X-Git-Url: http://git.equinoxoli.org/?p=migration-tools.git;a=blobdiff_plain;f=extract_holdings;h=58b757577e04980e9819dbca33734a7ab9f0fbfd;hp=05574163e69da994b93c6d3774e1bfc6f5a429ad;hb=76574ae0d301dd1441181e47540c3e274caf4609;hpb=31dfec235ae3eb0f0aada9a48fdf025449e3f290

diff --git a/extract_holdings b/extract_holdings
index 0557416..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  => $c->{map},
                                                            verbose  => 1,
@@ -36,20 +37,25 @@ sub extract_holdings {
     # open main holdings file
     open HOLDINGS, '>', ($c->{prefix} . "-HOLDINGS.pg");
     # create multi files
-    my @multifiles = <*HOLDINGS-MULT*>;
-    unlink @multifiles;
+    my $multis = $m->get_multis;
+    my %MULTIFILE = ();
+    for my $t ( keys %{$multis} ) {
+        for my $s ( keys %{$multis->{$t}}) 
+          { open my $fh, ">", ($c->{prefix} . "-HOLDINGS-MULT-$t$s.pg"); $MULTIFILE{"$t$s"} = $fh }
+    }
 
     my $i = 0; # record counter
     my $j = 0; # holdings counter
-    my $multis = $m->get_multis;
 
     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...
-        for my $holdidx ( @{$rec->{tmap}{ $c->{holdings} }} ) {
+        for my $holdidx ( @{ $rec->{tmap}{ $c->{holdings} } } ) {
+            # for each holdings tag in the record...
             my $tagid = $rec->{tags}[$holdidx]{tag};
+            $k++;
 
             my @out = ();            # clear the output buffer
             push @out, $rec->{egid}; # slug in the egid first thing
@@ -58,15 +64,15 @@ 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
             for my $sub ( sort keys %{$multis->{$tagid}} ) {
-                open MULT, ">>", ($c->{prefix} . "HOLDINGS-MULT" . "$tagid$sub.pg");
-                for my $value ( @{$rec->{tags}[$holdidx]{multi}{$sub}} ) 
-                  { print MULT join("\t", $rec->{egid}, $j, $value), "\n" }
-                close MULT;
+                for my $value ( @{$rec->{tags}[$holdidx]{multi}{$sub}} ) {
+                  my $fh = $MULTIFILE{"$tagid$sub"};
+                  print $fh join("\t", $rec->{egid}, $j, $value), "\n";
+              }
             }
 
 
@@ -74,11 +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 "l_", $m->name($rec->{tags}[$idx]{tag}, $sub), "\t"
+                    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
@@ -88,7 +104,7 @@ sub extract_holdings {
             $j++;
         }
         $i++;
-        print "\r$i";
+        print "\r$i $j";
     }
     print "\n";
 }
@@ -182,8 +198,8 @@ sub initialize {
                          'sample|s',
                          'samplemap|sm=s',
                          'samplestr|ss=s',
-                         'marcfile|m=s',
-                         'map=s',
+                         'marcfile=s',
+                         'map|m=s',
                          'holdings|h=i',
                          'copyid|c=s',
                          'prefix|p=s',
@@ -198,6 +214,10 @@ sub initialize {
       if ($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} }
@@ -218,7 +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
@@ -231,7 +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