3 # Library functions for the NewsStats package.
5 # Copyright (c) 2010 Thomas Hochstein <thh@inter.net>
7 # This module can be redistributed and/or modified under the same terms under
8 # which Perl itself is published.
14 our ($VERSION, @ISA, @EXPORT, @EXPORT_OK, %EXPORT_TAGS);
41 %EXPORT_TAGS = ( TimePeriods => [qw(GetTimePeriod LastMonth CheckMonth SplitPeriod ListMonth)],
42 Output => [qw(OutputData FormatOutput)],
43 SQLHelper => [qw(SQLHierarchies SQLGroupList GetMaxLenght)]);
45 our $PackageVersion = '0.01';
54 #####-------------------------------- Vars --------------------------------#####
56 our $MySelf = fileparse($0, '.pl');
57 our $MyVersion = "$MySelf $::VERSION (NewsStats.pm $VERSION)";
59 #####------------------------------- Basics -------------------------------#####
61 ################################################################################
63 ################################################################################
64 ### read commandline options and act on standard options -h and -V
65 ### IN : $Params: list of legal commandline paramaters (without -h and -V)
66 ### OUT: a hash containing the commandline options
67 $Getopt::Std::STANDARD_HELP_VERSION = 1;
72 getopts('Vh'.$Params, \%Options);
75 &ShowVersion if ($Options{'V'});
77 # -h: feed myself to perldoc
78 &ShowPOD if ($Options{'h'});
82 ################################################################################
84 ################################################################################
86 ################################################################################
87 ### display version and exit
88 print "NewsStats v$PackageVersion\n$MyVersion\nCopyright (c) 2010 Thomas Hochstein <thh\@inter.net>\n";
89 print "This program is free software; you may redistribute it and/or modify it under the same terms as Perl itself.\n";
92 ################################################################################
94 ################################################################################
96 ################################################################################
97 ### feed myself to perldoc and exit
101 ################################################################################
103 ################################################################################
105 ################################################################################
106 ### read config via Config::Auto
107 ### IN : $ConfFile: config filename
108 ### OUT: reference to a hash containing the configuration
110 return Config::Auto::parse($ConfFile, format => 'equal');
112 ################################################################################
114 ################################################################################
116 ################################################################################
117 ### override configuration values
118 ### IN : $ConfigR : reference to configuration hash
119 ### $OverrideR: reference to a hash containing overrides
120 my ($ConfigR,$OverrideR) = @_;
121 my %Override = %$OverrideR;
123 warn "$MySelf W: Empty configuration hash passed to OverrideConfig().\n" if ( keys %$ConfigR < 1);
124 # return if no overrides
125 return if (keys %Override < 1 or keys %$ConfigR < 1);
126 foreach my $Key (keys %Override) {
127 $$ConfigR{$Key} = $Override{$Key};
130 ################################################################################
132 ################################################################################
134 ################################################################################
135 ### initialise database connection
136 ### IN : $ConfigR: reference to configuration hash
137 ### $Die : if TRUE, die if connection fails
139 my ($ConfigR,$Die) = @_;
140 my %Conf = %$ConfigR;
141 my $DBHandle = DBI->connect(sprintf('DBI:%s:database=%s;host=%s',$Conf{'DBDriver'},$Conf{'DBDatabase'},$Conf{'DBHost'}), $Conf{'DBUser'}, $Conf{'DBPw'}, { PrintError => 0 });
143 die ("$MySelf: E: $DBI::errstr\n") if (defined($Die) and $Die);
144 warn("$MySelf: W: $DBI::errstr\n");
148 ################################################################################
150 #####------------------------------ GetStats ------------------------------#####
152 ################################################################################
154 ################################################################################
155 ### explode a (scalar) list of newsgroup names to a list of newsgroup and
156 ### hierarchy names where every newsgroup and hierarchy appears only once:
157 ### de.alt.test,de.alt.admin -> de.ALL, de.alt.ALL, de.alt.test, de.alt.admin
158 ### IN : $Newsgroups : a list of newsgroups (content of Newsgroups: header)
159 ### $TLH : top level hierarchy (all other newsgroups are ignored)
160 ### $ValidGroupsR: reference to a hash containing all valid newsgroups
162 ### OUT: %Newsgroups : hash containing all newsgroup and hierarchy names as keys
163 my ($Newsgroups,$TLH,$ValidGroupsR) = @_;
164 my %ValidGroups = %{$ValidGroupsR} if $ValidGroupsR;
167 # remove whitespace from contents of Newsgroups:
168 $Newsgroups =~ s/\s//;
169 # call &HierarchyCount for each newsgroup in $Newsgroups:
170 for (split /,/, $Newsgroups) {
171 # don't count newsgroup/hierarchy in wrong TLH
172 next if($TLH and !/^$TLH/);
173 # don't count invalid newsgroups
174 if(%ValidGroups and !defined($ValidGroups{$_})) {
175 warn (sprintf("DROPPED: %s\n",$_));
178 # add original newsgroup to %Newsgroups
180 # add all hierarchy elements to %Newsgroups, amended by '.ALL',
181 # i.e. de.alt.ALL and de.ALL
182 foreach (ParseHierarchies($_)) {
183 $Newsgroups{$_.'.ALL'} = 1;
189 ################################################################################
190 sub ParseHierarchies {
191 ################################################################################
192 ### return a list of all hierarchy levels a newsgroup belongs to
193 ### (for de.alt.test.moderated that would be de/de.alt/de.alt.test)
194 ### IN : $Newsgroup : a newsgroup name
195 ### OUT: @Hierarchies: array containing all hierarchies the newsgroup belongs to
196 my ($Newsgroup) = @_;
198 # strip trailing dots
199 $Newsgroup =~ s/(.+)\.+$/$1/;
200 # butcher newsgroup name by "." and add each hierarchy to @Hierarchies
201 # i.e. de.alt.test: "de.alt" and "de"
202 while ($Newsgroup =~ /\./) {
203 $Newsgroup =~ s/^((?:\.?[^.]+)*)\.[^.]+$/$1/;
204 push @Hierarchies, $Newsgroup;
209 ################################################################################
211 ################################################################################
212 ### read a list of valid newsgroups from file (each group on one line,
213 ### ignoring everything after the first whitespace and so accepting files
214 ### in checkgroups format as well as (parts of) an INN active file)
215 ### IN : $Filename : file to read
216 ### OUT: \%ValidGroups: hash containing all valid newsgroups
219 open (my $LIST,"<$Filename") or die "$MySelf: E: Cannot read $Filename: $!\n";
223 $ValidGroups{$_} = '1';
226 return \%ValidGroups;
229 ################################################################################
231 #####----------------------------- TimePeriods ----------------------------#####
233 ################################################################################
235 ################################################################################
236 ### get a time period to act on, in order of preference: by default the
237 ### last month; or a month submitted by -m YYYY-MM; or a time period submitted
238 ### by -p YYYY-MM:YYYY-MM
239 ### IN : $Month,$Period: contents of -m and -p
240 ### OUT: $StartMonth, $EndMonth (identical if period is just one month)
241 my ($Month,$Period) = @_;
242 # exit if -m is set and not like YYYY-MM
243 die "$MySelf: E: Wrong date format - use '$MySelf -m YYYY-MM'!\n" if not &CheckMonth($Month);
244 # warn if -m and -p is set
245 warn "$MySelf: W: Time period assigned by '-p' takes precendece over month assigned by '-m'.\n" if ($Month && $Period);
246 # default: set -m to last month
247 $Month = &LastMonth if (!defined($Month) and !defined($Period));
248 # set $StartMonth, $EndMonth
249 my ($StartMonth, $EndMonth);
252 ($StartMonth, $EndMonth) = &SplitPeriod($Period);
253 die "$MySelf: E: Wrong format for time period - use '$MySelf -p YYYY-MM:YYYY-MM'!\n" if !defined($StartMonth);
255 # set $StartMonth = $EndMonth = $Month if -p is not set
256 $StartMonth = $EndMonth = $Month;
258 return ($StartMonth, $EndMonth);
261 ################################################################################
263 ################################################################################
264 ### get last month from todays date in YYYY-MM format
265 ### OUT: last month as YYYY-MM
267 my (undef,undef,undef,undef,$Month,$Year,undef,undef,undef) = localtime(time);
268 # $Month is already defined from 0 to 11, so no need to decrease it by 1
275 return sprintf('%4d-%02d',$Year,$Month);
278 ################################################################################
280 ################################################################################
281 ### check if input is a valid month in YYYY-MM form
282 ### IN : $Month: month
283 ### OUT: TRUE / FALSE
285 return 0 if (defined($Month) and $Month !~ /^\d{4}-\d{2}$/);
289 ################################################################################
291 ################################################################################
292 ### split a time period denoted by YYYY-MM:YYYY-MM into start and end month
293 ### IN : $Period: time period
294 ### OUT: $StartMonth, Â$EndMonth
296 return (undef,undef) if $Period !~ /^\d{4}-\d{2}:\d{4}-\d{2}$/;
297 my ($StartMonth, $EndMonth) = split /:/, $Period;
298 # switch parameters as necessary
299 if ($EndMonth gt $StartMonth) {
300 return ($StartMonth, $EndMonth);
302 return ($EndMonth, $StartMonth);
306 ################################################################################
308 ################################################################################
309 ### return a list of months (YYYY-MM) between start and end month
310 ### IN : $StartMonth, $EndMonth
311 ### OUT: @Months: array containing all months from $StartMonth to $EndMonth
312 my ($StartMonth, $EndMonth) = @_;
313 return (undef,undef) if ($StartMonth !~ /^\d{4}-\d{2}$/ or $EndMonth !~ /^\d{4}-\d{2}$/);
314 # return if $StartMonth = $EndMonth
315 return ($StartMonth) if ($StartMonth eq $EndMonth);
316 # set $Year, $Month from $StartMonth
317 my ($Year, $Month) = split /-/, $StartMonth;
320 until ("$Year-$Month" gt $EndMonth) {
321 push @Months, "$Year-$Month";
322 $Month = "$Month"; # force string context
332 #####---------------------------- OutputFormats ---------------------------#####
334 ################################################################################
336 ################################################################################
337 ### read database query results from DBHandle and print results with formatting
338 ### IN : $Format : format specifier
339 ### $FileName: file name template (-f): filename-YYYY-MM
340 ### $DBQuery : database query handle with executed query,
341 ### containing $Month, $Key, $Value
342 ### $PadGroup: padding length for newsgroups field (optional) for 'pretty'
343 my ($Format, $FileName, $DBQuery, $PadGroup) = @_;
346 while (my ($Month, $Key, $Value) = $DBQuery->fetchrow_array) {
347 # set output file handle
349 $Handle = *STDOUT{IO}; # set $Handle to a reference to STDOUT
350 } elsif (!defined($LastIteration) or $LastIteration ne $Month) {
351 close $OUT if ($LastIteration);
352 open ($OUT,sprintf('>%s-%s',$FileName,$Month)) or die sprintf("$MySelf: E: Cannot open output file '%s-%s': $!\n",$FileName,$Month);
355 print $Handle &FormatOutput($Format, $Month, $Key, $Value, $PadGroup);
356 $LastIteration = $Month;
358 close $OUT if ($FileName);
361 ################################################################################
363 ################################################################################
364 ### format information for output according to format specifier
365 ### IN : $Format : format specifier
366 ### $Month : month (as YYYY-MM)
367 ### $Key : newsgroup, client, ...
368 ### $Value : number of postings with that attribute
369 ### $PadGroup: padding length for key field (optional) for 'pretty'
370 ### OUT: $Output: formatted output
371 my ($Format, $Month, $Key, $Value, $PadGroup) = @_;
373 # define output types
375 @LegalOutput{('dump','dumpgroup','list','pretty')} = ();
376 # bail out if format is unknown
377 die "$MySelf: E: Unknown output type '$Format'!\n" if !exists($LegalOutput{$Format});
380 # keep last month in mind
381 our ($LastIteration);
382 if ($Format eq 'dump') {
383 # output as dump (ng nnnnn)
384 $Output = sprintf ("%s %u\n",$Key,$Value);
385 } elsif ($Format eq 'dumpgroup') {
386 # output as dump (YYYY-NN: nnnnn)
387 $Output = sprintf ("%s: %5u\n",$Month,$Value);
388 } elsif ($Format eq 'list') {
389 # output as list (YYYY-NN: ng nnnnn)
390 $Output = sprintf ("%s: %s %u\n",$Month,$Key,$Value);
391 } elsif ($Format eq 'pretty') {
393 $Output = sprintf ("----- %s:\n",$Month) if (!defined($LastIteration) or $Month ne $LastIteration);
394 $LastIteration = $Month;
395 $Output .= sprintf ($PadGroup ? sprintf("%%-%us %%5u\n",$PadGroup) : "%s %u\n",$Key,$Value);
400 #####------------------------- QueryModifications -------------------------#####
402 ################################################################################
404 ################################################################################
405 ### add exclusion of hierarchy levels (de.alt.ALL) from SQL query by
406 ### amending the WHERE clause if $ShowHierarchies is false (or don't, if it is
407 ### true, accordingly)
408 ### IN : $ShowHierarchies: boolean value
410 my ($ShowHierarchies) = @_;
411 return $ShowHierarchies ? '' : "AND newsgroup NOT LIKE '%.ALL'";
414 ################################################################################
416 ################################################################################
417 ### get length of longest field in future query result
418 ### IN : $DBHandle : database handel
419 ### $Table : table to query
420 ### $Field : field to check
421 ### $WhereClause: WHERE clause
422 ### @BindVars : bind variables for WHERE clause
423 ### OUT: $Length: length of longest instnace of $Field
424 my ($DBHandle,$Table,$Field,$WhereClause,@BindVars) = @_;
425 my $DBQuery = $DBHandle->prepare(sprintf("SELECT MAX(LENGTH(%s)) FROM %s WHERE %s",$Field,$Table,$WhereClause));
426 $DBQuery->execute(@BindVars) or warn sprintf("$MySelf: W: Can't get field length for %s from table %s: $DBI::errstr\n",$Field,$Table);
427 my ($Length) = $DBQuery->fetchrow_array;
431 ################################################################################
433 ################################################################################
434 ### explode list of newsgroups separated by : (with wildcards) to a SQL WHERE
436 ### IN : $Newsgroups: list of newsgroups (group.one.*:group.two:group.three.*)
437 ### OUT: SQL code, list of newsgroups
438 my ($Newsgroups) = @_;
439 $Newsgroups =~ s/\*/%/g;
440 return ('newsgroup LIKE ?', $Newsgroups) if $Newsgroups !~ /:/;
442 my @GroupList = split /:/, $Newsgroups;
443 foreach (@GroupList) {
444 $SQL .= ' OR ' if $SQL gt '(';
445 $SQL .= 'newsgroup LIKE ?';
448 return ($SQL,@GroupList);
451 #####------------------------------- done ---------------------------------#####