################################################################################
sub ReadOptions {
################################################################################
-### read commandline options and act on standard options
-### IN : $Params: containing list of commandline paramaters (without -h and -V)
+### read commandline options and act on standard options -h and -V
+### IN : $Params: list of legal commandline paramaters (without -h and -V)
### OUT: a hash containing the commandline options
$Getopt::Std::STANDARD_HELP_VERSION = 1;
### $OverrideR: reference to a hash containing overrides
my ($ConfigR,$OverrideR) = @_;
my %Override = %$OverrideR;
+ # Config hash empty?
warn "$MySelf W: Empty configuration hash passed to OverrideConfig().\n" if ( keys %$ConfigR < 1);
+ # return if no overrides
return if (keys %Override < 1 or keys %$ConfigR < 1);
foreach my $Key (keys %Override) {
$$ConfigR{$Key} = $Override{$Key};
################################################################################
### initialise database connection
### IN : $ConfigR: reference to configuration hash
-### $Die : if TRUE, die if connection failed
+### $Die : if TRUE, die if connection fails
### OUT: DBHandle
my ($ConfigR,$Die) = @_;
my %Conf = %$ConfigR;
################################################################################
sub ListNewsgroups {
################################################################################
-### count each newsgroup and each hierarchy level, but only once
-### IN : $Newsgroups: a list of newsgroups (content of Newsgroups:)
+### explode a (scalar) list of newsgroup names to a list of newsgroup and
+### hierarchy names where every newsgroup and hierarchy appears only once:
+### de.alt.test,de.alt.admin -> de.ALL, de.alt.ALL, de.alt.test, de.alt.admin
+### IN : $Newsgroups: a list of newsgroups (content of Newsgroups: header)
### OUT: %Newsgroups: hash containing all newsgroup and hierarchy names as keys
my ($Newsgroups) = @_;
my %Newsgroups;
################################################################################
sub ParseHierarchies {
################################################################################
-### get all hierarchies a newsgroup belongs to
+### return a list of all hierarchy levels a newsgroup belongs to
+### (for de.alt.test.moderated that would be de/de.alt/de.alt.test)
### IN : $Newsgroup : a newsgroup name
### OUT: @Hierarchies: array containing all hierarchies the newsgroup belongs to
my ($Newsgroup) = @_;
################################################################################
sub GetTimePeriod {
################################################################################
-### get time period using -m / -p
+### get a time period to act on, in order of preference: by default the
+### last month; or a month submitted by -m YYYY-MM; or a time period submitted
+### by -p YYYY-MM:YYYY-MM
### IN : $Month,$Period: contents of -m and -p
-### OUT: $StartMonth, $EndMonth
+### OUT: $StartMonth, $EndMonth (identical if period is just one month)
my ($Month,$Period) = @_;
# exit if -m is set and not like YYYY-MM
die "$MySelf: E: Wrong date format - use '$MySelf -m YYYY-MM'!\n" if not &CheckMonth($Month);
################################################################################
sub LastMonth {
################################################################################
-### get last month from today in YYYY-MM format
+### get last month from todays date in YYYY-MM format
### OUT: last month as YYYY-MM
# get today's date
my (undef,undef,undef,undef,$Month,$Year,undef,undef,undef) = localtime(time);
################################################################################
sub CheckMonth {
################################################################################
-### check for valid month
+### check if input is a valid month in YYYY-MM form
### IN : $Month: month
### OUT: TRUE / FALSE
my ($Month) = @_;
################################################################################
sub SplitPeriod {
################################################################################
-### split a time period YYYY-MM:YYYY-MM into start and end month
+### split a time period denoted by YYYY-MM:YYYY-MM into start and end month
### IN : $Period: time period
### OUT: $StartMonth, Â$EndMonth
my ($Period) = @_;
################################################################################
sub ListMonth {
################################################################################
-### return a list of month (YYYY-MM) between start and end month
+### return a list of months (YYYY-MM) between start and end month
### IN : $StartMonth, $EndMonth
### OUT: @Months: array containing all months from $StartMonth to $EndMonth
my ($StartMonth, $EndMonth) = @_;
################################################################################
sub OutputData {
################################################################################
-### output information with formatting from DBHandle
+### read database query results from DBHandle and print results with formatting
### IN : $Format : format specifier
### $DBQuery: database query handle with executed query,
### containing $Month, $Key, $Value
### $PadGroup: padding length for newsgroups field (optional) for 'pretty'
-### OUT: $Output: formatted output
my ($Format, $DBQuery,$PadGroup) = @_;
while (my ($Month, $Key, $Value) = $DBQuery->fetchrow_array) {
print &FormatOutput($Format, $Month, $Key, $Value, $PadGroup);
################################################################################
sub FormatOutput {
################################################################################
-### format information for output
+### format information for output according to format specifier
### IN : $Format : format specifier
-### $PadGroup: padding length for newsgroups field (optional) for 'pretty'
### $Month : month (as YYYY-MM)
### $Key : newsgroup, client, ...
### $Value : number of postings with that attribute
+### $PadGroup: padding length for key field (optional) for 'pretty'
### OUT: $Output: formatted output
my ($Format, $Month, $Key, $Value, $PadGroup) = @_;
die "$MySelf: E: Unknown output type '$Format'!\n" if !exists($LegalOutput{$Format});
my ($Output);
+ # keep last month in mind
our ($LastIteration);
if ($Format eq 'dump') {
# output as dump (ng nnnnn)
################################################################################
sub SQLHierarchies {
################################################################################
-### amend WHERE clause to include hierarchies
+### add exclusion of hierarchy levels (de.alt.ALL) from SQL query by
+### amending the WHERE clause if $ShowHierarchies is false (or don't, if it is
+### true, accordingly)
### IN : $ShowHierarchies: boolean value
### OUT: SQL code
my ($ShowHierarchies) = @_;
################################################################################
sub GetMaxLenght {
################################################################################
-### get length of longest field in query
+### get length of longest field in future query result
### IN : $DBHandle : database handel
### $Table : table to query
### $Field : field to check
################################################################################
sub SQLGroupList {
################################################################################
-### create part of WHERE clause for list of newsgroups separated by :
+### explode list of newsgroups separated by : (with wildcards) to a SQL WHERE
+### clause
### IN : $Newsgroups: list of newsgroups (group.one.*:group.two:group.three.*)
### OUT: SQL code, list of newsgroups
my ($Newsgroups) = @_;
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
$ConfOverride{'TLH'} = $Options{'n'} if $Options{'n'};\r
&OverrideConfig(\%Conf,\%ConfOverride);\r
\r
-### get type of information to gather, default to 'all'\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{$Options{'t'}});\r
\r
-### get time period\r
+### get time period (-m or -p)\r
my ($StartMonth,$EndMonth) = &GetTimePeriod($Options{'m'},$Options{'p'});\r
\r
### init database\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
\r
use DBI;\r
\r
-################################# Definitions ##################################\r
-\r
-# ...\r
-\r
################################# Main program #################################\r
\r
### read commandline options\r
warn ("$MySelf: W: Output type forced to '-o pretty' due to usage of '-l'.\n");\r
};\r
\r
-### get query type, default to 'postings'\r
-#die "$MySelf: E: Unknown query type -q $Options{'q'}!\n" if ($Options{'q'} and !exists($LegalTypes{$Options{'q'}}));\r
-#die "$MySelf: E: You must submit a threshold ('-t') for query type '-q $Options{'q'}'!\n" if ($Options{'q'} and !$Options{'t'});\r
-\r
### get time period\r
my ($StartMonth,$EndMonth) = &GetTimePeriod($Options{'m'},$Options{'p'});\r
-# reset to one month for 'dump' type\r
+# reset to one month for 'dump' output type\r
if ($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
### init database\r
my $DBHandle = InitDB(\%Conf,1);\r
\r
-### get data\r
+### create report\r
# get list of newsgroups (-n)\r
my ($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) defined?\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
- # do query: get number of postings per group from groups table for given months and newsgroups\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
$OrderClause = 'postings DESC';\r
};\r
- # push LIMIT to GroupList to match number of binding vars\r
+ # push LIMIT to GroupList to match number of binding vars for DBQuery->execute\r
push @GroupList,$Options{'b'};\r
- # do query: get sum of postings per group from groups table for given months and newsgroups with LIMIT\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,$OrderClause));\r
-} else { # -l\r
+} else {\r
+ # -l must be set now, as all other cases have been taken care of\r
# set sorting order (-i)\r
if ($Options{'i'}) {\r
$OrderClause = '<';\r
} else {\r
$OrderClause = '>';\r
};\r
- # push level and $StartMonth,$EndMonth - again - to GroupList to match number of binding vars\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
- # do query: get number of postings per group from groups table for given months and \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,$EndMonth,@GroupList) or die sprintf("$MySelf: E: Can't get groups data for %s to %s from %s.%s: %s\n",$StartMonth,$EndMonth,$Conf{'DBDatabase'},$Conf{'DBTableGrps'},$DBI::errstr);\r
+$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 result\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($Options{'l'})) {\r
- &OutputData($Options{'o'},$DBQuery,$MaxLength);\r
+ # default: neither -b nor -l\r
+ &OutputData($Options{'o'},$DBQuery,$MaxLength);\r
} elsif ($Options{'b'}) {\r
- while (my ($Newsgroup,$Postings) = $DBQuery->fetchrow_array) {\r
+ # -b is set (then -l can't be!)\r
+ # we have to read in the query results ourselves, as they do not have standard layout\r
+ while (my ($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, $MaxLength);\r
};\r
-} else { # -l\r
- while (my ($Month,$Newsgroup,$Postings) = $DBQuery->fetchrow_array) {\r
+} else {\r
+ # -l must be set now, as all other cases have been taken care of\r
+ # we have to read in the query results ourselves, as they do not have standard layout\r
+ while (my ($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
### 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
\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