- improved debugging

- fixed xslt for response to marc transformation

bin/slubsipbuilder.pl

@@ -52,7 +52,7 @@ sub patch_mods($) {
     xmlns:xs="http://www.w3.org/2001/XMLSchema"
     xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
     xmlns:mods="http://www.loc.gov/mods/v3"
-    xsi:schemaLocation="http://www.loc.gov/mods/v3 http://www.loc.gov/standards/mods/v3/mods-3-5.xsd"
+    xsi:schemaLocation="http://www.loc.gov/mods/v3 http://www.loc.gov/standards/mods/v3/mods-3-6.xsd"
     exclude-result-prefixes="xs"
     version="1.0">
     <xsl:output encoding="UTF-8" indent="yes" method="xml"/>
@@ -79,12 +79,13 @@ sub patch_marc_response($) {
     my $xslt_patch_string =<<PATCH2;
 <?xml version="1.0" encoding="UTF-8"?>
 <xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
-    xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns="http://www.loc.gov/MARC21/slim"
-    exclude-result-prefixes="" version="1.0">
+    xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns="http://www.loc.gov/MARC21/slim" xmlns:srw="http://www.loc.gov/zing/srw/"
+    exclude-result-prefixes="srw" version="1.0">
     <xsl:template match="/record">
         <xsl:element name="collection">
-            <xsl:element name="{local-name()}" namespace="http://www.loc.gov/MARC21/slim">
-            <xsl:apply-templates select="node() | @*"/>
+            <xsl:element name="record" namespace="http://www.loc.gov/MARC21/slim">
+                <xsl:apply-templates select="@*"/>
+                <xsl:apply-templates select="node()"/>
             </xsl:element>
         </xsl:element>
     </xsl:template>
@@ -93,6 +94,11 @@ sub patch_marc_response($) {
             <xsl:apply-templates select="node() | @*"/>
         </xsl:element>
     </xsl:template>
+    <xsl:template match="@*">
+        <xsl:attribute name="{local-name()}">
+            <xsl:value-of select="."/>
+        </xsl:attribute>
+    </xsl:template>
 </xsl:stylesheet>
 PATCH2
     my $xslt = XML::LibXSLT->new();
@@ -107,6 +113,8 @@ PATCH2
 # the "old" approach does not handle umlauts or UTF8-chars above ASCII table
 # old: http://swb2.bsz-bw.de/sru/DB=2.1/username=/password=/?query=pica.ppn+%3D+"494384174"&startRecord=1&maximumRecords=10&recordSchema=marcxml
 # new: http://swb2.bsz-bw.de/sru/DB=2.1/username=/password=/?query=pica.ppn+%3D+"494384174"&startRecord=1&maximumRecords=10&recordSchema=marc21&recordPacking=xml&version=1.1
+# with stylesheet:
+# http://swb2.bsz-bw.de/sru/DB=2.1/username=/password=/?query=pica.ppn+%3D+%22494384174%22&version=1.1&operation=searchRetrieve&stylesheet=http%3A%2F%2Fswb2.bsz-bw.de%2Fsru%2FDB%3D2.1%2F%3Fxsl%3DsearchRetrieveResponse&recordSchema=marc21&maximumRecords=10&startRecord=1&recordPacking=xml&sortKeys=none&x-info-5-mg-requestGroupings=none
 sub get_mods_from ($$) { # $mods = ($url, $ppn)
     my $url = shift;
     my $ppn = shift; # example: "457035137"  for "Der Fichtelberg"
@@ -124,7 +132,7 @@ sub get_mods_from ($$) { # $mods = ($url, $ppn)
     #my $sruschema="marcxml";
     my $sruschema="marc21";
     #my $sru = "${srubase}?query=${srusearchkey}+%3D+%22${sruvalue}%22&startRecord=1&maximumRecords=${srumaxrecords}&recordSchema=${sruschema}";
-    my $sru = "${srubase}?query=${srusearchkey}+%3D+%22${sruvalue}%22&startRecord=1&maximumRecords=${srumaxrecords}&recordSchema=${sruschema}&recordPacking=xml&version=1.1";
+    my $sru = "${srubase}?query=${srusearchkey}+%3D+%22${sruvalue}%22&startRecord=1&maximumRecords=${srumaxrecords}&recordSchema=${sruschema}&recordPacking=xml&version=1.1&stylesheet=http%3A%2F%2Fswb2.bsz-bw.de%2Fsru%2FDB%3D2.1%2F%3Fxsl%3DsearchRetrieveResponse";
     #p ($sru); # debug output
     my $record = $ua->get($sru); # ask SWB for given PPN
     if ($record->is_success) {
@@ -132,6 +140,7 @@ sub get_mods_from ($$) { # $mods = ($url, $ppn)
         my $xp = XML::XPath->new( $record->decoded_content );
         my $parser = XML::LibXML->new();
         if ($with_debug) {
+            say "write DEBUG_${ppn}_response.xml";
             write_file("DEBUG_${ppn}_response.xml", {binmode => ':utf8'}, $record->decoded_content);
         }
         my $marcblob = $parser->parse_string(
@@ -139,7 +148,9 @@ sub get_mods_from ($$) { # $mods = ($url, $ppn)
         );
         my $marcblob_patched = patch_marc_response( $marcblob );
         if ($with_debug) {
+            say "write DEBUG_${ppn}_marc_unpatched.xml";
             write_file("DEBUG_${ppn}_marc_unpatched.xml", {binmode => ':utf8'}, $marcblob);
+            say "write DEBUG_${ppn}_marc.xml";
             write_file("DEBUG_${ppn}_marc.xml", {binmode => ':utf8'}, $marcblob_patched);
         }
         my $xslt = XML::LibXSLT->new();
@@ -148,6 +159,7 @@ sub get_mods_from ($$) { # $mods = ($url, $ppn)
         my $marc = $parser->parse_string( $marcblob_patched );
         my $result = $stylesheet->transform( $marc);
         if ($with_debug) {
+            say "write DEBUG_${ppn}_unpatched_mods.xml";
             write_file("DEBUG_${ppn}_unpatched_mods.xml", {binmode => ':utf8'}, $stylesheet->output_string( $result ));
         }
         $result = patch_mods( $result);
@@ -247,7 +259,7 @@ my $mods;
 if (defined $ppn) {
     $mods = get_mods_from($url, $ppn);
     if (1 == $with_debug) {
-        write_file("DEBUG_${ppn}_mods.xml", $mods);
+        write_file("DEBUG_${ppn}_mods.xml", {binmode => ':utf8'}, $mods);
     }
 # remove the <xml /> from beginning of the answer
     $mods=~ s#<\?xml version="1.0" encoding="UTF-8"\?>#<!-- removed xml header from mods part -->#;
@@ -349,7 +361,7 @@ METS
 
 # compress if needed
 if (!defined $as_zip) {
-    write_file( "${output}/${sip_root_dir}/sip.xml", $sip );
+    write_file( "${output}/${sip_root_dir}/sip.xml",{binmode => ':utf8'},  $sip );
     # copy source to target
     foreach my $source (sort keys (%filecopyhash)) {
         my $target = $filecopyhash{$source}->{"target"};