misc/migration_tools/rebuild_zebra.pl

   1 #!/usr/bin/perl
   2
   3 use strict;
   4 #use warnings; FIXME - Bug 2505
   5
   6 use C4::Context;
   7 use Getopt::Long;
   8 use File::Temp qw/ tempdir /;
   9 use File::Path;
  10 use C4::Biblio;
  11 use C4::AuthoritiesMarc;
  12 use C4::Items;
  13
  14 #
  15 # script that checks zebradir structure & create directories & mandatory files if needed
  16 #
  17 #
  18
  19 $|=1; # flushes output
  20 # If the cron job starts us in an unreadable dir, we will break without
  21 # this.
  22 chdir $ENV{HOME} if (!(-r '.'));
  23 my $directory;
  24 my $nosanitize;
  25 my $skip_export;
  26 my $keep_export;
  27 my $skip_index;
  28 my $reset;
  29 my $biblios;
  30 my $authorities;
  31 my $noxml;
  32 my $noshadow;
  33 my $do_munge;
  34 my $want_help;
  35 my $as_xml;
  36 my $process_zebraqueue;
  37 my $do_not_clear_zebraqueue;
  38 my $length;
  39 my $where;
  40 my $offset;
  41 my $verbose_logging = 0;
  42 my $zebraidx_log_opt = " -v none,fatal,warn ";
  43 my $result = GetOptions(
  44     'd:s'           => \$directory,
  45     'r|reset'       => \$reset,
  46     's'             => \$skip_export,
  47     'k'             => \$keep_export,
  48     'I|skip-index'    => \$skip_index,
  49     'nosanitize'    => \$nosanitize,
  50     'b'             => \$biblios,
  51     'noxml'         => \$noxml,
  52     'w'             => \$noshadow,
  53     'munge-config'  => \$do_munge,
  54     'a'             => \$authorities,
  55     'h|help'        => \$want_help,
  56         'x'                             => \$as_xml,
  57     'y'             => \$do_not_clear_zebraqueue,
  58     'z'             => \$process_zebraqueue,
  59     'where:s'        => \$where,
  60     'length:i'        => \$length,
  61     'offset:i'      => \$offset,
  62     'v+'             => \$verbose_logging,
  63 );
  64
  65
  66 if (not $result or $want_help) {
  67     print_usage();
  68     exit 0;
  69 }
  70
  71 if (not $biblios and not $authorities) {
  72     my $msg = "Must specify -b or -a to reindex bibs or authorities\n";
  73     $msg   .= "Please do '$0 --help' to see usage.\n";
  74     die $msg;
  75 }
  76
  77 if ( !$as_xml and $nosanitize ) {
  78     my $msg = "Cannot specify both -no_xml and -nosanitize\n";
  79     $msg   .= "Please do '$0 --help' to see usage.\n";
  80     die $msg;
  81 }
  82
  83 if ($process_zebraqueue and ($skip_export or $reset)) {
  84     my $msg = "Cannot specify -r or -s if -z is specified\n";
  85     $msg   .= "Please do '$0 --help' to see usage.\n";
  86     die $msg;
  87 }
  88
  89 if ($process_zebraqueue and $do_not_clear_zebraqueue) {
  90     my $msg = "Cannot specify both -y and -z\n";
  91     $msg   .= "Please do '$0 --help' to see usage.\n";
  92     die $msg;
  93 }
  94
  95 if ($noshadow) {
  96     $noshadow = ' -n ';
  97 }
  98
  99 #  -v is for verbose, which seems backwards here because of how logging is set
 100 #    on the CLI of zebraidx.  It works this way.  The default is to not log much
 101 if ($verbose_logging >= 2) {
 102     $zebraidx_log_opt = '-v none,fatal,warn,all';
 103 }
 104
 105 my $use_tempdir = 0;
 106 unless ($directory) {
 107     $use_tempdir = 1;
 108     $directory = tempdir(CLEANUP => ($keep_export ? 0 : 1));
 109 }
 110
 111
 112 my $biblioserverdir = C4::Context->zebraconfig('biblioserver')->{directory};
 113 my $authorityserverdir = C4::Context->zebraconfig('authorityserver')->{directory};
 114
 115 my $kohadir = C4::Context->config('intranetdir');
 116 my $bib_index_mode = C4::Context->config('zebra_bib_index_mode') || 'grs1';
 117 my $auth_index_mode = C4::Context->config('zebra_auth_index_mode') || 'dom';
 118
 119 my $dbh = C4::Context->dbh;
 120 my ($biblionumbertagfield,$biblionumbertagsubfield) = &GetMarcFromKohaField("biblio.biblionumber","");
 121 my ($biblioitemnumbertagfield,$biblioitemnumbertagsubfield) = &GetMarcFromKohaField("biblioitems.biblioitemnumber","");
 122
 123 if ( $verbose_logging ) {
 124     print "Zebra configuration information\n";
 125     print "================================\n";
 126     print "Zebra biblio directory      = $biblioserverdir\n";
 127     print "Zebra authorities directory = $authorityserverdir\n";
 128     print "Koha directory              = $kohadir\n";
 129     print "BIBLIONUMBER in :     $biblionumbertagfield\$$biblionumbertagsubfield\n";
 130     print "BIBLIOITEMNUMBER in : $biblioitemnumbertagfield\$$biblioitemnumbertagsubfield\n";
 131     print "================================\n";
 132 }
 133
 134 if ($do_munge) {
 135     munge_config();
 136 }
 137
 138 if ($authorities) {
 139     index_records('authority', $directory, $skip_export, $skip_index, $process_zebraqueue, $as_xml, $noxml, $nosanitize, $do_not_clear_zebraqueue, $verbose_logging, $zebraidx_log_opt, $authorityserverdir);
 140 } else {
 141     print "skipping authorities\n" if ( $verbose_logging );
 142 }
 143
 144 if ($biblios) {
 145     index_records('biblio', $directory, $skip_export, $skip_index, $process_zebraqueue, $as_xml, $noxml, $nosanitize, $do_not_clear_zebraqueue, $verbose_logging, $zebraidx_log_opt, $biblioserverdir);
 146 } else {
 147     print "skipping biblios\n" if ( $verbose_logging );
 148 }
 149
 150
 151 if ( $verbose_logging ) {
 152     print "====================\n";
 153     print "CLEANING\n";
 154     print "====================\n";
 155 }
 156 if ($keep_export) {
 157     print "NOTHING cleaned : the export $directory has been kept.\n";
 158     print "You can re-run this script with the -s ";
 159     if ($use_tempdir) {
 160         print " and -d $directory parameters";
 161     } else {
 162         print "parameter";
 163     }
 164     print "\n";
 165     print "if you just want to rebuild zebra after changing the record.abs\n";
 166     print "or another zebra config file\n";
 167 } else {
 168     unless ($use_tempdir) {
 169         # if we're using a temporary directory
 170         # created by File::Temp, it will be removed
 171         # automatically.
 172         rmtree($directory, 0, 1);
 173         print "directory $directory deleted\n";
 174     }
 175 }
 176
 177 # This checks to see if the zebra directories exist under the provided path.
 178 # If they don't, then zebra is likely to spit the dummy. This returns true
 179 # if the directories had to be created, false otherwise.
 180 sub check_zebra_dirs {
 181         my ($base) = shift() . '/';
 182         my $needed_repairing = 0;
 183         my @dirs = ( '', 'key', 'register', 'shadow', 'tmp' );
 184         foreach my $dir (@dirs) {
 185                 my $bdir = $base . $dir;
 186         if (! -d $bdir) {
 187                 $needed_repairing = 1;
 188                 mkdir $bdir || die "Unable to create '$bdir': $!\n";
 189                 print "$0: needed to create '$bdir'\n";
 190         }
 191     }
 192     return $needed_repairing;
 193 }       # ----------  end of subroutine check_zebra_dirs  ----------
 194
 195 sub index_records {
 196     my ($record_type, $directory, $skip_export, $skip_index, $process_zebraqueue, $as_xml, $noxml, $nosanitize, $do_not_clear_zebraqueue, $verbose_logging, $zebraidx_log_opt, $server_dir) = @_;
 197
 198     my $num_records_exported = 0;
 199     my $records_deleted;
 200     my $need_reset = check_zebra_dirs($server_dir);
 201     if ($need_reset) {
 202         print "$0: found broken zebra server directories: forcing a rebuild\n";
 203         $reset = 1;
 204     }
 205     if ($skip_export && $verbose_logging) {
 206         print "====================\n";
 207         print "SKIPPING $record_type export\n";
 208         print "====================\n";
 209     } else {
 210         if ( $verbose_logging ) {
 211             print "====================\n";
 212             print "exporting $record_type\n";
 213             print "====================\n";
 214         }
 215         mkdir "$directory" unless (-d $directory);
 216         mkdir "$directory/$record_type" unless (-d "$directory/$record_type");
 217         if ($process_zebraqueue) {
 218             my $entries = select_zebraqueue_records($record_type, 'deleted');
 219             mkdir "$directory/del_$record_type" unless (-d "$directory/del_$record_type");
 220             $records_deleted = generate_deleted_marc_records($record_type, $entries, "$directory/del_$record_type", $as_xml);
 221             mark_zebraqueue_batch_done($entries);
 222             $entries = select_zebraqueue_records($record_type, 'updated');
 223             mkdir "$directory/upd_$record_type" unless (-d "$directory/upd_$record_type");
 224             $num_records_exported = export_marc_records_from_list($record_type,
 225                                                                   $entries, "$directory/upd_$record_type", $as_xml, $noxml, $records_deleted);
 226             mark_zebraqueue_batch_done($entries);
 227         } else {
 228             my $sth = select_all_records($record_type);
 229             $num_records_exported = export_marc_records_from_sth($record_type, $sth, "$directory/$record_type", $as_xml, $noxml, $nosanitize);
 230             unless ($do_not_clear_zebraqueue) {
 231                 mark_all_zebraqueue_done($record_type);
 232             }
 233         }
 234     }
 235
 236     #
 237     # and reindexing everything
 238     #
 239     if ($skip_index) {
 240         if ($verbose_logging) {
 241             print "====================\n";
 242             print "SKIPPING $record_type indexing\n";
 243             print "====================\n";
 244         }
 245     } else {
 246         if ( $verbose_logging ) {
 247             print "====================\n";
 248             print "REINDEXING zebra\n";
 249             print "====================\n";
 250         }
 251         my $record_fmt = ($as_xml) ? 'marcxml' : 'iso2709' ;
 252         if ($process_zebraqueue) {
 253             do_indexing($record_type, 'delete', "$directory/del_$record_type", $reset, $noshadow, $record_fmt, $zebraidx_log_opt)
 254                 if %$records_deleted;
 255             do_indexing($record_type, 'update', "$directory/upd_$record_type", $reset, $noshadow, $record_fmt, $zebraidx_log_opt)
 256                 if $num_records_exported;
 257         } else {
 258             do_indexing($record_type, 'update', "$directory/$record_type", $reset, $noshadow, $record_fmt, $zebraidx_log_opt)
 259                 if ($num_records_exported or $skip_export);
 260         }
 261     }
 262 }
 263
 264
 265 sub select_zebraqueue_records {
 266     my ($record_type, $update_type) = @_;
 267
 268     my $server = ($record_type eq 'biblio') ? 'biblioserver' : 'authorityserver';
 269     my $op = ($update_type eq 'deleted') ? 'recordDelete' : 'specialUpdate';
 270
 271     my $sth = $dbh->prepare("SELECT id, biblio_auth_number
 272                              FROM zebraqueue
 273                              WHERE server = ?
 274                              AND   operation = ?
 275                              AND   done = 0
 276                              ORDER BY id DESC");
 277     $sth->execute($server, $op);
 278     my $entries = $sth->fetchall_arrayref({});
 279 }
 280
 281 sub mark_all_zebraqueue_done {
 282     my ($record_type) = @_;
 283
 284     my $server = ($record_type eq 'biblio') ? 'biblioserver' : 'authorityserver';
 285
 286     my $sth = $dbh->prepare("UPDATE zebraqueue SET done = 1
 287                              WHERE server = ?
 288                              AND done = 0");
 289     $sth->execute($server);
 290 }
 291
 292 sub mark_zebraqueue_batch_done {
 293     my ($entries) = @_;
 294
 295     $dbh->{AutoCommit} = 0;
 296     my $sth = $dbh->prepare("UPDATE zebraqueue SET done = 1 WHERE id = ?");
 297     $dbh->commit();
 298     foreach my $id (map { $_->{id} } @$entries) {
 299         $sth->execute($id);
 300     }
 301     $dbh->{AutoCommit} = 1;
 302 }
 303
 304 sub select_all_records {
 305     my $record_type = shift;
 306     return ($record_type eq 'biblio') ? select_all_biblios() : select_all_authorities();
 307 }
 308
 309 sub select_all_authorities {
 310     my $strsth=qq{SELECT authid FROM auth_header};
 311     $strsth.=qq{ WHERE $where } if ($where);
 312     $strsth.=qq{ LIMIT $length } if ($length && !$offset);
 313     $strsth.=qq{ LIMIT $offset,$length } if ($length && $offset);
 314     my $sth = $dbh->prepare($strsth);
 315     $sth->execute();
 316     return $sth;
 317 }
 318
 319 sub select_all_biblios {
 320     my $strsth = qq{ SELECT biblionumber FROM biblioitems };
 321     $strsth.=qq{ WHERE $where } if ($where);
 322     $strsth.=qq{ LIMIT $length } if ($length && !$offset);
 323     $strsth.=qq{ LIMIT $offset,$length } if ($offset);
 324     my $sth = $dbh->prepare($strsth);
 325     $sth->execute();
 326     return $sth;
 327 }
 328
 329 sub include_xml_wrapper {
 330     my $as_xml = shift;
 331     my $record_type = shift;
 332
 333     return 0 unless $as_xml;
 334     return 1 if $record_type eq 'biblio' and $bib_index_mode eq 'dom';
 335     return 1 if $record_type eq 'authority' and $auth_index_mode eq 'dom';
 336     return 0;
 337
 338 }
 339
 340 sub export_marc_records_from_sth {
 341     my ($record_type, $sth, $directory, $as_xml, $noxml, $nosanitize) = @_;
 342
 343     my $num_exported = 0;
 344     open my $fh, '>:encoding(UTF-8) ', "$directory/exported_records" or die $!;
 345     if (include_xml_wrapper($as_xml, $record_type)) {
 346         # include XML declaration and root element
 347         print {$fh} '<?xml version="1.0" encoding="UTF-8"?><collection>';
 348     }
 349     my $i = 0;
 350     my ( $itemtag, $itemsubfield ) = GetMarcFromKohaField("items.itemnumber",'');
 351     while (my ($record_number) = $sth->fetchrow_array) {
 352         print "." if ( $verbose_logging );
 353         print "\r$i" unless ($i++ %100 or !$verbose_logging);
 354         if ( $nosanitize ) {
 355             my $marcxml = $record_type eq 'biblio'
 356                           ? GetXmlBiblio( $record_number )
 357                           : GetAuthorityXML( $record_number );
 358             if ($record_type eq 'biblio'){
 359                 my @items = GetItemsInfo($record_number);
 360                 if (@items){
 361                     my $record = MARC::Record->new;
 362                     $record->encoding('UTF-8');
 363                     my @itemsrecord;
 364                     foreach my $item (@items){
 365                         my $record = Item2Marc($item, $record_number);
 366                         push @itemsrecord, $record->field($itemtag);
 367                     }
 368                     $record->insert_fields_ordered(@itemsrecord);
 369                     my $itemsxml = $record->as_xml_record();
 370                     $marcxml =
 371                         substr($marcxml, 0, length($marcxml)-10) .
 372                         substr($itemsxml, index($itemsxml, "</leader>\n", 0) + 10);
 373                 }
 374             }
 375             if ( $marcxml ) {
 376                 print {$fh} $marcxml if $marcxml;
 377                 $num_exported++;
 378             }
 379             next;
 380         }
 381         my ($marc) = get_corrected_marc_record($record_type, $record_number, $noxml);
 382         if (defined $marc) {
 383             eval {
 384                 my $rec;
 385                 if ($as_xml) {
 386                     $rec = $marc->as_xml_record(C4::Context->preference('marcflavour'));
 387                     $rec =~ s!<\?xml version="1.0" encoding="UTF-8"\?>\n!!;
 388                 } else {
 389                     $rec = $marc->as_usmarc();
 390                 }
 391                 print {$fh} $rec;
 392                 $num_exported++;
 393             };
 394             if ($@) {
 395               warn "Error exporting record $record_number ($record_type) ".($noxml ? "not XML" : "XML");
 396             }
 397         }
 398     }
 399     print "\nRecords exported: $num_exported\n" if ( $verbose_logging );
 400     print {$fh} '</collection>' if (include_xml_wrapper($as_xml, $record_type));
 401     close $fh;
 402     return $num_exported;
 403 }
 404
 405 sub export_marc_records_from_list {
 406     my ($record_type, $entries, $directory, $as_xml, $noxml, $records_deleted) = @_;
 407
 408     my $num_exported = 0;
 409     open my $fh, '>:encoding(UTF-8)', "$directory/exported_records" or die $!;
 410     if (include_xml_wrapper($as_xml, $record_type)) {
 411         # include XML declaration and root element
 412         print {$fh} '<?xml version="1.0" encoding="UTF-8"?><collection>';
 413     }
 414     my $i = 0;
 415
 416     # Skip any deleted records. We check for this anyway, but this reduces error spam
 417     my %found = %$records_deleted;
 418     foreach my $record_number ( map { $_->{biblio_auth_number} }
 419                                 grep { !$found{ $_->{biblio_auth_number} }++ }
 420                                 @$entries ) {
 421         print "." if ( $verbose_logging );
 422         print "\r$i" unless ($i++ %100 or !$verbose_logging);
 423         my ($marc) = get_corrected_marc_record($record_type, $record_number, $noxml);
 424         if (defined $marc) {
 425             eval {
 426                 my $rec;
 427                 if ($as_xml) {
 428                     $rec = $marc->as_xml_record(C4::Context->preference('marcflavour'));
 429                     $rec =~ s!<\?xml version="1.0" encoding="UTF-8"\?>\n!!;
 430                 } else {
 431                     $rec = $marc->as_usmarc();
 432                 }
 433                 print {$fh} $rec;
 434                 $num_exported++;
 435             };
 436             if ($@) {
 437               warn "Error exporting record $record_number ($record_type) ".($noxml ? "not XML" : "XML");
 438             }
 439             $num_exported++;
 440         }
 441     }
 442     print "\nRecords exported: $num_exported\n" if ( $verbose_logging );
 443     print {$fh} '</collection>' if (include_xml_wrapper($as_xml, $record_type));
 444     close $fh;
 445     return $num_exported;
 446 }
 447
 448 sub generate_deleted_marc_records {
 449     my ($record_type, $entries, $directory, $as_xml) = @_;
 450
 451     my $records_deleted = {};
 452     open my $fh, '>:encoding(UTF-8)', "$directory/exported_records" or die $!;
 453     if (include_xml_wrapper($as_xml, $record_type)) {
 454         # include XML declaration and root element
 455         print {$fh} '<?xml version="1.0" encoding="UTF-8"?><collection>';
 456     }
 457     my $i = 0;
 458     foreach my $record_number (map { $_->{biblio_auth_number} } @$entries ) {
 459         print "\r$i" unless ($i++ %100 or !$verbose_logging);
 460         print "." if ( $verbose_logging );
 461
 462         my $marc = MARC::Record->new();
 463         if ($record_type eq 'biblio') {
 464             fix_biblio_ids($marc, $record_number, $record_number);
 465         } else {
 466             fix_authority_id($marc, $record_number);
 467         }
 468         if (C4::Context->preference("marcflavour") eq "UNIMARC") {
 469             fix_unimarc_100($marc);
 470         }
 471
 472         my $rec;
 473         if ($as_xml) {
 474             $rec = $marc->as_xml_record(C4::Context->preference('marcflavour'));
 475             $rec =~ s!<\?xml version="1.0" encoding="UTF-8"\?>\n!!;
 476         } else {
 477             $rec = $marc->as_usmarc();
 478         }
 479         print {$fh} $rec;
 480
 481         $records_deleted->{$record_number} = 1;
 482     }
 483     print "\nRecords exported: $i\n" if ( $verbose_logging );
 484     print {$fh} '</collection>' if (include_xml_wrapper($as_xml, $record_type));
 485     close $fh;
 486     return $records_deleted;
 487
 488
 489 }
 490
 491 sub get_corrected_marc_record {
 492     my ($record_type, $record_number, $noxml) = @_;
 493
 494     my $marc = get_raw_marc_record($record_type, $record_number, $noxml);
 495
 496     if (defined $marc) {
 497         fix_leader($marc);
 498         if ($record_type eq 'authority') {
 499             fix_authority_id($marc, $record_number);
 500         }
 501         if (C4::Context->preference("marcflavour") eq "UNIMARC") {
 502             fix_unimarc_100($marc);
 503         }
 504     }
 505
 506     return $marc;
 507 }
 508
 509 sub get_raw_marc_record {
 510     my ($record_type, $record_number, $noxml) = @_;
 511
 512     my $marc;
 513     if ($record_type eq 'biblio') {
 514         if ($noxml) {
 515             my $fetch_sth = $dbh->prepare_cached("SELECT marc FROM biblioitems WHERE biblionumber = ?");
 516             $fetch_sth->execute($record_number);
 517             if (my ($blob) = $fetch_sth->fetchrow_array) {
 518                 $marc = MARC::Record->new_from_usmarc($blob);
 519                 unless ($marc) {
 520                     warn "error creating MARC::Record from $blob";
 521                 }
 522             }
 523             # failure to find a bib is not a problem -
 524             # a delete could have been done before
 525             # trying to process a record update
 526
 527             $fetch_sth->finish();
 528             return unless $marc;
 529         } else {
 530             eval { $marc = GetMarcBiblio($record_number, 1); };
 531             if ($@ || !$marc) {
 532                 # here we do warn since catching an exception
 533                 # means that the bib was found but failed
 534                 # to be parsed
 535                 warn "error retrieving biblio $record_number";
 536                 return;
 537             }
 538         }
 539     } else {
 540         eval { $marc = GetAuthority($record_number); };
 541         if ($@) {
 542             warn "error retrieving authority $record_number";
 543             return;
 544         }
 545     }
 546     return $marc;
 547 }
 548
 549 sub fix_leader {
 550     # FIXME - this routine is suspect
 551     # It blanks the Leader/00-05 and Leader/12-16 to
 552     # force them to be recalculated correct when
 553     # the $marc->as_usmarc() or $marc->as_xml() is called.
 554     # But why is this necessary?  It would be a serious bug
 555     # in MARC::Record (definitely) and MARC::File::XML (arguably)
 556     # if they are emitting incorrect leader values.
 557     my $marc = shift;
 558
 559     my $leader = $marc->leader;
 560     substr($leader,  0, 5) = '     ';
 561     substr($leader, 10, 7) = '22     ';
 562     $marc->leader(substr($leader, 0, 24));
 563 }
 564
 565 sub fix_biblio_ids {
 566     # FIXME - it is essential to ensure that the biblionumber is present,
 567     #         otherwise, Zebra will choke on the record.  However, this
 568     #         logic belongs in the relevant C4::Biblio APIs.
 569     my $marc = shift;
 570     my $biblionumber = shift;
 571     my $biblioitemnumber;
 572     if (@_) {
 573         $biblioitemnumber = shift;
 574     } else {
 575         my $sth = $dbh->prepare(
 576             "SELECT biblioitemnumber FROM biblioitems WHERE biblionumber=?");
 577         $sth->execute($biblionumber);
 578         ($biblioitemnumber) = $sth->fetchrow_array;
 579         $sth->finish;
 580         unless ($biblioitemnumber) {
 581             warn "failed to get biblioitemnumber for biblio $biblionumber";
 582             return 0;
 583         }
 584     }
 585
 586     # FIXME - this is cheating on two levels
 587     # 1. C4::Biblio::_koha_marc_update_bib_ids is meant to be an internal function
 588     # 2. Making sure that the biblionumber and biblioitemnumber are correct and
 589     #    present in the MARC::Record object ought to be part of GetMarcBiblio.
 590     #
 591     # On the other hand, this better for now than what rebuild_zebra.pl used to
 592     # do, which was duplicate the code for inserting the biblionumber
 593     # and biblioitemnumber
 594     C4::Biblio::_koha_marc_update_bib_ids($marc, '', $biblionumber, $biblioitemnumber);
 595
 596     return 1;
 597 }
 598
 599 sub fix_authority_id {
 600     # FIXME - as with fix_biblio_ids, the authid must be present
 601     #         for Zebra's sake.  However, this really belongs
 602     #         in C4::AuthoritiesMarc.
 603     my ($marc, $authid) = @_;
 604     unless ($marc->field('001') and $marc->field('001')->data() eq $authid){
 605         $marc->delete_field($marc->field('001'));
 606         $marc->insert_fields_ordered(MARC::Field->new('001',$authid));
 607     }
 608 }
 609
 610 sub fix_unimarc_100 {
 611     # FIXME - again, if this is necessary, it belongs in C4::AuthoritiesMarc.
 612     my $marc = shift;
 613
 614     my $string;
 615     if ( length($marc->subfield( 100, "a" )) == 36 ) {
 616         $string = $marc->subfield( 100, "a" );
 617         my $f100 = $marc->field(100);
 618         $marc->delete_field($f100);
 619     }
 620     else {
 621         $string = POSIX::strftime( "%Y%m%d", localtime );
 622         $string =~ s/\-//g;
 623         $string = sprintf( "%-*s", 35, $string );
 624     }
 625     substr( $string, 22, 6, "frey50" );
 626     unless ( length($marc->subfield( 100, "a" )) == 36 ) {
 627         $marc->delete_field($marc->field(100));
 628         $marc->insert_grouped_field(MARC::Field->new( 100, "", "", "a" => $string ));
 629     }
 630 }
 631
 632 sub do_indexing {
 633     my ($record_type, $op, $record_dir, $reset_index, $noshadow, $record_format, $zebraidx_log_opt) = @_;
 634
 635     my $zebra_server  = ($record_type eq 'biblio') ? 'biblioserver' : 'authorityserver';
 636     my $zebra_db_name = ($record_type eq 'biblio') ? 'biblios' : 'authorities';
 637     my $zebra_config  = C4::Context->zebraconfig($zebra_server)->{'config'};
 638     my $zebra_db_dir  = C4::Context->zebraconfig($zebra_server)->{'directory'};
 639
 640     system("zebraidx -c $zebra_config $zebraidx_log_opt -g $record_format -d $zebra_db_name init") if $reset_index;
 641     system("zebraidx -c $zebra_config $zebraidx_log_opt $noshadow -g $record_format -d $zebra_db_name $op $record_dir");
 642     system("zebraidx -c $zebra_config $zebraidx_log_opt -g $record_format -d $zebra_db_name commit") unless $noshadow;
 643
 644 }
 645
 646 sub print_usage {
 647     print <<_USAGE_;
 648 $0: reindex MARC bibs and/or authorities in Zebra.
 649
 650 Use this batch job to reindex all biblio or authority
 651 records in your Koha database.  This job is useful
 652 only if you are using Zebra; if you are using the 'NoZebra'
 653 mode, this job should not be used.
 654
 655 Parameters:
 656     -b                      index bibliographic records
 657
 658     -a                      index authority records
 659
 660     -z                      select only updated and deleted
 661                             records marked in the zebraqueue
 662                             table.  Cannot be used with -r
 663                             or -s.
 664
 665     -r                      clear Zebra index before
 666                             adding records to index
 667
 668     -d                      Temporary directory for indexing.
 669                             If not specified, one is automatically
 670                             created.  The export directory
 671                             is automatically deleted unless
 672                             you supply the -k switch.
 673
 674     -k                      Do not delete export directory.
 675
 676     -s                      Skip export.  Used if you have
 677                             already exported the records
 678                             in a previous run.
 679
 680     -noxml                  index from ISO MARC blob
 681                             instead of MARC XML.  This
 682                             option is recommended only
 683                             for advanced user.
 684
 685     -x                      export and index as xml instead of is02709 (biblios only).
 686                             use this if you might have records > 99,999 chars,
 687
 688     -nosanitize             export biblio/authority records directly from DB marcxml
 689                             field without sanitizing records. It speed up
 690                             dump process but could fail if DB contains badly
 691                             encoded records. Works only with -x,
 692
 693     -w                      skip shadow indexing for this batch
 694
 695     -y                      do NOT clear zebraqueue after indexing; normally,
 696                             after doing batch indexing, zebraqueue should be
 697                             marked done for the affected record type(s) so that
 698                             a running zebraqueue_daemon doesn't try to reindex
 699                             the same records - specify -y to override this.
 700                             Cannot be used with -z.
 701
 702     -v                      increase the amount of logging.  Normally only
 703                             warnings and errors from the indexing are shown.
 704                             Use log level 2 (-v -v) to include all Zebra logs.
 705
 706     --length   1234         how many biblio you want to export
 707     --offset 1243           offset you want to start to
 708                                 example: --offset 500 --length=500 will result in a LIMIT 500,1000 (exporting 1000 records, starting by the 500th one)
 709                                 note that the numbers are NOT related to biblionumber, that's the intended behaviour.
 710     --where                 let you specify a WHERE query, like itemtype='BOOK'
 711                             or something like that
 712
 713     --munge-config          Deprecated option to try
 714                             to fix Zebra config files.
 715     --help or -h            show this message.
 716 _USAGE_
 717 }
 718
 719 # FIXME: the following routines are deprecated and
 720 # will be removed once it is determined whether
 721 # a script to fix Zebra configuration files is
 722 # actually needed.
 723 sub munge_config {
 724 #
 725 # creating zebra-biblios.cfg depending on system
 726 #
 727
 728 # getting zebraidx directory
 729 my $zebraidxdir;
 730 foreach (qw(/usr/local/bin/zebraidx
 731         /opt/bin/zebraidx
 732         /usr/bin/zebraidx
 733         )) {
 734     if ( -f $_ ) {
 735         $zebraidxdir=$_;
 736     }
 737 }
 738
 739 unless ($zebraidxdir) {
 740     print qq|
 741     ERROR: could not find zebraidx directory
 742     ERROR: Either zebra is not installed,
 743     ERROR: or it's in a directory I don't checked.
 744     ERROR: do a which zebraidx and edit this file to add the result you get
 745 |;
 746     exit;
 747 }
 748 $zebraidxdir =~ s/\/bin\/.*//;
 749 print "Info : zebra is in $zebraidxdir \n";
 750
 751 # getting modules directory
 752 my $modulesdir;
 753 foreach (qw(/usr/local/lib/idzebra-2.0/modules/mod-grs-xml.so
 754             /usr/local/lib/idzebra/modules/mod-grs-xml.so
 755             /usr/lib/idzebra/modules/mod-grs-xml.so
 756             /usr/lib/idzebra-2.0/modules/mod-grs-xml.so
 757         )) {
 758     if ( -f $_ ) {
 759         $modulesdir=$_;
 760     }
 761 }
 762
 763 unless ($modulesdir) {
 764     print qq|
 765     ERROR: could not find mod-grs-xml.so directory
 766     ERROR: Either zebra is not properly compiled (libxml2 is not setup and you don t have mod-grs-xml.so,
 767     ERROR: or it's in a directory I don't checked.
 768     ERROR: find where mod-grs-xml.so is and edit this file to add the result you get
 769 |;
 770     exit;
 771 }
 772 $modulesdir =~ s/\/modules\/.*//;
 773 print "Info: zebra modules dir : $modulesdir\n";
 774
 775 # getting tab directory
 776 my $tabdir;
 777 foreach (qw(/usr/local/share/idzebra/tab/explain.att
 778             /usr/local/share/idzebra-2.0/tab/explain.att
 779             /usr/share/idzebra/tab/explain.att
 780             /usr/share/idzebra-2.0/tab/explain.att
 781         )) {
 782     if ( -f $_ ) {
 783         $tabdir=$_;
 784     }
 785 }
 786
 787 unless ($tabdir) {
 788     print qq|
 789     ERROR: could not find explain.att directory
 790     ERROR: Either zebra is not properly compiled,
 791     ERROR: or it's in a directory I don't checked.
 792     ERROR: find where explain.att is and edit this file to add the result you get
 793 |;
 794     exit;
 795 }
 796 $tabdir =~ s/\/tab\/.*//;
 797 print "Info: tab dir : $tabdir\n";
 798
 799 #
 800 # AUTHORITIES creating directory structure
 801 #
 802 my $created_dir_or_file = 0;
 803 if ($authorities) {
 804     if ( $verbose_logging ) {
 805         print "====================\n";
 806         print "checking directories & files for authorities\n";
 807         print "====================\n";
 808     }
 809     unless (-d "$authorityserverdir") {
 810         system("mkdir -p $authorityserverdir");
 811         print "Info: created $authorityserverdir\n";
 812         $created_dir_or_file++;
 813     }
 814     unless (-d "$authorityserverdir/lock") {
 815         mkdir "$authorityserverdir/lock";
 816         print "Info: created $authorityserverdir/lock\n";
 817         $created_dir_or_file++;
 818     }
 819     unless (-d "$authorityserverdir/register") {
 820         mkdir "$authorityserverdir/register";
 821         print "Info: created $authorityserverdir/register\n";
 822         $created_dir_or_file++;
 823     }
 824     unless (-d "$authorityserverdir/shadow") {
 825         mkdir "$authorityserverdir/shadow";
 826         print "Info: created $authorityserverdir/shadow\n";
 827         $created_dir_or_file++;
 828     }
 829     unless (-d "$authorityserverdir/tab") {
 830         mkdir "$authorityserverdir/tab";
 831         print "Info: created $authorityserverdir/tab\n";
 832         $created_dir_or_file++;
 833     }
 834     unless (-d "$authorityserverdir/key") {
 835         mkdir "$authorityserverdir/key";
 836         print "Info: created $authorityserverdir/key\n";
 837         $created_dir_or_file++;
 838     }
 839
 840     unless (-d "$authorityserverdir/etc") {
 841         mkdir "$authorityserverdir/etc";
 842         print "Info: created $authorityserverdir/etc\n";
 843         $created_dir_or_file++;
 844     }
 845
 846     #
 847     # AUTHORITIES : copying mandatory files
 848     #
 849     # the record model, depending on marc flavour
 850     unless (-f "$authorityserverdir/tab/record.abs") {
 851         if (C4::Context->preference("marcflavour") eq "UNIMARC") {
 852             system("cp -f $kohadir/etc/zebradb/marc_defs/unimarc/authorities/record.abs $authorityserverdir/tab/record.abs");
 853             print "Info: copied record.abs for UNIMARC\n";
 854         } else {
 855             system("cp -f $kohadir/etc/zebradb/marc_defs/marc21/authorities/record.abs $authorityserverdir/tab/record.abs");
 856             print "Info: copied record.abs for USMARC\n";
 857         }
 858         $created_dir_or_file++;
 859     }
 860     unless (-f "$authorityserverdir/tab/sort-string-utf.chr") {
 861         system("cp -f $kohadir/etc/zebradb/lang_defs/fr/sort-string-utf.chr $authorityserverdir/tab/sort-string-utf.chr");
 862         print "Info: copied sort-string-utf.chr\n";
 863         $created_dir_or_file++;
 864     }
 865     unless (-f "$authorityserverdir/tab/word-phrase-utf.chr") {
 866         system("cp -f $kohadir/etc/zebradb/lang_defs/fr/sort-string-utf.chr $authorityserverdir/tab/word-phrase-utf.chr");
 867         print "Info: copied word-phase-utf.chr\n";
 868         $created_dir_or_file++;
 869     }
 870     unless (-f "$authorityserverdir/tab/auth1.att") {
 871         system("cp -f $kohadir/etc/zebradb/authorities/etc/bib1.att $authorityserverdir/tab/auth1.att");
 872         print "Info: copied auth1.att\n";
 873         $created_dir_or_file++;
 874     }
 875     unless (-f "$authorityserverdir/tab/default.idx") {
 876         system("cp -f $kohadir/etc/zebradb/etc/default.idx $authorityserverdir/tab/default.idx");
 877         print "Info: copied default.idx\n";
 878         $created_dir_or_file++;
 879     }
 880
 881     unless (-f "$authorityserverdir/etc/ccl.properties") {
 882 #         system("cp -f $kohadir/etc/zebradb/ccl.properties ".C4::Context->zebraconfig('authorityserver')->{ccl2rpn});
 883         system("cp -f $kohadir/etc/zebradb/ccl.properties $authorityserverdir/etc/ccl.properties");
 884         print "Info: copied ccl.properties\n";
 885         $created_dir_or_file++;
 886     }
 887     unless (-f "$authorityserverdir/etc/pqf.properties") {
 888 #         system("cp -f $kohadir/etc/zebradb/pqf.properties ".C4::Context->zebraconfig('authorityserver')->{ccl2rpn});
 889         system("cp -f $kohadir/etc/zebradb/pqf.properties $authorityserverdir/etc/pqf.properties");
 890         print "Info: copied pqf.properties\n";
 891         $created_dir_or_file++;
 892     }
 893
 894     #
 895     # AUTHORITIES : copying mandatory files
 896     #
 897     unless (-f C4::Context->zebraconfig('authorityserver')->{config}) {
 898     open my $zd, '>:encoding(UTF-8)' ,C4::Context->zebraconfig('authorityserver')->{config};
 899     print {$zd} "
 900 # generated by KOHA/misc/migration_tools/rebuild_zebra.pl
 901 profilePath:\${srcdir:-.}:$authorityserverdir/tab/:$tabdir/tab/:\${srcdir:-.}/tab/
 902
 903 encoding: UTF-8
 904 # Files that describe the attribute sets supported.
 905 attset: auth1.att
 906 attset: explain.att
 907 attset: gils.att
 908
 909 modulePath:$modulesdir/modules/
 910 # Specify record type
 911 iso2709.recordType:grs.marcxml.record
 912 recordType:grs.xml
 913 recordId: (auth1,Local-Number)
 914 storeKeys:1
 915 storeData:1
 916
 917
 918 # Lock File Area
 919 lockDir: $authorityserverdir/lock
 920 perm.anonymous:r
 921 perm.kohaadmin:rw
 922 register: $authorityserverdir/register:4G
 923 shadow: $authorityserverdir/shadow:4G
 924
 925 # Temp File area for result sets
 926 setTmpDir: $authorityserverdir/tmp
 927
 928 # Temp File area for index program
 929 keyTmpDir: $authorityserverdir/key
 930
 931 # Approx. Memory usage during indexing
 932 memMax: 40M
 933 rank:rank-1
 934     ";
 935         print "Info: creating zebra-authorities.cfg\n";
 936         $created_dir_or_file++;
 937     }
 938
 939     if ($created_dir_or_file) {
 940         print "Info: created : $created_dir_or_file directories & files\n";
 941     } else {
 942         print "Info: file & directories OK\n";
 943     }
 944
 945 }
 946 if ($biblios) {
 947     if ( $verbose_logging ) {
 948         print "====================\n";
 949         print "checking directories & files for biblios\n";
 950         print "====================\n";
 951     }
 952
 953     #
 954     # BIBLIOS : creating directory structure
 955     #
 956     unless (-d "$biblioserverdir") {
 957         system("mkdir -p $biblioserverdir");
 958         print "Info: created $biblioserverdir\n";
 959         $created_dir_or_file++;
 960     }
 961     unless (-d "$biblioserverdir/lock") {
 962         mkdir "$biblioserverdir/lock";
 963         print "Info: created $biblioserverdir/lock\n";
 964         $created_dir_or_file++;
 965     }
 966     unless (-d "$biblioserverdir/register") {
 967         mkdir "$biblioserverdir/register";
 968         print "Info: created $biblioserverdir/register\n";
 969         $created_dir_or_file++;
 970     }
 971     unless (-d "$biblioserverdir/shadow") {
 972         mkdir "$biblioserverdir/shadow";
 973         print "Info: created $biblioserverdir/shadow\n";
 974         $created_dir_or_file++;
 975     }
 976     unless (-d "$biblioserverdir/tab") {
 977         mkdir "$biblioserverdir/tab";
 978         print "Info: created $biblioserverdir/tab\n";
 979         $created_dir_or_file++;
 980     }
 981     unless (-d "$biblioserverdir/key") {
 982         mkdir "$biblioserverdir/key";
 983         print "Info: created $biblioserverdir/key\n";
 984         $created_dir_or_file++;
 985     }
 986     unless (-d "$biblioserverdir/etc") {
 987         mkdir "$biblioserverdir/etc";
 988         print "Info: created $biblioserverdir/etc\n";
 989         $created_dir_or_file++;
 990     }
 991
 992     #
 993     # BIBLIOS : copying mandatory files
 994     #
 995     # the record model, depending on marc flavour
 996     unless (-f "$biblioserverdir/tab/record.abs") {
 997         if (C4::Context->preference("marcflavour") eq "UNIMARC") {
 998             system("cp -f $kohadir/etc/zebradb/marc_defs/unimarc/biblios/record.abs $biblioserverdir/tab/record.abs");
 999             print "Info: copied record.abs for UNIMARC\n";
1000         } else {
1001             system("cp -f $kohadir/etc/zebradb/marc_defs/marc21/biblios/record.abs $biblioserverdir/tab/record.abs");
1002             print "Info: copied record.abs for USMARC\n";
1003         }
1004         $created_dir_or_file++;
1005     }
1006     unless (-f "$biblioserverdir/tab/sort-string-utf.chr") {
1007         system("cp -f $kohadir/etc/zebradb/lang_defs/fr/sort-string-utf.chr $biblioserverdir/tab/sort-string-utf.chr");
1008         print "Info: copied sort-string-utf.chr\n";
1009         $created_dir_or_file++;
1010     }
1011     unless (-f "$biblioserverdir/tab/word-phrase-utf.chr") {
1012         system("cp -f $kohadir/etc/zebradb/lang_defs/fr/sort-string-utf.chr $biblioserverdir/tab/word-phrase-utf.chr");
1013         print "Info: copied word-phase-utf.chr\n";
1014         $created_dir_or_file++;
1015     }
1016     unless (-f "$biblioserverdir/tab/bib1.att") {
1017         system("cp -f $kohadir/etc/zebradb/biblios/etc/bib1.att $biblioserverdir/tab/bib1.att");
1018         print "Info: copied bib1.att\n";
1019         $created_dir_or_file++;
1020     }
1021     unless (-f "$biblioserverdir/tab/default.idx") {
1022         system("cp -f $kohadir/etc/zebradb/etc/default.idx $biblioserverdir/tab/default.idx");
1023         print "Info: copied default.idx\n";
1024         $created_dir_or_file++;
1025     }
1026     unless (-f "$biblioserverdir/etc/ccl.properties") {
1027 #         system("cp -f $kohadir/etc/zebradb/ccl.properties ".C4::Context->zebraconfig('biblioserver')->{ccl2rpn});
1028         system("cp -f $kohadir/etc/zebradb/ccl.properties $biblioserverdir/etc/ccl.properties");
1029         print "Info: copied ccl.properties\n";
1030         $created_dir_or_file++;
1031     }
1032     unless (-f "$biblioserverdir/etc/pqf.properties") {
1033 #         system("cp -f $kohadir/etc/zebradb/pqf.properties ".C4::Context->zebraconfig('biblioserver')->{ccl2rpn});
1034         system("cp -f $kohadir/etc/zebradb/pqf.properties $biblioserverdir/etc/pqf.properties");
1035         print "Info: copied pqf.properties\n";
1036         $created_dir_or_file++;
1037     }
1038
1039     #
1040     # BIBLIOS : copying mandatory files
1041     #
1042     unless (-f C4::Context->zebraconfig('biblioserver')->{config}) {
1043     open my $zd, '>:encoding(UTF-8)', C4::Context->zebraconfig('biblioserver')->{config};
1044     print {$zd} "
1045 # generated by KOHA/misc/migrtion_tools/rebuild_zebra.pl
1046 profilePath:\${srcdir:-.}:$biblioserverdir/tab/:$tabdir/tab/:\${srcdir:-.}/tab/
1047
1048 encoding: UTF-8
1049 # Files that describe the attribute sets supported.
1050 attset:bib1.att
1051 attset:explain.att
1052 attset:gils.att
1053
1054 modulePath:$modulesdir/modules/
1055 # Specify record type
1056 iso2709.recordType:grs.marcxml.record
1057 recordType:grs.xml
1058 recordId: (bib1,Local-Number)
1059 storeKeys:1
1060 storeData:1
1061
1062
1063 # Lock File Area
1064 lockDir: $biblioserverdir/lock
1065 perm.anonymous:r
1066 perm.kohaadmin:rw
1067 register: $biblioserverdir/register:4G
1068 shadow: $biblioserverdir/shadow:4G
1069
1070 # Temp File area for result sets
1071 setTmpDir: $biblioserverdir/tmp
1072
1073 # Temp File area for index program
1074 keyTmpDir: $biblioserverdir/key
1075
1076 # Approx. Memory usage during indexing
1077 memMax: 40M
1078 rank:rank-1
1079     ";
1080         print "Info: creating zebra-biblios.cfg\n";
1081         $created_dir_or_file++;
1082     }
1083
1084     if ($created_dir_or_file) {
1085         print "Info: created : $created_dir_or_file directories & files\n";
1086     } else {
1087         print "Info: file & directories OK\n";
1088     }
1089
1090 }
1091 }