misc/cronjobs/sitemap.pl

   1 #!/usr/bin/perl
   2
   3 # Copyright 2015 Tamil s.a.r.l.
   4 #
   5 # This file is part of Koha.
   6 #
   7 # Koha is free software; you can redistribute it and/or modify it under the
   8 # terms of the GNU General Public License as published by the Free Software
   9 # Foundation; either version 3 of the License, or (at your option) any later
  10 # version.
  11 #
  12 # Koha is distributed in the hope that it will be useful, but WITHOUT ANY
  13 # WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
  14 # A PARTICULAR PURPOSE.  See the GNU General Public License for more details.
  15 #
  16 # You should have received a copy of the GNU General Public License along
  17 # with Koha; if not, write to the Free Software Foundation, Inc.,
  18 # 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
  19
  20 package Main;
  21
  22 use Modern::Perl;
  23 use utf8;
  24 use Pod::Usage;
  25 use Getopt::Long;
  26 use C4::Biblio;
  27 use Koha::Sitemapper;
  28
  29
  30 my ($verbose, $help, $url, $dir, $short) = (0, 0, '', '.', 1);
  31 GetOptions(
  32     'verbose'   => \$verbose,
  33     'help'      => \$help,
  34     'url=s'     => \$url,
  35     'dir=s'     => \$dir,
  36     'short!'    => \$short,
  37 );
  38
  39 sub usage {
  40     pod2usage( -verbose => 2 );
  41     exit;
  42 }
  43
  44 usage() if $help;
  45
  46 unless ($url) {
  47     $url = C4::Context->preference("OPACBaseURL");
  48     unless ($url) {
  49         say "OPACBaseURL syspref isn't defined. You can use --url parameter.";
  50         exit;
  51     }
  52 }
  53 $url =~ s/\/*$//g;
  54
  55 my $sitemapper = Koha::Sitemapper->new(
  56     verbose => $verbose,
  57     url     => $url,
  58     dir     => $dir,
  59     short   => $short,
  60 );
  61 $sitemapper->run();
  62
  63
  64 =head1 USAGE
  65
  66 =over
  67
  68 =item sitemap.pl [--verbose|--help|--short|--noshort|--url|--dir]
  69
  70 =back
  71
  72 =head1 SYNOPSIS
  73
  74   sitemap.pl --verbose
  75   sitemap.pl --noshort --dir /home/koha/mylibrary/www
  76   sitemap.pl --url opac.myDNSname.org
  77
  78 =head1 DESCRIPTION
  79
  80 Process all biblio records from a Koha instance and generate Sitemap files
  81 complying with this protocol as described on L<http://sitemaps.org>. The goal of
  82 this script is to be able to provide to search engines direct access to biblio
  83 records. It avoid leaving search engine browsing Koha OPAC and so generating
  84 a lot of traffic, and workload, for a bad result.
  85
  86 A file name F<sitemapindex.xml> is generated. It contains references to Sitemap
  87 multiples files. Each file contains at most 50,000 urls, and is named
  88 F<sitemapXXXX.xml>.
  89
  90 The files must be stored on Koha OPAC root directory, ie
  91 F<<koha-root>/koha-tmpl/>. Place also in this directory a F<robots.txt> file
  92 like this one:
  93
  94  Sitemap: sitemapindex.xml
  95  User-agent: *
  96  Disallow: /cgi-bin/
  97
  98 =head1 PARAMETERS
  99
 100 =over
 101
 102 =item B<--url=Koha OPAC base URL>
 103
 104 If omitted, OPACBaseURL syspref is used.
 105
 106 =item B<--short|noshort>
 107
 108 By default, --short. With --short, URL to bib record ends with
 109 /bib/biblionumber. With --noshort, URL ends with
 110 /cgi-bin/koha/opac-detail.pl?biblionumber=bibnum
 111
 112 =item B<--dir>
 113
 114 Directory where to write sitemap files. By default, the current directory.
 115
 116 =item B<--verbose|-v>
 117
 118 Enable script verbose mode: a message is displayed for each 10,000 biblio
 119 records processed.
 120
 121 =item B<--help|-h>
 122
 123 Print this help page.
 124
 125 =back
 126
 127 =cut