From 9629f1b4347cebc0af040e89dde5c739e5edd3d9 Mon Sep 17 00:00:00 2001
From: Shawn Boyette <sboyette@esilibrary.com>
Date: Mon, 10 Nov 2008 17:25:52 +0000
Subject: [PATCH] miker_filter integration work finally proceeding

---
 filter_record.ids |  114 ++++++++++++++++++++++++++++++-----------------------
 1 files changed, 65 insertions(+), 49 deletions(-)

diff --git a/filter_record.ids b/filter_record.ids
index 99c6171..1f2b754 100644
--- a/filter_record.ids
+++ b/filter_record.ids
@@ -3,53 +3,43 @@ use warnings;
 use strict;
 
 use Getopt::Long;
-use Time::HiRes qw/time/;
-use MARC::Record;
-use MARC::File::XML ( BinaryEncoding => 'utf-8' );
-
-# THIS FILE EXTRACTS NONMATCHING RECORDS
+#use MARC::Record;
+#use MARC::File::XML ( BinaryEncoding => 'utf-8' );
 
 # configuration hashref
-my $conf  = ();
-#initialize($conf);
-
-my $idfile = shift;
-my $marcfile = shift;
-my $import = shift;
-my $shelve = shift;
+my $conf  = {};
+initialize($conf);
 
 my %id;
 
-open F, "<$idfile";
+open F, "<", $conf->{idfile};
 while (<F>) {
 	chomp;
 	$id{$_} = 1;
 }
-
 close F;
 
 my $M; my $I; my $S;
-open $M, '<:utf8', $marcfile;
-open $I, '>:utf8', $import;
-open $S, '>:utf8', $shelve;
-
-my $starttime = time;
-my $count = 0;
-my $icount = 0;
-my $scount = 0;
+open $M, '<:utf8', $conf->{marcfile};
+open $I, '>:utf8', $conf->{'output-import'};
+open $S, '>:utf8', $conf->{'output-shelve'};
+
 while (<$M>) {
-    /tag="903" ind1=" " ind2=" ">.*?<subfield code="a">(\d+)</;
-    if ( $id{$1} ) {
-        print $S $_;
-        $scount++;
+    my $tag = $conf->{tag};
+    my $sub = $conf->{subfield};
+
+    /tag="$tag" ind1=" " ind2=" ">.*?<subfield code="$sub">(\d+)</;
+    if ($conf->{incoming}) {
+        print $S $_ if ($id{$1});
+        print $I $_ unless ($id{$1});;
     } else {
-        print $I $_;
-        $icount++;
+        print $S $_ unless ($id{$1});
+        print $I $_ if ($id{$1});;
     }
-    $count++;
+    $conf->{count}++;
 
-    unless ($count && $count % 100) {
-        print STDERR "\r$count\t(shelved: $scount, import: $icount)\t". $count / (time - $starttime);
+    unless ($conf->{count} % 100) {
+        print STDERR "\rProcessed: ",$conf->{count};
     }
 }
 
@@ -70,11 +60,12 @@ sub initialize {
     my $rc = GetOptions( $c,
                          'incoming',
                          'incumbent',
-                         'incoming-tag|incot=i',
-                         'incoming-subfield|incos=s',
-                         'incumbent-tag|incut=i',
-                         'incumbent-subfield|incus=s',
-                         'output|o=s',
+                         'tag|t=i',
+                         'subfield|s=s',
+                         'idfile|i=s',
+                         'marcfile|m=s',
+                         'outputimport|oi=s',
+                         'outputshelved|os=s',
                          'help|h',
                        );
     show_help() unless $rc;
@@ -82,15 +73,18 @@ sub initialize {
 
     $c->{'incoming-tag'}         = 903;
     $c->{'incoming-subfield'}    = 'a';
-    $c->{'incoming-matchfile'}   = '';
-    $c->{'incoming-nomatchfile'} = '';
     $c->{'incumbent-tag'}         = 901;
-    $c->{'incumbent-subfield'}    = 'a';
-    $c->{'incumbent-matchfile'}   = '';
-    $c->{'incumbent-nomatchfile'} = '';
+    $c->{'incumbent-subfield'}    = 'c';
     my @keys = keys %{$c};
-    show_help() unless (@ARGV and @keys);
-    for my $key ('renumber-from', 'tag', 'subfield', 'output')
+    unless ($c->{incoming} or $c->{incumbent}) {
+        print "One of --incoming or --incumbent is required.\n";
+        show_help();
+    }
+    if ($c->{incoming} and $c->{incumbent}) {
+        print "Only one of --incoming or --incumbent can be specified.\n";
+        show_help();
+    }
+    for my $key ('idfile', 'marcfile', 'output-import', 'output-shelved')
       { push @missing, $key unless $c->{$key} }
     if (@missing) {
         print "Required option: ", join(', ', @missing), " missing!\n";
@@ -108,14 +102,36 @@ Display usage message when things go wrong
 
 sub show_help {
 print <<HELP;
+
+The purpose of this utility is to split a MARCXML file in twain,
+producing a set of records which will imported into Evergreen, and a
+set of records which will not.
+
 Usage is: $0 [REQUIRED ARGS]
 Req'd Arguments
-  --renumber-from=N        -rf First id# of new sequence
-  --tag=N                  -t  Which tag to use
-  --subfield=X             -s  Which subfield to use
-  --output=<file>          -o  Output filename
-
-Any number of input files may be specified; one output file will result.
+  --incoming  \\___ One (and only one) of these two must
+  --incumbent /    be specified
+
+  If --incoming is specified, the record ids in the file specified by
+  --idfile will be used as EXCLUSION data. That is, the given record
+  ids will be treated as records which match incumbent records and are
+  being compressed into existing data, and so WILL NOT be
+  imported. The --output-import file will contain records whose ids DO
+  NOT occur in --idfile; --output-shelve will contain the records
+  which DO occur.
+
+  If --incumbent is specified, the reverse occurs.
+
+  --idfile         -i   File of record ids to use as source for matchpoints
+  --marcfile       -m   MARCXML source file
+  --output-import  -oi  Output MARCXML file for records to be imported
+  --output-shelve  -os  Output MARCXML file for records to be ignored
+
+Optional Arguments
+  --tag            -t   MARC tag to use as matchpoint (default 903 for incoming,
+                        901 for incumbent)
+  --subfield       -s   Subfield of tag to use (default 'c' for incoming, 'a'
+                        for incumbent)
 HELP
 exit 1;
 }
-- 
1.7.2.5