misc/migration_tools/remove_unused_authorities.pl

   1 #!/usr/bin/perl
   2
   3 #script to administer Authorities without biblio
   4
   5 # Copyright 2009 BibLibre
   6 # written 2009-05-04 by paul dot poulain at biblibre.com
   7 #
   8 # This file is part of Koha.
   9 #
  10 # Koha is free software; you can redistribute it and/or modify it
  11 # under the terms of the GNU General Public License as published by
  12 # the Free Software Foundation; either version 3 of the License, or
  13 # (at your option) any later version.
  14 #
  15 # Koha is distributed in the hope that it will be useful, but
  16 # WITHOUT ANY WARRANTY; without even the implied warranty of
  17 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  18 # GNU General Public License for more details.
  19 #
  20 # You should have received a copy of the GNU General Public License
  21 # along with Koha; if not, see <http://www.gnu.org/licenses>.
  22
  23 use strict;
  24 use warnings;
  25
  26 use C4::Context;
  27 use C4::AuthoritiesMarc;
  28 use Getopt::Long;
  29
  30 use Koha::SearchEngine::Search;
  31
  32 my @authtypes;
  33 my $want_help = 0;
  34 my $test = 0;
  35 GetOptions(
  36     'aut|authtypecode:s' => \@authtypes,
  37     't|test'             => \$test,
  38     'h|help'             => \$want_help
  39 );
  40
  41 if ($want_help) {
  42     print_usage();
  43     exit 0;
  44 }
  45
  46 if ($test) {
  47     print "testing only, authorities will not be deleted.\n";
  48 }
  49
  50 my $errZebraConnection = C4::Context->Zconn("biblioserver",0)->errcode();
  51 if ( $errZebraConnection == 10000 ) {
  52     die "Zebra server seems not to be available. This script needs Zebra runs."
  53 } elsif ( $errZebraConnection ) {
  54     die "Error from Zebra: $errZebraConnection";
  55 }
  56
  57 my $dbh=C4::Context->dbh;
  58 my $thresholdmin=0;
  59 my $thresholdmax=0;
  60 my @results;
  61 # prepare the request to retrieve all authorities of the requested types
  62 my $rqsql = "SELECT * from auth_header where 1";
  63 $rqsql .= " AND authtypecode IN (".join(",",map{$dbh->quote($_)}@authtypes).")" if @authtypes;
  64 my $rqselect = $dbh->prepare($rqsql);
  65 $|=1;
  66
  67 $rqselect->execute;
  68 my $counter=0;
  69 my $totdeleted=0;
  70 my $totundeleted=0;
  71 my $searcher = Koha::SearchEngine::Search->new({index => 'biblios'});
  72 while (my $data=$rqselect->fetchrow_hashref){
  73     my $query;
  74     $query= "an=".$data->{'authid'};
  75     # search for biblios mapped
  76     my ($err,$res,$used) = $searcher->simple_search_compat($query,0,10);
  77     if (defined $err) {
  78         warn "error: $err on search $query\n";
  79         next;
  80     }
  81     print ".";
  82     print "$counter\n" unless $counter++ % 100;
  83     # if found, delete, otherwise, just count
  84     if ($used>=$thresholdmin and $used<=$thresholdmax){
  85         DelAuthority({ authid => $data->{'authid'} }) unless $test;
  86         $totdeleted++;
  87     } else {
  88         $totundeleted++;
  89     }
  90 }
  91
  92 print "$counter authorities parsed, $totdeleted deleted and $totundeleted unchanged because used\n";
  93
  94
  95 sub print_usage {
  96     print <<_USAGE_;
  97 $0: Remove unused authority records
  98
  99 This script removes authority records that do not have any biblio
 100 records attached to them.
 101
 102 If the --aut option is supplied, only authority records of that
 103 particular type will be checked for usage.  --aut can be repeated.
 104
 105 If --aut is not supplied, all authority records will be checked.
 106
 107 Use --test to perform a test run.  This script does not ask the
 108 operator to confirm the deletion of each authority record.
 109
 110 parameters
 111     --aut|authtypecode TYPE       the list of authtypes to check
 112     --test or -t                  test mode, don't delete really, just count
 113     --help or -h                  show this message.
 114
 115 _USAGE_
 116 }