7 |
_debug |
_debug |
8 |
_pack_subfields_hash |
_pack_subfields_hash |
9 |
|
|
10 |
tag search display |
search_display search display |
11 |
|
|
12 |
marc marc_indicators marc_repeatable_subfield |
marc marc_indicators marc_repeatable_subfield |
13 |
marc_compose marc_leader |
marc_compose marc_leader marc_fixed |
14 |
marc_duplicate marc_remove |
marc_duplicate marc_remove marc_count |
15 |
marc_original_order |
marc_original_order |
16 |
|
|
17 |
rec1 rec2 rec |
rec1 rec2 rec |
20 |
save_into_lookup |
save_into_lookup |
21 |
|
|
22 |
split_rec_on |
split_rec_on |
23 |
|
|
24 |
|
get set |
25 |
|
count |
26 |
/; |
/; |
27 |
|
|
28 |
use warnings; |
use warnings; |
43 |
|
|
44 |
=head1 VERSION |
=head1 VERSION |
45 |
|
|
46 |
Version 0.23 |
Version 0.30 |
47 |
|
|
48 |
=cut |
=cut |
49 |
|
|
50 |
our $VERSION = '0.23'; |
our $VERSION = '0.30'; |
51 |
|
|
52 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
53 |
|
|
60 |
C<perl -c normalize.pl>. |
C<perl -c normalize.pl>. |
61 |
|
|
62 |
Normalisation can generate multiple output normalized data. For now, supported output |
Normalisation can generate multiple output normalized data. For now, supported output |
63 |
types (on the left side of definition) are: C<tag>, C<display>, C<search> and |
types (on the left side of definition) are: C<search_display>, C<display>, C<search> and |
64 |
C<marc>. |
C<marc>. |
65 |
|
|
66 |
=head1 FUNCTIONS |
=head1 FUNCTIONS |
168 |
|
|
169 |
=cut |
=cut |
170 |
|
|
171 |
my ($out, $marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $leader); |
my ($out, $marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $marc_leader); |
172 |
my ($marc_record_offset, $marc_fetch_offset) = (0, 0); |
my ($marc_record_offset, $marc_fetch_offset) = (0, 0); |
173 |
|
|
174 |
sub _get_ds { |
sub _get_ds { |
185 |
|
|
186 |
sub _clean_ds { |
sub _clean_ds { |
187 |
my $a = {@_}; |
my $a = {@_}; |
188 |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $leader) = (); |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $marc_leader) = (); |
189 |
($marc_record_offset, $marc_fetch_offset) = (0,0); |
($marc_record_offset, $marc_fetch_offset) = (0,0); |
190 |
$marc_encoding = $a->{marc_encoding}; |
$marc_encoding = $a->{marc_encoding}; |
191 |
} |
} |
290 |
|
|
291 |
=cut |
=cut |
292 |
|
|
293 |
|
my $fetch_pos; |
294 |
|
|
295 |
sub _get_marc_fields { |
sub _get_marc_fields { |
296 |
|
|
297 |
my $arg = {@_}; |
my $arg = {@_}; |
298 |
warn "### _get_marc_fields arg: ", dump($arg), $/ if ($debug > 2); |
warn "### _get_marc_fields arg: ", dump($arg), $/ if ($debug > 2); |
299 |
my $offset = $marc_fetch_offset; |
$fetch_pos = $marc_fetch_offset; |
300 |
if ($arg->{offset}) { |
if ($arg->{offset}) { |
301 |
$offset = $arg->{offset}; |
$fetch_pos = $arg->{offset}; |
302 |
} elsif($arg->{fetch_next}) { |
} elsif($arg->{fetch_next}) { |
303 |
$marc_fetch_offset++; |
$marc_fetch_offset++; |
304 |
} |
} |
307 |
|
|
308 |
warn "### full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 2); |
warn "### full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 2); |
309 |
|
|
310 |
my $marc_rec = $marc_record->[ $offset ]; |
my $marc_rec = $marc_record->[ $fetch_pos ]; |
311 |
|
|
312 |
warn "## _get_marc_fields (at offset: $offset) -- marc_record = ", dump( @$marc_rec ), $/ if ($debug > 1); |
warn "## _get_marc_fields (at offset: $fetch_pos) -- marc_record = ", dump( @$marc_rec ), $/ if ($debug > 1); |
313 |
|
|
314 |
return if (! $marc_rec || ref($marc_rec) ne 'ARRAY' || $#{ $marc_rec } < 0); |
return if (! $marc_rec || ref($marc_rec) ne 'ARRAY' || $#{ $marc_rec } < 0); |
315 |
|
|
330 |
|
|
331 |
if ($debug) { |
if ($debug) { |
332 |
warn "## marc_repeatable_subfield = ", dump( $marc_repeatable_subfield ), $/ if ( $marc_repeatable_subfield ); |
warn "## marc_repeatable_subfield = ", dump( $marc_repeatable_subfield ), $/ if ( $marc_repeatable_subfield ); |
333 |
warn "## marc_record[$offset] = ", dump( $marc_rec ), $/; |
warn "## marc_record[$fetch_pos] = ", dump( $marc_rec ), $/; |
334 |
warn "## sorted_marc_record = ", dump( \@sorted_marc_record ), $/; |
warn "## sorted_marc_record = ", dump( \@sorted_marc_record ), $/; |
335 |
warn "## subfield count = ", dump( $u ), $/; |
warn "## subfield count = ", dump( $u ), $/; |
336 |
} |
} |
411 |
return \@m; |
return \@m; |
412 |
} |
} |
413 |
|
|
414 |
|
=head2 _get_marc_leader |
415 |
|
|
416 |
|
Return leader from currently fetched record by L</_get_marc_fields> |
417 |
|
|
418 |
|
print WebPAC::Normalize::_get_marc_leader(); |
419 |
|
|
420 |
|
=cut |
421 |
|
|
422 |
|
sub _get_marc_leader { |
423 |
|
die "no fetch_pos, did you called _get_marc_fields first?" unless ( defined( $fetch_pos ) ); |
424 |
|
return $marc_leader->[ $fetch_pos ]; |
425 |
|
} |
426 |
|
|
427 |
=head2 _debug |
=head2 _debug |
428 |
|
|
429 |
Change level of debug warnings |
Change level of debug warnings |
443 |
|
|
444 |
Those functions generally have to first in your normalization file. |
Those functions generally have to first in your normalization file. |
445 |
|
|
446 |
=head2 tag |
=head2 search_display |
447 |
|
|
448 |
Define new tag for I<search> and I<display>. |
Define output for L<search> and L<display> at the same time |
449 |
|
|
450 |
tag('Title', rec('200','a') ); |
search_display('Title', rec('200','a') ); |
451 |
|
|
452 |
|
|
453 |
=cut |
=cut |
454 |
|
|
455 |
sub tag { |
sub search_display { |
456 |
my $name = shift or die "tag needs name as first argument"; |
my $name = shift or die "search_display needs name as first argument"; |
457 |
my @o = grep { defined($_) && $_ ne '' } @_; |
my @o = grep { defined($_) && $_ ne '' } @_; |
458 |
return unless (@o); |
return unless (@o); |
|
$out->{$name}->{tag} = $name; |
|
459 |
$out->{$name}->{search} = \@o; |
$out->{$name}->{search} = \@o; |
460 |
$out->{$name}->{display} = \@o; |
$out->{$name}->{display} = \@o; |
461 |
} |
} |
462 |
|
|
463 |
=head2 display |
=head2 display |
464 |
|
|
465 |
Define tag just for I<display> |
Define output just for I<display> |
466 |
|
|
467 |
@v = display('Title', rec('200','a') ); |
@v = display('Title', rec('200','a') ); |
468 |
|
|
472 |
my $name = shift or die "display needs name as first argument"; |
my $name = shift or die "display needs name as first argument"; |
473 |
my @o = grep { defined($_) && $_ ne '' } @_; |
my @o = grep { defined($_) && $_ ne '' } @_; |
474 |
return unless (@o); |
return unless (@o); |
|
$out->{$name}->{tag} = $name; |
|
475 |
$out->{$name}->{display} = \@o; |
$out->{$name}->{display} = \@o; |
476 |
} |
} |
477 |
|
|
487 |
my $name = shift or die "search needs name as first argument"; |
my $name = shift or die "search needs name as first argument"; |
488 |
my @o = grep { defined($_) && $_ ne '' } @_; |
my @o = grep { defined($_) && $_ ne '' } @_; |
489 |
return unless (@o); |
return unless (@o); |
|
$out->{$name}->{tag} = $name; |
|
490 |
$out->{$name}->{search} = \@o; |
$out->{$name}->{search} = \@o; |
491 |
} |
} |
492 |
|
|
503 |
my ($offset,$value) = @_; |
my ($offset,$value) = @_; |
504 |
|
|
505 |
if ($offset) { |
if ($offset) { |
506 |
$leader->{ $offset } = $value; |
$marc_leader->[ $marc_record_offset ]->{ $offset } = $value; |
507 |
} else { |
} else { |
508 |
return $leader; |
|
509 |
|
if (defined($marc_leader)) { |
510 |
|
die "marc_leader not array = ", dump( $marc_leader ) unless (ref($marc_leader) eq 'ARRAY'); |
511 |
|
return $marc_leader->[ $marc_record_offset ]; |
512 |
|
} else { |
513 |
|
return; |
514 |
|
} |
515 |
|
} |
516 |
|
} |
517 |
|
|
518 |
|
=head2 marc_fixed |
519 |
|
|
520 |
|
Create control/indentifier fields with values in fixed positions |
521 |
|
|
522 |
|
marc_fixed('008', 00, '070402'); |
523 |
|
marc_fixed('008', 39, '|'); |
524 |
|
|
525 |
|
Positions not specified will be filled with spaces (C<0x20>). |
526 |
|
|
527 |
|
There will be no effort to extend last specified value to full length of |
528 |
|
field in standard. |
529 |
|
|
530 |
|
=cut |
531 |
|
|
532 |
|
sub marc_fixed { |
533 |
|
my ($f, $pos, $val) = @_; |
534 |
|
die "need marc(field, position, value)" unless defined($f) && defined($pos); |
535 |
|
|
536 |
|
confess "need val" unless defined $val; |
537 |
|
|
538 |
|
my $update = 0; |
539 |
|
|
540 |
|
map { |
541 |
|
if ($_->[0] eq $f) { |
542 |
|
my $old = $_->[1]; |
543 |
|
if (length($old) <= $pos) { |
544 |
|
$_->[1] .= ' ' x ( $pos - length($old) ) . $val; |
545 |
|
warn "## marc_fixed($f,$pos,'$val') append '$old' -> '$_->[1]'\n" if ($debug > 1); |
546 |
|
} else { |
547 |
|
$_->[1] = substr($old, 0, $pos) . $val . substr($old, $pos + length($val)); |
548 |
|
warn "## marc_fixed($f,$pos,'$val') update '$old' -> '$_->[1]'\n" if ($debug > 1); |
549 |
|
} |
550 |
|
$update++; |
551 |
|
} |
552 |
|
} @{ $marc_record->[ $marc_record_offset ] }; |
553 |
|
|
554 |
|
if (! $update) { |
555 |
|
my $v = ' ' x $pos . $val; |
556 |
|
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $v ]; |
557 |
|
warn "## marc_fixed($f,$pos,'val') created '$v'\n" if ($debug > 1); |
558 |
} |
} |
559 |
} |
} |
560 |
|
|
686 |
my $m = $marc_record->[ -1 ]; |
my $m = $marc_record->[ -1 ]; |
687 |
die "can't duplicate record which isn't defined" unless ($m); |
die "can't duplicate record which isn't defined" unless ($m); |
688 |
push @{ $marc_record }, dclone( $m ); |
push @{ $marc_record }, dclone( $m ); |
689 |
warn "## marc_duplicate = ", dump(@$marc_record), $/ if ($debug > 1); |
push @{ $marc_leader }, dclone( marc_leader() ); |
690 |
|
warn "## marc_duplicate = ", dump(@$marc_leader, @$marc_record), $/ if ($debug > 1); |
691 |
$marc_record_offset = $#{ $marc_record }; |
$marc_record_offset = $#{ $marc_record }; |
692 |
warn "## marc_record_offset = $marc_record_offset", $/ if ($debug > 1); |
warn "## marc_record_offset = $marc_record_offset", $/ if ($debug > 1); |
693 |
|
|
694 |
} |
} |
695 |
|
|
696 |
=head2 marc_remove |
=head2 marc_remove |
702 |
|
|
703 |
This will erase field C<200> or C<200^a> from current MARC record. |
This will erase field C<200> or C<200^a> from current MARC record. |
704 |
|
|
705 |
|
marc_remove('*'); |
706 |
|
|
707 |
|
Will remove all fields in current MARC record. |
708 |
|
|
709 |
This is useful after calling C<marc_duplicate> or on it's own (but, you |
This is useful after calling C<marc_duplicate> or on it's own (but, you |
710 |
should probably just remove that subfield definition if you are not |
should probably just remove that subfield definition if you are not |
711 |
using C<marc_duplicate>). |
using C<marc_duplicate>). |
723 |
|
|
724 |
warn "### marc_remove before = ", dump( $marc ), $/ if ($debug > 2); |
warn "### marc_remove before = ", dump( $marc ), $/ if ($debug > 2); |
725 |
|
|
726 |
my $i = 0; |
if ($f eq '*') { |
727 |
foreach ( 0 .. $#{ $marc } ) { |
|
728 |
last unless (defined $marc->[$i]); |
delete( $marc_record->[ $marc_record_offset ] ); |
729 |
warn "#### working on ",dump( @{ $marc->[$i] }), $/ if ($debug > 3); |
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
730 |
if ($marc->[$i]->[0] eq $f) { |
|
731 |
if (! defined $sf) { |
} else { |
732 |
# remove whole field |
|
733 |
splice @$marc, $i, 1; |
my $i = 0; |
734 |
warn "#### slice \@\$marc, $i, 1 = ",dump( @{ $marc }), $/ if ($debug > 3); |
foreach ( 0 .. $#{ $marc } ) { |
735 |
$i--; |
last unless (defined $marc->[$i]); |
736 |
} else { |
warn "#### working on ",dump( @{ $marc->[$i] }), $/ if ($debug > 3); |
737 |
foreach my $j ( 0 .. (( $#{ $marc->[$i] } - 3 ) / 2) ) { |
if ($marc->[$i]->[0] eq $f) { |
738 |
my $o = ($j * 2) + 3; |
if (! defined $sf) { |
739 |
if ($marc->[$i]->[$o] eq $sf) { |
# remove whole field |
740 |
# remove subfield |
splice @$marc, $i, 1; |
741 |
splice @{$marc->[$i]}, $o, 2; |
warn "#### slice \@\$marc, $i, 1 = ",dump( @{ $marc }), $/ if ($debug > 3); |
742 |
warn "#### slice \@{\$marc->[$i]}, $o, 2 = ", dump( @{ $marc }), $/ if ($debug > 3); |
$i--; |
743 |
# is record now empty? |
} else { |
744 |
if ($#{ $marc->[$i] } == 2) { |
foreach my $j ( 0 .. (( $#{ $marc->[$i] } - 3 ) / 2) ) { |
745 |
splice @$marc, $i, 1; |
my $o = ($j * 2) + 3; |
746 |
warn "#### slice \@\$marc, $i, 1 = ", dump( @{ $marc }), $/ if ($debug > 3); |
if ($marc->[$i]->[$o] eq $sf) { |
747 |
$i--; |
# remove subfield |
748 |
}; |
splice @{$marc->[$i]}, $o, 2; |
749 |
|
warn "#### slice \@{\$marc->[$i]}, $o, 2 = ", dump( @{ $marc }), $/ if ($debug > 3); |
750 |
|
# is record now empty? |
751 |
|
if ($#{ $marc->[$i] } == 2) { |
752 |
|
splice @$marc, $i, 1; |
753 |
|
warn "#### slice \@\$marc, $i, 1 = ", dump( @{ $marc }), $/ if ($debug > 3); |
754 |
|
$i--; |
755 |
|
}; |
756 |
|
} |
757 |
} |
} |
758 |
} |
} |
759 |
} |
} |
760 |
|
$i++; |
761 |
} |
} |
|
$i++; |
|
|
} |
|
762 |
|
|
763 |
warn "### marc_remove($f", $sf ? ",$sf" : "", ") after = ", dump( $marc ), $/ if ($debug > 2); |
warn "### marc_remove($f", $sf ? ",$sf" : "", ") after = ", dump( $marc ), $/ if ($debug > 2); |
764 |
|
|
765 |
$marc_record->[ $marc_record_offset ] = $marc; |
$marc_record->[ $marc_record_offset ] = $marc; |
766 |
|
} |
767 |
|
|
768 |
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
769 |
} |
} |
838 |
warn "## marc_record = ", dump( $marc_record ),$/ if ($debug > 1); |
warn "## marc_record = ", dump( $marc_record ),$/ if ($debug > 1); |
839 |
} |
} |
840 |
|
|
841 |
|
=head2 marc_count |
842 |
|
|
843 |
|
Return number of MARC records created using L</marc_duplicate>. |
844 |
|
|
845 |
|
print "created ", marc_count(), " records"; |
846 |
|
|
847 |
|
=cut |
848 |
|
|
849 |
|
sub marc_count { |
850 |
|
return $#{ $marc_record }; |
851 |
|
} |
852 |
|
|
853 |
|
|
854 |
=head1 Functions to extract data from input |
=head1 Functions to extract data from input |
855 |
|
|
872 |
|
|
873 |
my ($h,$include_subfields) = @_; |
my ($h,$include_subfields) = @_; |
874 |
|
|
875 |
|
# sanity and ease of use |
876 |
|
return $h if (ref($h) ne 'HASH'); |
877 |
|
|
878 |
if ( defined($h->{subfields}) ) { |
if ( defined($h->{subfields}) ) { |
879 |
my $sfs = delete $h->{subfields} || die "no subfields?"; |
my $sfs = delete $h->{subfields} || die "no subfields?"; |
880 |
my @out; |
my @out; |
1024 |
=cut |
=cut |
1025 |
|
|
1026 |
sub prefix { |
sub prefix { |
1027 |
my $p = shift or return; |
my $p = shift; |
1028 |
|
return @_ unless defined( $p ); |
1029 |
return map { $p . $_ } grep { defined($_) } @_; |
return map { $p . $_ } grep { defined($_) } @_; |
1030 |
} |
} |
1031 |
|
|
1038 |
=cut |
=cut |
1039 |
|
|
1040 |
sub suffix { |
sub suffix { |
1041 |
my $s = shift or die "suffix needs string as first argument"; |
my $s = shift; |
1042 |
|
return @_ unless defined( $s ); |
1043 |
return map { $_ . $s } grep { defined($_) } @_; |
return map { $_ . $s } grep { defined($_) } @_; |
1044 |
} |
} |
1045 |
|
|
1052 |
=cut |
=cut |
1053 |
|
|
1054 |
sub surround { |
sub surround { |
1055 |
my $p = shift or die "surround need prefix as first argument"; |
my $p = shift; |
1056 |
my $s = shift or die "surround needs suffix as second argument"; |
my $s = shift; |
1057 |
|
$p = '' unless defined( $p ); |
1058 |
|
$s = '' unless defined( $s ); |
1059 |
return map { $p . $_ . $s } grep { defined($_) } @_; |
return map { $p . $_ . $s } grep { defined($_) } @_; |
1060 |
} |
} |
1061 |
|
|
1120 |
sub lookup { |
sub lookup { |
1121 |
my ($what, $database, $input, $key, $having) = @_; |
my ($what, $database, $input, $key, $having) = @_; |
1122 |
|
|
1123 |
confess "lookup needs 5 arguments: what, database, input, key, having" unless ($#_ == 4); |
confess "lookup needs 5 arguments: what, database, input, key, having\n" unless ($#_ == 4); |
1124 |
|
|
1125 |
warn "## lookup ($database, $input, $key)", $/ if ($debug > 1); |
warn "## lookup ($database, $input, $key)", $/ if ($debug > 1); |
1126 |
return unless (defined($lookup->{$database}->{$input}->{$key})); |
return unless (defined($lookup->{$database}->{$input}->{$key})); |
1230 |
$database_code = config(); # use _ from hash |
$database_code = config(); # use _ from hash |
1231 |
$database_name = config('name'); |
$database_name = config('name'); |
1232 |
$database_input_name = config('input name'); |
$database_input_name = config('input name'); |
|
$tag = config('input normalize tag'); |
|
1233 |
|
|
1234 |
Up to three levels are supported. |
Up to three levels are supported. |
1235 |
|
|
1345 |
} |
} |
1346 |
} |
} |
1347 |
|
|
1348 |
|
my $hash; |
1349 |
|
|
1350 |
|
=head2 set |
1351 |
|
|
1352 |
|
set( key => 'value' ); |
1353 |
|
|
1354 |
|
=cut |
1355 |
|
|
1356 |
|
sub set { |
1357 |
|
my ($k,$v) = @_; |
1358 |
|
warn "## set ( $k => ", dump($v), " )", $/ if ( $debug ); |
1359 |
|
$hash->{$k} = $v; |
1360 |
|
}; |
1361 |
|
|
1362 |
|
=head2 get |
1363 |
|
|
1364 |
|
get( 'key' ); |
1365 |
|
|
1366 |
|
=cut |
1367 |
|
|
1368 |
|
sub get { |
1369 |
|
my $k = shift || return; |
1370 |
|
my $v = $hash->{$k}; |
1371 |
|
warn "## get $k = ", dump( $v ), $/ if ( $debug ); |
1372 |
|
return $v; |
1373 |
|
} |
1374 |
|
|
1375 |
|
=head2 count |
1376 |
|
|
1377 |
|
if ( count( @result ) == 1 ) { |
1378 |
|
# do something if only 1 result is there |
1379 |
|
} |
1380 |
|
|
1381 |
|
=cut |
1382 |
|
|
1383 |
|
sub count { |
1384 |
|
warn "## count ",dump(@_),$/ if ( $debug ); |
1385 |
|
return @_ . ''; |
1386 |
|
} |
1387 |
|
|
1388 |
# END |
# END |
1389 |
1; |
1; |