From 0c3d7b1c7283f9df91e36cc33fdaa967cd7d8762 Mon Sep 17 00:00:00 2001
From: Santiago Lo Coco <santilococo.01@gmail.com>
Date: Sun, 23 Jul 2023 13:47:47 +0200
Subject: [PATCH] Add initial .pl files

---
 build.pl   | 79 +++++++++++++++++++++++++++++++++++++++++++++
 marc2dc.pl | 95 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 174 insertions(+)
 create mode 100755 build.pl
 create mode 100755 marc2dc.pl

diff --git a/build.pl b/build.pl
new file mode 100755
index 0000000..3e9c9f7
--- /dev/null
+++ b/build.pl
@@ -0,0 +1,79 @@
+#!/usr/bin/perl -w
+
+use utf8;
+binmode(STDOUT, ":utf8");
+
+$/ = "</dublin_core>\n"; # record separator
+
+$what = 1000; # dummy id for when there’s no file
+
+while (<>) {
+    $cul = "";
+    $tac = "";
+
+    # discard the top and bottom tags
+    s/<collection>\n//;
+    s/<\/collection>\n//;
+
+    # extract the file path from the identifier
+    # use the file name as an id
+    # note that identifier element is discarded!
+    #if (s!<dcvalue element="relation" qualifier="uri">\s*(.*?)\s*<\/dcvalue>\n!!s) {
+    if (/<dcvalue element="relation" qualifier="uri">\s*(.*?)\s*<\/dcvalue>\n/) {
+	$cul = $1;
+	if ($cul =~ /docs/) {
+		#$cul =~ s@.*(folders/|d/)([^?&/]+).*@\2@;
+		$cul =~ s@.*(d/)([^?&/]+).*@$2@;
+		$tac = "https://drive.google.com/uc?export=download&id=$cul";
+    		print "$tac\n";
+    		print "\n";
+	}
+        $path = '';
+	$id = $what++;
+    } else {
+        $path = '';
+        $id = $what++;
+    }
+
+    # let the operator know where we’re up to
+    print "$path/$id\n";
+
+    # create the item directory
+    mkdir "import/$id", 0755;
+
+    # create the dublin_core.xml file
+    open DC, ">import/$id/dublin_core.xml"
+      or die "Cannot open dublin core for $id, $!\n";
+    print DC $_;
+    close DC;
+
+    # assuming we have a file ...
+    if ($tac) {
+	print "$tac\n";
+	print "$cul\n";
+
+	system "curl -L \"$tac\" -o import/$id/$cul";
+	#system "curl -s -L \"$tac\" -o import/$id/$cul";
+
+	my $output = `file import/$id/$cul`;
+	print $output;
+
+	if ($output =~ /PDF/) {
+		$ext = "pdf";
+	}
+	if ($output =~ /Word/) {
+		system "soffice --headless --convert-to pdf import/$id/$cul";
+		$ext = "pdf";
+	}
+
+	system "mv import/$id/$cul import/$id/$cul.$ext";
+
+        # ... create the contents file ...
+        open OUT, ">import/$id/contents"
+          or die "Cannot open contents for $id, $!\n";
+        print OUT "$cul.$ext";
+        close OUT;
+    }
+}
+
+__END__
diff --git a/marc2dc.pl b/marc2dc.pl
new file mode 100755
index 0000000..82b9ecc
--- /dev/null
+++ b/marc2dc.pl
@@ -0,0 +1,95 @@
+#!/usr/bin/perl -w
+
+use MARC::Crosswalk::DublinCore;
+use MARC::File::USMARC;
+use utf8;
+binmode(STDOUT, ":utf8");
+
+$/ = chr(29); # MARC record separator
+
+print qq|<collection>\n|;
+
+while (my $blob = <>) { # suck in one MARC record at a time
+
+        print qq|<dublin_core>\n|;
+
+        # convert the MARC to DC
+        my $marc = MARC::Record->new_from_usmarc( $blob );
+        my $crosswalk = MARC::Crosswalk::DublinCore->new( qualified => 0 );
+        my $dc        = $crosswalk->as_dublincore( $marc );
+
+        # output the DC as XML
+        for( $dc->elements ) {
+		my $element     = lc($_->name);
+                my $qualifier   = lc($_->qualifier);
+                my $scheme      = lc($_->scheme);
+                my $content     = $_->content;
+
+		##print $_->content;
+
+		#print qq|$element\n|;
+		#print qq|$qualifier\n|;
+		#print qq|$scheme\n|;
+		#print qq|$content\n|;
+
+                # escape reserved characters
+                $content =~ s/&/&amp;/gs;
+                $content =~ s/</&lt;/gs;
+                $content =~ s/>/&gt;/gs;
+
+                # munge attributes for DSpace compatibility
+                if ($element eq 'creator') {
+                        $element = 'contributor';
+                        $qualifier = 'author';
+                	$content =~ s/[0-9]//g;
+			$content =~ s/^\s+|\s+$//g;
+		}
+                if ($element eq 'format') {
+                        $element = 'description';
+                        $qualifier = '';
+                }
+                if ($element eq 'language') {
+                        $qualifier = 'iso';
+                }
+                if ($element eq 'type') {
+                        $element = 'type';
+                        $qualifier = '';
+                }
+                if ($element eq 'subject') {
+                	$content =~ s/[0-9]//g;
+			$content =~ s/^\s+|\s+$//g;
+                }
+                if ($element eq 'title') {
+                	$content =~ s/\[[^][]*\]//g;
+			$content =~ s/^\s+|\s+$//g;
+                }
+                if ($element eq 'identifier') {
+                        $element = 'relation';
+                        $qualifier = 'uri';
+                }
+
+                printf qq|  <dcvalue element="%s"|, $element;
+                printf qq| qualifier="%s"|, $qualifier if $qualifier;
+                printf qq| qualifier="%s"|, $scheme if $scheme and !$qualifier;
+
+		if ($element eq 'title' || $element eq 'description') {
+			#if ($element eq 'title' || $element eq 'description' || $element eq 'subject') {
+                	printf qq| language="es_AR">%s</dcvalue>\n|, $content;
+		} else {
+                	printf qq|>%s</dcvalue>\n|, $content;
+		}
+        }
+
+	#	if ($ARGV[0]) {
+	#                printf qq|  <dcvalue element="dspace"|, $element;
+	#                printf qq| qualifier="entity"|, $qualifier if $qualifier;
+	#                printf qq| qualifier="%s"|, $scheme if $scheme and !$qualifier;
+	#                printf qq|>%s</dcvalue>\n|, $content;
+	#	}
+
+        print qq|</dublin_core>\n|;
+}
+
+print qq|</collection>\n|;
+
+exit;