X-Git-Url: http://git.equinoxoli.org/?p=migration-tools.git;a=blobdiff_plain;f=extract_holdings;h=58b757577e04980e9819dbca33734a7ab9f0fbfd;hp=fadcf8dfe47cacc854cb7851ecdf43a6e0d6d9cd;hb=76574ae0d301dd1441181e47540c3e274caf4609;hpb=3bd5dd6ddf3857cf8c54cfe10df9fef27e2478d0

diff --git a/extract_holdings b/extract_holdings
index fadcf8d..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,
@@ -47,13 +48,12 @@ sub extract_holdings {
     my $j = 0; # holdings counter
 
     while (  $m->{data}{recs}[$i] ) {
-        print HOLDINGS "BEGIN;\n\negid\thseq\t" unless $j;
+        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 ( $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++;
 
@@ -64,7 +64,7 @@ sub extract_holdings {
             # grab the unary mappings and slug 'em in
             for my $sub ( sort keys %{$rec->{tags}[$holdidx]{uni}} ) {
                 push @out, $rec->{tags}[$holdidx]{uni}{$sub};
-                print HOLDINGS "l_", $m->name($tagid, $sub),"\t" unless $j;
+                print HOLDINGS "l_", $m->name($tagid, $sub),", " unless $j;
             }
 
             # handle holdings multis
@@ -80,9 +80,20 @@ 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 "l_", $m->name($rec->{tags}[$idx]{tag}, $sub), "\t" unless $j;
+                    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";
+                    }
                 }
             }
 
@@ -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 $j";
     }
     print "\n";
 }
@@ -189,8 +198,8 @@ sub initialize {
                          'sample|s',
                          'samplemap|sm=s',
                          'samplestr|ss=s',
-                         'marcfile|m=s',
-                         'map=s',
+                         'marcfile=s',
+                         'map|m=s',
                          'holdings|h=i',
                          'copyid|c=s',
                          'prefix|p=s',
@@ -205,6 +214,10 @@ sub initialize {
       if ($c->{map} and !($c->{holdings} and $c->{copyid}));
     show_version() if $c->{version};
 
+    if ($c->{prefix} and !$c->{marcfile}) {
+        $c->{marcfile} = $c->{prefix} . ".clean.marc.xml";
+    }
+
     my @keys = keys %{$c};
     for my $key ('prefix', 'marcfile')
       { push @missing, $key unless $c->{$key} }
@@ -225,7 +238,8 @@ 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
   --sample    -s   Generate a report of all tags in the MARC data
@@ -238,7 +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