my $term = new Term::ReadLine 'yaz-cleanup';
my $OUT = $term->OUT || \*STDOUT;
+my $conf = {};
+
my $count = 0;
my $reccount = 0;
my $oreccount = 0;
my $line = '';
my %trash = (); # hash for tags to be dumped
+# initialization and setup
+initialize($conf);
+
# read in trash tags file if it exists
-populate_trash() if (-e '.trashtags');
+populate_trash() if ($conf->{trash});
my @record = (); # current record storage
my %recmeta = (); # metadata about current record
my $input = shift || 'incoming.marc.xml';
-open MARC, '<', $input;
-open my $NUMARC, '>', 'incoming.clean.marc.xml';
+open MARC, '<:utf8', $input;
+open my $NUMARC, '>:utf8', $conf->{output};
print $NUMARC '<?xml version="1.0" encoding="UTF-8"?>',"\n";
print $NUMARC '<collection xmlns="http://www.loc.gov/MARC21/slim">',"\n";
-open my $EXMARC, '>', 'incoming.exceptions.marc.xml';
+open my $EXMARC, '>:utf8', $conf->{exception};
print $EXMARC '<?xml version="1.0" encoding="UTF-8"?>',"\n";
print $EXMARC '<collection xmlns="http://www.loc.gov/MARC21/slim">',"\n";
open MARC2, '<', $input;
}
update_linecontext();
+ next if ($line =~ m|</record>|);
+
# catch empty datafield elements
if ($line =~ m|</datafield>|) {
if ($record[-2] =~ m/<datafield tag="..." ind1="." ind2=".">/) {
$recmeta{ind2} = $3;
}
+ # automatable subfield maladies
+ $line =~ s/code=" ">c/code="c">/;
+ $line =~ s/code=" ">\$/code="c"$>/;
+
# and stow line back in record
$record[-1] = $line;
}
# subfields can't be non-alphanumeric
- if ($line =~ /<subfield code="(.+?)"/) {
+ if ($line =~ /<subfield code="(.*?)"/) {
my $match = $1;
- if ($match =~ /\P{IsAlnum}/) {
- edit("Junk in subfield code", $line);
+ if ($match =~ /\P{IsAlnum}/ or $match eq '') {
+ edit("Junk in subfield code/Null subfield code", $line);
next;
}
}
%recmeta = ();
$reccount++;
} elsif ($l =~ m|</record>|) {
- push @record, $l;
write_record($NUMARC) if $reccount;
} else {
push @record, $l;
$oreccount++ if ($FH eq $NUMARC);
print $FH '<!-- ', $recmeta{explanation}, " -->\n"
if(defined $recmeta{explanation});
+
+ # excise unwanted tags
+ if (keys %trash or $conf->{autoscrub}) {
+ my @trimmed = ();
+ my $istrash = 0;
+ for my $line (@record) {
+ if ($istrash) {
+ $istrash = 0 if $line =~ m|</datafield|;
+ next;
+ }
+ if ($line =~ m/<datafield tag="(.{3})"/) {
+ my $tag = $1;
+ if ($trash{$tag} or ($conf->{autoscrub} and $tag =~ /\D/)) {
+ $istrash = 1;
+ next
+ }
+ }
+ push @trimmed, $line;
+ }
+ @record = @trimmed;
+ }
+
+ # scrub newlines
+ unless ($conf->{nocollapse}) {
+ s/\n// for (@record);
+ }
+
+ # add 903(?) with new record id
+ my $renumber = '';
+ if ($conf->{'renumber-from'}) {
+ $renumber = join('', '<datafield tag="', $conf->{'renumber-tag'},
+ '" ind1=" " ind2=" ">',
+ '<subfield code="', $conf->{'renumber-subfield'}, '">',
+ $conf->{'renumber-from'}, '</subfield></datafield>');
+ $renumber .= "\n" if $conf->{nocollapse};
+ push @record, $renumber;
+ $conf->{'renumber-from'}++;
+ }
+
print $FH @record;
+ print $FH "</record>\n";
}
sub update_linecontext {
sub dump_record {
my ($line_in, @explanation) = @_;
- $recmeta{explanation} = join(' ', @explanation);
+ $recmeta{explanation} = join(' ', 'Tag', $recmeta{tag}, @explanation);
my $line = <MARC>; $count++;
update_linecontext();
until ($line =~ m|</record>|) {
# specifying a tag twice is an error, to help prevent typos
sub populate_trash {
- open TRASH, '<', '.trashtags';
+ print $OUT ">>> TRASHTAGS FILE FOUND. LOADING TAGS TO BE STRIPPED FROM OUTPUT\n";
+ open TRASH, '<', $conf->{trash}
+ or die "Can't open trash tags file!\n";
while (<TRASH>) {
my $lastwasrange = 0;
my %lastrange = ( high => 0, low => 0);
die "Unknown chunk $chunk in .trashtags file (line $.)\n";
}
}
- print $OUT join ",", (sort keys %trash);
- exit
+
+ # remove original id sequence tag from trash hash if we know it
+ trash_add($conf->{'original-tag'}, 1)
+ if ($conf->{'original-tag'} and $trash{$conf->{'original-tag'}});
}
sub trash_add_range {
$trash{$tag} = 1;
}
}
+
+#-----------------------------------------------------------------------
+
+=head2 initialize
+
+Performs boring script initialization. Handles argument parsing,
+mostly.
+
+=cut
+
+sub initialize {
+ my ($c) = @_;
+ my @missing = ();
+
+ # set mode on existing filehandles
+ binmode(STDIN, ':utf8');
+
+ my $rc = GetOptions( $c,
+ 'autoscrub|a',
+ 'exception|x=s',
+ 'output|o=s',
+ 'nocollapse|n',
+ 'renumber-from|rf=i',
+ 'original-tag|ot=i',
+ 'renumber-tag|rt=i',
+ 'renumber-subfield|rt=i',
+ 'trash|t=s',
+ 'help|h',
+ );
+ show_help() unless $rc;
+ show_help() if ($c->{help});
+
+ # defaults
+ $c->{output} = 'incoming.cleaned.marc.xml' unless defined $c->{output};
+ $c->{exception} = 'incoming.exception.marc.xml' unless defined $c->{exception};
+ $c->{'renumber-tag'} = 903 unless defined $c->{'renumber-tag'};
+ $c->{'renumber-subfield'} = 'a' unless defined $c->{'renumber-subfield'};
+
+ my @keys = keys %{$c};
+ show_help() unless (@ARGV and @keys);
+ #for my $key ('runtype', 'tag', 'subfield', 'output', 'exception')
+ # { push @missing, $key unless $c->{$key} }
+ #if (@missing) {
+ # print "Required option: ", join(', ', @missing), " missing!\n";
+ # show_help();
+ #}
+}
+
+sub show_help {
+ print <<HELP;
+Usage is: $0 [OPTIONS] <filelist>
+Options
+ --output -o Cleaned MARCXML output filename (default: incoming.cleaned.marc.xml)
+ --exception -x Exception (dumped records) MARCXML filename (incoming.exception.marc.xml)
+HELP
+exit;
+}