5 # This script will get statistical data on newgroup usage
8 # It is part of the NewsStats package.
10 # Copyright (c) 2010-2012 Thomas Hochstein <thh@inter.net>
12 # It can be redistributed and/or modified under the same terms under
13 # which Perl itself is published.
16 our $VERSION = "0.01";
18 push(@INC, dirname($0));
22 use NewsStats qw(:DEFAULT :TimePeriods :Output :SQLHelper ReadGroupList);
25 use Getopt::Long qw(GetOptions);
26 Getopt::Long::config ('bundling');
28 ################################# Main program #################################
30 ### read commandline options
31 my ($OptBoundType,$OptCaptions,$OptCheckgroupsFile,$OptComments,
32 $OptFileTemplate,$OptFormat,$OptGroupBy,$OptGroupsDB,$LowBound,$OptMonth,
33 $OptNewsgroups,$OptOrderBy,$OptReportType,$OptSums,$UppBound);
34 GetOptions ('b|boundary=s' => \$OptBoundType,
35 'c|captions!' => \$OptCaptions,
36 'checkgroups=s' => \$OptCheckgroupsFile,
37 'comments!' => \$OptComments,
38 'filetemplate=s' => \$OptFileTemplate,
39 'f|format=s' => \$OptFormat,
40 'g|group-by=s' => \$OptGroupBy,
41 'groupsdb=s' => \$OptGroupsDB,
42 'l|lower=i' => \$LowBound,
43 'm|month=s' => \$OptMonth,
44 'n|newsgroups=s' => \$OptNewsgroups,
45 'o|order-by=s' => \$OptOrderBy,
46 'r|report=s' => \$OptReportType,
47 's|sums!' => \$OptSums,
48 'u|upper=i' => \$UppBound,
49 'h|help' => \&ShowPOD,
50 'V|version' => \&ShowVersion) or exit 1;
52 # $OptComments defaults to TRUE
53 $OptComments = 1 if (!defined($OptComments));
54 # force --nocomments when --filetemplate is used
55 $OptComments = 0 if ($OptFileTemplate);
58 if ($OptBoundType =~ /level/i) {
59 $OptBoundType = 'level';
60 } elsif ($OptBoundType =~ /av(era)?ge?/i) {
61 $OptBoundType = 'average';
62 } elsif ($OptBoundType =~ /sums?/i) {
63 $OptBoundType = 'sum';
65 $OptBoundType = 'default';
68 # parse $OptReportType
70 if ($OptReportType =~ /av(era)?ge?/i) {
71 $OptReportType = 'average';
72 } elsif ($OptReportType =~ /sums?/i) {
73 $OptReportType = 'sum';
75 $OptReportType = 'default';
78 # read list of newsgroups from --checkgroups
79 # into a hash reference
80 my $ValidGroups = &ReadGroupList($OptCheckgroupsFile) if $OptCheckgroupsFile;
82 ### read configuration
83 my %Conf = %{ReadConfig($HomePath.'/newsstats.conf')};
85 ### override configuration via commandline options
87 $ConfOverride{'DBTableGrps'} = $OptGroupsDB if $OptGroupsDB;
88 &OverrideConfig(\%Conf,\%ConfOverride);
91 my $DBHandle = InitDB(\%Conf,1);
93 ### get time period and newsgroups, prepare SQL 'WHERE' clause
95 # and set caption for output and expression for SQL 'WHERE' clause
96 my ($CaptionPeriod,$SQLWherePeriod) = &GetTimePeriod($OptMonth);
97 # bail out if --month is invalid
98 &Bleat(2,"--month option has an invalid format - ".
99 "please use 'YYYY-MM', 'YYYY-MM:YYYY-MM' or 'ALL'!") if !$CaptionPeriod;
100 # get list of newsgroups and set expression for SQL 'WHERE' clause
101 # with placeholders as well as a list of newsgroup to bind to them
102 my ($SQLWhereNewsgroups,@SQLBindNewsgroups) = &SQLGroupList($OptNewsgroups)
105 ### build SQL WHERE clause (and HAVING clause, if needed)
106 my ($SQLWhereClause,$SQLHavingClause);
107 # $OptBoundType 'level'
108 if ($OptBoundType and $OptBoundType ne 'default') {
109 $SQLWhereClause = SQLBuildClause('where',$SQLWherePeriod,
110 $SQLWhereNewsgroups,&SQLHierarchies($OptSums));
111 $SQLHavingClause = SQLBuildClause('having',&SQLSetBounds($OptBoundType,
112 $LowBound,$UppBound));
113 # $OptBoundType 'threshold' / 'default' or none
115 $SQLWhereClause = SQLBuildClause('where',$SQLWherePeriod,
116 $SQLWhereNewsgroups,&SQLHierarchies($OptSums),
117 &SQLSetBounds('default',$LowBound,$UppBound));
120 ### get sort order and build SQL 'ORDER BY' clause
121 # default to 'newsgroup' for $OptBoundType 'level' or 'average'
122 $OptGroupBy = 'newsgroup' if (!$OptGroupBy and
123 $OptBoundType and $OptBoundType ne 'default');
124 # force to 'month' for $OptReportType 'average' or 'sum'
125 $OptGroupBy = 'month' if ($OptReportType and $OptReportType ne 'default');
126 # parse $OptGroupBy to $GroupBy, create ORDER BY clause $SQLOrderClause
127 my ($GroupBy,$SQLOrderClause) = SQLSortOrder($OptGroupBy, $OptOrderBy);
128 # $GroupBy will contain 'month' or 'newsgroup' (parsed result of $OptGroupBy)
129 # set it to 'month' or 'key' for OutputData()
130 $GroupBy = ($GroupBy eq 'month') ? 'month' : 'key';
132 ### get report type and build SQL 'SELECT' query
134 my $SQLGroupClause = '';
135 my $Precision = 0; # number of digits right of decimal point for output
136 if ($OptReportType and $OptReportType ne 'default') {
137 $SQLGroupClause = 'GROUP BY newsgroup';
138 # change $SQLOrderClause: replace everything before 'postings'
139 $SQLOrderClause =~ s/BY.+postings/BY postings/;
140 if ($OptReportType eq 'average') {
141 $SQLSelect = "'All months',newsgroup,AVG(postings)";
143 # change $SQLOrderClause: replace 'postings' with 'AVG(postings)'
144 $SQLOrderClause =~ s/postings/AVG(postings)/;
145 } elsif ($OptReportType eq 'sum') {
146 $SQLSelect = "'All months',newsgroup,SUM(postings)";
147 # change $SQLOrderClause: replace 'postings' with 'SUM(postings)'
148 $SQLOrderClause =~ s/postings/SUM(postings)/;
151 $SQLSelect = 'month,newsgroup,postings';
154 ### get length of longest newsgroup name delivered by query
155 ### for formatting purposes
156 my $Field = ($GroupBy eq 'month') ? 'newsgroup' : 'month';
157 my ($MaxLength,$MaxValLength) = &GetMaxLength($DBHandle,$Conf{'DBTableGrps'},
158 $Field,'postings',$SQLWhereClause,
162 ### build and execute SQL query
164 # special query preparation for $OptBoundType 'level', 'average' or 'sums'
165 if ($OptBoundType and $OptBoundType ne 'default') {
166 # prepare and execute first query:
167 # get list of newsgroups meeting level conditions
168 $DBQuery = $DBHandle->prepare(sprintf('SELECT newsgroup FROM %s.%s %s '.
169 'GROUP BY newsgroup %s',
170 $Conf{'DBDatabase'},$Conf{'DBTableGrps'},
171 $SQLWhereClause,$SQLHavingClause));
172 $DBQuery->execute(@SQLBindNewsgroups)
173 or &Bleat(2,sprintf("Can't get groups data for %s from %s.%s: %s\n",
174 $CaptionPeriod,$Conf{'DBDatabase'},$Conf{'DBTableGrps'},
176 # add newsgroups to a comma-seperated list ready for IN(...) query
178 while (my ($Newsgroup) = $DBQuery->fetchrow_array) {
179 $GroupList .= ',' if $GroupList;
180 $GroupList .= "'$Newsgroup'";
182 # enhance $WhereClause
184 $SQLWhereClause = SQLBuildClause('where',$SQLWhereClause,
185 sprintf('newsgroup IN (%s)',$GroupList));
187 # condition cannot be satisfied;
188 # force query to fail by adding '0=1'
189 $SQLWhereClause = SQLBuildClause('where',$SQLWhereClause,'0=1');
194 $DBQuery = $DBHandle->prepare(sprintf('SELECT %s FROM %s.%s %s %s %s',
196 $Conf{'DBDatabase'},$Conf{'DBTableGrps'},
197 $SQLWhereClause,$SQLGroupClause,$
201 $DBQuery->execute(@SQLBindNewsgroups)
202 or &Bleat(2,sprintf("Can't get groups data for %s from %s.%s: %s\n",
203 $CaptionPeriod,$Conf{'DBDatabase'},$Conf{'DBTableGrps'},
207 # set default to 'pretty'
208 $OptFormat = 'pretty' if !$OptFormat;
209 # print captions if --caption is set
210 if ($OptCaptions && $OptComments) {
211 # print time period with report type
212 my $CaptionReportType= '(number of postings for each month)';
213 if ($OptReportType and $OptReportType ne 'default') {
214 $CaptionReportType= '(average number of postings for each month)'
215 if $OptReportType eq 'average';
216 $CaptionReportType= '(number of all postings for that time period)'
217 if $OptReportType eq 'sum';
219 printf("# ----- Report for %s %s\n",$CaptionPeriod,$CaptionReportType);
220 # print newsgroup list if --newsgroups is set
221 printf("# ----- Newsgroups: %s\n",join(',',split(/:/,$OptNewsgroups)))
223 # print boundaries, if set
224 my $CaptionBoundary= '(counting only month fulfilling this condition)';
225 if ($OptBoundType and $OptBoundType ne 'default') {
226 $CaptionBoundary= '(every single month)' if $OptBoundType eq 'level';
227 $CaptionBoundary= '(on average)' if $OptBoundType eq 'average';
228 $CaptionBoundary= '(all month summed up)' if $OptBoundType eq 'sum';
230 printf("# ----- Threshold: %s %s x %s %s %s\n",
231 $LowBound ? $LowBound : '',$LowBound ? '=>' : '',
232 $UppBound ? '<=' : '',$UppBound ? $UppBound : '',$CaptionBoundary)
233 if ($LowBound or $UppBound);
234 # print primary and secondary sort order
235 printf("# ----- Grouped by %s (%s), sorted %s%s\n",
236 ($GroupBy eq 'month') ? 'Months' : 'Newsgroups',
237 ($OptGroupBy and $OptGroupBy =~ /-?desc$/i) ? 'descending' : 'ascending',
238 ($OptOrderBy and $OptOrderBy =~ /posting/i) ? 'by number of postings ' : '',
239 ($OptOrderBy and $OptOrderBy =~ /-?desc$/i) ? 'descending' : 'ascending');
243 &OutputData($OptFormat,$OptComments,$GroupBy,$Precision,
244 $OptCheckgroupsFile ? $ValidGroups : '',
245 $OptFileTemplate,$DBQuery,$MaxLength,$MaxValLength);
248 $DBHandle->disconnect;
252 ################################ Documentation #################################
256 groupstats - create reports on newsgroup usage
260 B<groupstats> [B<-Vhcs> B<--comments>] [B<-m> I<YYYY-MM>[:I<YYYY-MM>] | I<all>] [B<-n> I<newsgroup(s)>] [B<--checkgroups> I<checkgroups file>] [B<-r> I<report type>] [B<-l> I<lower boundary>] [B<-u> I<upper boundary>] [B<-b> I<boundary type>] [B<-g> I<group by>] [B<-o> I<order by>] [B<-f> I<output format>] [B<--filetemplate> I<filename template>] [B<--groupsdb> I<database table>]
268 This script create reports on newsgroup usage (number of postings per
269 group per month) taken from result tables created by
272 =head2 Features and options
274 =head3 Time period and newsgroups
276 The time period to act on defaults to last month; you can assign another
277 time period or a single month (or drop all time constraints) via the
278 B<--month> option (see below).
280 B<groupstats> will process all newsgroups by default; you can limit
281 processing to only some newsgroups by supplying a list of those groups via
282 B<--newsgroups> option (see below). You can include hierarchy levels in
283 the output by adding the B<--sums> switch (see below). Optionally
284 newsgroups not present in a checkgroups file can be excluded from output,
285 sse B<--checkgroups> below.
289 You can choose between different B<--report> types: postings per month,
290 average postings per month or all postings summed up; for details, see
293 =head3 Upper and lower boundaries
295 Furthermore you can set an upper and/or lower boundary to exclude some
296 results from output via the B<--lower> and B<--upper> options,
297 respectively. By default, all newsgroups with more and/or less postings
298 per month will be excluded from the result set (i.e. not shown and not
299 considered for average and sum reports). You can change the meaning of
300 those boundaries with the B<--boundary> option. For details, please see
303 =head3 Sorting and formatting the output
305 By default, all results are grouped by month; you can group results by
306 newsgroup instead via the B<--groupy-by> option. Within those groups, the
307 list of newsgroups (or months) is sorted alphabetically (or
308 chronologically, respectively) ascending. You can change that order (and
309 sort by number of postings) with the B<--order-by> option. For details and
310 exceptions, please see below.
312 The results will be formatted as a kind of table; you can change the
313 output format to a simple list or just a list of newsgroups and number of
314 postings with the B<--format> option. Captions will be added by means of
315 the B<--caption> option; all comments (and captions) can be supressed by
316 using B<--nocomments>.
318 Last but not least you can redirect all output to a number of files, e.g.
319 one for each month, by submitting the B<--filetemplate> option, see below.
320 Captions and comments are automatically disabled in this case.
324 B<groupstats> will read its configuration from F<newsstats.conf>
325 which should be present in the same directory via Config::Auto.
327 See doc/INSTALL for an overview of possible configuration options.
329 You can override some configuration options via the B<--groupsdb> option.
335 =item B<-V>, B<--version>
337 Print out version and copyright information and exit.
339 =item B<-h>, B<--help>
341 Print this man page and exit.
343 =item B<-m>, B<--month> I<YYYY-MM[:YYYY-MM]|all>
345 Set processing period to a single month in YYYY-MM format or to a time
346 period between two month in YYYY-MM:YYYY-MM format (two month, separated
347 by a colon). By using the keyword I<all> instead, you can set no
348 processing period to process the whole database.
350 =item B<-n>, B<--newsgroups> I<newsgroup(s)>
352 Limit processing to a certain set of newsgroups. I<newsgroup(s)> can
353 be a single newsgroup name (de.alt.test), a newsgroup hierarchy
354 (de.alt.*) or a list of either of these, separated by colons, for
357 de.test:de.alt.test:de.newusers.*
359 =item B<-s>, B<--sums|--nosums> (sum per hierarchy level)
361 Include "virtual" groups for every hierarchy level in output, for
368 See the B<gatherstats> man page for details.
370 =item B<--checkgroups> I<filename>
372 Restrict output to those newgroups present in a file in checkgroups format
373 (one newgroup name per line; everything after the first whitespace on each
374 line is ignored). All other newsgroups will be removed from output.
376 =item B<-r>, B<--report> I<default|average|sums>
378 Choose the report type: I<default>, I<average> or I<sums>
380 By default, B<groupstats> will report the number of postings for each
381 newsgroup in each month. But it can also report the average number of
382 postings per group for all months or the total sum of postings per group
385 For report types I<average> and I<sums>, the B<group-by> option has no
386 meaning and will be silently ignored (see below).
388 =item B<-l>, B<--lower> I<lower boundary>
390 Set the lower boundary. See B<--boundary> below.
392 =item B<-l>, B<--upper> I<upper boundary>
394 Set the upper boundary. See B<--boundary> below.
396 =item B<-b>, B<--boundary> I<boundary type>
398 Set the boundary type to one of I<default>, I<level>, I<average> or
401 By default, all newsgroups with more postings per month than the upper
402 boundary and/or less postings per month than the lower boundary will be
403 excluded from further processing. For the default report that means each
404 month only newsgroups with a number of postings between the boundaries
405 will be displayed. For the other report types, newsgroups with a number of
406 postings exceeding the boundaries in all (!) months will not be
409 For example, lets take a list of newsgroups like this:
412 de.comp.datenbanken.misc 6
413 de.comp.datenbanken.ms-access 84
414 de.comp.datenbanken.mysql 88
416 de.comp.datenbanken.misc 8
417 de.comp.datenbanken.ms-access 126
418 de.comp.datenbanken.mysql 21
420 de.comp.datenbanken.misc 24
421 de.comp.datenbanken.ms-access 83
422 de.comp.datenbanken.mysql 36
424 With C<groupstats --month 2012-01:2012-03 --lower 25 --report sums>,
425 you'll get the following result:
428 de.comp.datenbanken.ms-access 293
429 de.comp.datenbanken.mysql 124
431 de.comp.datenbanken.misc has not been considered even though it has 38
432 postings in total, because it has less than 25 postings in every single
433 month. If you want to list all newsgroups with more than 25 postings U<in
434 total>, you'll have to set the boundary type to I<sum>, see below.
436 A boundary type of I<level> will show only those newsgroups - at all -
437 that satisfy the boundaries in each and every single month. With the above
438 list of newsgroups and
439 C<groupstats --month 2012-01:2012-03 --lower 25 --boundary level --report sums>,
440 you'll get this result:
443 de.comp.datenbanken.ms-access 293
445 de.comp.datenbanken.mysql has not been considered because it had less than
446 25 postings in 2012-02.
448 You can use that to get a list of newsgroups that have more (or less) then
449 x postings during the whole reporting period.
451 A boundary type of I<average> will show only those newsgroups - at all -that
452 satisfy the boundaries on average. With the above list of newsgroups and
453 C<groupstats --month 2012-01:2012-03 --lower 25 --boundary avg --report sums>,
454 you'll get this result:
457 de.comp.datenbanken.ms-access 293
458 de.comp.datenbanken.mysql 145
460 The average number of postings in the three groups is:
462 de.comp.datenbanken.misc 12.67
463 de.comp.datenbanken.ms-access 97.67
464 de.comp.datenbanken.mysql 48.33
466 Last but not least, a boundary type of I<sums> will show only those
467 newsgroups - at all - that satisfy the boundaries with the total sum of
468 all postings during the reporting period. With the above list of
470 C<groupstats --month 2012-01:2012-03 --lower 25 --boundary sum --report sums>,
471 you'll finally get this result:
474 de.comp.datenbanken.misc 38
475 de.comp.datenbanken.ms-access 293
476 de.comp.datenbanken.mysql 145
479 =item B<-g>, B<--group-by> I<month[-desc]|newsgroups[-desc]>
481 By default, all results are grouped by month, sorted chronologically in
482 ascending order, like this:
485 de.comp.datenbanken.ms-access 84
486 de.comp.datenbanken.mysql 88
488 de.comp.datenbanken.ms-access 126
489 de.comp.datenbanken.mysql 21
491 The results can be grouped by newsgroups instead via
492 B<--group-by> I<newsgroup>:
494 ----- de.comp.datenbanken.ms-access:
497 ----- de.comp.datenbanken.mysql:
501 By appending I<-desc> to the group-by option parameter, you can reverse
502 the sort order - e.g. B<--group-by> I<month-desc> will give:
505 de.comp.datenbanken.ms-access 126
506 de.comp.datenbanken.mysql 21
508 de.comp.datenbanken.ms-access 84
509 de.comp.datenbanken.mysql 88
511 Average and sums reports (see above) will always be grouped by months;
512 this option will therefore be ignored.
514 =item B<-o>, B<--order-by> I<default[-desc]|postings[-desc]>
516 Within each group (a single month or single newsgroup, see above), the
517 report will be sorted by newsgroup names in ascending alphabetical order
518 by default. You can change the sort order to descending or sort by number
521 =item B<-f>, B<--format> I<pretty|list|dump>
523 Select the output format, I<pretty> being the default:
526 de.comp.datenbanken.ms-access 84
527 de.comp.datenbanken.mysql 88
529 de.comp.datenbanken.ms-access 126
530 de.comp.datenbanken.mysql 21
532 I<list> format looks like this:
534 2012-01 de.comp.datenbanken.ms-access 84
535 2012-01 de.comp.datenbanken.mysql 88
536 2012-02 de.comp.datenbanken.ms-access 126
537 2012-02 de.comp.datenbanken.mysql 21
539 And I<dump> format looks like this:
542 de.comp.datenbanken.ms-access 84
543 de.comp.datenbanken.mysql 88
545 de.comp.datenbanken.ms-access 126
546 de.comp.datenbanken.mysql 21
548 You can remove the comments by using B<--nocomments>, see below.
550 =item B<-c>, B<--captions|--nocaptions>
552 Add captions to output, like this:
554 ----- Report for 2012-01 to 2012-02 (number of postings for each month)
555 ----- Newsgroups: de.comp.datenbanken.*
556 ----- Threshold: 10 => x <= 20 (on average)
557 ----- Grouped by Newsgroups (ascending), sorted by number of postings descending
561 =item B<--comments|--nocomments>
563 Add comments (group headers) to I<dump> and I<pretty> output. True by default.
565 Use I<--nocomments> to suppress anything except newsgroup names/months and
566 numbers of postings. This is enforced when using B<--filetemplate>, see below.
568 =item B<--filetemplate> I<filename template>
570 Save output to file(s) instead of dumping it to STDOUT. B<groupstats> will
571 create one file for each month (or each newsgroup, accordant to the
572 setting of B<--group-by>, see above), with filenames composed by adding
573 year and month (or newsgroup names) to the I<filename template>, for
574 example with B<--filetemplate> I<stats>:
580 B<--nocomments> is enforced, see above.
582 =item B<--groupsdb> I<database table>
584 Override I<DBTableGrps> from F<newsstats.conf>.
594 Show number of postings per group for lasth month in I<pretty> format:
598 Show that report for January of 2010 and de.alt.* plus de.test,
599 including display of hierarchy levels:
601 groupstats --month 2010-01 --newsgroups de.alt.*:de.test --sums
603 Only show newsgroups with 30 postings or less last month, ordered
604 by number of postings, descending, in I<pretty> format:
606 groupstats --upper 30 --order-by postings-desc
608 Show the total of all postings for the year of 2010 for all groups that
609 had 30 postings or less in every single month in that year, ordered by
610 number of postings in descending order:
612 groupstats -m 2010-01:2010-12 -u 30 -b level -r sums -o postings-desc
614 The same for the average number of postings in the year of 2010:
616 groupstats -m 2010-01:2010-12 -u 30 -b level -r avg -o postings-desc
618 List number of postings per group for eacht month of 2010 and redirect
619 output to one file for each month, namend stats-2010-01 and so on, in
620 machine-readable form (without formatting):
622 groupstats -m 2010-01:2010-12 -f dump --filetemplate stats
629 =item F<groupstats.pl>
633 =item F<NewsStats.pm>
635 Library functions for the NewsStats package.
637 =item F<newsstats.conf>
639 Runtime configuration file.
645 Please report any bugs or feature requests to the author or use the
646 bug tracker at L<http://bugs.th-h.de/>!
666 This script is part of the B<NewsStats> package.
670 Thomas Hochstein <thh@inter.net>
672 =head1 COPYRIGHT AND LICENSE
674 Copyright (c) 2010-2012 Thomas Hochstein <thh@inter.net>
676 This program is free software; you may redistribute it and/or modify it
677 under the same terms as Perl itself.