8 |
tag search display |
tag search display |
9 |
marc marc_indicators marc_repeatable_subfield |
marc marc_indicators marc_repeatable_subfield |
10 |
marc_compose marc_leader |
marc_compose marc_leader |
11 |
|
marc_duplicate marc_remove |
12 |
|
|
13 |
rec1 rec2 rec |
rec1 rec2 rec |
14 |
regex prefix suffix surround |
regex prefix suffix surround |
23 |
#use base qw/WebPAC::Common/; |
#use base qw/WebPAC::Common/; |
24 |
use Data::Dump qw/dump/; |
use Data::Dump qw/dump/; |
25 |
use Encode qw/from_to/; |
use Encode qw/from_to/; |
26 |
|
use Storable qw/dclone/; |
27 |
|
|
28 |
# debugging warn(s) |
# debugging warn(s) |
29 |
my $debug = 0; |
my $debug = 0; |
35 |
|
|
36 |
=head1 VERSION |
=head1 VERSION |
37 |
|
|
38 |
Version 0.09 |
Version 0.14 |
39 |
|
|
40 |
=cut |
=cut |
41 |
|
|
42 |
our $VERSION = '0.09'; |
our $VERSION = '0.14'; |
43 |
|
|
44 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
45 |
|
|
69 |
row => $row, |
row => $row, |
70 |
rules => $normalize_pl_config, |
rules => $normalize_pl_config, |
71 |
marc_encoding => 'utf-8', |
marc_encoding => 'utf-8', |
72 |
|
config => $config, |
73 |
); |
); |
74 |
|
|
75 |
Options C<lookup>, C<row>, C<rules> and C<log> are mandatory while all |
Options C<lookup>, C<row>, C<rules> and C<log> are mandatory while all |
91 |
no strict 'subs'; |
no strict 'subs'; |
92 |
_set_lookup( $arg->{lookup} ); |
_set_lookup( $arg->{lookup} ); |
93 |
_set_rec( $arg->{row} ); |
_set_rec( $arg->{row} ); |
94 |
|
_set_config( $arg->{config} ); |
95 |
_clean_ds( %{ $arg } ); |
_clean_ds( %{ $arg } ); |
96 |
eval "$arg->{rules}"; |
eval "$arg->{rules}"; |
97 |
die "error evaling $arg->{rules}: $@\n" if ($@); |
die "error evaling $arg->{rules}: $@\n" if ($@); |
113 |
$rec = shift or die "no record hash"; |
$rec = shift or die "no record hash"; |
114 |
} |
} |
115 |
|
|
116 |
|
=head2 _set_config |
117 |
|
|
118 |
|
Set current config hash |
119 |
|
|
120 |
|
_set_config( $config ); |
121 |
|
|
122 |
|
Magic keys are: |
123 |
|
|
124 |
|
=over 4 |
125 |
|
|
126 |
|
=item _ |
127 |
|
|
128 |
|
Code of current database |
129 |
|
|
130 |
|
=item _mfn |
131 |
|
|
132 |
|
Current MFN |
133 |
|
|
134 |
|
=back |
135 |
|
|
136 |
|
=cut |
137 |
|
|
138 |
|
my $config; |
139 |
|
|
140 |
|
sub _set_config { |
141 |
|
$config = shift; |
142 |
|
} |
143 |
|
|
144 |
=head2 _get_ds |
=head2 _get_ds |
145 |
|
|
146 |
Return hash formatted as data structure |
Return hash formatted as data structure |
149 |
|
|
150 |
=cut |
=cut |
151 |
|
|
152 |
my ($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators); |
my ($out, $marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators); |
153 |
|
my ($marc_record_offset, $marc_fetch_offset) = (0, 0); |
154 |
|
|
155 |
sub _get_ds { |
sub _get_ds { |
156 |
return $out; |
return $out; |
167 |
sub _clean_ds { |
sub _clean_ds { |
168 |
my $a = {@_}; |
my $a = {@_}; |
169 |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators) = (); |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators) = (); |
170 |
|
($marc_record_offset, $marc_fetch_offset) = (0,0); |
171 |
$marc_encoding = $a->{marc_encoding}; |
$marc_encoding = $a->{marc_encoding}; |
172 |
} |
} |
173 |
|
|
216 |
which might not be what you have in mind. If you need repeatable subfield, |
which might not be what you have in mind. If you need repeatable subfield, |
217 |
define it using C<marc_repeatable_subfield> like this: |
define it using C<marc_repeatable_subfield> like this: |
218 |
|
|
219 |
.... |
marc_repeatable_subfield('900','a'); |
220 |
|
marc('900','a', rec('200','a') ); |
221 |
|
marc('900','b', rec('200','b') ); |
222 |
|
marc('900','c', rec('200','c') ); |
223 |
|
|
224 |
|
will create: |
225 |
|
|
226 |
|
900a-1 900a-2 900a-3 900b-1 900c-1 |
227 |
|
900b-2 |
228 |
|
|
229 |
|
There is also support for returning next or specific using: |
230 |
|
|
231 |
|
while (my $mf = WebPAC::Normalize:_get_marc_fields( fetch_next => 1 ) ) { |
232 |
|
# do something with $mf |
233 |
|
} |
234 |
|
|
235 |
|
will always return fields from next MARC record or |
236 |
|
|
237 |
|
my $mf = WebPAC::Normalize::_get_marc_fields( offset => 42 ); |
238 |
|
|
239 |
|
will return 42th copy record (if it exists). |
240 |
|
|
241 |
=cut |
=cut |
242 |
|
|
243 |
sub _get_marc_fields { |
sub _get_marc_fields { |
244 |
|
|
245 |
return if (! $marc_record || ref($marc_record) ne 'ARRAY' || $#{ $marc_record } < 0); |
my $arg = {@_}; |
246 |
|
warn "### _get_marc_fields arg: ", dump($arg), $/ if ($debug > 2); |
247 |
|
my $offset = $marc_fetch_offset; |
248 |
|
if ($arg->{offset}) { |
249 |
|
$offset = $arg->{offset}; |
250 |
|
} elsif($arg->{fetch_next}) { |
251 |
|
$marc_fetch_offset++; |
252 |
|
} |
253 |
|
|
254 |
|
return if (! $marc_record || ref($marc_record) ne 'ARRAY'); |
255 |
|
|
256 |
|
warn "### full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 2); |
257 |
|
|
258 |
|
my $marc_rec = $marc_record->[ $offset ]; |
259 |
|
|
260 |
|
warn "## _get_marc_fields (at offset: $offset) -- marc_record = ", dump( @$marc_rec ), $/ if ($debug > 1); |
261 |
|
|
262 |
|
return if (! $marc_rec || ref($marc_rec) ne 'ARRAY' || $#{ $marc_rec } < 0); |
263 |
|
|
264 |
# first, sort all existing fields |
# first, sort all existing fields |
265 |
# XXX might not be needed, but modern perl might randomize elements in hash |
# XXX might not be needed, but modern perl might randomize elements in hash |
266 |
my @sorted_marc_record = sort { |
my @sorted_marc_record = sort { |
267 |
$a->[0] . ( $a->[3] || '' ) cmp $b->[0] . ( $b->[3] || '') |
$a->[0] . ( $a->[3] || '' ) cmp $b->[0] . ( $b->[3] || '') |
268 |
} @{ $marc_record }; |
} @{ $marc_rec }; |
269 |
|
|
270 |
@sorted_marc_record = @{ $marc_record }; ### FIXME disable sorting |
@sorted_marc_record = @{ $marc_rec }; ### FIXME disable sorting |
271 |
|
|
272 |
# output marc fields |
# output marc fields |
273 |
my @m; |
my @m; |
277 |
map { $u->{ $_->[0] . ( $_->[3] || '') }++ } @sorted_marc_record; |
map { $u->{ $_->[0] . ( $_->[3] || '') }++ } @sorted_marc_record; |
278 |
|
|
279 |
if ($debug) { |
if ($debug) { |
280 |
warn "## marc_repeatable_subfield ", dump( $marc_repeatable_subfield ), $/; |
warn "## marc_repeatable_subfield = ", dump( $marc_repeatable_subfield ), $/ if ( $marc_repeatable_subfield ); |
281 |
warn "## marc_record ", dump( $marc_record ), $/; |
warn "## marc_record[$offset] = ", dump( $marc_rec ), $/; |
282 |
warn "## sorted_marc_record ", dump( \@sorted_marc_record ), $/; |
warn "## sorted_marc_record = ", dump( \@sorted_marc_record ), $/; |
283 |
warn "## subfield count ", dump( $u ), $/; |
warn "## subfield count = ", dump( $u ), $/; |
284 |
} |
} |
285 |
|
|
286 |
my $len = $#sorted_marc_record; |
my $len = $#sorted_marc_record; |
298 |
# mark it visited |
# mark it visited |
299 |
$visited->{$i}++; |
$visited->{$i}++; |
300 |
|
|
301 |
my $row = $sorted_marc_record[$i]; |
my $row = dclone( $sorted_marc_record[$i] ); |
302 |
|
|
303 |
# field and subfield which is key for |
# field and subfield which is key for |
304 |
# marc_repeatable_subfield and u |
# marc_repeatable_subfield and u |
356 |
warn "## saved/3 ", dump( $field ),$/ if ($debug); |
warn "## saved/3 ", dump( $field ),$/ if ($debug); |
357 |
} |
} |
358 |
|
|
359 |
return @m; |
return \@m; |
360 |
} |
} |
361 |
|
|
362 |
=head2 _debug |
=head2 _debug |
471 |
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
472 |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
473 |
if (defined $sf) { |
if (defined $sf) { |
474 |
push @{ $marc_record }, [ $f, $i1, $i2, $sf => $v ]; |
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $i1, $i2, $sf => $v ]; |
475 |
} else { |
} else { |
476 |
push @{ $marc_record }, [ $f, $v ]; |
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $v ]; |
477 |
} |
} |
478 |
} |
} |
479 |
} |
} |
534 |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
535 |
my $m = [ $f, $i1, $i2 ]; |
my $m = [ $f, $i1, $i2 ]; |
536 |
|
|
537 |
|
warn "### marc_compose input subfields = ", dump(@_),$/ if ($debug > 2); |
538 |
|
|
539 |
while (@_) { |
while (@_) { |
540 |
my $sf = shift or die "marc_compose $f needs subfield"; |
my $sf = shift or die "marc_compose $f needs subfield"; |
541 |
my $v = shift; |
my $v = shift; |
546 |
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
547 |
} |
} |
548 |
|
|
549 |
warn "## marc_compose(d) ", dump( $m ),$/ if ($debug > 1); |
warn "## marc_compose current marc = ", dump( $m ),$/ if ($debug > 1); |
550 |
|
|
551 |
push @{ $marc_record }, $m if ($#{$m} > 2); |
push @{ $marc_record->[ $marc_record_offset ] }, $m if ($#{$m} > 2); |
552 |
} |
} |
553 |
|
|
554 |
|
=head2 marc_duplicate |
555 |
|
|
556 |
|
Generate copy of current MARC record and continue working on copy |
557 |
|
|
558 |
|
marc_duplicate(); |
559 |
|
|
560 |
|
Copies can be accessed using C<< _get_marc_fields( fetch_next => 1 ) >> or |
561 |
|
C<< _get_marc_fields( offset => 42 ) >>. |
562 |
|
|
563 |
|
=cut |
564 |
|
|
565 |
|
sub marc_duplicate { |
566 |
|
my $m = $marc_record->[ -1 ]; |
567 |
|
die "can't duplicate record which isn't defined" unless ($m); |
568 |
|
push @{ $marc_record }, dclone( $m ); |
569 |
|
warn "## marc_duplicate = ", dump(@$marc_record), $/ if ($debug > 1); |
570 |
|
$marc_record_offset = $#{ $marc_record }; |
571 |
|
warn "## marc_record_offset = $marc_record_offset", $/ if ($debug > 1); |
572 |
|
} |
573 |
|
|
574 |
|
=head2 marc_remove |
575 |
|
|
576 |
|
Remove some field or subfield from MARC record. |
577 |
|
|
578 |
|
marc_remove('200'); |
579 |
|
marc_remove('200','a'); |
580 |
|
|
581 |
|
This will erase field C<200> or C<200^a> from current MARC record. |
582 |
|
|
583 |
|
This is useful after calling C<marc_duplicate> or on it's own (but, you |
584 |
|
should probably just remove that subfield definition if you are not |
585 |
|
using C<marc_duplicate>). |
586 |
|
|
587 |
|
FIXME: support fields < 10. |
588 |
|
|
589 |
|
=cut |
590 |
|
|
591 |
|
sub marc_remove { |
592 |
|
my ($f, $sf) = @_; |
593 |
|
|
594 |
|
die "marc_remove needs record number" unless defined($f); |
595 |
|
|
596 |
|
my $marc = $marc_record->[ $marc_record_offset ]; |
597 |
|
|
598 |
|
warn "### marc_remove before = ", dump( $marc ), $/ if ($debug > 2); |
599 |
|
|
600 |
|
my $i = 0; |
601 |
|
foreach ( 0 .. $#{ $marc } ) { |
602 |
|
last unless (defined $marc->[$i]); |
603 |
|
warn "#### working on ",dump( @{ $marc->[$i] }), $/ if ($debug > 3); |
604 |
|
if ($marc->[$i]->[0] eq $f) { |
605 |
|
if (! defined $sf) { |
606 |
|
# remove whole field |
607 |
|
splice @$marc, $i, 1; |
608 |
|
warn "#### slice \@\$marc, $i, 1 = ",dump( @{ $marc }), $/ if ($debug > 3); |
609 |
|
$i--; |
610 |
|
} else { |
611 |
|
foreach my $j ( 0 .. (( $#{ $marc->[$i] } - 3 ) / 2) ) { |
612 |
|
my $o = ($j * 2) + 3; |
613 |
|
if ($marc->[$i]->[$o] eq $sf) { |
614 |
|
# remove subfield |
615 |
|
splice @{$marc->[$i]}, $o, 2; |
616 |
|
warn "#### slice \@{\$marc->[$i]}, $o, 2 = ", dump( @{ $marc }), $/ if ($debug > 3); |
617 |
|
# is record now empty? |
618 |
|
if ($#{ $marc->[$i] } == 2) { |
619 |
|
splice @$marc, $i, 1; |
620 |
|
warn "#### slice \@\$marc, $i, 1 = ", dump( @{ $marc }), $/ if ($debug > 3); |
621 |
|
$i--; |
622 |
|
}; |
623 |
|
} |
624 |
|
} |
625 |
|
} |
626 |
|
} |
627 |
|
$i++; |
628 |
|
} |
629 |
|
|
630 |
|
warn "### marc_remove($f", $sf ? ",$sf" : "", ") after = ", dump( $marc ), $/ if ($debug > 2); |
631 |
|
|
632 |
|
$marc_record->[ $marc_record_offset ] = $marc; |
633 |
|
|
634 |
|
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
635 |
|
} |
636 |
|
|
637 |
=head1 Functions to extract data from input |
=head1 Functions to extract data from input |
638 |
|
|
679 |
my $f = shift; |
my $f = shift; |
680 |
return unless (defined($rec && $rec->{$f})); |
return unless (defined($rec && $rec->{$f})); |
681 |
my $sf = shift; |
my $sf = shift; |
682 |
return map { $_->{$sf} } grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} }; |
return map { |
683 |
|
if (ref($_->{$sf}) eq 'ARRAY') { |
684 |
|
@{ $_->{$sf} }; |
685 |
|
} else { |
686 |
|
$_->{$sf}; |
687 |
|
} |
688 |
|
} grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} }; |
689 |
} |
} |
690 |
|
|
691 |
=head2 rec |
=head2 rec |
698 |
=cut |
=cut |
699 |
|
|
700 |
sub rec { |
sub rec { |
701 |
|
my @out; |
702 |
if ($#_ == 0) { |
if ($#_ == 0) { |
703 |
return rec1(@_); |
@out = rec1(@_); |
704 |
} elsif ($#_ == 1) { |
} elsif ($#_ == 1) { |
705 |
return rec2(@_); |
@out = rec2(@_); |
706 |
|
} |
707 |
|
if (@out) { |
708 |
|
return @out; |
709 |
|
} else { |
710 |
|
return ''; |
711 |
} |
} |
712 |
} |
} |
713 |
|
|
740 |
=cut |
=cut |
741 |
|
|
742 |
sub prefix { |
sub prefix { |
743 |
my $p = shift or die "prefix needs string as first argument"; |
my $p = shift or return; |
744 |
return map { $p . $_ } grep { defined($_) } @_; |
return map { $p . $_ } grep { defined($_) } @_; |
745 |
} |
} |
746 |
|
|
803 |
} |
} |
804 |
} |
} |
805 |
|
|
806 |
|
=head2 config |
807 |
|
|
808 |
|
Consult config values stored in C<config.yml> |
809 |
|
|
810 |
|
# return database code (key under databases in yaml) |
811 |
|
$database_code = config(); # use _ from hash |
812 |
|
$database_name = config('name'); |
813 |
|
$database_input_name = config('input name'); |
814 |
|
$tag = config('input normalize tag'); |
815 |
|
|
816 |
|
Up to three levels are supported. |
817 |
|
|
818 |
|
=cut |
819 |
|
|
820 |
|
sub config { |
821 |
|
return unless ($config); |
822 |
|
|
823 |
|
my $p = shift; |
824 |
|
|
825 |
|
$p ||= ''; |
826 |
|
|
827 |
|
my $v; |
828 |
|
|
829 |
|
warn "### getting config($p)\n" if ($debug > 1); |
830 |
|
|
831 |
|
my @p = split(/\s+/,$p); |
832 |
|
if ($#p < 0) { |
833 |
|
$v = $config->{ '_' }; # special, database code |
834 |
|
} else { |
835 |
|
|
836 |
|
my $c = dclone( $config ); |
837 |
|
|
838 |
|
foreach my $k (@p) { |
839 |
|
warn "### k: $k c = ",dump($c),$/ if ($debug > 1); |
840 |
|
if (ref($c) eq 'ARRAY') { |
841 |
|
$c = shift @$c; |
842 |
|
warn "config($p) taking first occurence of '$k', probably not what you wanted!\n"; |
843 |
|
last; |
844 |
|
} |
845 |
|
|
846 |
|
if (! defined($c->{$k}) ) { |
847 |
|
$c = undef; |
848 |
|
last; |
849 |
|
} else { |
850 |
|
$c = $c->{$k}; |
851 |
|
} |
852 |
|
} |
853 |
|
$v = $c if ($c); |
854 |
|
|
855 |
|
} |
856 |
|
|
857 |
|
warn "## config( '$p' ) = ",dump( $v ),$/ if ($v && $debug); |
858 |
|
warn "config( '$p' ) is empty\n" if (! $v); |
859 |
|
|
860 |
|
return $v; |
861 |
|
} |
862 |
|
|
863 |
|
=head2 id |
864 |
|
|
865 |
|
Returns unique id of this record |
866 |
|
|
867 |
|
$id = id(); |
868 |
|
|
869 |
|
Returns C<42/2> for 2nd occurence of MFN 42. |
870 |
|
|
871 |
|
=cut |
872 |
|
|
873 |
|
sub id { |
874 |
|
my $mfn = $config->{_mfn} || die "no _mfn in config data"; |
875 |
|
return $mfn . $#{$marc_record} ? $#{$marc_record} + 1 : ''; |
876 |
|
} |
877 |
|
|
878 |
=head2 join_with |
=head2 join_with |
879 |
|
|
880 |
Joins walues with some delimiter |
Joins walues with some delimiter |
885 |
|
|
886 |
sub join_with { |
sub join_with { |
887 |
my $d = shift; |
my $d = shift; |
888 |
return join($d, grep { defined($_) && $_ ne '' } @_); |
warn "### join_with('$d',",dump(@_),")\n" if ($debug > 2); |
889 |
|
my $v = join($d, grep { defined($_) && $_ ne '' } @_); |
890 |
|
return '' unless defined($v); |
891 |
|
return $v; |
892 |
} |
} |
893 |
|
|
894 |
=head2 split_rec_on |
=head2 split_rec_on |