do not let MARC::Batch open MARC files

The version of MARC::Batch->new() distributed with version
2.0.0 of MARC::Record, if given a file name, will
open it using the ':utf8' layer.  This results in an
incorrect character conversion when processing records
in the MARC-8 character encoding.

To avoid this, batch jobs that use MARC::Batch now
open the file themselves, then pass the file handle
to MARC::Batch->new().

Signed-off-by: Joshua Ferraro <jmf@liblime.com>

diff --git a/misc/batchCompareMARCvsFrameworks.pl b/misc/batchCompareMARCvsFrameworks.pl
index 0b806f0..e255602 100755 (executable)
--- a/misc/batchCompareMARCvsFrameworks.pl
+++ b/misc/batchCompareMARCvsFrameworks.pl
@@ -17,6 +17,8 @@ use MARC::Record;
 use MARC::Batch;
 
 use Getopt::Long;
+use IO::File;
+
 my ( $input_marc_file,$number,$nowarning,$frameworkcode) = ('',0);
 my $version;
 GetOptions(
@@ -49,7 +51,8 @@ EOF
 die;
 }#/
 
-my $batch = MARC::Batch->new( 'USMARC', $input_marc_file );
+my $fh = IO::File->new($input_marc_file); # don't let MARC::Batch open the file, as it applies the ':utf8' IO layer
+my $batch = MARC::Batch->new( 'USMARC', $fh );
 $batch->warnings_off() unless $nowarning;
 $batch->strict_off() unless $nowarning;
 my $dbh=C4::Context->dbh;

diff --git a/misc/batchImportMARCWithBiblionumbers.pl b/misc/batchImportMARCWithBiblionumbers.pl
index 893d988..a1bf26e 100755 (executable)
--- a/misc/batchImportMARCWithBiblionumbers.pl
+++ b/misc/batchImportMARCWithBiblionumbers.pl
@@ -19,6 +19,8 @@ use MARC::File::XML;
 use MARC::Batch;
 use Time::HiRes qw(gettimeofday);
 use Getopt::Long;
+use IO::File;
+
 my  $input_marc_file = '';
 my ($version);
 GetOptions(
@@ -45,7 +47,8 @@ my $timeneeded;
 my $dbh = C4::Context->dbh;
 
 my $sth2=$dbh->prepare("update biblioitems  set marc=? where biblionumber=?");
-my $batch = MARC::Batch->new( 'USMARC', $input_marc_file );
+my $fh = IO::File->new($input_marc_file); # don't let MARC::Batch open the file, as it applies the ':utf8' IO layer
+my $batch = MARC::Batch->new( 'USMARC', $fh );
 $batch->warnings_off();
 $batch->strict_off();
 my ($tagfield,$biblionumtagsubfield) = &GetMarcFromKohaField("biblio.biblionumber","");

diff --git a/misc/migration_tools/bulkauthimport.pl b/misc/migration_tools/bulkauthimport.pl
index 85233ba..76aa6eb 100755 (executable)
--- a/misc/migration_tools/bulkauthimport.pl
+++ b/misc/migration_tools/bulkauthimport.pl
@@ -19,8 +19,9 @@ use C4::Context;
 use C4::Charset;
 use C4::AuthoritiesMarc;
 use Time::HiRes qw(gettimeofday);
-
 use Getopt::Long;
+use IO::File;
+
 my ( $input_marc_file, $number) = ('',0);
 my ($version, $delete, $test_parameter,$char_encoding, $verbose, $format, $commit);
 $| = 1;
@@ -72,6 +73,7 @@ my $marcFlavour = C4::Context->preference('marcflavour') || 'MARC21';
 $char_encoding = 'MARC21' unless ($char_encoding);
 print "CHAR : $char_encoding\n" if $verbose;
 my $starttime = gettimeofday;
+my $fh = IO::File->new($input_marc_file); # don't let MARC::Batch open the file, as it applies the ':utf8' IO layer
 my $batch;
 if ($format =~ /XML/i) {
     # ugly hack follows -- MARC::File::XML, when used by MARC::Batch,
@@ -84,9 +86,9 @@ if ($format =~ /XML/i) {
     #       extract the records, not using regexes to look
     #       for <record>.*</record>.
     $MARC::File::XML::_load_args{BinaryEncoding} = 'utf-8';
-    $batch = MARC::Batch->new( 'XML', $input_marc_file );
+    $batch = MARC::Batch->new( 'XML', $fh );
 } else {
-    $batch = MARC::Batch->new( 'USMARC', $input_marc_file );
+    $batch = MARC::Batch->new( 'USMARC', $fh );
 }
 $batch->warnings_off();
 $batch->strict_off();

diff --git a/misc/migration_tools/bulkmarcimport.pl b/misc/migration_tools/bulkmarcimport.pl
index d6b8d40..019167f 100755 (executable)
--- a/misc/migration_tools/bulkmarcimport.pl
+++ b/misc/migration_tools/bulkmarcimport.pl
@@ -24,9 +24,9 @@ use C4::Items;
 use Unicode::Normalize;
 use Time::HiRes qw(gettimeofday);
 use Getopt::Long;
-binmode(STDOUT, ":utf8");
+use IO::File;
 
-use Getopt::Long;
+binmode(STDOUT, ":utf8");
 
 my ( $input_marc_file, $number) = ('',0);
 my ($version, $delete, $test_parameter, $skip_marc8_conversion, $char_encoding, $verbose, $commit, $fk_off,$format);
@@ -107,6 +107,7 @@ my $marcFlavour = C4::Context->preference('marcflavour') || 'MARC21';
 print "Characteristic MARC flavour: $marcFlavour\n" if $verbose;
 my $starttime = gettimeofday;
 my $batch;
+my $fh = IO::File->new($input_marc_file); # don't let MARC::Batch open the file, as it applies the ':utf8' IO layer
 if ($format =~ /XML/i) {
     # ugly hack follows -- MARC::File::XML, when used by MARC::Batch,
     # appears to try to convert incoming XML records from MARC-8
@@ -118,9 +119,9 @@ if ($format =~ /XML/i) {
     #       extract the records, not using regexes to look
     #       for <record>.*</record>.
     $MARC::File::XML::_load_args{BinaryEncoding} = 'utf-8';
-    $batch = MARC::Batch->new( 'XML', $input_marc_file );
+    $batch = MARC::Batch->new( 'XML', $fh );
 } else {
-    $batch = MARC::Batch->new( 'USMARC', $input_marc_file );
+    $batch = MARC::Batch->new( 'USMARC', $fh );
 }
 $batch->warnings_off();
 $batch->strict_off();