Attachment 151672 Details for Bug 33404 – Bug 33404: Z3950SearchAuth: Save in UTF-8 encoding

[patch] Bug 33404: Z3950SearchAuth: Save in UTF-8 encoding

Bug-33404-Z3950SearchAuth-Save-in-UTF-8-encoding.patch (text/plain), 4.55 KB, created by Marcel de Rooy on 2023-05-25 12:27:30 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Marcel de Rooy

Created: 2023-05-25 12:27:30 UTC

Size: 4.55 KB

patch

obsolete

>From 097ac93bf6180e416fe5298a8c8c0d13bae364d9 Mon Sep 17 00:00:00 2001
>From: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
>Date: Wed, 24 May 2023 14:57:46 +0000
>Subject: [PATCH] Bug 33404: Z3950SearchAuth: Save in UTF-8 encoding
>Content-Type: text/plain; charset=utf-8
>
>Test plan:
>[1] If you have access to a Z3950 MARC8 auth server, search
>    for an authority record and import it.
>[2] If you have access to a Z3950 UTF8 auth server, search
>    for an authority record and import it.
>
>Signed-off-by: Nick Clemens <nick@bywatersolutions.com>
>
>Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
>---
> C4/Breeding.pm | 39 ++++++++++++++++++++++-----------------
> 1 file changed, 22 insertions(+), 17 deletions(-)
>
>diff --git a/C4/Breeding.pm b/C4/Breeding.pm
>index efc9029f48..45a0563dcc 100644
>--- a/C4/Breeding.pm
>+++ b/C4/Breeding.pm
>@@ -478,9 +478,6 @@ sub ImportBreedingAuth {
> 
>     my $controlnumber = $marcrecord->field('001')->data;
> 
>-    # Normalize the record so it doesn't have separated diacritics
>-    SetUTF8Flag($marcrecord);
>-
>     $searchbreeding->execute($controlnumber,$heading);
>     my ($breedingid) = $searchbreeding->fetchrow;
> 
>@@ -578,31 +575,34 @@ sub Z3950SearchAuth {
>                     for ($i = ($page-1)*20; $i < (($numresults < ($page*20)) ? $numresults : ($page*20)); $i++) {
>                         my $rec = $oResult[$k]->record($i);
>                         if ($rec) {
>+                            my $marcdata = $rec->raw();
>                             my $marcrecord;
>-                            my $marcdata;
>-                            $marcdata   = $rec->raw();
>-
>-                            my ($charset_result, $charset_errors);
>                             if( $servers[$k]->{servertype} eq 'sru' ) {
>-                                $marcrecord = MARC::Record->new_from_xml( $marcdata, 'UTF-8', $servers[$k]->{syntax} );
>-                                $marcrecord->encoding('UTF-8');
>+                                $marcrecord = eval { MARC::Record->new_from_xml( $marcdata, 'UTF-8', $servers[$k]->{syntax} ) };
>+                                if( !$marcrecord || $@ ) {
>+                                    _dump_conversion_error( $servers[$k]->{servername}, $marcdata, $@ );
>+                                    next; # skip this one
>+                                }
>                             } else {
>+                                my ($charset_result, $charset_errors);
>                                 ( $marcrecord, $charset_result, $charset_errors ) = MarcToUTF8Record( $marcdata, $marc_type, $encoding[$k] );
>+                                if( !$marcrecord || @$charset_errors ) {
>+                                    _dump_conversion_error( $servers[$k]->{servername}, $marcdata, $charset_result, $charset_errors );
>+                                    next; # skip this one
>+                                }
>                             }
>-                            my $heading;
>-                            my $heading_authtype_code;
>-                            $heading_authtype_code = GuessAuthTypeCode($marcrecord);
>-                            next if ( not defined $heading_authtype_code ) ;
>+                            $marcrecord->encoding('UTF-8');
>+                            SetUTF8Flag($marcrecord);
> 
>-                            $heading = GetAuthorizedHeading({ record => $marcrecord });
>-
>-                            my $breedingid = ImportBreedingAuth( $marcrecord, $serverhost[$k], $encoding[$k], $heading );
>+                            my $heading_authtype_code = GuessAuthTypeCode($marcrecord) or next;
>+                            my $heading = GetAuthorizedHeading({ record => $marcrecord });
>+                            my $breedingid = ImportBreedingAuth( $marcrecord, $serverhost[$k], 'UTF-8', $heading );
>                             my %row_data;
>                             $row_data{server}       = $servers[$k]->{'servername'};
>                             $row_data{breedingid}   = $breedingid;
>                             $row_data{heading}      = $heading;
>                             $row_data{authid}       = $authid;
>-                            $row_data{heading_code}      = $heading_authtype_code;
>+                            $row_data{heading_code} = $heading_authtype_code;
>                             push( @breeding_loop, \%row_data );
>                         }
>                         else {
>@@ -639,6 +639,11 @@ sub Z3950SearchAuth {
>     );
> }
> 
>+sub _dump_conversion_error {
>+    require Data::Dumper;
>+    warn Data::Dumper->new([ 'Z3950SearchAuth conversion error', @_ ])->Indent(0)->Dump;
>+}
>+
> 1;
> __END__
> 
>-- 
>2.30.2

Actions: View | Diff | Splinter Review

Attachments on bug 33404: 151510 | 151624 | 151636 | 151641 | 151644 | 151672 | 151673 | 151674 | 151868 | 151869 | 151870 | 153014 | 153015 | 153016