eprints-dev: /home/dpavlin/tsv2xp-xml.pl [commit]

diff --git a/tsv2xp-xml.pl b/tsv2xp-xml.pl
new file mode 100755 (executable)
index 0000000..2991632
--- /dev/null
+++ b/tsv2xp-xml.pl
@@ -0,0 +1,88 @@
+#!/usr/bin/perl
+
+# sudo -u eprints /usr/share/eprints3/bin/import --verbose --migration ffzg eprint XML dipl.xml
+
+use warnings;
+use strict;
+
+use File::Slurp;
+use Data::Dump qw(dump);
+
+my $tsv_file = 'items.mentor_ime.mentor_perzime.IME_FILE_a.PREZIME.IME.NAZIV_RADA.MENTOR.GOD_OBR63a8814e199c9b969dbe8251fdef0fa2';
+
+our $eprintsid = 700;
+
+my $xml = read_file 'ep-xml.xml';
+my @files = read_file "files.txt";
+
+my $file2path;
+foreach my $full ( @files ) {
+       chomp $full;
+       my $file = $1 if $full =~ m{/([^/]+)$};
+       $file =~ s/\.\w+$//;
+       $file2path->{ lc $file } = $full;
+}
+#warn "# file2path ",dump($file2path);
+
+print qq{<?xml version="1.0" encoding="utf-8" ?>
+<eprints>
+};
+
+my @header;
+my $header2col;
+my $col = 0;
+
+our @v;
+sub interpolate {
+       my $f = shift;
+       my $optional = $1 if $f =~ s{(\?)$}{};
+       my $i = $header2col->{$f};
+       die "no $f in ", dump( $header2col ) if not defined $i and not $optional;
+       my $v = $v[$i];
+       warn "# $f $i = $v\n";
+       return $v;
+}
+
+
+open(my $tsv, '<', $tsv_file) || die "$tsv_file: $!";
+while(<$tsv>) {
+       chomp;
+       if ( m/#(.+)/ ) {
+               @header = split(/\t/, $1);
+               warn "# header ",dump( @header );
+               my $i = 0;
+               $header2col->{$_} = $col++ foreach @header;
+               warn "# header2col ",dump( $header2col );
+               next;
+       }
+
+       @v = map { s/\\N//g; $_ } split(/\t/, $_);
+       warn "# v = ", dump(@v);
+
+       my $file = interpolate 'IME FILE-a';
+       $file =~ s/ //g;
+       my $full_path;
+       if ( my $full = $file2path->{ lc $file } ) {
+               $full_path = $full;
+               warn "# file $file -> $full_path\n";
+       }
+
+       my $c = $col;
+       $header2col->{'eprintsid'} = $c; $v[$c++] = $eprintsid++;
+       $header2col->{'file'} = $c; $v[$c++] = $file;
+       $header2col->{'full_path'} = $c; $v[$c++] = $full_path;
+
+       my $eprints = $xml;
+       while ( $eprints =~ s/<!-- "(.+?)" -->/interpolate($1)/seg ) {
+               warn "# replaced $1\n";
+       }
+
+       $eprints =~ s{<documents>.+</documents>}{<!-- no documents -->}s if ! $full_path;
+
+       print $eprints;
+}
+
+print qq{
+</eprints>
+};
+