misc/cronjobs/cloud-kw.pl

   1 #!/usr/bin/perl
   2
   3 #
   4 # Copyright 2008 Tamil s.a.r.l.
   5 #
   6 # This file is part of Koha.
   7 #
   8 # Koha is free software; you can redistribute it and/or modify it
   9 # under the terms of the GNU General Public License as published by
  10 # the Free Software Foundation; either version 3 of the License, or
  11 # (at your option) any later version.
  12 #
  13 # Koha is distributed in the hope that it will be useful, but
  14 # WITHOUT ANY WARRANTY; without even the implied warranty of
  15 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  16 # GNU General Public License for more details.
  17 #
  18 # You should have received a copy of the GNU General Public License
  19 # along with Koha; if not, see <http://www.gnu.org/licenses>.
  20
  21 use strict;
  22 use warnings;
  23 use diagnostics;
  24 use Carp;
  25 use YAML::Syck;
  26 use Pod::Usage;
  27 use Getopt::Long;
  28 use C4::Context;
  29 use C4::Log;
  30
  31 my $verbose     = 0;
  32 my $help        = 0;
  33 my $conf        = '';
  34 GetOptions(
  35     'verbose'   => \$verbose,
  36     'help'      => \$help,
  37     'conf=s'    => \$conf,
  38 );
  39
  40 sub usage {
  41     pod2usage( -verbose => 2 );
  42     exit;
  43 }
  44
  45 usage() if $help || !$conf;
  46
  47 cronlogaction();
  48
  49 my @clouds;
  50 print "Reading configuration file: $conf\n" if $verbose;
  51 eval {
  52     @clouds = LoadFile( $conf );
  53 };
  54 croak "Unable to read configuration file: $conf\n" if $@;
  55
  56 for my $cloud ( @clouds ) {
  57     print "Create a cloud\n",
  58           "  Koha conf file:  ", $cloud->{KohaConf} ? $cloud->{KohaConf} : "default", "\n",
  59           "  Zebra Index:     ", $cloud->{ZebraIndex}, "\n",
  60           "  Koha Keyword:    ", $cloud->{KohaIndex}, "\n",
  61           "  Count:           ", $cloud->{Count}, "\n",
  62           "  Withcss:         ", $cloud->{Withcss}, "\n",
  63           "  Output:          ", $cloud->{Output}, "\n",
  64       if $verbose;
  65
  66     # Set Koha context if KohaConf is present
  67     my $set_new_context = 0;
  68     if ( $cloud->{KohaConf} ) {
  69         if ( -e $cloud->{KohaConf} ) {
  70             my $context = C4::Context->new( $cloud->{KohaConf} );
  71             $context->set_context();
  72             $set_new_context = 1;
  73         }
  74         else {
  75             carp "Koha conf file doesn't exist: ", $cloud->{KohaConf}, " ; use KOHA_CONF\n";
  76         }
  77     }
  78
  79     my $index = new ZebraIndex( $cloud->{ZebraIndex} );
  80     $index->scan( $cloud->{Count} );
  81
  82     open my $fh, ">", $cloud->{Output}
  83         or croak "Unable to create file ", $cloud->{Output};
  84
  85     my $withcss = $cloud->{Withcss} =~ /^yes/i;
  86     print $fh $index->html_cloud( $cloud->{KohaIndex}, $withcss );
  87     close $fh;
  88     $set_new_context && restore_context C4::Context;
  89 }
  90
  91
  92
  93 package ZebraIndex;
  94
  95 use strict;
  96 use warnings;
  97 use diagnostics;
  98 use Carp;
  99
 100 sub new {
 101     my $self = {};
 102     my $class = shift;
 103     $self->{ zebra_index  } = shift;
 104     $self->{ top_terms    } = undef;
 105     $self->{ levels_cloud } = 24;
 106     bless $self, $class;
 107
 108     # Test Zebra index
 109     my $zbiblio = C4::Context->Zconn( "biblioserver" );
 110     eval {
 111         my $ss = $zbiblio->scan_pqf(
 112             '@attr 1=' . $self->{ zebra_index } . ' @attr 4=1 @attr 6=3 "a"'
 113         );
 114     };
 115     croak "Invalid Zebra index: ", $self->{ zebra_index } if $@;
 116
 117     return $self;
 118 }
 119
 120
 121 #
 122 # scan
 123 #   Scan zebra index and populate an array of top terms
 124 #
 125 # PARAMETERS:
 126 #   $max_terms    Max number of top terms
 127 #
 128 # RETURN:
 129 #   A 4-dimensionnal array in $self->{top_terms}
 130 #   [0] term
 131 #   [1] term number of occurrences
 132 #   [2] term proportional relative weight in terms set E[0-1]
 133 #   [3] term logarithmic relative weight E [0-levels_cloud]
 134 #
 135 #   This array is sorted alphabetically by terms ([0])
 136 #   It can be easily sorted by occurrences:
 137 #     @t = sort { $a[1] <=> $a[1] } @{$self->{top_terms}};
 138 #
 139 sub scan {
 140     my $self       = shift;
 141     my $index_name = $self->{ zebra_index };
 142     my $max_terms  = shift;
 143
 144     my $MAX_OCCURENCE = 1000000000;
 145
 146     my $zbiblio = C4::Context->Zconn( "biblioserver" );
 147     my $number_of_terms = 0;
 148     my @terms;      # 2 dimensions array
 149     my $min_occurence_index = -1;
 150     my $min_occurence;
 151     my $from = '0';
 152
 153     while (1) {
 154         my $ss;
 155         eval {
 156             print "$from\n" if $verbose;
 157             $from =~ s/\"/\\\"/g;
 158             my $query = '@attr 1=' . $index_name . ' @attr 4=1 @attr 6=3 "'
 159                         . $from . 'a"';
 160             $ss = $zbiblio->scan_pqf( $query );
 161         };
 162         if ($@) {
 163             chop $from;
 164             next;
 165         }
 166         $ss->option( rpnCharset => 'UTF-8' );
 167         last if $ss->size() == 0;
 168         my $term = '';
 169         my $occ = 0;
 170         for my $index ( 0..$ss->size()-1 ) {
 171             ($term, $occ) = $ss->display_term($index);
 172             #print "$term:$occ\n";
 173             if ( $number_of_terms < $max_terms ) {
 174                 push( @terms, [ $term, $occ ] );
 175                 ++$number_of_terms;
 176                 if ( $number_of_terms == $max_terms ) {
 177                     $min_occurence = $MAX_OCCURENCE;
 178                     for (0..$number_of_terms-1) {
 179                         my @term = @{ $terms[$_] };
 180                         if ( $term[1] <= $min_occurence ) {
 181                             $min_occurence       = $term[1];
 182                             $min_occurence_index = $_;
 183                         }
 184                     }
 185                 }
 186             }
 187             else {
 188                 if ( $occ > $min_occurence) {
 189                     @{ $terms[$min_occurence_index] }[0] = $term;
 190                     @{ $terms[$min_occurence_index] }[1] = $occ;
 191                     $min_occurence = $MAX_OCCURENCE;
 192                     for (0..$max_terms-1) {
 193                         my @term = @{ $terms[$_] };
 194                         if ( $term[1] <= $min_occurence ) {
 195                             $min_occurence       = $term[1];
 196                             $min_occurence_index = $_;
 197                         }
 198                     }
 199                 }
 200             }
 201         }
 202         $from = $term;
 203     }
 204
 205     # Sort array of array by terms weight
 206     @terms = sort { @{$a}[1] <=> @{$b}[1] } @terms;
 207
 208     # A relatif weight to other set terms is added to each term
 209     my $min     = $terms[0][1];
 210     my $log_min = log( $min );
 211     my $max     = $terms[$#terms][1];
 212     my $log_max = log( $max );
 213     my $delta   = $max - $min;
 214     $delta = 1 if $delta == 0; # Very unlikely
 215     my $factor;
 216     if ($log_max - $log_min == 0) {
 217         $log_min = $log_min - $self->{levels_cloud};
 218         $factor = 1;
 219     }
 220     else {
 221         $factor = $self->{levels_cloud} / ($log_max - $log_min);
 222     }
 223
 224     foreach (0..$#terms) {
 225         my $count = @{ $terms[$_] }[1];
 226         my $weight = ( $count - $min ) / $delta;
 227         my $log_weight = int( (log($count) - $log_min) * $factor);
 228         push( @{ $terms[$_] }, $weight );
 229         push( @{ $terms[$_] }, $log_weight );
 230     }
 231     $self->{ top_terms } = \@terms;
 232
 233     # Sort array of array by terms alphabetical order
 234     @terms = sort { @{$a}[0] cmp @{$b}[0] } @terms;
 235 }
 236
 237
 238 #
 239 # Returns a HTML version of index top terms formatted
 240 # as a 'tag cloud'.
 241 #
 242 sub html_cloud {
 243     my $self = shift;
 244     my $koha_index = shift;
 245     my $withcss = shift;
 246     my @terms = @{ $self->{top_terms} };
 247     my $html = '';
 248     if ( $withcss ) {
 249         $html = <<EOS;
 250 <style>
 251 .subjectcloud {
 252     text-align:  center;
 253     line-height: 16px;
 254     margin: 20px;
 255     background: #f0f0f0;
 256     padding: 3%;
 257 }
 258 .subjectcloud a {
 259     font-weight: lighter;
 260     text-decoration: none;
 261 }
 262 span.tagcloud0  { font-size: 12px;}
 263 span.tagcloud1  { font-size: 13px;}
 264 span.tagcloud2  { font-size: 14px;}
 265 span.tagcloud3  { font-size: 15px;}
 266 span.tagcloud4  { font-size: 16px;}
 267 span.tagcloud5  { font-size: 17px;}
 268 span.tagcloud6  { font-size: 18px;}
 269 span.tagcloud7  { font-size: 19px;}
 270 span.tagcloud8  { font-size: 20px;}
 271 span.tagcloud9  { font-size: 21px;}
 272 span.tagcloud10 { font-size: 22px;}
 273 span.tagcloud11 { font-size: 23px;}
 274 span.tagcloud12 { font-size: 24px;}
 275 span.tagcloud13 { font-size: 25px;}
 276 span.tagcloud14 { font-size: 26px;}
 277 span.tagcloud15 { font-size: 27px;}
 278 span.tagcloud16 { font-size: 28px;}
 279 span.tagcloud17 { font-size: 29px;}
 280 span.tagcloud18 { font-size: 30px;}
 281 span.tagcloud19 { font-size: 31px;}
 282 span.tagcloud20 { font-size: 32px;}
 283 span.tagcloud21 { font-size: 33px;}
 284 span.tagcloud22 { font-size: 34px;}
 285 span.tagcloud23 { font-size: 35px;}
 286 span.tagcloud24 { font-size: 36px;}
 287 </style>
 288 <div class="subjectcloud">
 289 EOS
 290     }
 291     for (0..$#terms) {
 292         my @term = @{ $terms[$_] };
 293         my $uri = $term[0];
 294         $uri =~ s/\(//g;
 295         #print "  0=", $term[0]," - 1=", $term[1], " - 2=", $term[2], " - 3=", $term[3],"\n";
 296         $html = $html
 297             . '<span class="tagcloud'
 298             . $term[3]
 299             . '">'
 300             . '<a href="/cgi-bin/koha/opac-search.pl?q='
 301             . $koha_index
 302             . '%3A'
 303             . $uri
 304             . '">'
 305             . $term[0]
 306             . "</a></span>\n";
 307     }
 308     $html .= "</div>\n";
 309     return $html;
 310 }
 311
 312
 313 =head1 NAME
 314
 315 cloud-kw.pl - Creates HTML keywords clouds from Koha Zebra Indexes
 316
 317 =head1 USAGE
 318
 319 =over
 320
 321 =item cloud-kw.pl [--verbose|--help] --conf=F<cloud.conf>
 322
 323 Creates multiple HTML files containing kewords cloud with top terms sorted
 324 by their logarithmic weight.
 325 F<cloud.conf> is a YAML configuration file driving cloud generation
 326 process.
 327
 328 =back
 329
 330 =head1 PARAMETERS
 331
 332 =over
 333
 334 =item B<--conf=configuration file>
 335
 336 Specify configuration file name
 337
 338 =item B<--verbose|-v>
 339
 340 Enable script verbose mode.
 341
 342 =item B<--help|-h>
 343
 344 Print this help page.
 345
 346 =back
 347
 348 =head1 CONFIGURATION
 349
 350 Configuration file looks like that:
 351
 352  ---
 353   # Koha configuration file for a specific installation
 354   # If not present, defaults to KOHA_CONF
 355   KohaConf: /home/koha/mylibray/etc/koha-conf.xml
 356   # Zebra index to scan
 357   ZebraIndex: Author
 358   # Koha index used to link found kewords with an opac search URL
 359   KohaIndex: au
 360   # Number of top keyword to use for the cloud
 361   Count: 50
 362   # Include CSS style directives with the cloud
 363   # This could be used as a model and then CSS directives are
 364   # put in the appropriate CSS file directly.
 365   Withcss: Yes
 366   # HTML file where to output the cloud
 367   Output: /home/koha/mylibrary/koharoot/koha-tmpl/cloud-author.html
 368  ---
 369   KohaConf: /home/koha/yourlibray/etc/koha-conf.xml
 370   ZebraIndex: Subject
 371   KohaIndex: su
 372   Count: 200
 373   Withcss: no
 374   Output: /home/koha/yourlibrary/koharoot/koha-tmpl/cloud-subject.html
 375
 376 =head1 IMPROVEMENTS
 377
 378 Generated top terms have more informations than those outputted from
 379 the time being. Some parameters could be easily added to improve
 380 this script:
 381
 382 =over
 383
 384 =item B<WithCount>
 385
 386 In order to output terms with the number of occurrences they
 387 have been found in Koha Catalogue by Zebra.
 388
 389 =item B<CloudLevels>
 390
 391 Number of levels in the cloud. Now 24 levels are hardcoded.
 392
 393 =item B<Weithing>
 394
 395 Weighting method used to distribute terms in the cloud. We could have two
 396 values: Logarithmic and Linear. Now it's Logarithmic by default.
 397
 398 =item B<Order>
 399
 400 Now terms are outputted in the lexical order. They could be sorted
 401 by their weight.
 402
 403 =back
 404
 405 =cut
 406
 407