use Getopt::Long;
use Term::ReadLine;
+use Equinox::Migration::SimpleTagList;
binmode STDOUT, ":utf8";
my $term = new Term::ReadLine 'yaz-cleanup';
# initialization and setup
my $conf = {};
initialize($conf);
-populate_trash() if ($conf->{trashfile});
# set up files, since everything appears to be in order
my $marcfile = shift || 'incoming.marc.xml';
sub edit {
my ($msg) = @_;
- return if $conf->{trash}{ $recmeta{tag} };
+ return if $conf->{trash}->has( $recmeta{tag} );
$conf->{editmsg} = $msg;
print_fullcontext();
%recmeta = ();
$conf->{ricount}++;
- until ($l =~ m|</record>|)
+ until ($l =~ m|</record>|)
{ push @record, $l; $l = <MARC>; }
push @record, $l;
return 1;
if(defined $recmeta{explanation});
# excise unwanted tags
- if (keys %{$trash} or $conf->{autoscrub}) {
+ if (defined $trash or $conf->{autoscrub}) {
my @trimmed = ();
my $istrash = 0;
for my $line (@record) {
}
if ($line =~ m/<datafield tag="(.{3})"/) {
my $tag = $1;
- if ($trash->{$tag} or ($conf->{autoscrub} and $tag =~ /\D/)) {
+ if ($trash->has($tag) or ($conf->{autoscrub} and $tag =~ /\D/)) {
$istrash = 1;
next
}
sub quit { exit }
-#-----------------------------------------------------------------------------------
-# populate_trash
-#-----------------------------------------------------------------------------------
-# defined a domain-specific language for specifying MARC tags to be dropped from
-# records during processing. it is line oriented, and is specified as follows:
-#
-# each line may specify any number of tags to be included, either singly (\d{1,3})
-# or as a range (\d{1,3}\.\.\d{1,3}
-#
-# if a single number is given, it must be between '000' and '999', inclusive.
-#
-# ranges obey the previous rule, and also the first number of the range must be less
-# than the second number
-#
-# finally, any single range in a line may be followed by the keyword 'except'. every
-# number or range after 'except' is excluded from the range specified. all these
-# numbers must actually be within the range.
-#
-# specifying a tag twice is an error, to help prevent typos
-
-sub populate_trash {
- print $OUT ">>> TRASHTAGS FILE FOUND. LOADING TAGS TO BE STRIPPED FROM OUTPUT\n";
- open TRASH, '<', $conf->{trashfile}
- or die "Can't open trash tags file!\n";
- while (<TRASH>) {
- my $lastwasrange = 0;
- my %lastrange = ( high => 0, low => 0);
- my $except = 0;
-
- my @chunks = split /\s+/;
- while (my $chunk = shift @chunks) {
-
- # single values
- if ($chunk =~ /^\d{1,3}$/) {
- trash_add($chunk, $except);
- $lastwasrange = 0;
- next;
- }
-
- # ranges
- if ($chunk =~ /^\d{1,3}\.\.\d{1,3}$/) {
- my ($low, $high) = trash_add_range($chunk, $except, \%lastrange);
- $lastwasrange = 1;
- %lastrange = (low => $low, high => $high)
- unless $except;
- next;
- }
-
- # 'except'
- if ($chunk eq 'except') {
- die "Keyword 'except' can only follow a range (line $.)\n"
- unless $lastwasrange;
- die "Keyword 'except' may only occur once per line (line $.)\n"
- if $except;
- $except = 1;
- next;
- }
-
- die "Unknown chunk $chunk in .trashtags file (line $.)\n";
- }
- }
-
- # remove original id sequence tag from trash hash if we know it
- trash_add($conf->{'original-tag'}, 1)
- if ($conf->{'original-tag'} and $conf->{trash}{ $conf->{'original-tag'} });
-}
-
-sub trash_add_range {
- my ($chunk, $except, $range) = @_;
- my ($low,$high) = split /\.\./, $chunk;
- die "Ranges must be 'low..high' ($low is greater than $high on line $.)\n"
- if ($low > $high);
- if ($except) {
- die "Exception ranges must be within last addition range (line $.)\n"
- if ($low < $range->{low} or $high > $range->{high});
- }
- for my $tag ($low..$high) {
- trash_add($tag, $except)
- }
- return $low, $high;
-}
-
-sub trash_add {
- my ($tag, $except) = @_;
- my $trash = $conf->{trash};
-
- die "Trash values must be valid tags (000-999)\n"
- unless ($tag >= 0 and $tag <= 999);
-
- if ($except) {
- delete $trash->{$tag};
- } else {
- die "Trash tag '$tag' specified twice (line $.)\n"
- if $trash->{$tag};
- $trash->{$tag} = 1;
- }
-}
-
#-----------------------------------------------------------------------
=head2 initialize
$c->{'renumber-subfield'} = 'a' unless defined $c->{'renumber-subfield'};
$c->{window} = 5;
+ if ($c->{trashfile}) {
+ $c->{trash} = Equinox::Migration::SimpleTagList->new($conf->{trashfile})
+ } else {
+ $c->{trash} = Equinox::Migration::SimpleTagList->new;
+ }
+ # remove original id sequence tag from trash hash if we know it
+ $c->{trash}->remove_tag($c->{'original-tag'})
+ if ( $c->{'original-tag'} and $c->{trash}->has($c->{'original-tag'}) );
+
# autotrash 901, 903 if strip-nines
if ($c->{'strip9'}) {
- $c->{trash}{901} = 1;
- $c->{trash}{903} = 1;
+ $c->{trash}->add_tag(901);
+ $c->{trash}->add_tag(903);
}
my @keys = keys %{$c};
--autoscrub -a Automatically remove non-numeric tags in data
--nocollapse -n Don't compress records to one line on output
- --strip9 Automatically remove any existing 901/903 tags in data (reversible)
+ --strip9 Automatically remove any existing 901/903 tags in data
+ --no-strip9 Don't complain about 901/903 tags in data
--trashfile -t File containing trash tag data (see --trashhelp)
sub show_trashhelp {
print <<HELP;
-The marc-cleanup trash tags file is a simple plaintext file. It is a
-line oriented format. There are three basic tokens:
-
- * The tag
- * The tag range
- * The "except" clause
-
-Any number of tags and/or tag ranges can appear on a single line. A
-tag cannot appear twice in the file, either alone or as part of a
-range. This is to prevent errors in the trash tag listing. Items do
-not have to be sorted within a line. These following lines are valid:
-
- 850 852 870..879 886 890 896..899
- 214 696..699 012
-
-Ranges must be ordered internally. That is, "870..879" is valid while
-"879..870" is not.
-
-Finally, there can be only one "except" clause on a line. It is
-composed of the word "except" followed by one or more tags or
-ranges. Except clauses must follow a range, and all tags within the
-clause must be within the range which the clause follows.
+See
- 900..997 except 935 950..959 987 994
+http://intra.lan.hq.esilibrary.com/dokuwiki/doku.php?id=migration:tag_files
-is a valid example.
+for tag file syntax information.
HELP
exit;
}
projects / migration-tools.git / blobdiff