<table> (client data table)
+
+Override I<DBTableClnts> from F<newsstats.conf>.
+
+=item B<-s> I<table> (server/host data table)
+
+Override I<DBTableHosts> from F<newsstats.conf>.
+
+=back
+
+=head1 INSTALLATION
+
+See doc/INSTALL.
+
+=head1 EXAMPLES
+
+Process all types of information for lasth month:
+
+ gatherstats
+
+Do a dry run, showing results of processing:
+
+ gatherstats -do
+
+Process all types of information for January of 2010:
+
+ gatherstats -m 2010-01
+
+Process only number of postings for the year of 2010:
+
+ gatherstats -p 2010-01:2010-12 -t groups
+
+=head1 FILES
+
+=over 4
+
+=item F<gatherstats.pl>
+
+The script itself.
+
+=item F<NewsStats.pm>
+
+Library functions for the NewsStats package.
+
+=item F<newsstats.conf>
+
+Runtime configuration file for B<yapfaq>.
+
+=back
+
+=head1 BUGS
+
+Please report any bugs or feature requests to the author or use the
+bug tracker at L<http://bugs.th-h.de/>!
+
+=head1 SEE ALSO
+
+=over 2
+
+=item -
+
+doc/README
+
+=item -
+
+doc/INSTALL
+
+=back
+
+This script is part of the B<NewsStats> package.
+
+=head1 AUTHOR
+
+Thomas Hochstein <thh@inter.net>
+
+=head1 COPYRIGHT AND LICENSE
+
+Copyright (c) 2010 Thomas Hochstein <thh@inter.net>
+
+This program is free software; you may redistribute it and/or modify it
+under the same terms as Perl itself.
+
+=cut
-#! /usr/bin/perl -W\r
-#\r
-# groupstats.pl\r
-#\r
-# This script will get statistical data on newgroup usage\r
-# form a database.\r
-# \r
-# It is part of the NewsStats package.\r
-#\r
-# Copyright (c) 2010 Thomas Hochstein <thh@inter.net>\r
-#\r
-# It can be redistributed and/or modified under the same terms under \r
-# which Perl itself is published.\r
-\r
-BEGIN {\r
- our $VERSION = "0.01";\r
- use File::Basename;\r
- push(@INC, dirname($0));\r
-}\r
-use strict;\r
-\r
-use NewsStats qw(:DEFAULT :TimePeriods :Output :SQLHelper);\r
-\r
-use DBI;\r
-\r
-################################# Main program #################################\r
-\r
-### read commandline options\r
-my %Options = &ReadOptions('m:p:n:o:t:l:b:iscqdg:');\r
-\r
-### read configuration\r
-my %Conf = %{ReadConfig('newsstats.conf')};\r
-\r
-### override configuration via commandline options\r
-my %ConfOverride;\r
-$ConfOverride{'DBTableGrps'} = $Options{'g'} if $Options{'g'};\r
-&OverrideConfig(\%Conf,\%ConfOverride);\r
-\r
-### check for incompatible command line options\r
-# you can't mix '-t', '-b' and '-l'\r
-# -b/-l take preference over -t, and -b takes preference over -l\r
-if ($Options{'b'} or $Options{'l'}) {\r
- if ($Options{'t'}) {\r
- # drop -t\r
- warn ("$MySelf: W: You cannot combine thresholds (-t) and top lists (-b) or levels (-l). Threshold '-t $Options{'t'}' was ignored.\n");\r
- undef($Options{'t'});\r
- };\r
- if ($Options{'b'} and $Options{'l'}) {\r
- # drop -l\r
- warn ("$MySelf: W: You cannot combine top lists (-b) and levels (-l). Level '-l $Options{'l'}' was ignored.\n");\r
- undef($Options{'l'});\r
- };\r
- # -q/-d don't work with -b or -l\r
- warn ("$MySelf: W: Sorting by number of postings (-q) ignored due to top list mode (-b) / levels (-l).\n") if $Options{'q'};\r
- warn ("$MySelf: W: Reverse sorting (-d) ignored due to top list mode (-b) / levels (-l).\n") if $Options{'d'};\r
-};\r
-\r
-### check output type\r
-# default output type to 'dump'\r
-$Options{'o'} = 'dump' if !$Options{'o'};\r
-# fail if more than one newsgroup is combined with 'dumpgroup' type\r
-die ("$MySelf: E: You cannot combine newsgroup lists (-n) with more than one group with '-o dumpgroup'!\n") if ($Options{'o'} eq 'dumpgroup' and defined($Options{'n'}) and $Options{'n'} =~ /:|\*/);\r
-# accept 'dumpgroup' only with -n\r
-if ($Options{'o'} eq 'dumpgroup' and !defined($Options{'n'})) {\r
- $Options{'o'} = 'dump';\r
- warn ("$MySelf: W: You must submit exactly one newsgroup ('-n news.group') for '-o dumpgroup'. Output type was set to 'dump'.\n");\r
-};\r
-# set output type to 'pretty' for -l\r
-if ($Options{'l'}) {\r
- $Options{'o'} = 'pretty';\r
- warn ("$MySelf: W: Output type forced to '-o pretty' due to usage of '-l'.\n");\r
-};\r
-\r
-### get time period\r
-my ($StartMonth,$EndMonth) = &GetTimePeriod($Options{'m'},$Options{'p'});\r
-# reset to one month for 'dump' output type\r
-if ($Options{'o'} eq 'dump' and $Options{'p'}) {\r
- $StartMonth = $EndMonth;\r
- warn ("$MySelf: W: You cannot combine time periods (-p) with '-o dump'. Month was set to $StartMonth.\n");\r
-};\r
-\r
-### init database\r
-my $DBHandle = InitDB(\%Conf,1);\r
-\r
-### create report\r
-# get list of newsgroups (-n)\r
-my ($QueryPart,@GroupList);\r
-my $Newsgroups = $Options{'n'};\r
-if ($Newsgroups) {\r
- # explode list of newsgroups for WHERE clause\r
- ($QueryPart,@GroupList) = &SQLGroupList($Newsgroups);\r
-} else {\r
- # set to dummy value (always true)\r
- $QueryPart = 1;\r
-};\r
-\r
-# manage thresholds\r
-if (defined($Options{'t'})) {\r
- if ($Options{'i'}) {\r
- # -i: list groups below threshold\r
- $QueryPart .= ' AND postings < ?';\r
- } else {\r
- # default: list groups above threshold\r
- $QueryPart .= ' AND postings > ?';\r
- };\r
- # push threshold to GroupList to match number of binding vars for DBQuery->execute\r
- push @GroupList,$Options{'t'};\r
-}\r
-\r
-# construct WHERE clause\r
-# $QueryPart is "list of newsgroup" (or 1),\r
-# &SQLHierarchies() takes care of the exclusion of hierarchy levels (.ALL)\r
-# according to setting of -s\r
-my $WhereClause = sprintf('month BETWEEN ? AND ? AND %s %s',$QueryPart,&SQLHierarchies($Options{'s'}));\r
-\r
-# get lenght of longest newsgroup delivered by query for formatting purposes\r
-# FIXME\r
-my $MaxLength = &GetMaxLenght($DBHandle,$Conf{'DBTableGrps'},'newsgroup',$WhereClause,$StartMonth,$EndMonth,@GroupList);\r
-\r
-my ($OrderClause,$DBQuery);\r
-# -b (best of / top list) defined?\r
-if (!defined($Options{'b'}) and !defined($Options{'l'})) {\r
- # default: neither -b nor -l\r
- # set ordering (ORDER BY) to "newsgroups" or "postings", "ASC" or "DESC"\r
- # according to -q and -d\r
- $OrderClause = 'newsgroup';\r
- $OrderClause = 'postings' if $Options{'q'};\r
- $OrderClause .= ' DESC' if $Options{'d'};\r
- # prepare query: get number of postings per group from groups table for given months and newsgroups\r
- $DBQuery = $DBHandle->prepare(sprintf("SELECT month,newsgroup,postings FROM %s.%s WHERE %s ORDER BY month,%s",$Conf{'DBDatabase'},$Conf{'DBTableGrps'},$WhereClause,$OrderClause));\r
-} elsif ($Options{'b'}) {\r
- # -b is set (then -l can't be!)\r
- # set sorting order (-i)\r
- if ($Options{'i'}) {\r
- $OrderClause = 'postings';\r
- } else {\r
- $OrderClause = 'postings DESC';\r
- };\r
- # push LIMIT to GroupList to match number of binding vars for DBQuery->execute\r
- push @GroupList,$Options{'b'};\r
- # prepare query: get sum of postings per group from groups table for given months and newsgroups with LIMIT\r
- $DBQuery = $DBHandle->prepare(sprintf("SELECT newsgroup,SUM(postings) AS postings FROM %s.%s WHERE %s GROUP BY newsgroup ORDER BY %s,newsgroup LIMIT ?",$Conf{'DBDatabase'},$Conf{'DBTableGrps'},$WhereClause,$OrderClause));\r
-} else {\r
- # -l must be set now, as all other cases have been taken care of\r
- # set sorting order (-i)\r
- if ($Options{'i'}) {\r
- $OrderClause = '<';\r
- } else {\r
- $OrderClause = '>';\r
- };\r
- # push level and $StartMonth,$EndMonth - again - to GroupList to match number of binding vars for DBQuery->execute\r
- # FIXME -- together with the query (see below)\r
- push @GroupList,$Options{'l'};\r
- push @GroupList,$StartMonth,$EndMonth;\r
- # prepare query: get number of postings per group from groups table for given months and \r
- # FIXME -- this query is ... in dire need of impromevent\r
- $DBQuery = $DBHandle->prepare(sprintf("SELECT month,newsgroup,postings FROM %s.%s WHERE newsgroup IN (SELECT newsgroup FROM %s.%s WHERE %s GROUP BY newsgroup HAVING MAX(postings) %s ?) AND %s ORDER BY newsgroup,month",$Conf{'DBDatabase'},$Conf{'DBTableGrps'},$Conf{'DBDatabase'},$Conf{'DBTableGrps'},$WhereClause,$OrderClause,$WhereClause));\r
-};\r
-\r
-# execute query\r
-$DBQuery->execute($StartMonth,$EndMonth,@GroupList)\r
- or die sprintf("$MySelf: E: Can't get groups data for %s to %s from %s.%s: %s\n",$StartMonth,$EndMonth,$Conf{'DBDatabase'},$Conf{'DBTableGrps'},$DBI::errstr);\r
-\r
-# output results\r
-# print caption (-c) with time period if -m or -p is set\r
-# FIXME - month or period should handled differently\r
-printf ("----- Report from %s to %s\n",$StartMonth,$EndMonth) if $Options{'c'} and ($Options{'m'} or $Options{'p'});\r
-# print caption (-c) with newsgroup list if -n is set\r
-printf ("----- Newsgroups: %s\n",join(',',split(/:/,$Newsgroups))) if $Options{'c'} and $Options{'n'};\r
-# print caption (-c) with threshold if -t is set, taking -i in account\r
-printf ("----- Threshold: %s %u\n",$Options{'i'} ? '<' : '>',$Options{'t'}) if $Options{'c'} and $Options{'t'};\r
-if (!defined($Options{'b'}) and !defined($Options{'l'})) {\r
- # default: neither -b nor -l\r
- &OutputData($Options{'o'},$DBQuery,$MaxLength);\r
-} elsif ($Options{'b'}) {\r
- # -b is set (then -l can't be!)\r
- # we have to read in the query results ourselves, as they do not have standard layout\r
- while (my ($Newsgroup,$Postings) = $DBQuery->fetchrow_array) {\r
- # we just assign "top x" or "bottom x" instead of a month for the caption\r
- # FIXME\r
- print &FormatOutput($Options{'o'}, ($Options{'i'} ? 'Bottom ' : 'Top ').$Options{'b'}, $Newsgroup, $Postings, $MaxLength);\r
- };\r
-} else {\r
- # -l must be set now, as all other cases have been taken care of\r
- # we have to read in the query results ourselves, as they do not have standard layout\r
- while (my ($Month,$Newsgroup,$Postings) = $DBQuery->fetchrow_array) {\r
- # we just switch $Newsgroups and $Month for output generation\r
- # FIXME\r
- print &FormatOutput($Options{'o'}, $Newsgroup, $Month, $Postings, 7);\r
- };\r
-};\r
-\r
-### close handles\r
-$DBHandle->disconnect;\r
-\r
-__END__\r
-\r
-################################ Documentation #################################\r
-\r
-=head1 NAME\r
-\r
-groupstats - create reports on newsgroup usage\r
-\r
-=head1 SYNOPSIS\r
-\r
-B<groupstats> [B<-Vhiscqd>] [B<-m> I<YYYY-MM>] [B<-p> I<YYYY-MM:YYYY-MM>] [B<-n> I<newsgroup(s)>] [B<-t> I<threshold>] [B<-l> I<level>] [B<-b> I<number>] [B<-o> I<output type>] [B<-g> I<database table>]\r
-\r
-=head1 REQUIREMENTS\r
-\r
-See doc/README: Perl 5.8.x itself and the following modules from CPAN:\r
-\r
-=over 2\r
-\r
-=item -\r
-\r
-Config::Auto\r
-\r
-=item -\r
-\r
-DBI\r
-\r
-=back\r
-\r
-=head1 DESCRIPTION\r
-\r
-This script create reports on newsgroup usage (number of postings per\r
-group per month) taken from result tables created by\r
-F<gatherstats.pl>.\r
-\r
-The time period to act on defaults to last month; you can assign\r
-another month via the B<-m> switch or a time period via the B<-p>\r
-switch; the latter takes preference.\r
-\r
-B<groupstats> will process all newsgroups by default; you can limit\r
-that to only some newsgroups by supplying a list of those groups via\r
-B<-n> (see below). You can include hierarchy levels in the output by\r
-adding the B<-s> switch (see below).\r
-\r
-Furthermore you can set a threshold via B<-t> so that only newsgroups\r
-with more postings per month will be included in the report. You can\r
-invert that by the B<-i> switch so only newsgroups with less than\r
-I<threshold> postings per month will be included.\r
-\r
-You can sort the output by number of postings per month instead of the\r
-default (alphabetical list of newsgroups) by using B<-q>; you can\r
-reverse the sorting order (from highest to lowest or in reversed\r
-alphabetical order) by using B<-d>.\r
-\r
-Furthermore, you can create a list of newsgroups that had consistently\r
-more (or less) than x postings per month during the whole report\r
-period by using B<-l> (together with B<i> as needed).\r
-\r
-Last but not least you can create a "best of" list of the top x\r
-newsgroups via B<-b> (or a "worst of" list by adding B<i>).\r
-\r
-By default, B<groupstats> will dump a very simple alphabetical list of\r
-newsgroups, one per line, followed by the number of postings in that\r
-month. This output format of course cannot sensibly be combined with\r
-time periods, so you can set the output format by using B<-o> (see\r
-below). Captions can be added by setting the B<-c> switch.\r
-\r
-=head2 Configuration\r
-\r
-F<groupstats.pl> will read its configuration from F<newsstats.conf>\r
-which should be present in the same directory via Config::Auto.\r
-\r
-See doc/INSTALL for an overview of possible configuration options.\r
-\r
-You can override configuration options via the B<-g> switch.\r
-\r
-=head1 OPTIONS\r
-\r
-=over 3\r
-\r
-=item B<-V> (version)\r
-\r
-Print out version and copyright information on B<yapfaq> and exit.\r
-\r
-=item B<-h> (help)\r
-\r
-Print this man page and exit.\r
-\r
-=item B<-m> I<YYYY-MM> (month)\r
-\r
-Set processing period to a month in YYYY-MM format. Ignored if B<-p>\r
-is set.\r
-\r
-=item B<-p> I<YYYY-MM:YYYY-MM> (period)\r
-\r
-Set processing period to a time period between two month, each in\r
-YYYY-MM format, separated by a colon. Overrides B<-m>.\r
-\r
-=item B<-n> I<newsgroup(s)> (newsgroups)\r
-\r
-Limit processing to a certain set of newsgroups. I<newsgroup(s)> can\r
-be a single newsgroup name (de.alt.test), a newsgroup hierarchy\r
-(de.alt.*) or a list of either of these, separated by colons, for\r
-example\r
-\r
- de.test:de.alt.test:de.newusers.*\r
-\r
-=item B<-t> I<threshold> (threshold)\r
-\r
-Only include newsgroups with more than I<threshold> postings per\r
-month. Can be inverted by the B<-i> switch so that only newsgroups\r
-with less than I<threshold> postings will be included.\r
-\r
-This setting will be ignored if B<-l> or B<-b> is set.\r
-\r
-=item B<-l> I<level> (level)\r
-\r
-Only include newsgroups with more than I<level> postings per\r
-month, every month during the whole reporting period. Can be inverted\r
-by the B<-i> switch so that only newsgroups with less than I<level>\r
-postings every single month will be included. Output will be ordered\r
-by newsgroup name, followed by month.\r
-\r
-This setting will be ignored if B<-b> is set. Overrides B<-t> and\r
-can't be used together with B<-q> or B<-d>.\r
-\r
-=item B<-b> I<n> (best of)\r
-\r
-Create a list of the I<n> newsgroups with the most postings over the\r
-whole reporting period. Can be inverted by the B<-i> switch so that a\r
-list of the I<n> newsgroups with the least postings over the whole\r
-period is generated. Output will be ordered by sum of postings.\r
-\r
-Overrides B<-t> and B<-l> and can't be used together with B<-q> or\r
-B<-d>. Output format is set to I<pretty> (see below).\r
-\r
-=item B<-i> (invert)\r
-\r
-Used in conjunction with B<-t>, B<-l> or B<-b> to set a lower\r
-threshold or level or generate a "bottom list" instead of a top list.\r
-\r
-=item B<-s> (sum per hierarchy level)\r
-\r
-Include "virtual" groups for every hierarchy level in output, for\r
-example:\r
-\r
- de.alt.ALL 10\r
- de.alt.test 5\r
- de.alt.admin 7\r
-\r
-See the B<gatherstats> man page for details.\r
-\r
-=item B<-o> I<output type> (output format)\r
-\r
-Set output format. Default is I<dump>, consisting of an alphabetical\r
-list of newsgroups, each on a new line, followed by the number of\r
-postings in that month. This default format can't be used with time\r
-periods of more than one month.\r
-\r
-I<list> format is like I<dump>, but will print the month in front of\r
-the newsgroup name.\r
-\r
-I<dumpgroup> format can only be use with a group list (see B<-n>) of\r
-exactly one newsgroup and is like I<dump>, but will output months,\r
-followed by the number of postings.\r
-\r
-If you don't need easily parsable output, you'll mostly use I<pretty>\r
-format, which will print a header for each new month and try to align\r
-newsgroup names and posting counts. Usage of B<-b> will force this\r
-format.\r
-\r
-=item B<-c> (captions)\r
-\r
-Add captions to output (reporting period, newsgroups list, threshold).\r
-\r
-=item B<-q> (quantity of postings)\r
-\r
-Sort by number of postings instead of by newsgroup names.\r
-\r
-Cannot be used with B<-l> or B<-b>.\r
-\r
-=item B<-d> (descending)\r
-\r
-Change sort order to descending.\r
-\r
-Cannot be used with B<-l> or B<-b>.\r
-\r
-=item B<-g> I<table> (postings per group table)\r
-\r
-Override I<DBTableGrps> from F<newsstats.conf>.\r
-\r
-=back\r
-\r
-=head1 INSTALLATION\r
-\r
-See doc/INSTALL.\r
-\r
-=head1 EXAMPLES\r
-\r
-Show number of postings per group for lasth month in I<dump> format:\r
-\r
- groupstats\r
-\r
-Show that report for January of 2010 and de.alt.* plus de.test,\r
-including display of hierarchy levels:\r
-\r
- groupstats -m 2010-01 -n de.alt.*:de.test -s\r
-\r
-Show that report for the year of 2010 in I<pretty> format:\r
-\r
- groupstats -p 2010-01:2010-12 -o pretty\r
-\r
-Only show newsgroups with less than 30 postings last month, ordered\r
-by number of postings, descending, in I<pretty> format:\r
-\r
- groupstats -iqdt 30 -o pretty\r
-\r
-Show top 10 for the first half-year of of 2010 in I<pretty> format:\r
-\r
- groupstats -p 2010-01:2010-06 -b 10 -o pretty\r
-\r
-Report all groups that had less than 30 postings every singele month\r
-in the year of 2010 (I<pretty> format is forced)\r
-\r
- groupstats -p 2010-01:2010-12 -il 30\r
-\r
-=head1 FILES\r
-\r
-=over 4\r
-\r
-=item F<groupstats.pl>\r
-\r
-The script itself.\r
-\r
-=item F<NewsStats.pm>\r
-\r
-Library functions for the NewsStats package.\r
-\r
-=item F<newsstats.conf>\r
-\r
-Runtime configuration file for B<yapfaq>.\r
-\r
-=back\r
-\r
-=head1 BUGS\r
-\r
-Please report any bugs or feature requests to the author or use the\r
-bug tracker at L<http://bugs.th-h.de/>!\r
-\r
-=head1 SEE ALSO\r
-\r
-=over 2\r
-\r
-=item -\r
-\r
-doc/README\r
-\r
-=item -\r
-\r
-doc/INSTALL\r
-\r
-=item -\r
-\r
-gatherstats -h\r
-\r
-=back\r
-\r
-This script is part of the B<NewsStats> package.\r
-\r
-=head1 AUTHOR\r
-\r
-Thomas Hochstein <thh@inter.net>\r
-\r
-=head1 COPYRIGHT AND LICENSE\r
-\r
-Copyright (c) 2010 Thomas Hochstein <thh@inter.net>\r
-\r
-This program is free software; you may redistribute it and/or modify it\r
-under the same terms as Perl itself.\r
-\r
-=cut\r
+#! /usr/bin/perl -W
+#
+# groupstats.pl
+#
+# This script will get statistical data on newgroup usage
+# form a database.
+#
+# It is part of the NewsStats package.
+#
+# Copyright (c) 2010 Thomas Hochstein <thh@inter.net>
+#
+# It can be redistributed and/or modified under the same terms under
+# which Perl itself is published.
+
+BEGIN {
+ our $VERSION = "0.01";
+ use File::Basename;
+ push(@INC, dirname($0));
+}
+use strict;
+
+use NewsStats qw(:DEFAULT :TimePeriods :Output :SQLHelper);
+
+use DBI;
+
+################################# Main program #################################
+
+### read commandline options
+my %Options = &ReadOptions('m:p:n:o:t:l:b:iscqdg:');
+
+### read configuration
+my %Conf = %{ReadConfig('newsstats.conf')};
+
+### override configuration via commandline options
+my %ConfOverride;
+$ConfOverride{'DBTableGrps'} = $Options{'g'} if $Options{'g'};
+&OverrideConfig(\%Conf,\%ConfOverride);
+
+### check for incompatible command line options
+# you can't mix '-t', '-b' and '-l'
+# -b/-l take preference over -t, and -b takes preference over -l
+if ($Options{'b'} or $Options{'l'}) {
+ if ($Options{'t'}) {
+ # drop -t
+ warn ("$MySelf: W: You cannot combine thresholds (-t) and top lists (-b) or levels (-l). Threshold '-t $Options{'t'}' was ignored.\n");
+ undef($Options{'t'});
+ };
+ if ($Options{'b'} and $Options{'l'}) {
+ # drop -l
+ warn ("$MySelf: W: You cannot combine top lists (-b) and levels (-l). Level '-l $Options{'l'}' was ignored.\n");
+ undef($Options{'l'});
+ };
+ # -q/-d don't work with -b or -l
+ warn ("$MySelf: W: Sorting by number of postings (-q) ignored due to top list mode (-b) / levels (-l).\n") if $Options{'q'};
+ warn ("$MySelf: W: Reverse sorting (-d) ignored due to top list mode (-b) / levels (-l).\n") if $Options{'d'};
+};
+
+### check output type
+# default output type to 'dump'
+$Options{'o'} = 'dump' if !$Options{'o'};
+# fail if more than one newsgroup is combined with 'dumpgroup' type
+die ("$MySelf: E: You cannot combine newsgroup lists (-n) with more than one group with '-o dumpgroup'!\n") if ($Options{'o'} eq 'dumpgroup' and defined($Options{'n'}) and $Options{'n'} =~ /:|\*/);
+# accept 'dumpgroup' only with -n
+if ($Options{'o'} eq 'dumpgroup' and !defined($Options{'n'})) {
+ $Options{'o'} = 'dump';
+ warn ("$MySelf: W: You must submit exactly one newsgroup ('-n news.group') for '-o dumpgroup'. Output type was set to 'dump'.\n");
+};
+# set output type to 'pretty' for -l
+if ($Options{'l'}) {
+ $Options{'o'} = 'pretty';
+ warn ("$MySelf: W: Output type forced to '-o pretty' due to usage of '-l'.\n");
+};
+
+### get time period
+my ($StartMonth,$EndMonth) = &GetTimePeriod($Options{'m'},$Options{'p'});
+# reset to one month for 'dump' output type
+if ($Options{'o'} eq 'dump' and $Options{'p'}) {
+ $StartMonth = $EndMonth;
+ warn ("$MySelf: W: You cannot combine time periods (-p) with '-o dump'. Month was set to $StartMonth.\n");
+};
+
+### init database
+my $DBHandle = InitDB(\%Conf,1);
+
+### create report
+# get list of newsgroups (-n)
+my ($QueryPart,@GroupList);
+my $Newsgroups = $Options{'n'};
+if ($Newsgroups) {
+ # explode list of newsgroups for WHERE clause
+ ($QueryPart,@GroupList) = &SQLGroupList($Newsgroups);
+} else {
+ # set to dummy value (always true)
+ $QueryPart = 1;
+};
+
+# manage thresholds
+if (defined($Options{'t'})) {
+ if ($Options{'i'}) {
+ # -i: list groups below threshold
+ $QueryPart .= ' AND postings < ?';
+ } else {
+ # default: list groups above threshold
+ $QueryPart .= ' AND postings > ?';
+ };
+ # push threshold to GroupList to match number of binding vars for DBQuery->execute
+ push @GroupList,$Options{'t'};
+}
+
+# construct WHERE clause
+# $QueryPart is "list of newsgroup" (or 1),
+# &SQLHierarchies() takes care of the exclusion of hierarchy levels (.ALL)
+# according to setting of -s
+my $WhereClause = sprintf('month BETWEEN ? AND ? AND %s %s',$QueryPart,&SQLHierarchies($Options{'s'}));
+
+# get lenght of longest newsgroup delivered by query for formatting purposes
+# FIXME
+my $MaxLength = &GetMaxLenght($DBHandle,$Conf{'DBTableGrps'},'newsgroup',$WhereClause,$StartMonth,$EndMonth,@GroupList);
+
+my ($OrderClause,$DBQuery);
+# -b (best of / top list) defined?
+if (!defined($Options{'b'}) and !defined($Options{'l'})) {
+ # default: neither -b nor -l
+ # set ordering (ORDER BY) to "newsgroups" or "postings", "ASC" or "DESC"
+ # according to -q and -d
+ $OrderClause = 'newsgroup';
+ $OrderClause = 'postings' if $Options{'q'};
+ $OrderClause .= ' DESC' if $Options{'d'};
+ # prepare query: get number of postings per group from groups table for given months and newsgroups
+ $DBQuery = $DBHandle->prepare(sprintf("SELECT month,newsgroup,postings FROM %s.%s WHERE %s ORDER BY month,%s",$Conf{'DBDatabase'},$Conf{'DBTableGrps'},$WhereClause,$OrderClause));
+} elsif ($Options{'b'}) {
+ # -b is set (then -l can't be!)
+ # set sorting order (-i)
+ if ($Options{'i'}) {
+ $OrderClause = 'postings';
+ } else {
+ $OrderClause = 'postings DESC';
+ };
+ # push LIMIT to GroupList to match number of binding vars for DBQuery->execute
+ push @GroupList,$Options{'b'};
+ # prepare query: get sum of postings per group from groups table for given months and newsgroups with LIMIT
+ $DBQuery = $DBHandle->prepare(sprintf("SELECT newsgroup,SUM(postings) AS postings FROM %s.%s WHERE %s GROUP BY newsgroup ORDER BY %s,newsgroup LIMIT ?",$Conf{'DBDatabase'},$Conf{'DBTableGrps'},$WhereClause,$OrderClause));
+} else {
+ # -l must be set now, as all other cases have been taken care of
+ # set sorting order (-i)
+ if ($Options{'i'}) {
+ $OrderClause = '<';
+ } else {
+ $OrderClause = '>';
+ };
+ # push level and $StartMonth,$EndMonth - again - to GroupList to match number of binding vars for DBQuery->execute
+ # FIXME -- together with the query (see below)
+ push @GroupList,$Options{'l'};
+ push @GroupList,$StartMonth,$EndMonth;
+ # prepare query: get number of postings per group from groups table for given months and
+ # FIXME -- this query is ... in dire need of impromevent
+ $DBQuery = $DBHandle->prepare(sprintf("SELECT month,newsgroup,postings FROM %s.%s WHERE newsgroup IN (SELECT newsgroup FROM %s.%s WHERE %s GROUP BY newsgroup HAVING MAX(postings) %s ?) AND %s ORDER BY newsgroup,month",$Conf{'DBDatabase'},$Conf{'DBTableGrps'},$Conf{'DBDatabase'},$Conf{'DBTableGrps'},$WhereClause,$OrderClause,$WhereClause));
+};
+
+# execute query
+$DBQuery->execute($StartMonth,$EndMonth,@GroupList)
+ or die sprintf("$MySelf: E: Can't get groups data for %s to %s from %s.%s: %s\n",$StartMonth,$EndMonth,$Conf{'DBDatabase'},$Conf{'DBTableGrps'},$DBI::errstr);
+
+# output results
+# print caption (-c) with time period if -m or -p is set
+# FIXME - month or period should handled differently
+printf ("----- Report from %s to %s\n",$StartMonth,$EndMonth) if $Options{'c'} and ($Options{'m'} or $Options{'p'});
+# print caption (-c) with newsgroup list if -n is set
+printf ("----- Newsgroups: %s\n",join(',',split(/:/,$Newsgroups))) if $Options{'c'} and $Options{'n'};
+# print caption (-c) with threshold if -t is set, taking -i in account
+printf ("----- Threshold: %s %u\n",$Options{'i'} ? '<' : '>',$Options{'t'}) if $Options{'c'} and $Options{'t'};
+if (!defined($Options{'b'}) and !defined($Options{'l'})) {
+ # default: neither -b nor -l
+ &OutputData($Options{'o'},$DBQuery,$MaxLength);
+} elsif ($Options{'b'}) {
+ # -b is set (then -l can't be!)
+ # we have to read in the query results ourselves, as they do not have standard layout
+ while (my ($Newsgroup,$Postings) = $DBQuery->fetchrow_array) {
+ # we just assign "top x" or "bottom x" instead of a month for the caption
+ # FIXME
+ print &FormatOutput($Options{'o'}, ($Options{'i'} ? 'Bottom ' : 'Top ').$Options{'b'}, $Newsgroup, $Postings, $MaxLength);
+ };
+} else {
+ # -l must be set now, as all other cases have been taken care of
+ # we have to read in the query results ourselves, as they do not have standard layout
+ while (my ($Month,$Newsgroup,$Postings) = $DBQuery->fetchrow_array) {
+ # we just switch $Newsgroups and $Month for output generation
+ # FIXME
+ print &FormatOutput($Options{'o'}, $Newsgroup, $Month, $Postings, 7);
+ };
+};
+
+### close handles
+$DBHandle->disconnect;
+
+__END__
+
+################################ Documentation #################################
+
+=head1 NAME
+
+groupstats - create reports on newsgroup usage
+
+=head1 SYNOPSIS
+
+B<groupstats> [B<-Vhiscqd>] [B<-m> I<YYYY-MM>] [B<-p> I<YYYY-MM:YYYY-MM>] [B<-n> I<newsgroup(s)>] [B<-t> I<threshold>] [B<-l> I<level>] [B<-b> I<number>] [B<-o> I<output type>] [B<-g> I<database table>]
+
+=head1 REQUIREMENTS
+
+See doc/README: Perl 5.8.x itself and the following modules from CPAN:
+
+=over 2
+
+=item -
+
+Config::Auto
+
+=item -
+
+DBI
+
+=back
+
+=head1 DESCRIPTION
+
+This script create reports on newsgroup usage (number of postings per
+group per month) taken from result tables created by
+F<gatherstats.pl>.
+
+The time period to act on defaults to last month; you can assign
+another month via the B<-m> switch or a time period via the B<-p>
+switch; the latter takes preference.
+
+B<groupstats> will process all newsgroups by default; you can limit
+that to only some newsgroups by supplying a list of those groups via
+B<-n> (see below). You can include hierarchy levels in the output by
+adding the B<-s> switch (see below).
+
+Furthermore you can set a threshold via B<-t> so that only newsgroups
+with more postings per month will be included in the report. You can
+invert that by the B<-i> switch so only newsgroups with less than
+I<threshold> postings per month will be included.
+
+You can sort the output by number of postings per month instead of the
+default (alphabetical list of newsgroups) by using B<-q>; you can
+reverse the sorting order (from highest to lowest or in reversed
+alphabetical order) by using B<-d>.
+
+Furthermore, you can create a list of newsgroups that had consistently
+more (or less) than x postings per month during the whole report
+period by using B<-l> (together with B<i> as needed).
+
+Last but not least you can create a "best of" list of the top x
+newsgroups via B<-b> (or a "worst of" list by adding B<i>).
+
+By default, B<groupstats> will dump a very simple alphabetical list of
+newsgroups, one per line, followed by the number of postings in that
+month. This output format of course cannot sensibly be combined with
+time periods, so you can set the output format by using B<-o> (see
+below). Captions can be added by setting the B<-c> switch.
+
+=head2 Configuration
+
+F<groupstats.pl> will read its configuration from F<newsstats.conf>
+which should be present in the same directory via Config::Auto.
+
+See doc/INSTALL for an overview of possible configuration options.
+
+You can override configuration options via the B<-g> switch.
+
+=head1 OPTIONS
+
+=over 3
+
+=item B<-V> (version)
+
+Print out version and copyright information on B<yapfaq> and exit.
+
+=item B<-h> (help)
+
+Print this man page and exit.
+
+=item B<-m> I<YYYY-MM> (month)
+
+Set processing period to a month in YYYY-MM format. Ignored if B<-p>
+is set.
+
+=item B<-p> I<YYYY-MM:YYYY-MM> (period)
+
+Set processing period to a time period between two month, each in
+YYYY-MM format, separated by a colon. Overrides B<-m>.
+
+=item B<-n> I<newsgroup(s)> (newsgroups)
+
+Limit processing to a certain set of newsgroups. I<newsgroup(s)> can
+be a single newsgroup name (de.alt.test), a newsgroup hierarchy
+(de.alt.*) or a list of either of these, separated by colons, for
+example
+
+ de.test:de.alt.test:de.newusers.*
+
+=item B<-t> I<threshold> (threshold)
+
+Only include newsgroups with more than I<threshold> postings per
+month. Can be inverted by the B<-i> switch so that only newsgroups
+with less than I<threshold> postings will be included.
+
+This setting will be ignored if B<-l> or B<-b> is set.
+
+=item B<-l> I<level> (level)
+
+Only include newsgroups with more than I<level> postings per
+month, every month during the whole reporting period. Can be inverted
+by the B<-i> switch so that only newsgroups with less than I<level>
+postings every single month will be included. Output will be ordered
+by newsgroup name, followed by month.
+
+This setting will be ignored if B<-b> is set. Overrides B<-t> and
+can't be used together with B<-q> or B<-d>.
+
+=item B<-b> I<n> (best of)
+
+Create a list of the I<n> newsgroups with the most postings over the
+whole reporting period. Can be inverted by the B<-i> switch so that a
+list of the I<n> newsgroups with the least postings over the whole
+period is generated. Output will be ordered by sum of postings.
+
+Overrides B<-t> and B<-l> and can't be used together with B<-q> or
+B<-d>. Output format is set to I<pretty> (see below).
+
+=item B<-i> (invert)
+
+Used in conjunction with B<-t>, B<-l> or B<-b> to set a lower
+threshold or level or generate a "bottom list" instead of a top list.
+
+=item B<-s> (sum per hierarchy level)
+
+Include "virtual" groups for every hierarchy level in output, for
+example:
+
+ de.alt.ALL 10
+ de.alt.test 5
+ de.alt.admin 7
+
+See the B<gatherstats> man page for details.
+
+=item B<-o> I<output type> (output format)
+
+Set output format. Default is I<dump>, consisting of an alphabetical
+list of newsgroups, each on a new line, followed by the number of
+postings in that month. This default format can't be used with time
+periods of more than one month.
+
+I<list> format is like I<dump>, but will print the month in front of
+the newsgroup name.
+
+I<dumpgroup> format can only be use with a group list (see B<-n>) of
+exactly one newsgroup and is like I<dump>, but will output months,
+followed by the number of postings.
+
+If you don't need easily parsable output, you'll mostly use I<pretty>
+format, which will print a header for each new month and try to align
+newsgroup names and posting counts. Usage of B<-b> will force this
+format.
+
+=item B<-c> (captions)
+
+Add captions to output (reporting period, newsgroups list, threshold).
+
+=item B<-q> (quantity of postings)
+
+Sort by number of postings instead of by newsgroup names.
+
+Cannot be used with B<-l> or B<-b>.
+
+=item B<-d> (descending)
+
+Change sort order to descending.
+
+Cannot be used with B<-l> or B<-b>.
+
+=item B<-g> I<table> (postings per group table)
+
+Override I<DBTableGrps> from F<newsstats.conf>.
+
+=back
+
+=head1 INSTALLATION
+
+See doc/INSTALL.
+
+=head1 EXAMPLES
+
+Show number of postings per group for lasth month in I<dump> format:
+
+ groupstats
+
+Show that report for January of 2010 and de.alt.* plus de.test,
+including display of hierarchy levels:
+
+ groupstats -m 2010-01 -n de.alt.*:de.test -s
+
+Show that report for the year of 2010 in I<pretty> format:
+
+ groupstats -p 2010-01:2010-12 -o pretty
+
+Only show newsgroups with less than 30 postings last month, ordered
+by number of postings, descending, in I<pretty> format:
+
+ groupstats -iqdt 30 -o pretty
+
+Show top 10 for the first half-year of of 2010 in I<pretty> format:
+
+ groupstats -p 2010-01:2010-06 -b 10 -o pretty
+
+Report all groups that had less than 30 postings every singele month
+in the year of 2010 (I<pretty> format is forced)
+
+ groupstats -p 2010-01:2010-12 -il 30
+
+=head1 FILES
+
+=over 4
+
+=item F<groupstats.pl>
+
+The script itself.
+
+=item F<NewsStats.pm>
+
+Library functions for the NewsStats package.
+
+=item F<newsstats.conf>
+
+Runtime configuration file for B<yapfaq>.
+
+=back
+
+=head1 BUGS
+
+Please report any bugs or feature requests to the author or use the
+bug tracker at L<http://bugs.th-h.de/>!
+
+=head1 SEE ALSO
+
+=over 2
+
+=item -
+
+doc/README
+
+=item -
+
+doc/INSTALL
+
+=item -
+
+gatherstats -h
+
+=back
+
+This script is part of the B<NewsStats> package.
+
+=head1 AUTHOR
+
+Thomas Hochstein <thh@inter.net>
+
+=head1 COPYRIGHT AND LICENSE
+
+Copyright (c) 2010 Thomas Hochstein <thh@inter.net>
+
+This program is free software; you may redistribute it and/or modify it
+under the same terms as Perl itself.
+
+=cut
-#! /usr/bin/perl -W\r
-#\r
-# install.pl\r
-#\r
-# This script will create database tables as necessary.\r
-# \r
-# It is part of the NewsStats package.\r
-#\r
-# Copyright (c) 2010 Thomas Hochstein <thh@inter.net>\r
-#\r
-# It can be redistributed and/or modified under the same terms under \r
-# which Perl itself is published.\r
-\r
-BEGIN {\r
- our $VERSION = "0.01";\r
- use File::Basename;\r
- # we're in .../install, so our module is in ..\r
- push(@INC, dirname($0).'/..');\r
-}\r
-use strict;\r
-\r
-use NewsStats qw(:DEFAULT);\r
-\r
-use Cwd;\r
-\r
-use DBI;\r
-\r
-################################# Main program #################################\r
-\r
-### read commandline options\r
-my %Options = &ReadOptions('');\r
-\r
-### change working directory to .. (as we're in .../install)\r
-chdir dirname($0).'/..';\r
-\r
-### read configuration\r
-print("Reading configuration.\n");\r
-my %Conf = %{ReadConfig('newsstats.conf')};\r
-\r
-##### --------------------------------------------------------------------------\r
-##### Database table definitions\r
-##### --------------------------------------------------------------------------\r
-\r
-my %DBCreate = ('DBTableRaw' => <<RAW, 'DBTableGrps' => <<GRPS);\r
--- \r
--- Table structure for table DBTableRaw\r
--- \r
-\r
-CREATE TABLE IF NOT EXISTS `$Conf{'DBTableRaw'}` (\r
- `id` bigint(20) unsigned NOT NULL auto_increment,\r
- `day` date NOT NULL,\r
- `mid` varchar(250) character set ascii NOT NULL,\r
- `date` datetime NOT NULL,\r
- `timestamp` bigint(20) NOT NULL,\r
- `token` varchar(80) character set ascii NOT NULL,\r
- `size` bigint(20) NOT NULL,\r
- `peer` varchar(250) NOT NULL,\r
- `path` varchar(1000) NOT NULL,\r
- `newsgroups` varchar(1000) NOT NULL,\r
- `headers` longtext NOT NULL,\r
- `disregard` tinyint(1) default '0',\r
- PRIMARY KEY (`id`),\r
- KEY `day` (`day`),\r
- KEY `mid` (`mid`),\r
- KEY `peer` (`peer`)\r
-) ENGINE=MyISAM DEFAULT CHARSET=utf8 COMMENT='Raw data';\r
-RAW\r
--- \r
--- Table structure for table DBTableGrps\r
--- \r
-\r
-CREATE TABLE IF NOT EXISTS `$Conf{'DBTableGrps'}` (\r
- `id` bigint(20) unsigned NOT NULL auto_increment,\r
- `month` varchar(7) character set ascii NOT NULL,\r
- `newsgroup` varchar(100) NOT NULL,\r
- `postings` int(11) NOT NULL,\r
- `revision` timestamp NOT NULL default CURRENT_TIMESTAMP on update CURRENT_TIMESTAMP,\r
- PRIMARY KEY (`id`),\r
- UNIQUE KEY `month_newsgroup` (`month`,`newsgroup`),\r
- KEY `newsgroup` (`newsgroup`),\r
- KEY `postings` (`postings`)\r
-) ENGINE=MyISAM DEFAULT CHARSET=utf8 COMMENT='Postings per newsgroup';\r
-GRPS\r
-\r
-##### --------------------------- End of definitions ---------------------------\r
-\r
-### create database tables\r
-print "-----\nStarting database table generation.\n";\r
-# DB init\r
-my $DBHandle = InitDB(\%Conf,1);\r
-\r
-# read tables\r
-my %TablesInDB = %{$DBHandle->table_info('%', '%', '%', 'TABLE')->fetchall_hashref('TABLE_NAME')};\r
-\r
-# check for tables and create them, if they don't exist yet\r
-foreach my $Table (keys %DBCreate) {\r
- if (defined($TablesInDB{$Conf{$Table}})) {\r
- printf("Database table %s.%s already exists, skipping ....\n",$Conf{'DBDatabase'},$Conf{$Table});\r
- next;\r
- };\r
- my $DBQuery = $DBHandle->prepare($DBCreate{$Table});\r
- $DBQuery->execute() or die sprintf("$MySelf: E: Can't create table %s in database %s: %s%\n",$Table,$Conf{'DBDatabase'},$DBI::errstr);\r
- printf("Database table %s.%s created succesfully.\n",$Conf{'DBDatabase'},$Conf{$Table});\r
-};\r
-\r
-# close handle\r
-$DBHandle->disconnect;\r
-print "Database table generation done.\n";\r
-\r
-### output information on other necessary steps\r
-my $Path = cwd();\r
-print <<TODO;\r
------\r
-Things left to do:\r
-\r
-1) Setup an INN feed to feedlog.pl\r
-\r
- a) Edit your 'newsfeeds' file and insert something like\r
-\r
- ## gather statistics for NewsStats\r
- newsstats!\\r
- :!*,de.*\\r
- :Tc,WmtfbsPNH,Ac:$Path/feedlog.pl\r
-\r
- Please\r
-\r
- * check that you got the path to feedlog.pl right\r
- * check that feedlog.pl can be executed by the news user\r
- * adapt the pattern (here: 'de.*') to your needs\r
-\r
- b) Check your 'newsfeeds' syntax:\r
-\r
- # ctlinnd checkfile\r
-\r
- and reload 'newsfeeds':\r
-\r
- # ctlinnd reload newsfeeds 'Adding newsstats! feed'\r
-\r
- c) Watch your 'news.notice' and 'errlog' files:\r
-\r
- # tail -f /var/log/news/news.notice\r
- ...\r
- # tail -f /var/log/news/errlog\r
-\r
-2) Watch your $Conf{'DBTableRaw'} table fill.\r
-\r
-3) Read the documentation. ;)\r
-\r
-Enjoy!\r
-\r
--thh <thh\@inter.net>\r
-TODO\r
-\r
-__END__\r
-\r
-################################ Documentation #################################\r
-\r
-=head1 NAME\r
-\r
-install - installation script\r
-\r
-=head1 SYNOPSIS\r
-\r
-B<install> [B<-Vh>]\r
-\r
-=head1 REQUIREMENTS\r
-\r
-See doc/README: Perl 5.8.x itself and the following modules from CPAN:\r
-\r
-=over 2\r
-\r
-=item -\r
-\r
-Config::Auto\r
-\r
-=item -\r
-\r
-DBI\r
-\r
-=back\r
-\r
-=head1 DESCRIPTION\r
-\r
-This script will create database tables as necessary and configured.\r
-\r
-=head2 Configuration\r
-\r
-F<install.pl> will read its configuration from F<newsstats.conf> via\r
-Config::Auto.\r
-\r
-See doc/INSTALL for an overview of possible configuration options.\r
-\r
-=head1 OPTIONS\r
-\r
-=over 3\r
-\r
-=item B<-V> (version)\r
-\r
-Print out version and copyright information on B<yapfaq> and exit.\r
-\r
-=item B<-h> (help)\r
-\r
-Print this man page and exit.\r
-\r
-=back\r
-\r
-=head1 FILES\r
-\r
-=over 4\r
-\r
-=item F<install.pl>\r
-\r
-The script itself.\r
-\r
-=item F<NewsStats.pm>\r
-\r
-Library functions for the NewsStats package.\r
-\r
-=item F<newsstats.conf>\r
-\r
-Runtime configuration file for B<yapfaq>.\r
-\r
-=back\r
-\r
-=head1 BUGS\r
-\r
-Please report any bugs or feature requests to the author or use the\r
-bug tracker at L<http://bugs.th-h.de/>!\r
-\r
-=head1 SEE ALSO\r
-\r
-=over 2\r
-\r
-=item -\r
-\r
-doc/README\r
-\r
-=item -\r
-\r
-doc/INSTALL\r
-\r
-=back\r
-\r
-This script is part of the B<NewsStats> package.\r
-\r
-=head1 AUTHOR\r
-\r
-Thomas Hochstein <thh@inter.net>\r
-\r
-=head1 COPYRIGHT AND LICENSE\r
-\r
-Copyright (c) 2010 Thomas Hochstein <thh@inter.net>\r
-\r
-This program is free software; you may redistribute it and/or modify it\r
-under the same terms as Perl itself.\r
-\r
-=cut\r
+#! /usr/bin/perl -W
+#
+# install.pl
+#
+# This script will create database tables as necessary.
+#
+# It is part of the NewsStats package.
+#
+# Copyright (c) 2010 Thomas Hochstein <thh@inter.net>
+#
+# It can be redistributed and/or modified under the same terms under
+# which Perl itself is published.
+
+BEGIN {
+ our $VERSION = "0.01";
+ use File::Basename;
+ # we're in .../install, so our module is in ..
+ push(@INC, dirname($0).'/..');
+}
+use strict;
+
+use NewsStats qw(:DEFAULT);
+
+use Cwd;
+
+use DBI;
+
+################################# Main program #################################
+
+### read commandline options
+my %Options = &ReadOptions('');
+
+### change working directory to .. (as we're in .../install)
+chdir dirname($0).'/..';
+
+### read configuration
+print("Reading configuration.\n");
+my %Conf = %{ReadConfig('newsstats.conf')};
+
+##### --------------------------------------------------------------------------
+##### Database table definitions
+##### --------------------------------------------------------------------------
+
+my %DBCreate = ('DBTableRaw' => <<RAW, 'DBTableGrps' => <<GRPS);
+--
+-- Table structure for table DBTableRaw
+--
+
+CREATE TABLE IF NOT EXISTS `$Conf{'DBTableRaw'}` (
+ `id` bigint(20) unsigned NOT NULL auto_increment,
+ `day` date NOT NULL,
+ `mid` varchar(250) character set ascii NOT NULL,
+ `date` datetime NOT NULL,
+ `timestamp` bigint(20) NOT NULL,
+ `token` varchar(80) character set ascii NOT NULL,
+ `size` bigint(20) NOT NULL,
+ `peer` varchar(250) NOT NULL,
+ `path` varchar(1000) NOT NULL,
+ `newsgroups` varchar(1000) NOT NULL,
+ `headers` longtext NOT NULL,
+ `disregard` tinyint(1) default '0',
+ PRIMARY KEY (`id`),
+ KEY `day` (`day`),
+ KEY `mid` (`mid`),
+ KEY `peer` (`peer`)
+) ENGINE=MyISAM DEFAULT CHARSET=utf8 COMMENT='Raw data';
+RAW
+--
+-- Table structure for table DBTableGrps
+--
+
+CREATE TABLE IF NOT EXISTS `$Conf{'DBTableGrps'}` (
+ `id` bigint(20) unsigned NOT NULL auto_increment,
+ `month` varchar(7) character set ascii NOT NULL,
+ `newsgroup` varchar(100) NOT NULL,
+ `postings` int(11) NOT NULL,
+ `revision` timestamp NOT NULL default CURRENT_TIMESTAMP on update CURRENT_TIMESTAMP,
+ PRIMARY KEY (`id`),
+ UNIQUE KEY `month_newsgroup` (`month`,`newsgroup`),
+ KEY `newsgroup` (`newsgroup`),
+ KEY `postings` (`postings`)
+) ENGINE=MyISAM DEFAULT CHARSET=utf8 COMMENT='Postings per newsgroup';
+GRPS
+
+##### --------------------------- End of definitions ---------------------------
+
+### create database tables
+print "-----\nStarting database table generation.\n";
+# DB init
+my $DBHandle = InitDB(\%Conf,1);
+
+# read tables
+my %TablesInDB = %{$DBHandle->table_info('%', '%', '%', 'TABLE')->fetchall_hashref('TABLE_NAME')};
+
+# check for tables and create them, if they don't exist yet
+foreach my $Table (keys %DBCreate) {
+ if (defined($TablesInDB{$Conf{$Table}})) {
+ printf("Database table %s.%s already exists, skipping ....\n",$Conf{'DBDatabase'},$Conf{$Table});
+ next;
+ };
+ my $DBQuery = $DBHandle->prepare($DBCreate{$Table});
+ $DBQuery->execute() or die sprintf("$MySelf: E: Can't create table %s in database %s: %s%\n",$Table,$Conf{'DBDatabase'},$DBI::errstr);
+ printf("Database table %s.%s created succesfully.\n",$Conf{'DBDatabase'},$Conf{$Table});
+};
+
+# close handle
+$DBHandle->disconnect;
+print "Database table generation done.\n";
+
+### output information on other necessary steps
+my $Path = cwd();
+print <<TODO;
+-----
+Things left to do:
+
+1) Setup an INN feed to feedlog.pl
+
+ a) Edit your 'newsfeeds' file and insert something like
+
+ ## gather statistics for NewsStats
+ newsstats!\
+ :!*,de.*\
+ :Tc,WmtfbsPNH,Ac:$Path/feedlog.pl
+
+ Please
+
+ * check that you got the path to feedlog.pl right
+ * check that feedlog.pl can be executed by the news user
+ * adapt the pattern (here: 'de.*') to your needs
+
+ b) Check your 'newsfeeds' syntax:
+
+ # ctlinnd checkfile
+
+ and reload 'newsfeeds':
+
+ # ctlinnd reload newsfeeds 'Adding newsstats! feed'
+
+ c) Watch your 'news.notice' and 'errlog' files:
+
+ # tail -f /var/log/news/news.notice
+ ...
+ # tail -f /var/log/news/errlog
+
+2) Watch your $Conf{'DBTableRaw'} table fill.
+
+3) Read the documentation. ;)
+
+Enjoy!
+
+-thh <thh\@inter.net>
+TODO
+
+__END__
+
+################################ Documentation #################################
+
+=head1 NAME
+
+install - installation script
+
+=head1 SYNOPSIS
+
+B<install> [B<-Vh>]
+
+=head1 REQUIREMENTS
+
+See doc/README: Perl 5.8.x itself and the following modules from CPAN:
+
+=over 2
+
+=item -
+
+Config::Auto
+
+=item -
+
+DBI
+
+=back
+
+=head1 DESCRIPTION
+
+This script will create database tables as necessary and configured.
+
+=head2 Configuration
+
+F<install.pl> will read its configuration from F<newsstats.conf> via
+Config::Auto.
+
+See doc/INSTALL for an overview of possible configuration options.
+
+=head1 OPTIONS
+
+=over 3
+
+=item B<-V> (version)
+
+Print out version and copyright information on B<yapfaq> and exit.
+
+=item B<-h> (help)
+
+Print this man page and exit.
+
+=back
+
+=head1 FILES
+
+=over 4
+
+=item F<install.pl>
+
+The script itself.
+
+=item F<NewsStats.pm>
+
+Library functions for the NewsStats package.
+
+=item F<newsstats.conf>
+
+Runtime configuration file for B<yapfaq>.
+
+=back
+
+=head1 BUGS
+
+Please report any bugs or feature requests to the author or use the
+bug tracker at L<http://bugs.th-h.de/>!
+
+=head1 SEE ALSO
+
+=over 2
+
+=item -
+
+doc/README
+
+=item -
+
+doc/INSTALL
+
+=back
+
+This script is part of the B<NewsStats> package.
+
+=head1 AUTHOR
+
+Thomas Hochstein <thh@inter.net>
+
+=head1 COPYRIGHT AND LICENSE
+
+Copyright (c) 2010 Thomas Hochstein <thh@inter.net>
+
+This program is free software; you may redistribute it and/or modify it
+under the same terms as Perl itself.
+
+=cut
projects / usenet / newsstats.git / commitdiff
