X-Git-Url: http://git.equinoxoli.org/?p=migration-tools.git;a=blobdiff_plain;f=extract_holdings;h=27cf58a419ad361aec61220d80d5d432a53eb4da;hp=913aba124f1d9f0184b80f590403c525ef4b93f8;hb=7a654b50c2edbb1f746895a2e73e6038e59d6e93;hpb=e7e57c0546ad29123502edb73a1e59fa8c6aa4b3

diff --git a/extract_holdings b/extract_holdings
index 913aba1..27cf58a 100755
--- a/extract_holdings
+++ b/extract_holdings
@@ -3,11 +3,19 @@ use strict;
 use warnings;
 
 use Getopt::Long;
-use Equinox::Migration::MapDrivenMARCXMLProc;
+use Equinox::Migration::MapDrivenMARCXMLProc 1.003;
 use Equinox::Migration::MARCXMLSampler;
 
 my $VERSION = '1.001';
 
+=pod
+
+TODO
+
+  * Have detail mode report on number of subfields per datafield
+
+=cut
+
 my $c = initialize();
 $| = 1;
 
@@ -21,40 +29,49 @@ sub extract_holdings {
     my ($c) = @_;
     print "Parsing records for extraction... ";
     my $m = Equinox::Migration::MapDrivenMARCXMLProc->new( marcfile => $c->{marcfile},
-                                                           mapfile  => $c->{map} );
+                                                           mapfile  => $c->{map},
+                                                           verbose  => 1,
+                                                         );
+    print "Writing holdings to output file(s)...\n";
     # open main holdings file
     open HOLDINGS, '>', ($c->{prefix} . "-HOLDINGS.pg");
-    # open the files for multi mappings
-    # FIXME DO THIS
-    open X, '>', ($c->{prefix} . "-HOLDINGS-privnotes.pg");
-    open Z, '>', ($c->{prefix} . "-HOLDINGS-pubnotes.pg");
-    select HOLDINGS;
+    # create multi files
+    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 }
+    }
 
-    while (my $rec = $m->parse_record) {
+    my $i = 0; # record counter
+    my $j = 0; # holdings counter
+
+    while (  $m->{data}{recs}[$i] ) {
+        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} }} ) {
+        while ( defined $rec->{tmap}{ $c->{holdings} }[$k] ) {
+            my $holdidx = $rec->{tmap}{ $c->{holdings} }[$k];
             my $tagid = $rec->{tags}[$holdidx]{tag};
+            $k++;
 
-            print STDOUT "\r", $m->recno;
             my @out = ();            # clear the output buffer
             push @out, $rec->{egid}; # slug in the egid first thing
-            print "BEGIN;\n\negid\t" if ($m->recno == 1);
+            push @out, $j;           # holding seq goes next
 
             # 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 "l_", $m->name($tagid, $sub),"\t"
-                  if ($m->recno == 1);
-            }
-            for my $x (@{$rec->{tags}[$holdidx]{multi}{x}} ) {
-                print X $rec->{egid}, "\t",
-                  $rec->{tags}[$holdidx]{uni}{ $c->{copyid} },
-                    "\t$x\n";
+                print HOLDINGS "l_", $m->name($tagid, $sub),", " unless $j;
             }
-            for my $z (@{$rec->{tags}[$holdidx]{multi}{z}} ) {
-                print Z $rec->{egid}, "\t",
-                  $rec->{tags}[$holdidx]{uni}{ $c->{copyid} },
-                    "\t$z\n";
+
+            # handle holdings multis
+            for my $sub ( sort keys %{$multis->{$tagid}} ) {
+                for my $value ( @{$rec->{tags}[$holdidx]{multi}{$sub}} ) {
+                  my $fh = $MULTIFILE{"$tagid$sub"};
+                  print $fh join("\t", $rec->{egid}, $j, $value), "\n";
+              }
             }
 
 
@@ -64,19 +81,22 @@ sub extract_holdings {
                 my $idx = $rec->{tmap}{$othertag}[0]; # get index into tags struct
                 for my $sub ( sort keys %{$rec->{tags}[$idx]{uni}} ) {
                     push @out, $rec->{tags}[$idx]{uni}{$sub};
-                    print "l_", $m->name($rec->{tags}[$idx]{tag}, $sub), "\t"
-                      if ($m->recno == 1);
+                    print HOLDINGS "l_", $m->name($rec->{tags}[$idx]{tag}, $sub), ", "
+                      unless $j;
                 }
             }
 
             # and dump it
-            print "\n" if ($m->recno == 1);
-            print join("\t", @out);
-            print "\n";
-        }
+            print HOLDINGS "\n" unless $j;
+            print HOLDINGS join("\t", @out);
+            print HOLDINGS "\n";
+            $j++;
 
+            @out = undef;
+        }
+        $i++;
+        print "\r$i";
     }
-    select STDOUT;
     print "\n";
 }
 
@@ -85,6 +105,7 @@ sub extract_holdings {
 sub run_samples {
     my ($c) = @_;
     my $s;
+    print "Parsing records for sampling... ";
     if ($c->{samplemap}) {
         $s = Equinox::Migration::MARCXMLSampler->new( marcfile => $c->{marcfile},
                                                       mapfile  => $c->{samplemap});
@@ -94,8 +115,6 @@ sub run_samples {
     } else {
         $s = Equinox::Migration::MARCXMLSampler->new( marcfile => $c->{marcfile} );
     }
-    print "Parsing records for sampling... ";
-    $s->parse_records;
 
     dump_sample_overview($c, $s) if $c->{sample};
     dump_sample_detail($c, $s) if ($c->{samplemap} or $c->{samplestr});
@@ -105,20 +124,29 @@ sub dump_sample_detail {
     my ($c, $s) = @_;
     my $tags = $s->{data}{samp};
     my $count = $s->{data}{rcnt};
+    my $scnt  = $s->{data}{scnt};
 
     open DETAIL, '>', ($c->{prefix} . "-HOLDINGS-DETAIL.txt");
     select DETAIL;
-    for my $tag (sort keys %{$tags}) {
+    for my $tag (sort keys %{ $tags }) {
         print ">>>>> TAG $tag\n\n";
-        for my $subkey (sort keys %{$tags->{$tag}}) {
+        for my $subkey (sort keys %{ $tags->{$tag} }) {
             my $sub = $tags->{$tag}{$subkey};
             print "|| $subkey | ", $sub->{value}, " | ", 
               $sub->{count}, "/", $sub->{tcnt}, " |  ||\n";
         }
+        print "\n";
     }
-    print "\n";
-    select STDOUT;
     close DETAIL;
+    open SCOUNT, '>', ($c->{prefix} . "-HOLDINGS-SUBCOUNTS.txt");
+    select SCOUNT;
+    for my $tag (sort keys %{ $scnt }) {
+        print ">>>>> TAG $tag\n\n";
+        for my $len (sort keys %{ $scnt->{$tag} }) 
+          { print "|| $len | ", $scnt->{$tag}{$len}, " ||\n" }
+        print "\n";
+    }
+    select STDOUT;
 }
 
 sub dump_sample_overview {
@@ -174,7 +202,7 @@ sub initialize {
     show_help("Nothing to do!")
       unless ($c->{map} or $c->{sample} or $c->{samplemap} or $c->{samplestr});
     show_help("map, holdings, and copyid must be specified together!")
-      if ($c->{map} and !$c->{holdings} and !$c->{copyid});
+      if ($c->{map} and !($c->{holdings} and $c->{copyid}));
     show_version() if $c->{version};
 
     my @keys = keys %{$c};