Bugzilla – Attachment 11161 Details for
Bug 7419
Add authority deduplication script
Home
|
New
|
Browse
|
Search
|
[?]
|
Reports
|
Help
|
New Account
|
Log In
[x]
|
Forgot Password
Login:
[x]
[patch]
Bug 7419: Move indexation subs in C4::Search
0001-Bug-7419-Move-indexation-subs-in-C4-Search.patch (text/plain), 33.01 KB, created by
Julian Maurice
on 2012-07-27 12:22:35 UTC
(
hide
)
Description:
Bug 7419: Move indexation subs in C4::Search
Filename:
MIME Type:
Creator:
Julian Maurice
Created:
2012-07-27 12:22:35 UTC
Size:
33.01 KB
patch
obsolete
>From 301b63ba974269e32661d83989766c08f1fde58a Mon Sep 17 00:00:00 2001 >From: Julian Maurice <julian.maurice@biblibre.com> >Date: Fri, 27 Jul 2012 08:41:14 +0200 >Subject: [PATCH 1/2] Bug 7419: Move indexation subs in C4::Search > >Move indexation-related subs from rebuild_zebra.pl to C4::Search and >create two new subs IndexRecord and DeleteRecordIndex >--- > C4/Search.pm | 367 +++++++++++++++++++++++++++++ > misc/migration_tools/rebuild_zebra.pl | 405 ++++----------------------------- > 2 files changed, 414 insertions(+), 358 deletions(-) > >diff --git a/C4/Search.pm b/C4/Search.pm >index fd81610..bad5f13 100644 >--- a/C4/Search.pm >+++ b/C4/Search.pm >@@ -34,6 +34,7 @@ use C4::Charset; > use YAML; > use URI::Escape; > use Business::ISBN; >+use File::Temp qw/tempdir/; > > use vars qw($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS $DEBUG); > >@@ -2712,6 +2713,372 @@ sub GetDistinctValues { > } > } > >+sub IndexRecord { >+ my ($recordtype, $recordids, $options) = @_; >+ >+ my $as_xml = ($options and $options->{as_xml}) ? 1 : 0; >+ my $noxml = ($options and $options->{noxml}) ? 1 : 0; >+ my $nosanitize = ($options and $options->{nosanitize}) ? $as_xml : 0; >+ my $reset_index = ($options and $options->{reset_index}) ? 1 : 0; >+ my $noshadow = ($options and $options->{noshadow}) ? " -n " : ""; >+ my $record_format = ($options and $options->{record_format}) >+ ? $options->{record_format} >+ : ($as_xml) >+ ? 'marcxml' >+ : 'iso2709'; >+ my $zebraidx_log_opt = ($options and defined $options->{zebraidx_log_opt}) >+ ? $options->{zebraidx_log_opt} >+ : ""; >+ my $verbose = ($options and $options->{verbose}) ? 1 : 0; >+ my $skip_export = ($options and $options->{skip_export}) ? 1 : 0; >+ my $keep_export = ($options and $options->{keep_export}) ? 1 : 0; >+ my $directory = ($options and $options->{directory}) >+ ? $options->{directory} >+ : tempdir(CLEANUP => ($keep_export ? 0 : 1)); >+ >+ my @recordids = (ref $recordids eq "ARRAY") ? @$recordids : split (/ /, $recordids); >+ @recordids = map { { biblio_auth_number => $_ } } @recordids; >+ >+ my $num_exported = 0; >+ unless ($skip_export) { >+ $num_exported = _export_marc_records_from_list($recordtype, \@recordids, >+ $directory, $as_xml, $noxml, $nosanitize, $verbose); >+ } >+ _do_indexing($recordtype, "update", $directory, $reset_index, $noshadow, >+ $record_format, $zebraidx_log_opt); >+ >+ return $num_exported; >+} >+ >+sub DeleteRecordIndex { >+ my ($recordtype, $recordids, $options) = @_; >+ >+ my $as_xml = ($options and $options->{as_xml}) ? 1 : 0; >+ my $noshadow = ($options and $options->{noshadow}) ? " -n " : ""; >+ my $record_format = ($options and $options->{record_format}) >+ ? $options->{record_format} >+ : ($as_xml) >+ ? 'marcxml' >+ : 'iso2709'; >+ my $zebraidx_log_opt = ($options and defined $options->{zebraidx_log_opt}) >+ ? $options->{zebraidx_log_opt} >+ : ""; >+ my $verbose = ($options and $options->{verbose}) ? 1 : 0; >+ my $skip_export = ($options and $options->{skip_export}) ? 1 : 0; >+ my $keep_export = ($options and $options->{keep_export}) ? 1 : 0; >+ my $directory = ($options and $options->{directory}) >+ ? $options->{directory} >+ : tempdir(CLEANUP => ($keep_export ? 0 : 1)); >+ >+ my @recordids = (ref $recordids eq "ARRAY") ? @$recordids : split (/ /, $recordids); >+ @recordids = map { { biblio_auth_number => $_ } } @recordids; >+ >+ my $num_exported = 0; >+ unless ($skip_export) { >+ $num_exported = _generate_deleted_marc_records($recordtype, \@recordids, >+ $directory, $as_xml, $verbose); >+ } >+ _do_indexing($recordtype, "delete", $directory, 0, $noshadow, >+ $record_format, $zebraidx_log_opt); >+ >+ return $num_exported; >+} >+ >+ >+sub _export_marc_records_from_list { >+ my ( $record_type, $entries, $directory, $as_xml, $noxml, $nosanitize, $verbose ) = @_; >+ >+ my $num_exported = 0; >+ open my $fh, ">:encoding(UTF-8) ", "$directory/exported_records" or die $!; >+ if (_include_xml_wrapper($as_xml, $record_type)) { >+ # include XML declaration and root element >+ print {$fh} '<?xml version="1.0" encoding="UTF-8"?><collection>'; >+ } >+ my $i = 0; >+ my %found = (); >+ my ($itemtag) = GetMarcFromKohaField("items.itemnumber",''); >+ foreach my $record_number ( >+ map { $_->{biblio_auth_number} } >+ grep { !$found{ $_->{biblio_auth_number} }++ } @$entries >+ ) { >+ if($nosanitize) { >+ my $marcxml = $record_type eq 'biblio' >+ ? GetXmlBiblio($record_number) >+ : GetAuthorityXML($record_number); >+ if ($record_type eq 'biblio'){ >+ my @items = GetItemsInfo($record_number); >+ if (@items){ >+ my $record = MARC::Record->new; >+ $record->encoding('UTF-8'); >+ my @itemsrecord; >+ foreach my $item (@items){ >+ my $record = Item2Marc($item, $record_number); >+ push @itemsrecord, $record->field($itemtag); >+ } >+ $record->insert_fields_ordered(@itemsrecord); >+ my $itemsxml = $record->as_xml_record(); >+ $marcxml = >+ substr($marcxml, 0, length($marcxml)-10) . >+ substr($itemsxml, index($itemsxml, "</leader>\n", 0) + 10); >+ } >+ } >+ if ($marcxml) { >+ print {$fh} $marcxml; >+ $num_exported++; >+ } >+ } else { >+ my ($marc) = _get_corrected_marc_record( $record_type, $record_number, $noxml ); >+ if ( defined $marc ) { >+ eval { >+ my $rec; >+ if ($as_xml) { >+ $rec = $marc->as_xml_record(C4::Context->preference('marcflavour')); >+ $rec =~ s!<\?xml version="1.0" encoding="UTF-8"\?>\n!!; >+ } else { >+ $rec = $marc->as_usmarc(); >+ } >+ print {$fh} $rec; >+ $num_exported++; >+ }; >+ if ($@) { >+ warn "Error exporting record $record_number ($record_type) ".($noxml ? "not XML" : "XML"); >+ } >+ } >+ } >+ $i++; >+ if($verbose) { >+ print "."; >+ print "$i\n" unless($i % 100); >+ } >+ } >+ print {$fh} '</collection>' if (_include_xml_wrapper($as_xml, $record_type)); >+ close $fh; >+ print "\n$num_exported records exported to $directory/exported_records\n" if $verbose; >+ return $num_exported; >+} >+ >+sub _generate_deleted_marc_records { >+ my ( $record_type, $entries, $directory, $as_xml, $verbose ) = @_; >+ >+ my $num_exported = 0; >+ open my $fh, ">:encoding(UTF-8)", "$directory/exported_records" or die $!; >+ if (_include_xml_wrapper($as_xml, $record_type)) { >+ # include XML declaration and root element >+ print {$fh} '<?xml version="1.0" encoding="UTF-8"?><collection>'; >+ } >+ my $i = 0; >+ foreach my $record_number ( map { $_->{biblio_auth_number} } @$entries ) { >+ my $marc = MARC::Record->new(); >+ if ( $record_type eq 'biblio' ) { >+ _fix_biblio_ids( $marc, $record_number, $record_number ); >+ } else { >+ _fix_authority_id( $marc, $record_number ); >+ } >+ if ( C4::Context->preference("marcflavour") eq "UNIMARC" ) { >+ _fix_unimarc_100($marc); >+ } >+ >+ my $rec; >+ if ($as_xml) { >+ $rec = $marc->as_xml_record(C4::Context->preference('marcflavour')); >+ $rec =~ s!<\?xml version="1.0" encoding="UTF-8"\?>\n!!; >+ } else { >+ $rec = $marc->as_usmarc(); >+ } >+ print {$fh} $rec; >+ >+ $num_exported++; >+ $i++; >+ if($verbose) { >+ print "."; >+ print "$i\n" unless($i % 100); >+ } >+ } >+ print {$fh} '</collection>' if (_include_xml_wrapper($as_xml, $record_type)); >+ close $fh; >+ >+ return $num_exported; >+} >+ >+sub _get_corrected_marc_record { >+ my ( $record_type, $record_number, $noxml ) = @_; >+ >+ my $marc = _get_raw_marc_record( $record_type, $record_number, $noxml ); >+ >+ if ( defined $marc ) { >+ _fix_leader($marc); >+ if ( $record_type eq 'biblio' ) { >+ my $succeeded = _fix_biblio_ids( $marc, $record_number ); >+ return unless $succeeded; >+ } else { >+ _fix_authority_id( $marc, $record_number ); >+ } >+ if ( C4::Context->preference("marcflavour") eq "UNIMARC" ) { >+ _fix_unimarc_100($marc); >+ } >+ } >+ >+ return $marc; >+} >+ >+sub _get_raw_marc_record { >+ my ( $record_type, $record_number, $noxml ) = @_; >+ >+ my $dbh = C4::Context->dbh; >+ my $marc; >+ if ( $record_type eq 'biblio' ) { >+ if ($noxml) { >+ my $fetch_sth = $dbh->prepare_cached("SELECT marc FROM biblioitems >+ WHERE biblionumber = ?"); >+ $fetch_sth->execute($record_number); >+ if ( my ($blob) = $fetch_sth->fetchrow_array ) { >+ $marc = MARC::Record->new_from_usmarc($blob); >+ unless ($marc) { >+ warn "error creating MARC::Record from $blob"; >+ } >+ } >+ # failure to find a bib is not a problem - >+ # a delete could have been done before >+ # trying to process a record update >+ >+ $fetch_sth->finish(); >+ return unless $marc; >+ } else { >+ eval { $marc = GetMarcBiblio($record_number, 1); }; >+ if ($@ || !$marc) { >+ # here we do warn since catching an exception >+ # means that the bib was found but failed >+ # to be parsed >+ warn "error retrieving biblio $record_number: $@"; >+ return; >+ } >+ } >+ } else { >+ eval { $marc = C4::AuthoritiesMarc::GetAuthority($record_number); }; >+ if ($@) { >+ warn "error retrieving authority $record_number: $@"; >+ return; >+ } >+ } >+ return $marc; >+} >+ >+sub _fix_leader { >+ >+ # FIXME - this routine is suspect >+ # It blanks the Leader/00-05 and Leader/12-16 to >+ # force them to be recalculated correct when >+ # the $marc->as_usmarc() or $marc->as_xml() is called. >+ # But why is this necessary? It would be a serious bug >+ # in MARC::Record (definitely) and MARC::File::XML (arguably) >+ # if they are emitting incorrect leader values. >+ my $marc = shift; >+ >+ my $leader = $marc->leader; >+ substr( $leader, 0, 5 ) = ' '; >+ substr( $leader, 10, 7 ) = '22 '; >+ $marc->leader( substr( $leader, 0, 24 ) ); >+} >+ >+ >+sub _fix_biblio_ids { >+ >+ # FIXME - it is essential to ensure that the biblionumber is present, >+ # otherwise, Zebra will choke on the record. However, this >+ # logic belongs in the relevant C4::Biblio APIs. >+ my $marc = shift; >+ my $biblionumber = shift; >+ my $dbh = C4::Context->dbh; >+ my $biblioitemnumber; >+ if (@_) { >+ $biblioitemnumber = shift; >+ } else { >+ my $sth = $dbh->prepare( >+ "SELECT biblioitemnumber FROM biblioitems WHERE biblionumber=?"); >+ $sth->execute($biblionumber); >+ ($biblioitemnumber) = $sth->fetchrow_array; >+ $sth->finish; >+ unless ($biblioitemnumber) { >+ warn "failed to get biblioitemnumber for biblio $biblionumber"; >+ return 0; >+ } >+ } >+ >+ # FIXME - this is cheating on two levels >+ # 1. C4::Biblio::_koha_marc_update_bib_ids is meant to be an internal function >+ # 2. Making sure that the biblionumber and biblioitemnumber are correct and >+ # present in the MARC::Record object ought to be part of GetMarcBiblio. >+ # >+ # On the other hand, this better for now than what rebuild_zebra.pl used to >+ # do, which was duplicate the code for inserting the biblionumber >+ # and biblioitemnumber >+ C4::Biblio::_koha_marc_update_bib_ids( $marc, '', $biblionumber, $biblioitemnumber ); >+ >+ return 1; >+} >+ >+sub _fix_authority_id { >+ >+ # FIXME - as with fix_biblio_ids, the authid must be present >+ # for Zebra's sake. However, this really belongs >+ # in C4::AuthoritiesMarc. >+ my ( $marc, $authid ) = @_; >+ unless ( $marc->field('001') and $marc->field('001')->data() eq $authid ) { >+ $marc->delete_field( $marc->field('001') ); >+ $marc->insert_fields_ordered( MARC::Field->new( '001', $authid ) ); >+ } >+} >+ >+sub _fix_unimarc_100 { >+ >+ # FIXME - again, if this is necessary, it belongs in C4::AuthoritiesMarc. >+ my $marc = shift; >+ >+ my $string; >+ if ( length( $marc->subfield( 100, "a" ) ) == 36 ) { >+ $string = $marc->subfield( 100, "a" ); >+ my $f100 = $marc->field(100); >+ $marc->delete_field($f100); >+ } else { >+ $string = POSIX::strftime( "%Y%m%d", localtime ); >+ $string =~ s/\-//g; >+ $string = sprintf( "%-*s", 35, $string ); >+ } >+ substr( $string, 22, 6, "frey50" ); >+ unless ( length( $marc->subfield( 100, "a" ) ) == 36 ) { >+ $marc->delete_field( $marc->field(100) ); >+ $marc->insert_grouped_field( MARC::Field->new( 100, "", "", "a" => $string ) ); >+ } >+} >+ >+sub _do_indexing { >+ my ( $record_type, $op, $record_dir, $reset_index, $noshadow, $record_format, $zebraidx_log_opt ) = @_; >+ >+ my $zebra_server = ( $record_type eq 'biblio' ) ? 'biblioserver' : 'authorityserver'; >+ my $zebra_db_name = ( $record_type eq 'biblio' ) ? 'biblios' : 'authorities'; >+ my $zebra_config = C4::Context->zebraconfig($zebra_server)->{'config'}; >+ my $zebra_db_dir = C4::Context->zebraconfig($zebra_server)->{'directory'}; >+ >+ system("zebraidx -c $zebra_config $zebraidx_log_opt -g $record_format -d $zebra_db_name init") if $reset_index; >+ system("zebraidx -c $zebra_config $zebraidx_log_opt $noshadow -g $record_format -d $zebra_db_name $op $record_dir"); >+ system("zebraidx -c $zebra_config $zebraidx_log_opt -g $record_format -d $zebra_db_name commit") unless $noshadow; >+ >+} >+ >+sub _include_xml_wrapper { >+ my $as_xml = shift; >+ my $record_type = shift; >+ >+ my $bib_index_mode = C4::Context->config('zebra_bib_index_mode') || 'grs1'; >+ my $auth_index_mode = C4::Context->config('zebra_auth_index_mode') || 'dom'; >+ >+ return 0 unless $as_xml; >+ return 1 if $record_type eq 'biblio' and $bib_index_mode eq 'dom'; >+ return 1 if $record_type eq 'authority' and $auth_index_mode eq 'dom'; >+ return 0; >+ >+} >+ > > END { } # module clean-up code here (global destructor) > >diff --git a/misc/migration_tools/rebuild_zebra.pl b/misc/migration_tools/rebuild_zebra.pl >index 1621e84..9bfb370 100755 >--- a/misc/migration_tools/rebuild_zebra.pl >+++ b/misc/migration_tools/rebuild_zebra.pl >@@ -10,6 +10,7 @@ use File::Path; > use C4::Biblio; > use C4::AuthoritiesMarc; > use C4::Items; >+use File::Temp qw(tempdir); > > # > # script that checks zebradir structure & create directories & mandatory files if needed >@@ -155,12 +156,7 @@ if ( $verbose_logging ) { > } > if ($keep_export) { > print "NOTHING cleaned : the export $directory has been kept.\n"; >- print "You can re-run this script with the -s "; >- if ($use_tempdir) { >- print " and -d $directory parameters"; >- } else { >- print "parameter"; >- } >+ print "You can re-run this script with the -s and -d $directory parameters"; > print "\n"; > print "if you just want to rebuild zebra after changing the record.abs\n"; > print "or another zebra config file\n"; >@@ -195,8 +191,6 @@ sub check_zebra_dirs { > sub index_records { > my ($record_type, $directory, $skip_export, $skip_index, $process_zebraqueue, $as_xml, $noxml, $nosanitize, $do_not_clear_zebraqueue, $verbose_logging, $zebraidx_log_opt, $server_dir) = @_; > >- my $num_records_exported = 0; >- my $records_deleted; > my $need_reset = check_zebra_dirs($server_dir); > if ($need_reset) { > print "$0: found broken zebra server directories: forcing a rebuild\n"; >@@ -206,36 +200,8 @@ sub index_records { > print "====================\n"; > print "SKIPPING $record_type export\n"; > print "====================\n"; >- } else { >- if ( $verbose_logging ) { >- print "====================\n"; >- print "exporting $record_type\n"; >- print "====================\n"; >- } >- mkdir "$directory" unless (-d $directory); >- mkdir "$directory/$record_type" unless (-d "$directory/$record_type"); >- if ($process_zebraqueue) { >- my $entries = select_zebraqueue_records($record_type, 'deleted'); >- mkdir "$directory/del_$record_type" unless (-d "$directory/del_$record_type"); >- $records_deleted = generate_deleted_marc_records($record_type, $entries, "$directory/del_$record_type", $as_xml); >- mark_zebraqueue_batch_done($entries); >- $entries = select_zebraqueue_records($record_type, 'updated'); >- mkdir "$directory/upd_$record_type" unless (-d "$directory/upd_$record_type"); >- $num_records_exported = export_marc_records_from_list($record_type, >- $entries, "$directory/upd_$record_type", $as_xml, $noxml, $records_deleted); >- mark_zebraqueue_batch_done($entries); >- } else { >- my $sth = select_all_records($record_type); >- $num_records_exported = export_marc_records_from_sth($record_type, $sth, "$directory/$record_type", $as_xml, $noxml, $nosanitize); >- unless ($do_not_clear_zebraqueue) { >- mark_all_zebraqueue_done($record_type); >- } >- } > } > >- # >- # and reindexing everything >- # > if ($skip_index) { > if ($verbose_logging) { > print "====================\n"; >@@ -250,13 +216,52 @@ sub index_records { > } > my $record_fmt = ($as_xml) ? 'marcxml' : 'iso2709' ; > if ($process_zebraqueue) { >- do_indexing($record_type, 'delete', "$directory/del_$record_type", $reset, $noshadow, $record_fmt, $zebraidx_log_opt) >- if %$records_deleted; >- do_indexing($record_type, 'update', "$directory/upd_$record_type", $reset, $noshadow, $record_fmt, $zebraidx_log_opt) >- if $num_records_exported; >+ # Process 'deleted' records >+ my $entries = select_zebraqueue_records( $record_type, 'deleted' ); >+ my @entries_to_delete = map { $_->{biblio_auth_id} } @$entries; >+ >+ C4::Search::DeleteRecordIndex($record_type, \@entries_to_delete, { >+ as_xml => $as_xml, noshadow => $noshadow, record_format => $record_fmt, >+ zebraidx_log_opt => $zebraidx_log_opt, verbose => $verbose_logging, >+ skip_export => $skip_export, keep_export => $keep_export, >+ directory => $directory >+ }); >+ >+ mark_zebraqueue_batch_done($entries); >+ >+ # Process 'updated' records' >+ $entries = select_zebraqueue_records( $record_type, 'updated' ); >+ my @entries_to_update = map { >+ my $id = $_->{biblio_auth_id}; >+ # do not try to update deleted records >+ (0 == grep { $id == $_ } @entries_to_delete) ? $id : () >+ } @$entries; >+ >+ C4::Search::IndexRecord($record_type, \@entries_to_update, { >+ as_xml => $as_xml, noxml => $noxml, reset_index => $reset, >+ noshadow => $noshadow, record_format => $record_fmt, >+ zebraidx_log_opt => $zebraidx_log_opt, verbose => $verbose_logging, >+ skip_export => $skip_export, keep_export => $keep_export, >+ directory => $directory >+ }); >+ >+ mark_zebraqueue_batch_done($entries); > } else { >- do_indexing($record_type, 'update', "$directory/$record_type", $reset, $noshadow, $record_fmt, $zebraidx_log_opt) >- if ($num_records_exported or $skip_export); >+ my $sth = select_all_records($record_type); >+ my $entries = $sth->fetchall_arrayref([]); >+ my @entries_to_update = map { $_->[0] } @$entries; >+ >+ C4::Search::IndexRecord($record_type, \@entries_to_update, { >+ as_xml => $as_xml, noxml => $noxml, reset_index => $reset, >+ noshadow => $noshadow, record_format => $record_fmt, >+ zebraidx_log_opt => $zebraidx_log_opt, nosanitize => $nosanitize, >+ verbose => $verbose_logging, skip_export => $skip_export, >+ keep_export => $keep_export, directory => $directory >+ }); >+ >+ unless ($do_not_clear_zebraqueue) { >+ mark_all_zebraqueue_done($record_type); >+ } > } > } > } >@@ -326,322 +331,6 @@ sub select_all_biblios { > return $sth; > } > >-sub include_xml_wrapper { >- my $as_xml = shift; >- my $record_type = shift; >- >- return 0 unless $as_xml; >- return 1 if $record_type eq 'biblio' and $bib_index_mode eq 'dom'; >- return 1 if $record_type eq 'authority' and $auth_index_mode eq 'dom'; >- return 0; >- >-} >- >-sub export_marc_records_from_sth { >- my ($record_type, $sth, $directory, $as_xml, $noxml, $nosanitize) = @_; >- >- my $num_exported = 0; >- open my $fh, '>:encoding(UTF-8) ', "$directory/exported_records" or die $!; >- if (include_xml_wrapper($as_xml, $record_type)) { >- # include XML declaration and root element >- print {$fh} '<?xml version="1.0" encoding="UTF-8"?><collection>'; >- } >- my $i = 0; >- my ( $itemtag, $itemsubfield ) = GetMarcFromKohaField("items.itemnumber",''); >- while (my ($record_number) = $sth->fetchrow_array) { >- print "." if ( $verbose_logging ); >- print "\r$i" unless ($i++ %100 or !$verbose_logging); >- if ( $nosanitize ) { >- my $marcxml = $record_type eq 'biblio' >- ? GetXmlBiblio( $record_number ) >- : GetAuthorityXML( $record_number ); >- if ($record_type eq 'biblio'){ >- my @items = GetItemsInfo($record_number); >- if (@items){ >- my $record = MARC::Record->new; >- $record->encoding('UTF-8'); >- my @itemsrecord; >- foreach my $item (@items){ >- my $record = Item2Marc($item, $record_number); >- push @itemsrecord, $record->field($itemtag); >- } >- $record->insert_fields_ordered(@itemsrecord); >- my $itemsxml = $record->as_xml_record(); >- $marcxml = >- substr($marcxml, 0, length($marcxml)-10) . >- substr($itemsxml, index($itemsxml, "</leader>\n", 0) + 10); >- } >- } >- if ( $marcxml ) { >- print {$fh} $marcxml if $marcxml; >- $num_exported++; >- } >- next; >- } >- my ($marc) = get_corrected_marc_record($record_type, $record_number, $noxml); >- if (defined $marc) { >- eval { >- my $rec; >- if ($as_xml) { >- $rec = $marc->as_xml_record(C4::Context->preference('marcflavour')); >- $rec =~ s!<\?xml version="1.0" encoding="UTF-8"\?>\n!!; >- } else { >- $rec = $marc->as_usmarc(); >- } >- print {$fh} $rec; >- $num_exported++; >- }; >- if ($@) { >- warn "Error exporting record $record_number ($record_type) ".($noxml ? "not XML" : "XML"); >- } >- } >- } >- print "\nRecords exported: $num_exported\n" if ( $verbose_logging ); >- print {$fh} '</collection>' if (include_xml_wrapper($as_xml, $record_type)); >- close $fh; >- return $num_exported; >-} >- >-sub export_marc_records_from_list { >- my ($record_type, $entries, $directory, $as_xml, $noxml, $records_deleted) = @_; >- >- my $num_exported = 0; >- open my $fh, '>:encoding(UTF-8)', "$directory/exported_records" or die $!; >- if (include_xml_wrapper($as_xml, $record_type)) { >- # include XML declaration and root element >- print {$fh} '<?xml version="1.0" encoding="UTF-8"?><collection>'; >- } >- my $i = 0; >- >- # Skip any deleted records. We check for this anyway, but this reduces error spam >- my %found = %$records_deleted; >- foreach my $record_number ( map { $_->{biblio_auth_number} } >- grep { !$found{ $_->{biblio_auth_number} }++ } >- @$entries ) { >- print "." if ( $verbose_logging ); >- print "\r$i" unless ($i++ %100 or !$verbose_logging); >- my ($marc) = get_corrected_marc_record($record_type, $record_number, $noxml); >- if (defined $marc) { >- eval { >- my $rec; >- if ($as_xml) { >- $rec = $marc->as_xml_record(C4::Context->preference('marcflavour')); >- $rec =~ s!<\?xml version="1.0" encoding="UTF-8"\?>\n!!; >- } else { >- $rec = $marc->as_usmarc(); >- } >- print {$fh} $rec; >- $num_exported++; >- }; >- if ($@) { >- warn "Error exporting record $record_number ($record_type) ".($noxml ? "not XML" : "XML"); >- } >- $num_exported++; >- } >- } >- print "\nRecords exported: $num_exported\n" if ( $verbose_logging ); >- print {$fh} '</collection>' if (include_xml_wrapper($as_xml, $record_type)); >- close $fh; >- return $num_exported; >-} >- >-sub generate_deleted_marc_records { >- my ($record_type, $entries, $directory, $as_xml) = @_; >- >- my $records_deleted = {}; >- open my $fh, '>:encoding(UTF-8)', "$directory/exported_records" or die $!; >- if (include_xml_wrapper($as_xml, $record_type)) { >- # include XML declaration and root element >- print {$fh} '<?xml version="1.0" encoding="UTF-8"?><collection>'; >- } >- my $i = 0; >- foreach my $record_number (map { $_->{biblio_auth_number} } @$entries ) { >- print "\r$i" unless ($i++ %100 or !$verbose_logging); >- print "." if ( $verbose_logging ); >- >- my $marc = MARC::Record->new(); >- if ($record_type eq 'biblio') { >- fix_biblio_ids($marc, $record_number, $record_number); >- } else { >- fix_authority_id($marc, $record_number); >- } >- if (C4::Context->preference("marcflavour") eq "UNIMARC") { >- fix_unimarc_100($marc); >- } >- >- my $rec; >- if ($as_xml) { >- $rec = $marc->as_xml_record(C4::Context->preference('marcflavour')); >- $rec =~ s!<\?xml version="1.0" encoding="UTF-8"\?>\n!!; >- } else { >- $rec = $marc->as_usmarc(); >- } >- print {$fh} $rec; >- >- $records_deleted->{$record_number} = 1; >- } >- print "\nRecords exported: $i\n" if ( $verbose_logging ); >- print {$fh} '</collection>' if (include_xml_wrapper($as_xml, $record_type)); >- close $fh; >- return $records_deleted; >- >- >-} >- >-sub get_corrected_marc_record { >- my ($record_type, $record_number, $noxml) = @_; >- >- my $marc = get_raw_marc_record($record_type, $record_number, $noxml); >- >- if (defined $marc) { >- fix_leader($marc); >- if ($record_type eq 'authority') { >- fix_authority_id($marc, $record_number); >- } >- if (C4::Context->preference("marcflavour") eq "UNIMARC") { >- fix_unimarc_100($marc); >- } >- } >- >- return $marc; >-} >- >-sub get_raw_marc_record { >- my ($record_type, $record_number, $noxml) = @_; >- >- my $marc; >- if ($record_type eq 'biblio') { >- if ($noxml) { >- my $fetch_sth = $dbh->prepare_cached("SELECT marc FROM biblioitems WHERE biblionumber = ?"); >- $fetch_sth->execute($record_number); >- if (my ($blob) = $fetch_sth->fetchrow_array) { >- $marc = MARC::Record->new_from_usmarc($blob); >- unless ($marc) { >- warn "error creating MARC::Record from $blob"; >- } >- } >- # failure to find a bib is not a problem - >- # a delete could have been done before >- # trying to process a record update >- >- $fetch_sth->finish(); >- return unless $marc; >- } else { >- eval { $marc = GetMarcBiblio($record_number, 1); }; >- if ($@ || !$marc) { >- # here we do warn since catching an exception >- # means that the bib was found but failed >- # to be parsed >- warn "error retrieving biblio $record_number"; >- return; >- } >- } >- } else { >- eval { $marc = GetAuthority($record_number); }; >- if ($@) { >- warn "error retrieving authority $record_number"; >- return; >- } >- } >- return $marc; >-} >- >-sub fix_leader { >- # FIXME - this routine is suspect >- # It blanks the Leader/00-05 and Leader/12-16 to >- # force them to be recalculated correct when >- # the $marc->as_usmarc() or $marc->as_xml() is called. >- # But why is this necessary? It would be a serious bug >- # in MARC::Record (definitely) and MARC::File::XML (arguably) >- # if they are emitting incorrect leader values. >- my $marc = shift; >- >- my $leader = $marc->leader; >- substr($leader, 0, 5) = ' '; >- substr($leader, 10, 7) = '22 '; >- $marc->leader(substr($leader, 0, 24)); >-} >- >-sub fix_biblio_ids { >- # FIXME - it is essential to ensure that the biblionumber is present, >- # otherwise, Zebra will choke on the record. However, this >- # logic belongs in the relevant C4::Biblio APIs. >- my $marc = shift; >- my $biblionumber = shift; >- my $biblioitemnumber; >- if (@_) { >- $biblioitemnumber = shift; >- } else { >- my $sth = $dbh->prepare( >- "SELECT biblioitemnumber FROM biblioitems WHERE biblionumber=?"); >- $sth->execute($biblionumber); >- ($biblioitemnumber) = $sth->fetchrow_array; >- $sth->finish; >- unless ($biblioitemnumber) { >- warn "failed to get biblioitemnumber for biblio $biblionumber"; >- return 0; >- } >- } >- >- # FIXME - this is cheating on two levels >- # 1. C4::Biblio::_koha_marc_update_bib_ids is meant to be an internal function >- # 2. Making sure that the biblionumber and biblioitemnumber are correct and >- # present in the MARC::Record object ought to be part of GetMarcBiblio. >- # >- # On the other hand, this better for now than what rebuild_zebra.pl used to >- # do, which was duplicate the code for inserting the biblionumber >- # and biblioitemnumber >- C4::Biblio::_koha_marc_update_bib_ids($marc, '', $biblionumber, $biblioitemnumber); >- >- return 1; >-} >- >-sub fix_authority_id { >- # FIXME - as with fix_biblio_ids, the authid must be present >- # for Zebra's sake. However, this really belongs >- # in C4::AuthoritiesMarc. >- my ($marc, $authid) = @_; >- unless ($marc->field('001') and $marc->field('001')->data() eq $authid){ >- $marc->delete_field($marc->field('001')); >- $marc->insert_fields_ordered(MARC::Field->new('001',$authid)); >- } >-} >- >-sub fix_unimarc_100 { >- # FIXME - again, if this is necessary, it belongs in C4::AuthoritiesMarc. >- my $marc = shift; >- >- my $string; >- if ( length($marc->subfield( 100, "a" )) == 36 ) { >- $string = $marc->subfield( 100, "a" ); >- my $f100 = $marc->field(100); >- $marc->delete_field($f100); >- } >- else { >- $string = POSIX::strftime( "%Y%m%d", localtime ); >- $string =~ s/\-//g; >- $string = sprintf( "%-*s", 35, $string ); >- } >- substr( $string, 22, 6, "frey50" ); >- unless ( length($marc->subfield( 100, "a" )) == 36 ) { >- $marc->delete_field($marc->field(100)); >- $marc->insert_grouped_field(MARC::Field->new( 100, "", "", "a" => $string )); >- } >-} >- >-sub do_indexing { >- my ($record_type, $op, $record_dir, $reset_index, $noshadow, $record_format, $zebraidx_log_opt) = @_; >- >- my $zebra_server = ($record_type eq 'biblio') ? 'biblioserver' : 'authorityserver'; >- my $zebra_db_name = ($record_type eq 'biblio') ? 'biblios' : 'authorities'; >- my $zebra_config = C4::Context->zebraconfig($zebra_server)->{'config'}; >- my $zebra_db_dir = C4::Context->zebraconfig($zebra_server)->{'directory'}; >- >- system("zebraidx -c $zebra_config $zebraidx_log_opt -g $record_format -d $zebra_db_name init") if $reset_index; >- system("zebraidx -c $zebra_config $zebraidx_log_opt $noshadow -g $record_format -d $zebra_db_name $op $record_dir"); >- system("zebraidx -c $zebra_config $zebraidx_log_opt -g $record_format -d $zebra_db_name commit") unless $noshadow; >- >-} > > sub print_usage { > print <<_USAGE_; >-- >1.7.10.4 >
<b>You cannot view the attachment while viewing its details because your browser does not support IFRAMEs. <a href="attachment.cgi?id=11161">View the attachment on a separate page</a>.</b>
View Attachment As Diff
View Attachment As Raw
Actions:
View
|
Diff
|
Splinter Review
Attachments on
bug 7419
:
11161
|
11162
|
11412
|
11413
|
11486
|
11487
|
11488
|
13073
|
13074
|
13629
|
13630
|
17790
|
17791
|
17792
|
18712
|
18713
|
18714
|
20783
|
20784
|
20785
|
22150
|
22907
|
25087
|
25088
|
25089
|
35886
|
35887
|
35888
|
35889
|
42972