X-Git-Url: http://git.equinoxoli.org/?p=migration-tools.git;a=blobdiff_plain;f=extract_holdings;h=58b757577e04980e9819dbca33734a7ab9f0fbfd;hp=051175a7cafb2f9f0bced983f4a31edfd7aae0f3;hb=76574ae0d301dd1441181e47540c3e274caf4609;hpb=32ca9a96e2ddb2823cbe90c525593bef3bfd2c76 diff --git a/extract_holdings b/extract_holdings index 051175a..58b7575 100755 --- a/extract_holdings +++ b/extract_holdings @@ -3,7 +3,7 @@ use strict; use warnings; use Getopt::Long; -use Equinox::Migration::MapDrivenMARCXMLProc 1.003; +use Equinox::Migration::MapDrivenMARCXMLProc 1.005; use Equinox::Migration::MARCXMLSampler; my $VERSION = '1.001'; @@ -22,12 +22,13 @@ $| = 1; # run samples if we've been asked for them run_samples($c) if ($c->{sample} or $c->{samplemap} or $c->{samplestr}); extract_holdings($c) if ($c->{map}); +print "\n"; #-------------------------- sub extract_holdings { my ($c) = @_; - print "Parsing records for extraction... "; + print "Parsing records for extraction:\n"; my $m = Equinox::Migration::MapDrivenMARCXMLProc->new( marcfile => $c->{marcfile}, mapfile => $c->{map}, verbose => 1, @@ -50,9 +51,9 @@ sub extract_holdings { print HOLDINGS "BEGIN;\n\negid, hseq, " unless $j; my $rec = $m->{data}{recs}[$i]; my $k = 0; # holding-within-record pointer - # for each holdings tag in the record... - while ( defined $rec->{tmap}{ $c->{holdings} }[$k] ) { - my $holdidx = $rec->{tmap}{ $c->{holdings} }[$k]; + + for my $holdidx ( @{ $rec->{tmap}{ $c->{holdings} } } ) { + # for each holdings tag in the record... my $tagid = $rec->{tags}[$holdidx]{tag}; $k++; @@ -79,11 +80,21 @@ sub extract_holdings { for my $othertag ( sort keys %{$rec->{tmap}} ) { next if $othertag eq $c->{holdings}; # ignoring the holdings, o'course my $idx = $rec->{tmap}{$othertag}[0]; # get index into tags struct + unless (defined $idx) { + push @out, ''; + next; + } for my $sub ( sort keys %{$rec->{tags}[$idx]{uni}} ) { push @out, $rec->{tags}[$idx]{uni}{$sub}; print HOLDINGS "l_", $m->name($rec->{tags}[$idx]{tag}, $sub), ", " unless $j; } + for my $sub ( sort keys %{$multis->{$othertag}} ) { + for my $value ( @{$rec->{tags}[$idx]{multi}{$sub}} ) { + my $fh = $MULTIFILE{"$othertag$sub"}; + print $fh join("\t", $rec->{egid}, $j, $value), "\n"; + } + } } # and dump it @@ -91,11 +102,9 @@ sub extract_holdings { print HOLDINGS join("\t", @out); print HOLDINGS "\n"; $j++; - - @out = undef; } $i++; - print "\r$i"; + print "\r$i $j"; } print "\n"; } @@ -189,8 +198,8 @@ sub initialize { 'sample|s', 'samplemap|sm=s', 'samplestr|ss=s', - 'marcfile|m=s', - 'map=s', + 'marcfile=s', + 'map|m=s', 'holdings|h=i', 'copyid|c=s', 'prefix|p=s', @@ -229,7 +238,7 @@ Usage is: extract_holdings -p PREFIX -m MARCFILE [ARGUMENTS] REQUIRED ARGUMENTS --prefix -p Prefix string for output filenames - --marcfile -m MARCXML to use as source data + --marcfile MARCXML to use as source data Defaults to 'PREFIX.clean.marc.xml' SAMPLING ARGUMENTS @@ -243,7 +252,7 @@ SAMPLING ARGUMENTS If --samplemap and --samplestr are both specified, --samplemap wins. HOLDINGS EXTRACTION ARGUMENTS - --map E::M::SM map file which will be used to extract holdings data + --map -m E::M::SM map file which will be used to extract holdings data from the input MARC file --holdings -h Specifies actual holdings tag --copyid -c Specifies subfield of holdings with unique copy identifier