X-Git-Url: http://git.rot13.org/?a=blobdiff_plain;f=C4%2FCharset.pm;h=f8ddd63627fb3fb817deec662d73df8e2dab0157;hb=568f32606c2c9c247b2b477193a2d6814f738fa6;hp=712bd85ca862d79fe0a16af0a19568df9b9893c8;hpb=f75f964ba40b7bdd1911842bc43b4df095aee583;p=koha.git

diff --git a/C4/Charset.pm b/C4/Charset.pm
index 712bd85ca8..f8ddd63627 100644
--- a/C4/Charset.pm
+++ b/C4/Charset.pm
@@ -29,7 +29,7 @@ use vars qw($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS);
 
 BEGIN {
     # set the version for version checking
-    $VERSION = 3.01;
+    $VERSION = 3.07.00.049;
     require Exporter;
     @ISA    = qw(Exporter);
     @EXPORT = qw(
@@ -39,6 +39,7 @@ BEGIN {
         SetUTF8Flag
         SetMarcUnicodeFlag
         StripNonXmlChars
+        nsb_clean
     );
 }
 
@@ -112,7 +113,7 @@ sub IsStringUTF8ish {
 
 =head2 SetUTF8Flag
 
-  my $marc_record = SetUTF8Flag($marc_record);
+  my $marc_record = SetUTF8Flag($marc_record, $nfd);
 
 This function sets the PERL UTF8 flag for data.
 It is required when using new_from_usmarc 
@@ -120,6 +121,8 @@ since MARC::File::USMARC does not handle PERL UTF8 setting.
 When editing unicode marc records fields and subfields, you
 would end up in double encoding without using this function. 
 
+If $nfd is set, string normalization will use NFD instead of NFC
+
 FIXME
 In my opinion, this function belongs to MARC::Record and not
 to this package.
@@ -128,13 +131,13 @@ But since it handles charset, and MARC::Record, it finds its way in that package
 =cut
 
 sub SetUTF8Flag{
-	my ($record)=@_;
+	my ($record, $nfd)=@_;
 	return unless ($record && $record->fields());
 	foreach my $field ($record->fields()){
 		if ($field->tag()>=10){
 			my @subfields;
 			foreach my $subfield ($field->subfields()){
-				push @subfields,($$subfield[0],NormalizeString($$subfield[1]));
+				push @subfields,($$subfield[0],NormalizeString($$subfield[1],$nfd));
 			}
 			my $newfield=MARC::Field->new(
 							$field->tag(),
@@ -248,20 +251,20 @@ sub MarcToUTF8Record {
     # If we do not know the source encoding, try some guesses
     # as follows:
     #   1. Record is UTF-8 already.
-    #   2. If MARC flavor is MARC21, then
+    #   2. If MARC flavor is MARC21 or NORMARC, then
     #      a. record is MARC-8
     #      b. record is ISO-8859-1
     #   3. If MARC flavor is UNIMARC, then
     if (not defined $source_encoding) {
         if ($marc_blob_is_utf8) {
-            # note that for MARC21 we are not bothering to check
+            # note that for MARC21/NORMARC we are not bothering to check
             # if the Leader/09 is set to 'a' or not -- because
             # of problems with various ILSs (including Koha in the
             # past, alas), this just is not trustworthy.
             SetMarcUnicodeFlag($marc_record, $marc_flavour);
             return $marc_record, 'UTF-8', [];
         } else {
-            if ($marc_flavour eq 'MARC21') {
+            if ($marc_flavour eq 'MARC21' || $marc_flavour eq 'NORMARC') {
                 return _default_marc21_charconv_to_utf8($marc_record, $marc_flavour);
             } elsif ($marc_flavour =~/UNIMARC/) {
                 return _default_unimarc_charconv_to_utf8($marc_record, $marc_flavour);
@@ -316,13 +319,15 @@ sub SetMarcUnicodeFlag {
     my $marc_flavour = shift; # || C4::Context->preference("marcflavour");
 
     $marc_record->encoding('UTF-8');
-    if ($marc_flavour eq 'MARC21') {
+    if ($marc_flavour eq 'MARC21' || $marc_flavour eq 'NORMARC') {
         my $leader = $marc_record->leader();
         substr($leader, 9, 1) = 'a';
         $marc_record->leader($leader); 
     } elsif ($marc_flavour =~/UNIMARC/) {
+	my $defaultlanguage = C4::Context->preference("UNIMARCField100Language");
+        $defaultlanguage = "fre" if (!$defaultlanguage || length($defaultlanguage) != 3);
         my $string; 
-		my ($subflength,$encodingposition)=($marc_flavour=~/AUTH/?(21,9):(36,22));
+		my ($subflength,$encodingposition)=($marc_flavour=~/AUTH/?(21,12):(36,25));
 		$string=$marc_record->subfield( 100, "a" );
         if (defined $string && length($string)==$subflength) { 
 			$string = substr $string, 0,$subflength if (length($string)>$subflength);
@@ -330,9 +335,10 @@ sub SetMarcUnicodeFlag {
         else { 
             $string = POSIX::strftime( "%Y%m%d", localtime ); 
             $string =~ s/\-//g; 
-            $string = sprintf( "%-*s", $subflength, $string ); 
+            $string = sprintf( "%-*s", $subflength, $string );
+	    substr ( $string, ($encodingposition - 3), 3, $defaultlanguage);
         } 
-        substr( $string, $encodingposition, 8, "frey50  " ); 
+        substr( $string, $encodingposition, 3, "y50" );
         if ( $marc_record->subfield( 100, "a" ) ) { 
 			$marc_record->field('100')->update(a=>$string);
 		}
@@ -340,7 +346,7 @@ sub SetMarcUnicodeFlag {
             $marc_record->insert_grouped_field( 
                 MARC::Field->new( 100, '', '', "a" => $string ) ); 
         }
-		$debug && warn "encodage: ", substr( $marc_record->subfield(100, 'a'), $encodingposition, 8 );
+		$debug && warn "encodage: ", substr( $marc_record->subfield(100, 'a'), $encodingposition, 3 );
     } else {
         warn "Unrecognized marcflavour: $marc_flavour";
     }
@@ -380,6 +386,40 @@ sub StripNonXmlChars {
     return $str;
 }
 
+
+
+=head2 nsb_clean
+
+=over 4
+
+nsb_clean($string);
+
+=back
+
+Removes Non Sorting Block characters
+
+=cut
+sub nsb_clean {
+    my $NSB  = '\x88' ;        # NSB : begin Non Sorting Block
+    my $NSE  = '\x89' ;        # NSE : Non Sorting Block end
+    my $NSB2 = '\x98' ;        # NSB : begin Non Sorting Block
+    my $NSE2 = '\x9C' ;        # NSE : Non Sorting Block end
+    my $C2   = '\xC2' ;        # What is this char ? It is sometimes left by the regexp after removing NSB / NSE
+
+    # handles non sorting blocks
+    my ($string) = @_ ;
+    $_ = $string ;
+    s/$NSB//g ;
+    s/$NSE//g ;
+    s/$NSB2//g ;
+    s/$NSE2//g ;
+    s/$C2//g ;
+    $string = $_ ;
+
+    return($string) ;
+}
+
+
 =head1 INTERNAL FUNCTIONS
 
 =head2 _default_marc21_charconv_to_utf8