1 |
package WebPAC::Normalize; |
package WebPAC::Normalize; |
2 |
use Exporter 'import'; |
use Exporter 'import'; |
3 |
@EXPORT = qw/ |
our @EXPORT = qw/ |
4 |
_set_rec _set_lookup |
_set_ds _set_lookup |
5 |
_set_load_row |
_set_load_row |
6 |
_get_ds _clean_ds |
_get_ds _clean_ds |
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_duplicate marc_remove marc_count |
15 |
marc_original_order |
marc_original_order |
16 |
|
marc_template |
17 |
|
|
18 |
rec1 rec2 rec |
rec1 rec2 rec |
19 |
|
frec frec_eq frec_ne |
20 |
regex prefix suffix surround |
regex prefix suffix surround |
21 |
first lookup join_with |
first lookup join_with |
22 |
save_into_lookup |
save_into_lookup |
23 |
|
|
24 |
split_rec_on |
split_rec_on |
25 |
|
|
26 |
|
get set |
27 |
|
count |
28 |
|
|
29 |
/; |
/; |
30 |
|
|
31 |
use warnings; |
use warnings; |
39 |
# debugging warn(s) |
# debugging warn(s) |
40 |
my $debug = 0; |
my $debug = 0; |
41 |
|
|
42 |
|
use WebPAC::Normalize::ISBN; |
43 |
|
push @EXPORT, ( 'isbn_10', 'isbn_13' ); |
44 |
|
|
45 |
=head1 NAME |
=head1 NAME |
46 |
|
|
47 |
WebPAC::Normalize - describe normalisaton rules using sets |
WebPAC::Normalize - describe normalisaton rules using sets |
48 |
|
|
|
=head1 VERSION |
|
|
|
|
|
Version 0.23 |
|
|
|
|
49 |
=cut |
=cut |
50 |
|
|
51 |
our $VERSION = '0.23'; |
our $VERSION = '0.34'; |
52 |
|
|
53 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
54 |
|
|
61 |
C<perl -c normalize.pl>. |
C<perl -c normalize.pl>. |
62 |
|
|
63 |
Normalisation can generate multiple output normalized data. For now, supported output |
Normalisation can generate multiple output normalized data. For now, supported output |
64 |
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 |
65 |
C<marc>. |
C<marc>. |
66 |
|
|
67 |
=head1 FUNCTIONS |
=head1 FUNCTIONS |
80 |
marc_encoding => 'utf-8', |
marc_encoding => 'utf-8', |
81 |
config => $config, |
config => $config, |
82 |
load_row_coderef => sub { |
load_row_coderef => sub { |
83 |
my ($database,$input,$mfn) = shift; |
my ($database,$input,$mfn) = @_; |
84 |
$store->load_row( database => $database, input => $input, id => $mfn ); |
$store->load_row( database => $database, input => $input, id => $mfn ); |
85 |
}, |
}, |
86 |
); |
); |
106 |
die "need row argument" unless ($arg->{row}); |
die "need row argument" unless ($arg->{row}); |
107 |
die "need normalisation argument" unless ($arg->{rules}); |
die "need normalisation argument" unless ($arg->{rules}); |
108 |
|
|
|
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_ds( $arg->{row} ); |
111 |
_set_config( $arg->{config} ) if defined($arg->{config}); |
_set_config( $arg->{config} ) if defined($arg->{config}); |
112 |
_clean_ds( %{ $arg } ); |
_clean_ds( %{ $arg } ); |
113 |
$load_row_coderef = $arg->{load_row_coderef}; |
$load_row_coderef = $arg->{load_row_coderef}; |
114 |
|
|
115 |
eval "$arg->{rules}"; |
no strict 'subs'; |
116 |
|
no warnings 'redefine'; |
117 |
|
eval "$arg->{rules};"; |
118 |
die "error evaling $arg->{rules}: $@\n" if ($@); |
die "error evaling $arg->{rules}: $@\n" if ($@); |
119 |
|
|
120 |
return _get_ds(); |
return _get_ds(); |
121 |
} |
} |
122 |
|
|
123 |
=head2 _set_rec |
=head2 _set_ds |
124 |
|
|
125 |
Set current record hash |
Set current record hash |
126 |
|
|
127 |
_set_rec( $rec ); |
_set_ds( $rec ); |
128 |
|
|
129 |
=cut |
=cut |
130 |
|
|
131 |
my $rec; |
my $rec; |
132 |
|
|
133 |
sub _set_rec { |
sub _set_ds { |
134 |
$rec = shift or die "no record hash"; |
$rec = shift or die "no record hash"; |
135 |
} |
} |
136 |
|
|
170 |
|
|
171 |
=cut |
=cut |
172 |
|
|
173 |
my ($out, $marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $leader); |
my ($out, $marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $marc_leader); |
174 |
my ($marc_record_offset, $marc_fetch_offset) = (0, 0); |
my ($marc_record_offset, $marc_fetch_offset) = (0, 0); |
175 |
|
|
176 |
sub _get_ds { |
sub _get_ds { |
177 |
|
#warn "## out = ",dump($out); |
178 |
return $out; |
return $out; |
179 |
} |
} |
180 |
|
|
188 |
|
|
189 |
sub _clean_ds { |
sub _clean_ds { |
190 |
my $a = {@_}; |
my $a = {@_}; |
191 |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $leader) = (); |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $marc_leader) = (); |
192 |
($marc_record_offset, $marc_fetch_offset) = (0,0); |
($marc_record_offset, $marc_fetch_offset) = (0,0); |
193 |
$marc_encoding = $a->{marc_encoding}; |
$marc_encoding = $a->{marc_encoding}; |
194 |
} |
} |
293 |
|
|
294 |
=cut |
=cut |
295 |
|
|
296 |
|
my $fetch_pos; |
297 |
|
|
298 |
sub _get_marc_fields { |
sub _get_marc_fields { |
299 |
|
|
300 |
my $arg = {@_}; |
my $arg = {@_}; |
301 |
warn "### _get_marc_fields arg: ", dump($arg), $/ if ($debug > 2); |
warn "### _get_marc_fields arg: ", dump($arg), $/ if ($debug > 2); |
302 |
my $offset = $marc_fetch_offset; |
$fetch_pos = $marc_fetch_offset; |
303 |
if ($arg->{offset}) { |
if ($arg->{offset}) { |
304 |
$offset = $arg->{offset}; |
$fetch_pos = $arg->{offset}; |
305 |
} elsif($arg->{fetch_next}) { |
} elsif($arg->{fetch_next}) { |
306 |
$marc_fetch_offset++; |
$marc_fetch_offset++; |
307 |
} |
} |
310 |
|
|
311 |
warn "### full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 2); |
warn "### full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 2); |
312 |
|
|
313 |
my $marc_rec = $marc_record->[ $offset ]; |
my $marc_rec = $marc_record->[ $fetch_pos ]; |
314 |
|
|
315 |
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); |
316 |
|
|
317 |
return if (! $marc_rec || ref($marc_rec) ne 'ARRAY' || $#{ $marc_rec } < 0); |
return if (! $marc_rec || ref($marc_rec) ne 'ARRAY' || $#{ $marc_rec } < 0); |
318 |
|
|
333 |
|
|
334 |
if ($debug) { |
if ($debug) { |
335 |
warn "## marc_repeatable_subfield = ", dump( $marc_repeatable_subfield ), $/ if ( $marc_repeatable_subfield ); |
warn "## marc_repeatable_subfield = ", dump( $marc_repeatable_subfield ), $/ if ( $marc_repeatable_subfield ); |
336 |
warn "## marc_record[$offset] = ", dump( $marc_rec ), $/; |
warn "## marc_record[$fetch_pos] = ", dump( $marc_rec ), $/; |
337 |
warn "## sorted_marc_record = ", dump( \@sorted_marc_record ), $/; |
warn "## sorted_marc_record = ", dump( \@sorted_marc_record ), $/; |
338 |
warn "## subfield count = ", dump( $u ), $/; |
warn "## subfield count = ", dump( $u ), $/; |
339 |
} |
} |
414 |
return \@m; |
return \@m; |
415 |
} |
} |
416 |
|
|
417 |
|
=head2 _get_marc_leader |
418 |
|
|
419 |
|
Return leader from currently fetched record by L</_get_marc_fields> |
420 |
|
|
421 |
|
print WebPAC::Normalize::_get_marc_leader(); |
422 |
|
|
423 |
|
=cut |
424 |
|
|
425 |
|
sub _get_marc_leader { |
426 |
|
die "no fetch_pos, did you called _get_marc_fields first?" unless ( defined( $fetch_pos ) ); |
427 |
|
return $marc_leader->[ $fetch_pos ]; |
428 |
|
} |
429 |
|
|
430 |
=head2 _debug |
=head2 _debug |
431 |
|
|
432 |
Change level of debug warnings |
Change level of debug warnings |
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 |
|
|
525 |
my ($offset,$value) = @_; |
my ($offset,$value) = @_; |
526 |
|
|
527 |
if ($offset) { |
if ($offset) { |
528 |
$leader->{ $offset } = $value; |
$marc_leader->[ $marc_record_offset ]->{ $offset } = $value; |
529 |
} else { |
} else { |
530 |
return $leader; |
|
531 |
|
if (defined($marc_leader)) { |
532 |
|
die "marc_leader not array = ", dump( $marc_leader ) unless (ref($marc_leader) eq 'ARRAY'); |
533 |
|
return $marc_leader->[ $marc_record_offset ]; |
534 |
|
} else { |
535 |
|
return; |
536 |
|
} |
537 |
|
} |
538 |
|
} |
539 |
|
|
540 |
|
=head2 marc_fixed |
541 |
|
|
542 |
|
Create control/indentifier fields with values in fixed positions |
543 |
|
|
544 |
|
marc_fixed('008', 00, '070402'); |
545 |
|
marc_fixed('008', 39, '|'); |
546 |
|
|
547 |
|
Positions not specified will be filled with spaces (C<0x20>). |
548 |
|
|
549 |
|
There will be no effort to extend last specified value to full length of |
550 |
|
field in standard. |
551 |
|
|
552 |
|
=cut |
553 |
|
|
554 |
|
sub marc_fixed { |
555 |
|
my ($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 |
|
|
708 |
my $m = $marc_record->[ -1 ]; |
my $m = $marc_record->[ -1 ]; |
709 |
die "can't duplicate record which isn't defined" unless ($m); |
die "can't duplicate record which isn't defined" unless ($m); |
710 |
push @{ $marc_record }, dclone( $m ); |
push @{ $marc_record }, dclone( $m ); |
711 |
warn "## marc_duplicate = ", dump(@$marc_record), $/ if ($debug > 1); |
push @{ $marc_leader }, dclone( marc_leader() ); |
712 |
|
warn "## marc_duplicate = ", dump(@$marc_leader, @$marc_record), $/ if ($debug > 1); |
713 |
$marc_record_offset = $#{ $marc_record }; |
$marc_record_offset = $#{ $marc_record }; |
714 |
warn "## marc_record_offset = $marc_record_offset", $/ if ($debug > 1); |
warn "## marc_record_offset = $marc_record_offset", $/ if ($debug > 1); |
715 |
|
|
716 |
} |
} |
717 |
|
|
718 |
=head2 marc_remove |
=head2 marc_remove |
724 |
|
|
725 |
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. |
726 |
|
|
727 |
|
marc_remove('*'); |
728 |
|
|
729 |
|
Will remove all fields in current MARC record. |
730 |
|
|
731 |
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 |
732 |
should probably just remove that subfield definition if you are not |
should probably just remove that subfield definition if you are not |
733 |
using C<marc_duplicate>). |
using C<marc_duplicate>). |
745 |
|
|
746 |
warn "### marc_remove before = ", dump( $marc ), $/ if ($debug > 2); |
warn "### marc_remove before = ", dump( $marc ), $/ if ($debug > 2); |
747 |
|
|
748 |
my $i = 0; |
if ($f eq '*') { |
749 |
foreach ( 0 .. $#{ $marc } ) { |
|
750 |
last unless (defined $marc->[$i]); |
delete( $marc_record->[ $marc_record_offset ] ); |
751 |
warn "#### working on ",dump( @{ $marc->[$i] }), $/ if ($debug > 3); |
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
752 |
if ($marc->[$i]->[0] eq $f) { |
|
753 |
if (! defined $sf) { |
} else { |
754 |
# remove whole field |
|
755 |
splice @$marc, $i, 1; |
my $i = 0; |
756 |
warn "#### slice \@\$marc, $i, 1 = ",dump( @{ $marc }), $/ if ($debug > 3); |
foreach ( 0 .. $#{ $marc } ) { |
757 |
$i--; |
last unless (defined $marc->[$i]); |
758 |
} else { |
warn "#### working on ",dump( @{ $marc->[$i] }), $/ if ($debug > 3); |
759 |
foreach my $j ( 0 .. (( $#{ $marc->[$i] } - 3 ) / 2) ) { |
if ($marc->[$i]->[0] eq $f) { |
760 |
my $o = ($j * 2) + 3; |
if (! defined $sf) { |
761 |
if ($marc->[$i]->[$o] eq $sf) { |
# remove whole field |
762 |
# remove subfield |
splice @$marc, $i, 1; |
763 |
splice @{$marc->[$i]}, $o, 2; |
warn "#### slice \@\$marc, $i, 1 = ",dump( @{ $marc }), $/ if ($debug > 3); |
764 |
warn "#### slice \@{\$marc->[$i]}, $o, 2 = ", dump( @{ $marc }), $/ if ($debug > 3); |
$i--; |
765 |
# is record now empty? |
} else { |
766 |
if ($#{ $marc->[$i] } == 2) { |
foreach my $j ( 0 .. (( $#{ $marc->[$i] } - 3 ) / 2) ) { |
767 |
splice @$marc, $i, 1; |
my $o = ($j * 2) + 3; |
768 |
warn "#### slice \@\$marc, $i, 1 = ", dump( @{ $marc }), $/ if ($debug > 3); |
if ($marc->[$i]->[$o] eq $sf) { |
769 |
$i--; |
# remove subfield |
770 |
}; |
splice @{$marc->[$i]}, $o, 2; |
771 |
|
warn "#### slice \@{\$marc->[$i]}, $o, 2 = ", dump( @{ $marc }), $/ if ($debug > 3); |
772 |
|
# is record now empty? |
773 |
|
if ($#{ $marc->[$i] } == 2) { |
774 |
|
splice @$marc, $i, 1; |
775 |
|
warn "#### slice \@\$marc, $i, 1 = ", dump( @{ $marc }), $/ if ($debug > 3); |
776 |
|
$i--; |
777 |
|
}; |
778 |
|
} |
779 |
} |
} |
780 |
} |
} |
781 |
} |
} |
782 |
|
$i++; |
783 |
} |
} |
|
$i++; |
|
|
} |
|
784 |
|
|
785 |
warn "### marc_remove($f", $sf ? ",$sf" : "", ") after = ", dump( $marc ), $/ if ($debug > 2); |
warn "### marc_remove($f", $sf ? ",$sf" : "", ") after = ", dump( $marc ), $/ if ($debug > 2); |
786 |
|
|
787 |
$marc_record->[ $marc_record_offset ] = $marc; |
$marc_record->[ $marc_record_offset ] = $marc; |
788 |
|
} |
789 |
|
|
790 |
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
791 |
} |
} |
815 |
return unless defined($rec->{$from}); |
return unless defined($rec->{$from}); |
816 |
|
|
817 |
my $r = $rec->{$from}; |
my $r = $rec->{$from}; |
818 |
die "record field $from isn't array\n" unless (ref($r) eq 'ARRAY'); |
die "record field $from isn't array ",dump( $rec ) unless (ref($r) eq 'ARRAY'); |
819 |
|
|
820 |
my ($i1,$i2) = defined($marc_indicators->{$to}) ? @{ $marc_indicators->{$to} } : (' ',' '); |
my ($i1,$i2) = defined($marc_indicators->{$to}) ? @{ $marc_indicators->{$to} } : (' ',' '); |
821 |
warn "## marc_original_order($to,$from) source = ", dump( $r ),$/ if ($debug > 1); |
warn "## marc_original_order($to,$from) source = ", dump( $r ),$/ if ($debug > 1); |
860 |
warn "## marc_record = ", dump( $marc_record ),$/ if ($debug > 1); |
warn "## marc_record = ", dump( $marc_record ),$/ if ($debug > 1); |
861 |
} |
} |
862 |
|
|
863 |
|
=head2 marc_template |
864 |
|
|
865 |
|
=cut |
866 |
|
|
867 |
|
sub marc_template { |
868 |
|
my $args = {@_}; |
869 |
|
warn "## marc_template(",dump($args),")"; |
870 |
|
|
871 |
|
foreach ( qw/subfields_rename marc_template/ ) { |
872 |
|
# warn "ref($_) = ",ref($args->{$_}); |
873 |
|
die "$_ not ARRAY" if ref($args->{$_}) ne 'ARRAY'; |
874 |
|
} |
875 |
|
|
876 |
|
my $r = $rec->{ $args->{from} } || return; |
877 |
|
die "record field ", $args->{from}, " isn't array ",dump( $rec ) unless (ref($r) eq 'ARRAY'); |
878 |
|
|
879 |
|
my @subfields_rename = @{ $args->{subfields_rename} }; |
880 |
|
# warn "### subfields_rename [$#subfields_rename] = ",dump( @subfields_rename ); |
881 |
|
|
882 |
|
confess "need mapping in pairs for subfields_rename" |
883 |
|
if $#subfields_rename % 2 != 1; |
884 |
|
|
885 |
|
my ( $subfields_rename, $from_subfields, $to_subfields ); |
886 |
|
while ( my ( $from, $to ) = splice(@subfields_rename, 0, 2) ) { |
887 |
|
my ( $f, $t ) = ( |
888 |
|
$from_subfields->{ $from }++, |
889 |
|
$to_subfields->{ $to }++ |
890 |
|
); |
891 |
|
$subfields_rename->{ $from }->[ $f ] = [ $to => $t ]; |
892 |
|
} |
893 |
|
warn "### subfields_rename = ",dump( $subfields_rename ),$/; |
894 |
|
warn "### from_subfields = ", dump( $from_subfields ),$/; |
895 |
|
warn "### to_subfields = ", dump( $to_subfields ),$/; |
896 |
|
|
897 |
|
my $fields_re = join('|', keys %$to_subfields ); |
898 |
|
|
899 |
|
my $pos_templates; |
900 |
|
my $count; |
901 |
|
my @marc_order; |
902 |
|
my $marc_template_order; |
903 |
|
my $fill_in; |
904 |
|
my @marc_out; |
905 |
|
|
906 |
|
foreach my $template ( @{ $args->{marc_template} } ) { |
907 |
|
$count = {}; |
908 |
|
@marc_order = (); |
909 |
|
sub my_count { |
910 |
|
my $sf = shift; |
911 |
|
my $nr = $count->{$sf}++; |
912 |
|
push @marc_order, [ $sf, $nr ]; |
913 |
|
return $sf . $nr; |
914 |
|
} |
915 |
|
my $pos_template = $template; |
916 |
|
$pos_template =~ s/($fields_re)/my_count($1)/ge; |
917 |
|
my $count_key = dump( $count ); |
918 |
|
warn "### template: |$template| -> |$pos_template| count = $count_key marc_order = ",dump( @marc_order ),$/; |
919 |
|
$pos_templates->{ $count_key } = $pos_template; |
920 |
|
$marc_template_order->{ $pos_template } = [ @marc_order ]; |
921 |
|
} |
922 |
|
warn "### from ",dump( $args->{marc_template} ), " created ", dump( $pos_templates ), " and ", dump( $marc_template_order ); |
923 |
|
|
924 |
|
my $m; |
925 |
|
|
926 |
|
foreach my $r ( @{ $rec->{ $args->{from} } } ) { |
927 |
|
|
928 |
|
my $i1 = $r->{i1} || ' '; |
929 |
|
my $i2 = $r->{i2} || ' '; |
930 |
|
$m = [ $args->{to}, $i1, $i2 ]; |
931 |
|
|
932 |
|
warn "### r = ",dump( $r ); |
933 |
|
|
934 |
|
my ( $new_r, $from_count, $to_count ); |
935 |
|
foreach my $sf ( keys %{$r} ) { |
936 |
|
my $nr = $from_count->{$sf}++; |
937 |
|
my $rename_to = $subfields_rename->{ $sf }; # || |
938 |
|
# die "can't find subfield rename for $sf/$nr in ", dump( $subfields_rename ); |
939 |
|
warn "### rename $sf/$nr to ", dump( $rename_to->[$nr] ), $/; |
940 |
|
my ( $to_sf, $to_nr ) = @{ $rename_to->[$nr] }; |
941 |
|
$new_r->{ $to_sf }->[ $to_nr ] = [ $sf => $nr ]; |
942 |
|
|
943 |
|
$to_count->{ $to_sf }++; |
944 |
|
} |
945 |
|
|
946 |
|
warn "### new_r = ",dump( $new_r ); |
947 |
|
|
948 |
|
my $from_count_key = dump( $to_count ); |
949 |
|
|
950 |
|
warn "### from_count = ",dump( $from_count ), $/; |
951 |
|
warn "### to_count = ",dump( $to_count ), $/; |
952 |
|
|
953 |
|
my $template = $pos_templates->{ $from_count_key } || |
954 |
|
die "I don't have template for:\n$from_count_key\n## available templates\n", dump( $pos_templates ); |
955 |
|
|
956 |
|
warn "### selected template: |$template|\n"; |
957 |
|
|
958 |
|
$fill_in = {}; |
959 |
|
|
960 |
|
foreach my $sf ( split(/\|/, $template ) ) { |
961 |
|
sub fill_in { |
962 |
|
my ( $r, $sf, $nr ) = @_; |
963 |
|
my ( $from_sf, $from_nr ) = @{ $new_r->{ $sf }->[ $nr ] }; |
964 |
|
my $v = $r->{ $from_sf }; # || die "no $from_sf/$from_nr"; |
965 |
|
warn "#### fill_in( $sf, $nr ) = $from_sf/$from_nr >>>> ",dump( $v ), $/; |
966 |
|
if ( ref( $v ) eq 'ARRAY' ) { |
967 |
|
$fill_in->{$sf}->[$nr] = $v->[$from_nr]; |
968 |
|
return $v->[$from_nr]; |
969 |
|
} elsif ( $from_nr == 0 ) { |
970 |
|
$fill_in->{$sf}->[$nr] = $v; |
971 |
|
return $v; |
972 |
|
} else { |
973 |
|
die "requested subfield $from_sf/$from_nr but it's ",dump( $v ); |
974 |
|
} |
975 |
|
} |
976 |
|
warn "#### $sf <<<< $fields_re\n"; |
977 |
|
$sf =~ s/($fields_re)(\d+)/fill_in($r,$1,$2)/ge; |
978 |
|
warn "#### >>>> $sf with fill_in = ",dump( $fill_in ),$/; |
979 |
|
} |
980 |
|
|
981 |
|
warn "## template: |$template|\n## marc_template_order = ",dump( $marc_template_order ); |
982 |
|
|
983 |
|
foreach my $sf ( @{ $marc_template_order->{$template} } ) { |
984 |
|
my ( $sf, $nr ) = @$sf; |
985 |
|
my $v = $fill_in->{$sf}->[$nr] || die "can't find fill_in $sf/$nr"; |
986 |
|
warn "++ $sf/$nr |$v|\n"; |
987 |
|
push @$m, ( $sf, $v ); |
988 |
|
} |
989 |
|
|
990 |
|
warn "#### >>>> created marc: ", dump( $m ); |
991 |
|
|
992 |
|
push @marc_out, $m; |
993 |
|
} |
994 |
|
|
995 |
|
warn "### marc_template produced: ",dump( @marc_out ); |
996 |
|
|
997 |
|
foreach my $marc ( @marc_out ) { |
998 |
|
warn "+++ ",dump( $marc ); |
999 |
|
push @{ $marc_record->[ $marc_record_offset ] }, $marc; |
1000 |
|
} |
1001 |
|
} |
1002 |
|
|
1003 |
|
=head2 marc_count |
1004 |
|
|
1005 |
|
Return number of MARC records created using L</marc_duplicate>. |
1006 |
|
|
1007 |
|
print "created ", marc_count(), " records"; |
1008 |
|
|
1009 |
|
=cut |
1010 |
|
|
1011 |
|
sub marc_count { |
1012 |
|
return $#{ $marc_record }; |
1013 |
|
} |
1014 |
|
|
1015 |
|
|
1016 |
=head1 Functions to extract data from input |
=head1 Functions to extract data from input |
1017 |
|
|
1034 |
|
|
1035 |
my ($h,$include_subfields) = @_; |
my ($h,$include_subfields) = @_; |
1036 |
|
|
1037 |
|
# sanity and ease of use |
1038 |
|
return $h if (ref($h) ne 'HASH'); |
1039 |
|
|
1040 |
if ( defined($h->{subfields}) ) { |
if ( defined($h->{subfields}) ) { |
1041 |
my $sfs = delete $h->{subfields} || die "no subfields?"; |
my $sfs = delete $h->{subfields} || die "no subfields?"; |
1042 |
my @out; |
my @out; |
1157 |
} |
} |
1158 |
} |
} |
1159 |
|
|
1160 |
|
=head2 frec |
1161 |
|
|
1162 |
|
Returns first value from field |
1163 |
|
|
1164 |
|
$v = frec('200'); |
1165 |
|
$v = frec('200','a'); |
1166 |
|
|
1167 |
|
=cut |
1168 |
|
|
1169 |
|
sub frec { |
1170 |
|
my @out = rec(@_); |
1171 |
|
warn "rec(",dump(@_),") has more than one return value, ignoring\n" if $#out > 0; |
1172 |
|
return shift @out; |
1173 |
|
} |
1174 |
|
|
1175 |
|
=head2 frec_eq |
1176 |
|
|
1177 |
|
=head2 frec_ne |
1178 |
|
|
1179 |
|
Check if first values from two fields are same or different |
1180 |
|
|
1181 |
|
if ( frec_eq( 900 => 'a', 910 => 'c' ) ) { |
1182 |
|
# values are same |
1183 |
|
} else { |
1184 |
|
# values are different |
1185 |
|
} |
1186 |
|
|
1187 |
|
Strictly speaking C<frec_eq> and C<frec_ne> wouldn't be needed if you |
1188 |
|
could write something like: |
1189 |
|
|
1190 |
|
if ( frec( '900','a' ) eq frec( '910','c' ) ) { |
1191 |
|
# yada tada |
1192 |
|
} |
1193 |
|
|
1194 |
|
but you can't since our parser L<WebPAC::Parser> will remove all whitespaces |
1195 |
|
in order to parse text and create invalid function C<eqfrec>. |
1196 |
|
|
1197 |
|
=cut |
1198 |
|
|
1199 |
|
sub frec_eq { |
1200 |
|
my ( $f1,$sf1, $f2, $sf2 ) = @_; |
1201 |
|
return (rec( $f1, $sf1 ))[0] eq (rec( $f2, $sf2 ))[0]; |
1202 |
|
} |
1203 |
|
|
1204 |
|
sub frec_ne { |
1205 |
|
return ! frec_eq( @_ ); |
1206 |
|
} |
1207 |
|
|
1208 |
=head2 regex |
=head2 regex |
1209 |
|
|
1210 |
Apply regex to some or all values |
Apply regex to some or all values |
1234 |
=cut |
=cut |
1235 |
|
|
1236 |
sub prefix { |
sub prefix { |
1237 |
my $p = shift or return; |
my $p = shift; |
1238 |
|
return @_ unless defined( $p ); |
1239 |
return map { $p . $_ } grep { defined($_) } @_; |
return map { $p . $_ } grep { defined($_) } @_; |
1240 |
} |
} |
1241 |
|
|
1248 |
=cut |
=cut |
1249 |
|
|
1250 |
sub suffix { |
sub suffix { |
1251 |
my $s = shift or die "suffix needs string as first argument"; |
my $s = shift; |
1252 |
|
return @_ unless defined( $s ); |
1253 |
return map { $_ . $s } grep { defined($_) } @_; |
return map { $_ . $s } grep { defined($_) } @_; |
1254 |
} |
} |
1255 |
|
|
1262 |
=cut |
=cut |
1263 |
|
|
1264 |
sub surround { |
sub surround { |
1265 |
my $p = shift or die "surround need prefix as first argument"; |
my $p = shift; |
1266 |
my $s = shift or die "surround needs suffix as second argument"; |
my $s = shift; |
1267 |
|
$p = '' unless defined( $p ); |
1268 |
|
$s = '' unless defined( $s ); |
1269 |
return map { $p . $_ . $s } grep { defined($_) } @_; |
return map { $p . $_ . $s } grep { defined($_) } @_; |
1270 |
} |
} |
1271 |
|
|
1330 |
sub lookup { |
sub lookup { |
1331 |
my ($what, $database, $input, $key, $having) = @_; |
my ($what, $database, $input, $key, $having) = @_; |
1332 |
|
|
1333 |
confess "lookup needs 5 arguments: what, database, input, key, having" unless ($#_ == 4); |
confess "lookup needs 5 arguments: what, database, input, key, having\n" unless ($#_ == 4); |
1334 |
|
|
1335 |
warn "## lookup ($database, $input, $key)", $/ if ($debug > 1); |
warn "## lookup ($database, $input, $key)", $/ if ($debug > 1); |
1336 |
return unless (defined($lookup->{$database}->{$input}->{$key})); |
return unless (defined($lookup->{$database}->{$input}->{$key})); |
1344 |
|
|
1345 |
foreach my $h ( @having ) { |
foreach my $h ( @having ) { |
1346 |
if (defined($lookup->{$database}->{$input}->{$key}->{$h})) { |
if (defined($lookup->{$database}->{$input}->{$key}->{$h})) { |
1347 |
warn "lookup for $database/$input/$key/$h return ",dump($lookup->{$database}->{$input}->{$key}->{$h}),"\n"; |
warn "lookup for $database/$input/$key/$h return ",dump($lookup->{$database}->{$input}->{$key}->{$h}),"\n" if ($debug); |
1348 |
$mfns->{$_}++ foreach keys %{ $lookup->{$database}->{$input}->{$key}->{$h} }; |
$mfns->{$_}++ foreach keys %{ $lookup->{$database}->{$input}->{$key}->{$h} }; |
1349 |
} |
} |
1350 |
} |
} |
1361 |
foreach my $mfn (@mfns) { |
foreach my $mfn (@mfns) { |
1362 |
$rec = $load_row_coderef->( $database, $input, $mfn ); |
$rec = $load_row_coderef->( $database, $input, $mfn ); |
1363 |
|
|
1364 |
warn "got $database/$input/$mfn = ", dump($rec), $/; |
warn "got $database/$input/$mfn = ", dump($rec), $/ if ($debug); |
1365 |
|
|
1366 |
my @vals = $what->(); |
my @vals = $what->(); |
1367 |
|
|
1368 |
push @out, ( @vals ); |
push @out, ( @vals ); |
1369 |
|
|
1370 |
warn "lookup for mfn $mfn returned ", dump(@vals), $/; |
warn "lookup for mfn $mfn returned ", dump(@vals), $/ if ($debug); |
1371 |
} |
} |
1372 |
|
|
1373 |
# if (ref($lookup->{$k}) eq 'ARRAY') { |
# if (ref($lookup->{$k}) eq 'ARRAY') { |
1440 |
$database_code = config(); # use _ from hash |
$database_code = config(); # use _ from hash |
1441 |
$database_name = config('name'); |
$database_name = config('name'); |
1442 |
$database_input_name = config('input name'); |
$database_input_name = config('input name'); |
|
$tag = config('input normalize tag'); |
|
1443 |
|
|
1444 |
Up to three levels are supported. |
Up to three levels are supported. |
1445 |
|
|
1555 |
} |
} |
1556 |
} |
} |
1557 |
|
|
1558 |
|
my $hash; |
1559 |
|
|
1560 |
|
=head2 set |
1561 |
|
|
1562 |
|
set( key => 'value' ); |
1563 |
|
|
1564 |
|
=cut |
1565 |
|
|
1566 |
|
sub set { |
1567 |
|
my ($k,$v) = @_; |
1568 |
|
warn "## set ( $k => ", dump($v), " )", $/ if ( $debug ); |
1569 |
|
$hash->{$k} = $v; |
1570 |
|
}; |
1571 |
|
|
1572 |
|
=head2 get |
1573 |
|
|
1574 |
|
get( 'key' ); |
1575 |
|
|
1576 |
|
=cut |
1577 |
|
|
1578 |
|
sub get { |
1579 |
|
my $k = shift || return; |
1580 |
|
my $v = $hash->{$k}; |
1581 |
|
warn "## get $k = ", dump( $v ), $/ if ( $debug ); |
1582 |
|
return $v; |
1583 |
|
} |
1584 |
|
|
1585 |
|
=head2 count |
1586 |
|
|
1587 |
|
if ( count( @result ) == 1 ) { |
1588 |
|
# do something if only 1 result is there |
1589 |
|
} |
1590 |
|
|
1591 |
|
=cut |
1592 |
|
|
1593 |
|
sub count { |
1594 |
|
warn "## count ",dump(@_),$/ if ( $debug ); |
1595 |
|
return @_ . ''; |
1596 |
|
} |
1597 |
|
|
1598 |
# END |
# END |
1599 |
1; |
1; |