Bug 10407: Add marcxml import (follow-up)

This patch makes the following changes:

[1] Based on the groundwork of the former patch, add call to
    RecordsFromMARCXMLFile in stage-marc-import. Use format param.
[2] Add format to the template. Use file extension to determine.
    If you use .xml or .marcxml as extension, MARCXML is selected.
[3] In stage-marc-import.tt mark UTF-8 encoding as UTF-8 not as utf8.
[4] BatchStageMarcRecords: do not call plugin if you have no records.
[5] RecordsFromISO2709File: also return errors in an array.
[6] In misc/stage_file.pl also use UTF-8. Handling of errors from [5].

Test plan:
[1] Import an empty file as MARC or MARCXML (with Tools/Stage..import).
[2] Import an non-empty file with invalid contents as MARC or MARCXML.
[3] Export a few records with Tools/Export as MARC and MARCXML.
[4] Import these two files. Check selected format versus file extension.
[5] Import a MARCXML file with misc/stage_file.pl.

Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>

Signed-off-by: Josef Moravec <josef.moravec@gmail.com>

Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>

Signed-off-by: Kyle M Hall <kyle@bywatersolutions.com>

diff --git a/C4/ImportBatch.pm b/C4/ImportBatch.pm
index 19853b3..51fcb76 100644 (file)
--- a/C4/ImportBatch.pm
+++ b/C4/ImportBatch.pm
@@ -405,7 +405,7 @@ sub BatchStageMarcRecords {
             method => 'to_marc',
             params => { data => $marc_records }
         }
-    ) if $to_marc_plugin;
+    ) if $to_marc_plugin && @$marc_records;
 
     my $marc_type = C4::Context->preference('marcflavour');
     $marc_type .= 'AUTH' if ($marc_type eq 'UNIMARC' && $record_type eq 'auth');
@@ -1492,11 +1492,13 @@ Reads ISO2709 binary porridge from the given file and creates MARC::Record-objec
 @PARAM2, String, see stage_file.pl
 @PARAM3, String, should be utf8
 
+Returns two array refs.
+
 =cut
 
 sub RecordsFromISO2709File {
     my ($input_file, $record_type, $encoding) = @_;
-    my $errors;
+    my @errors;
 
     my $marc_type = C4::Context->preference('marcflavour');
     $marc_type .= 'AUTH' if ($marc_type eq 'UNIMARC' && $record_type eq 'auth');
@@ -1512,40 +1514,37 @@ sub RecordsFromISO2709File {
         my ($marc_record, $charset_guessed, $char_errors) = MarcToUTF8Record($_, $marc_type, $encoding);
         push @marc_records, $marc_record;
         if ($charset_guessed ne $encoding) {
-            $errors = '' unless $errors;
-            $errors .= "Unexpected charset $charset_guessed, expecting $encoding\n";
+            push @errors,
+                "Unexpected charset $charset_guessed, expecting $encoding";
         }
     }
     close IN;
-    return ($errors, \@marc_records);
+    return ( \@errors, \@marc_records );
 }
 
 =head2 RecordsFromMARCXMLFile
 
     my ($errors, $records) = C4::ImportBatch::RecordsFromMARCXMLFile($input_file, $encoding);
 
-
-
 Creates MARC::Record-objects out of the given MARCXML-file.
 
 @PARAM1, String, absolute path to the ISO2709 file.
 @PARAM2, String, should be utf8
 
+Returns two array refs.
+
 =cut
 
 sub RecordsFromMARCXMLFile {
     my ( $filename, $encoding ) = @_;
     my $batch = MARC::File::XML->in( $filename );
-    my @marcRecords;
-    my @errors;
+    my ( @marcRecords, @errors, $record );
     do {
-        eval {
-            my $record = $batch->next($encoding);
-            push @marcRecords, $record if $record;
-        };
+        eval { $record = $batch->next( $encoding ); };
         if ($@) {
             push @errors, $@;
         }
+        push @marcRecords, $record if $record;
     } while( $record );
     return (\@errors, \@marcRecords);
 }

diff --git a/koha-tmpl/intranet-tmpl/prog/en/modules/tools/stage-marc-import.tt b/koha-tmpl/intranet-tmpl/prog/en/modules/tools/stage-marc-import.tt
index 45aeac9..4d57a13 100644 (file)
--- a/koha-tmpl/intranet-tmpl/prog/en/modules/tools/stage-marc-import.tt
+++ b/koha-tmpl/intranet-tmpl/prog/en/modules/tools/stage-marc-import.tt
@@ -66,6 +66,10 @@ function cbUpload( status, fileid ) {
         $('#fileuploadbutton').prop('disabled',true);
         $('#fileuploadbutton').show();
         $("#fileuploadcancel").hide();
+        var filename=$('#fileToUpload').prop('files')[0].name;
+        if( filename.match( new RegExp(/\.[^.]+xml$/) ) ) {
+            $('#format').val('MARCXML');
+        }
         $("#processfile").show();
     } else {
         $('#fileuploadbutton').show();
@@ -181,7 +185,7 @@ function cbUpload( status, fileid ) {
        <li>
                <label for="encoding">Character encoding: </label>
             <select name="encoding" id="encoding">
-                <option value="utf8" selected="selected">UTF-8 (Default)</option>
+                <option value="UTF-8" selected="selected">UTF-8 (Default)</option>
                 <option value="MARC-8">MARC 8</option>
                 <option value="ISO_5426">ISO 5426</option>
                 <option value="ISO_6937">ISO 6937</option>
@@ -189,6 +193,13 @@ function cbUpload( status, fileid ) {
                 <option value="EUC-KR">EUC-KR</option>
             </select>
        </li>
+    <li>
+        <label for='format'>Format:</label>
+        <select name='format' id='format'>
+            <option value='ISO2709'>MARC</option>
+            <option value='MARCXML'>MARCXML</option>
+        </select>
+    </li>
 </ol></fieldset>
 
   [% IF plugins %]

diff --git a/misc/stage_file.pl b/misc/stage_file.pl
index 95ac941..3e55dba 100755 (executable)
--- a/misc/stage_file.pl
+++ b/misc/stage_file.pl
@@ -37,7 +37,7 @@ $| = 1;
 
 # command-line parameters
 my $record_type = "biblio";
-my $encoding = "";
+my $encoding = "UTF-8";
 my $authorities = 0;
 my $match = 0;
 my $add_items = 0;
@@ -65,10 +65,6 @@ my $result = GetOptions(
 
 $record_type = 'auth' if ($authorities);
 
-if ($encoding eq "") {
-    $encoding = "utf8";
-}
-
 if (not $result or $input_file eq "" or $want_help) {
     print_usage();
     exit 0;
@@ -93,9 +89,10 @@ exit 0;
 sub process_batch {
     my ($format, $input_file, $record_type, $match, $add_items, $batch_comment) = @_;
 
-    my ($errors, $marc_records) = C4::ImportBatch::RecordsFromISO2709File($input_file, $record_type, $encoding) if $format eq 'ISO2709';
-    warn $errors if $errors;
-    $marc_records = C4::ImportBatch::RecordsFromMARCXMLFile($input_file, $encoding) if $format eq 'MARCXML';
+    my ( $errors, $marc_records );
+    ( $errors, $marc_records ) = C4::ImportBatch::RecordsFromISO2709File($input_file, $record_type, $encoding) if $format eq 'ISO2709';
+    ( $errors, $marc_records ) = C4::ImportBatch::RecordsFromMARCXMLFile($input_file, $encoding) if $format eq 'MARCXML';
+    warn ( join ',', @$errors ) if @$errors;
     my $num_input_records = ($marc_records) ? scalar(@$marc_records) : 0;
 
     print "... staging MARC records -- please wait\n";
@@ -176,7 +173,7 @@ records into the main Koha database.
 Parameters:
     --file <file_name>      name of input MARC bib file
     --authorities           stage authority records instead of bibs
-    --encoding <encoding>   encoding of MARC records, default is utf8.
+    --encoding <encoding>   encoding of MARC records, default is UTF-8.
                             Other possible options are: MARC-8,
                             ISO_5426, ISO_6937, ISO_8859-1, EUC-KR
     --format                The MARC transport format to use?

diff --git a/tools/stage-marc-import.pl b/tools/stage-marc-import.pl
index f353abd..a5f60bd 100755 (executable)
--- a/tools/stage-marc-import.pl
+++ b/tools/stage-marc-import.pl
@@ -56,7 +56,7 @@ my $parse_items                = $input->param('parse_items');
 my $item_action                = $input->param('item_action');
 my $comments                   = $input->param('comments');
 my $record_type                = $input->param('record_type');
-my $encoding                   = $input->param('encoding') || 'utf8';
+my $encoding                   = $input->param('encoding') || 'UTF-8';
 my $format                     = $input->param('format') || 'ISO2709';
 my $to_marc_plugin             = $input->param('to_marc_plugin');
 my $marc_modification_template = $input->param('marc_modification_template_id');
@@ -86,9 +86,16 @@ if ($completedJobID) {
     $template->param(map { $_ => $results->{$_} } keys %{ $results });
 } elsif ($fileID) {
     my $upload = Koha::Upload->new->get({ id => $fileID });
-    my $filename = $upload->{path};
-       my $marcrecord='';
-    my ($errors, $marcrecords) = C4::ImportBatch::RecordsFromISO2709File($uploaded_file->filename(), $record_type, $encoding);
+    my ( $file, $filename ) = ( $upload->{path}, $upload->{name} );
+    my ( $errors, $marcrecords );
+    if( $format eq 'MARCXML' ) {
+        ( $errors, $marcrecords ) = C4::ImportBatch::RecordsFromMARCXMLFile( $file, $encoding);
+    } else {
+        ( $errors, $marcrecords ) = C4::ImportBatch::RecordsFromISO2709File( $file, $record_type, $encoding );
+    }
+    warn "$filename: " . ( join ',', @$errors ) if @$errors;
+        # no need to exit if we have no records (or only errors) here
+        # BatchStageMarcRecords can handle that
 
     my $job = undef;
     my $dbh;