7 |
_debug |
_debug |
8 |
_pack_subfields_hash |
_pack_subfields_hash |
9 |
|
|
10 |
tag search display |
search_display search display sorted |
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_count |
marc_duplicate marc_remove marc_count |
15 |
marc_original_order |
marc_original_order |
16 |
|
|
17 |
rec1 rec2 rec |
rec1 rec2 rec |
18 |
|
frec |
19 |
regex prefix suffix surround |
regex prefix suffix surround |
20 |
first lookup join_with |
first lookup join_with |
21 |
save_into_lookup |
save_into_lookup |
42 |
|
|
43 |
WebPAC::Normalize - describe normalisaton rules using sets |
WebPAC::Normalize - describe normalisaton rules using sets |
44 |
|
|
|
=head1 VERSION |
|
|
|
|
|
Version 0.27 |
|
|
|
|
45 |
=cut |
=cut |
46 |
|
|
47 |
our $VERSION = '0.27'; |
our $VERSION = '0.31'; |
48 |
|
|
49 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
50 |
|
|
57 |
C<perl -c normalize.pl>. |
C<perl -c normalize.pl>. |
58 |
|
|
59 |
Normalisation can generate multiple output normalized data. For now, supported output |
Normalisation can generate multiple output normalized data. For now, supported output |
60 |
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 |
61 |
C<marc>. |
C<marc>. |
62 |
|
|
63 |
=head1 FUNCTIONS |
=head1 FUNCTIONS |
76 |
marc_encoding => 'utf-8', |
marc_encoding => 'utf-8', |
77 |
config => $config, |
config => $config, |
78 |
load_row_coderef => sub { |
load_row_coderef => sub { |
79 |
my ($database,$input,$mfn) = shift; |
my ($database,$input,$mfn) = @_; |
80 |
$store->load_row( database => $database, input => $input, id => $mfn ); |
$store->load_row( database => $database, input => $input, id => $mfn ); |
81 |
}, |
}, |
82 |
); |
); |
102 |
die "need row argument" unless ($arg->{row}); |
die "need row argument" unless ($arg->{row}); |
103 |
die "need normalisation argument" unless ($arg->{rules}); |
die "need normalisation argument" unless ($arg->{rules}); |
104 |
|
|
105 |
|
# FIXME load this conditionally |
106 |
|
# use WebPAC::Normalize::ISBN; |
107 |
|
|
108 |
no strict 'subs'; |
no strict 'subs'; |
109 |
_set_lookup( $arg->{lookup} ) if defined($arg->{lookup}); |
_set_lookup( $arg->{lookup} ) if defined($arg->{lookup}); |
110 |
_set_rec( $arg->{row} ); |
_set_rec( $arg->{row} ); |
172 |
my ($marc_record_offset, $marc_fetch_offset) = (0, 0); |
my ($marc_record_offset, $marc_fetch_offset) = (0, 0); |
173 |
|
|
174 |
sub _get_ds { |
sub _get_ds { |
175 |
|
|
176 |
|
warn "## out = ",dump($out); |
177 |
|
|
178 |
return $out; |
return $out; |
179 |
} |
} |
180 |
|
|
446 |
|
|
447 |
Those functions generally have to first in your normalization file. |
Those functions generally have to first in your normalization file. |
448 |
|
|
449 |
=head2 tag |
=head2 search_display |
450 |
|
|
451 |
Define new tag for I<search> and I<display>. |
Define output for L<search> and L<display> at the same time |
452 |
|
|
453 |
tag('Title', rec('200','a') ); |
search_display('Title', rec('200','a') ); |
454 |
|
|
455 |
|
|
456 |
=cut |
=cut |
457 |
|
|
458 |
sub tag { |
sub search_display { |
459 |
my $name = shift or die "tag needs name as first argument"; |
my $name = shift or die "search_display needs name as first argument"; |
460 |
my @o = grep { defined($_) && $_ ne '' } @_; |
my @o = grep { defined($_) && $_ ne '' } @_; |
461 |
return unless (@o); |
return unless (@o); |
|
$out->{$name}->{tag} = $name; |
|
462 |
$out->{$name}->{search} = \@o; |
$out->{$name}->{search} = \@o; |
463 |
$out->{$name}->{display} = \@o; |
$out->{$name}->{display} = \@o; |
464 |
} |
} |
465 |
|
|
466 |
|
=head2 tag |
467 |
|
|
468 |
|
Old name for L<search_display>, but supported |
469 |
|
|
470 |
|
=cut |
471 |
|
|
472 |
|
sub tag { |
473 |
|
search_display( @_ ); |
474 |
|
} |
475 |
|
|
476 |
=head2 display |
=head2 display |
477 |
|
|
478 |
Define tag just for I<display> |
Define output just for I<display> |
479 |
|
|
480 |
@v = display('Title', rec('200','a') ); |
@v = display('Title', rec('200','a') ); |
481 |
|
|
482 |
=cut |
=cut |
483 |
|
|
484 |
sub display { |
sub _field { |
485 |
my $name = shift or die "display needs name as first argument"; |
my $type = shift or confess "need type -- BUG?"; |
486 |
|
my $name = shift or confess "needs name as first argument"; |
487 |
my @o = grep { defined($_) && $_ ne '' } @_; |
my @o = grep { defined($_) && $_ ne '' } @_; |
488 |
return unless (@o); |
return unless (@o); |
489 |
$out->{$name}->{tag} = $name; |
$out->{$name}->{$type} = \@o; |
|
$out->{$name}->{display} = \@o; |
|
490 |
} |
} |
491 |
|
|
492 |
|
sub display { _field( 'display', @_ ) } |
493 |
|
|
494 |
=head2 search |
=head2 search |
495 |
|
|
496 |
Prepare values just for I<search> |
Prepare values just for I<search> |
499 |
|
|
500 |
=cut |
=cut |
501 |
|
|
502 |
sub search { |
sub search { _field( 'search', @_ ) } |
503 |
my $name = shift or die "search needs name as first argument"; |
|
504 |
my @o = grep { defined($_) && $_ ne '' } @_; |
=head2 sorted |
505 |
return unless (@o); |
|
506 |
$out->{$name}->{tag} = $name; |
Insert into lists which will be automatically sorted |
507 |
$out->{$name}->{search} = \@o; |
|
508 |
} |
sorted('Title', rec('200','a') ); |
509 |
|
|
510 |
|
=cut |
511 |
|
|
512 |
|
sub sorted { _field( 'sorted', @_ ) } |
513 |
|
|
514 |
|
|
515 |
=head2 marc_leader |
=head2 marc_leader |
516 |
|
|
537 |
} |
} |
538 |
} |
} |
539 |
|
|
540 |
|
=head2 marc_fixed |
541 |
|
|
542 |
|
Create control/indentifier fields with values in fixed positions |
543 |
|
|
544 |
|
marc_fixed('008', 00, '070402'); |
545 |
|
marc_fixed('008', 39, '|'); |
546 |
|
|
547 |
|
Positions not specified will be filled with spaces (C<0x20>). |
548 |
|
|
549 |
|
There will be no effort to extend last specified value to full length of |
550 |
|
field in standard. |
551 |
|
|
552 |
|
=cut |
553 |
|
|
554 |
|
sub marc_fixed { |
555 |
|
my ($f, $pos, $val) = @_; |
556 |
|
die "need marc(field, position, value)" unless defined($f) && defined($pos); |
557 |
|
|
558 |
|
confess "need val" unless defined $val; |
559 |
|
|
560 |
|
my $update = 0; |
561 |
|
|
562 |
|
map { |
563 |
|
if ($_->[0] eq $f) { |
564 |
|
my $old = $_->[1]; |
565 |
|
if (length($old) <= $pos) { |
566 |
|
$_->[1] .= ' ' x ( $pos - length($old) ) . $val; |
567 |
|
warn "## marc_fixed($f,$pos,'$val') append '$old' -> '$_->[1]'\n" if ($debug > 1); |
568 |
|
} else { |
569 |
|
$_->[1] = substr($old, 0, $pos) . $val . substr($old, $pos + length($val)); |
570 |
|
warn "## marc_fixed($f,$pos,'$val') update '$old' -> '$_->[1]'\n" if ($debug > 1); |
571 |
|
} |
572 |
|
$update++; |
573 |
|
} |
574 |
|
} @{ $marc_record->[ $marc_record_offset ] }; |
575 |
|
|
576 |
|
if (! $update) { |
577 |
|
my $v = ' ' x $pos . $val; |
578 |
|
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $v ]; |
579 |
|
warn "## marc_fixed($f,$pos,'val') created '$v'\n" if ($debug > 1); |
580 |
|
} |
581 |
|
} |
582 |
|
|
583 |
=head2 marc |
=head2 marc |
584 |
|
|
585 |
Save value for MARC field |
Save value for MARC field |
894 |
|
|
895 |
my ($h,$include_subfields) = @_; |
my ($h,$include_subfields) = @_; |
896 |
|
|
897 |
|
# sanity and ease of use |
898 |
|
return $h if (ref($h) ne 'HASH'); |
899 |
|
|
900 |
if ( defined($h->{subfields}) ) { |
if ( defined($h->{subfields}) ) { |
901 |
my $sfs = delete $h->{subfields} || die "no subfields?"; |
my $sfs = delete $h->{subfields} || die "no subfields?"; |
902 |
my @out; |
my @out; |
1001 |
|
|
1002 |
=cut |
=cut |
1003 |
|
|
1004 |
|
sub frec { |
1005 |
|
my @out = rec(@_); |
1006 |
|
warn "rec(",dump(@_),") has more than one return value, ignoring\n" if $#out > 0; |
1007 |
|
return shift @out; |
1008 |
|
} |
1009 |
|
|
1010 |
sub rec { |
sub rec { |
1011 |
my @out; |
my @out; |
1012 |
if ($#_ == 0) { |
if ($#_ == 0) { |
1052 |
=cut |
=cut |
1053 |
|
|
1054 |
sub prefix { |
sub prefix { |
1055 |
my $p = shift or return; |
my $p = shift; |
1056 |
|
return @_ unless defined( $p ); |
1057 |
return map { $p . $_ } grep { defined($_) } @_; |
return map { $p . $_ } grep { defined($_) } @_; |
1058 |
} |
} |
1059 |
|
|
1066 |
=cut |
=cut |
1067 |
|
|
1068 |
sub suffix { |
sub suffix { |
1069 |
my $s = shift or die "suffix needs string as first argument"; |
my $s = shift; |
1070 |
|
return @_ unless defined( $s ); |
1071 |
return map { $_ . $s } grep { defined($_) } @_; |
return map { $_ . $s } grep { defined($_) } @_; |
1072 |
} |
} |
1073 |
|
|
1080 |
=cut |
=cut |
1081 |
|
|
1082 |
sub surround { |
sub surround { |
1083 |
my $p = shift or die "surround need prefix as first argument"; |
my $p = shift; |
1084 |
my $s = shift or die "surround needs suffix as second argument"; |
my $s = shift; |
1085 |
|
$p = '' unless defined( $p ); |
1086 |
|
$s = '' unless defined( $s ); |
1087 |
return map { $p . $_ . $s } grep { defined($_) } @_; |
return map { $p . $_ . $s } grep { defined($_) } @_; |
1088 |
} |
} |
1089 |
|
|
1258 |
$database_code = config(); # use _ from hash |
$database_code = config(); # use _ from hash |
1259 |
$database_name = config('name'); |
$database_name = config('name'); |
1260 |
$database_input_name = config('input name'); |
$database_input_name = config('input name'); |
|
$tag = config('input normalize tag'); |
|
1261 |
|
|
1262 |
Up to three levels are supported. |
Up to three levels are supported. |
1263 |
|
|