2 |
use Exporter 'import'; |
use Exporter 'import'; |
3 |
@EXPORT = qw/ |
@EXPORT = qw/ |
4 |
_set_rec _set_lookup |
_set_rec _set_lookup |
5 |
|
_set_load_row |
6 |
_get_ds _clean_ds |
_get_ds _clean_ds |
7 |
_debug |
_debug |
8 |
|
_pack_subfields_hash |
9 |
|
|
10 |
tag search display |
tag search display |
11 |
marc marc_indicators marc_repeatable_subfield |
marc marc_indicators marc_repeatable_subfield |
12 |
marc_compose marc_leader |
marc_compose marc_leader marc_fixed |
13 |
marc_duplicate marc_remove |
marc_duplicate marc_remove marc_count |
14 |
marc_original_order |
marc_original_order |
15 |
|
|
16 |
rec1 rec2 rec |
rec1 rec2 rec |
17 |
regex prefix suffix surround |
regex prefix suffix surround |
18 |
first lookup join_with |
first lookup join_with |
19 |
|
save_into_lookup |
20 |
|
|
21 |
split_rec_on |
split_rec_on |
22 |
|
|
23 |
|
get set |
24 |
|
count |
25 |
/; |
/; |
26 |
|
|
27 |
use warnings; |
use warnings; |
29 |
|
|
30 |
#use base qw/WebPAC::Common/; |
#use base qw/WebPAC::Common/; |
31 |
use Data::Dump qw/dump/; |
use Data::Dump qw/dump/; |
|
use Encode qw/from_to/; |
|
32 |
use Storable qw/dclone/; |
use Storable qw/dclone/; |
33 |
|
use Carp qw/confess/; |
34 |
|
|
35 |
# debugging warn(s) |
# debugging warn(s) |
36 |
my $debug = 0; |
my $debug = 0; |
42 |
|
|
43 |
=head1 VERSION |
=head1 VERSION |
44 |
|
|
45 |
Version 0.15 |
Version 0.28 |
46 |
|
|
47 |
=cut |
=cut |
48 |
|
|
49 |
our $VERSION = '0.15'; |
our $VERSION = '0.28'; |
50 |
|
|
51 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
52 |
|
|
72 |
Return data structure |
Return data structure |
73 |
|
|
74 |
my $ds = WebPAC::Normalize::data_structure( |
my $ds = WebPAC::Normalize::data_structure( |
75 |
lookup => $lookup->lookup_hash, |
lookup => $lookup_hash, |
76 |
row => $row, |
row => $row, |
77 |
rules => $normalize_pl_config, |
rules => $normalize_pl_config, |
78 |
marc_encoding => 'utf-8', |
marc_encoding => 'utf-8', |
79 |
config => $config, |
config => $config, |
80 |
|
load_row_coderef => sub { |
81 |
|
my ($database,$input,$mfn) = shift; |
82 |
|
$store->load_row( database => $database, input => $input, id => $mfn ); |
83 |
|
}, |
84 |
); |
); |
85 |
|
|
86 |
Options C<lookup>, C<row>, C<rules> and C<log> are mandatory while all |
Options C<row>, C<rules> and C<log> are mandatory while all |
87 |
other are optional. |
other are optional. |
88 |
|
|
89 |
|
C<load_row_coderef> is closure only used when executing lookups, so they will |
90 |
|
die if it's not defined. |
91 |
|
|
92 |
This function will B<die> if normalizastion can't be evaled. |
This function will B<die> if normalizastion can't be evaled. |
93 |
|
|
94 |
Since this function isn't exported you have to call it with |
Since this function isn't exported you have to call it with |
96 |
|
|
97 |
=cut |
=cut |
98 |
|
|
99 |
|
my $load_row_coderef; |
100 |
|
|
101 |
sub data_structure { |
sub data_structure { |
102 |
my $arg = {@_}; |
my $arg = {@_}; |
103 |
|
|
105 |
die "need normalisation argument" unless ($arg->{rules}); |
die "need normalisation argument" unless ($arg->{rules}); |
106 |
|
|
107 |
no strict 'subs'; |
no strict 'subs'; |
108 |
_set_lookup( $arg->{lookup} ); |
_set_lookup( $arg->{lookup} ) if defined($arg->{lookup}); |
109 |
_set_rec( $arg->{row} ); |
_set_rec( $arg->{row} ); |
110 |
_set_config( $arg->{config} ); |
_set_config( $arg->{config} ) if defined($arg->{config}); |
111 |
_clean_ds( %{ $arg } ); |
_clean_ds( %{ $arg } ); |
112 |
|
$load_row_coderef = $arg->{load_row_coderef}; |
113 |
|
|
114 |
eval "$arg->{rules}"; |
eval "$arg->{rules}"; |
115 |
die "error evaling $arg->{rules}: $@\n" if ($@); |
die "error evaling $arg->{rules}: $@\n" if ($@); |
116 |
|
|
167 |
|
|
168 |
=cut |
=cut |
169 |
|
|
170 |
my ($out, $marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators); |
my ($out, $marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $marc_leader); |
171 |
my ($marc_record_offset, $marc_fetch_offset) = (0, 0); |
my ($marc_record_offset, $marc_fetch_offset) = (0, 0); |
172 |
|
|
173 |
sub _get_ds { |
sub _get_ds { |
184 |
|
|
185 |
sub _clean_ds { |
sub _clean_ds { |
186 |
my $a = {@_}; |
my $a = {@_}; |
187 |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators) = (); |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $marc_leader) = (); |
188 |
($marc_record_offset, $marc_fetch_offset) = (0,0); |
($marc_record_offset, $marc_fetch_offset) = (0,0); |
189 |
$marc_encoding = $a->{marc_encoding}; |
$marc_encoding = $a->{marc_encoding}; |
190 |
} |
} |
203 |
$lookup = shift; |
$lookup = shift; |
204 |
} |
} |
205 |
|
|
206 |
|
=head2 _get_lookup |
207 |
|
|
208 |
|
Get current lookup hash |
209 |
|
|
210 |
|
my $lookup = _get_lookup(); |
211 |
|
|
212 |
|
=cut |
213 |
|
|
214 |
|
sub _get_lookup { |
215 |
|
return $lookup; |
216 |
|
} |
217 |
|
|
218 |
|
=head2 _set_load_row |
219 |
|
|
220 |
|
Setup code reference which will return L<data_structure> from |
221 |
|
L<WebPAC::Store> |
222 |
|
|
223 |
|
_set_load_row(sub { |
224 |
|
my ($database,$input,$mfn) = @_; |
225 |
|
$store->load_row( database => $database, input => $input, id => $mfn ); |
226 |
|
}); |
227 |
|
|
228 |
|
=cut |
229 |
|
|
230 |
|
sub _set_load_row { |
231 |
|
my $coderef = shift; |
232 |
|
confess "argument isn't CODE" unless ref($coderef) eq 'CODE'; |
233 |
|
|
234 |
|
$load_row_coderef = $coderef; |
235 |
|
} |
236 |
|
|
237 |
=head2 _get_marc_fields |
=head2 _get_marc_fields |
238 |
|
|
239 |
Get all fields defined by calls to C<marc> |
Get all fields defined by calls to C<marc> |
289 |
|
|
290 |
=cut |
=cut |
291 |
|
|
292 |
|
my $fetch_pos; |
293 |
|
|
294 |
sub _get_marc_fields { |
sub _get_marc_fields { |
295 |
|
|
296 |
my $arg = {@_}; |
my $arg = {@_}; |
297 |
warn "### _get_marc_fields arg: ", dump($arg), $/ if ($debug > 2); |
warn "### _get_marc_fields arg: ", dump($arg), $/ if ($debug > 2); |
298 |
my $offset = $marc_fetch_offset; |
$fetch_pos = $marc_fetch_offset; |
299 |
if ($arg->{offset}) { |
if ($arg->{offset}) { |
300 |
$offset = $arg->{offset}; |
$fetch_pos = $arg->{offset}; |
301 |
} elsif($arg->{fetch_next}) { |
} elsif($arg->{fetch_next}) { |
302 |
$marc_fetch_offset++; |
$marc_fetch_offset++; |
303 |
} |
} |
306 |
|
|
307 |
warn "### full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 2); |
warn "### full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 2); |
308 |
|
|
309 |
my $marc_rec = $marc_record->[ $offset ]; |
my $marc_rec = $marc_record->[ $fetch_pos ]; |
310 |
|
|
311 |
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); |
312 |
|
|
313 |
return if (! $marc_rec || ref($marc_rec) ne 'ARRAY' || $#{ $marc_rec } < 0); |
return if (! $marc_rec || ref($marc_rec) ne 'ARRAY' || $#{ $marc_rec } < 0); |
314 |
|
|
329 |
|
|
330 |
if ($debug) { |
if ($debug) { |
331 |
warn "## marc_repeatable_subfield = ", dump( $marc_repeatable_subfield ), $/ if ( $marc_repeatable_subfield ); |
warn "## marc_repeatable_subfield = ", dump( $marc_repeatable_subfield ), $/ if ( $marc_repeatable_subfield ); |
332 |
warn "## marc_record[$offset] = ", dump( $marc_rec ), $/; |
warn "## marc_record[$fetch_pos] = ", dump( $marc_rec ), $/; |
333 |
warn "## sorted_marc_record = ", dump( \@sorted_marc_record ), $/; |
warn "## sorted_marc_record = ", dump( \@sorted_marc_record ), $/; |
334 |
warn "## subfield count = ", dump( $u ), $/; |
warn "## subfield count = ", dump( $u ), $/; |
335 |
} |
} |
410 |
return \@m; |
return \@m; |
411 |
} |
} |
412 |
|
|
413 |
|
=head2 _get_marc_leader |
414 |
|
|
415 |
|
Return leader from currently fetched record by L</_get_marc_fields> |
416 |
|
|
417 |
|
print WebPAC::Normalize::_get_marc_leader(); |
418 |
|
|
419 |
|
=cut |
420 |
|
|
421 |
|
sub _get_marc_leader { |
422 |
|
die "no fetch_pos, did you called _get_marc_fields first?" unless ( defined( $fetch_pos ) ); |
423 |
|
return $marc_leader->[ $fetch_pos ]; |
424 |
|
} |
425 |
|
|
426 |
=head2 _debug |
=head2 _debug |
427 |
|
|
428 |
Change level of debug warnings |
Change level of debug warnings |
505 |
my ($offset,$value) = @_; |
my ($offset,$value) = @_; |
506 |
|
|
507 |
if ($offset) { |
if ($offset) { |
508 |
$out->{' leader'}->{ $offset } = $value; |
$marc_leader->[ $marc_record_offset ]->{ $offset } = $value; |
509 |
} else { |
} else { |
510 |
return $out->{' leader'}; |
|
511 |
|
if (defined($marc_leader)) { |
512 |
|
die "marc_leader not array = ", dump( $marc_leader ) unless (ref($marc_leader) eq 'ARRAY'); |
513 |
|
return $marc_leader->[ $marc_record_offset ]; |
514 |
|
} else { |
515 |
|
return; |
516 |
|
} |
517 |
|
} |
518 |
|
} |
519 |
|
|
520 |
|
=head2 marc_fixed |
521 |
|
|
522 |
|
Create control/indentifier fields with values in fixed positions |
523 |
|
|
524 |
|
marc_fixed('008', 00, '070402'); |
525 |
|
marc_fixed('008', 39, '|'); |
526 |
|
|
527 |
|
Positions not specified will be filled with spaces (C<0x20>). |
528 |
|
|
529 |
|
There will be no effort to extend last specified value to full length of |
530 |
|
field in standard. |
531 |
|
|
532 |
|
=cut |
533 |
|
|
534 |
|
sub marc_fixed { |
535 |
|
my ($f, $pos, $val) = @_; |
536 |
|
die "need marc(field, position, value)" unless defined($f) && defined($pos); |
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 |
|
|
579 |
foreach (@_) { |
foreach (@_) { |
580 |
my $v = $_; # make var read-write for Encode |
my $v = $_; # make var read-write for Encode |
581 |
next unless (defined($v) && $v !~ /^\s*$/); |
next unless (defined($v) && $v !~ /^\s*$/); |
|
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
|
582 |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
583 |
if (defined $sf) { |
if (defined $sf) { |
584 |
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $i1, $i2, $sf => $v ]; |
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $i1, $i2, $sf => $v ]; |
649 |
|
|
650 |
warn "### marc_compose input subfields = ", dump(@_),$/ if ($debug > 2); |
warn "### marc_compose input subfields = ", dump(@_),$/ if ($debug > 2); |
651 |
|
|
652 |
|
if ($#_ % 2 != 1) { |
653 |
|
die "ERROR: marc_compose",dump($f,@_)," not valid (must be even).\nDo you need to add first() or join() around some argument?\n"; |
654 |
|
} |
655 |
|
|
656 |
while (@_) { |
while (@_) { |
657 |
my $sf = shift or die "marc_compose $f needs subfield"; |
my $sf = shift; |
658 |
my $v = shift; |
my $v = shift; |
659 |
|
|
660 |
next unless (defined($v) && $v !~ /^\s*$/); |
next unless (defined($v) && $v !~ /^\s*$/); |
|
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
|
661 |
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
662 |
if ($sf ne '+') { |
if ($sf ne '+') { |
663 |
push @$m, ( $sf, $v ); |
push @$m, ( $sf, $v ); |
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 |
} |
} |
772 |
|
|
773 |
Copy all subfields preserving original order to marc field. |
Copy all subfields preserving original order to marc field. |
774 |
|
|
775 |
marc_original_order(210, 260); |
marc_original_order( marc_field_number, original_input_field_number ); |
776 |
|
|
777 |
|
Please note that field numbers are consistent with other commands (marc |
778 |
|
field number first), but somewhat counter-intuitive (destination and then |
779 |
|
source). |
780 |
|
|
781 |
You might want to use this command if you are just renaming subfields or |
You might want to use this command if you are just renaming subfields or |
782 |
using pre-processing modify_record in C<config.yml> and don't need any |
using pre-processing modify_record in C<config.yml> and don't need any |
783 |
post-processing or want to preserve order of original subfields. |
post-processing or want to preserve order of original subfields. |
784 |
|
|
785 |
|
|
786 |
=cut |
=cut |
787 |
|
|
788 |
sub marc_original_order { |
sub marc_original_order { |
789 |
|
|
790 |
my ($from, $to) = @_; |
my ($to, $from) = @_; |
791 |
die "marc_original_order needs from and to fields\n" unless ($from && $to); |
die "marc_original_order needs from and to fields\n" unless ($from && $to); |
792 |
|
|
793 |
my $r = $rec->{$from} || return; |
return unless defined($rec->{$from}); |
794 |
|
|
795 |
|
my $r = $rec->{$from}; |
796 |
die "record field $from isn't array\n" unless (ref($r) eq 'ARRAY'); |
die "record field $from isn't array\n" unless (ref($r) eq 'ARRAY'); |
797 |
|
|
798 |
my ($i1,$i2) = defined($marc_indicators->{$to}) ? @{ $marc_indicators->{$to} } : (' ',' '); |
my ($i1,$i2) = defined($marc_indicators->{$to}) ? @{ $marc_indicators->{$to} } : (' ',' '); |
799 |
warn "## marc_original_order($from,$to) source = ", dump( $r ),$/ if ($debug > 1); |
warn "## marc_original_order($to,$from) source = ", dump( $r ),$/ if ($debug > 1); |
800 |
|
|
801 |
foreach my $d (@$r) { |
foreach my $d (@$r) { |
802 |
|
|
803 |
|
if (! defined($d->{subfields}) && ref($d->{subfields}) ne 'ARRAY') { |
804 |
|
warn "# marc_original_order($to,$from): field $from doesn't have subfields specification\n"; |
805 |
|
next; |
806 |
|
} |
807 |
|
|
808 |
my @sfs = @{ $d->{subfields} }; |
my @sfs = @{ $d->{subfields} }; |
809 |
|
|
|
die "field $from doesn't have subfields specification\n" unless(@sfs); |
|
810 |
die "field $from doesn't have even number of subfields specifications\n" unless($#sfs % 2 == 1); |
die "field $from doesn't have even number of subfields specifications\n" unless($#sfs % 2 == 1); |
811 |
|
|
812 |
warn "#--> d: ",dump($d), "\n#--> sfs: ",dump(@sfs),$/; |
warn "#--> d: ",dump($d), "\n#--> sfs: ",dump(@sfs),$/ if ($debug > 2); |
813 |
|
|
814 |
my $m = [ $to, $i1, $i2 ]; |
my $m = [ $to, $i1, $i2 ]; |
815 |
|
|
816 |
while (my $sf = shift @sfs) { |
while (my $sf = shift @sfs) { |
817 |
warn "#--> sf: ",dump($sf), $/; |
|
818 |
|
warn "#--> sf: ",dump($sf), $/ if ($debug > 2); |
819 |
my $offset = shift @sfs; |
my $offset = shift @sfs; |
820 |
die "corrupted sufields specification for field $from\n" unless defined($offset); |
die "corrupted sufields specification for field $from\n" unless defined($offset); |
821 |
|
|
836 |
} |
} |
837 |
|
|
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 |
warn "# marc_original_order is partly implemented"; |
sub marc_count { |
850 |
|
return $#{ $marc_record }; |
851 |
} |
} |
852 |
|
|
853 |
|
|
856 |
This function should be used inside functions to create C<data_structure> described |
This function should be used inside functions to create C<data_structure> described |
857 |
above. |
above. |
858 |
|
|
859 |
|
=head2 _pack_subfields_hash |
860 |
|
|
861 |
|
@subfields = _pack_subfields_hash( $h ); |
862 |
|
$subfields = _pack_subfields_hash( $h, 1 ); |
863 |
|
|
864 |
|
Return each subfield value in array or pack them all together and return scalar |
865 |
|
with subfields (denoted by C<^>) and values. |
866 |
|
|
867 |
|
=cut |
868 |
|
|
869 |
|
sub _pack_subfields_hash { |
870 |
|
|
871 |
|
warn "## _pack_subfields_hash( ",dump(@_), " )\n" if ($debug > 1); |
872 |
|
|
873 |
|
my ($h,$include_subfields) = @_; |
874 |
|
|
875 |
|
if ( defined($h->{subfields}) ) { |
876 |
|
my $sfs = delete $h->{subfields} || die "no subfields?"; |
877 |
|
my @out; |
878 |
|
while (@$sfs) { |
879 |
|
my $sf = shift @$sfs; |
880 |
|
push @out, '^' . $sf if ($include_subfields); |
881 |
|
my $o = shift @$sfs; |
882 |
|
if ($o == 0 && ref( $h->{$sf} ) ne 'ARRAY' ) { |
883 |
|
# single element subfields are not arrays |
884 |
|
#warn "====> $sf $o / $#$sfs ", dump( $sfs, $h->{$sf} ), "\n"; |
885 |
|
|
886 |
|
push @out, $h->{$sf}; |
887 |
|
} else { |
888 |
|
#warn "====> $sf $o / $#$sfs ", dump( $sfs, $h->{$sf} ), "\n"; |
889 |
|
push @out, $h->{$sf}->[$o]; |
890 |
|
} |
891 |
|
} |
892 |
|
if ($include_subfields) { |
893 |
|
return join('', @out); |
894 |
|
} else { |
895 |
|
return @out; |
896 |
|
} |
897 |
|
} else { |
898 |
|
if ($include_subfields) { |
899 |
|
my $out = ''; |
900 |
|
foreach my $sf (sort keys %$h) { |
901 |
|
if (ref($h->{$sf}) eq 'ARRAY') { |
902 |
|
$out .= '^' . $sf . join('^' . $sf, @{ $h->{$sf} }); |
903 |
|
} else { |
904 |
|
$out .= '^' . $sf . $h->{$sf}; |
905 |
|
} |
906 |
|
} |
907 |
|
return $out; |
908 |
|
} else { |
909 |
|
# FIXME this should probably be in alphabetical order instead of hash order |
910 |
|
values %{$h}; |
911 |
|
} |
912 |
|
} |
913 |
|
} |
914 |
|
|
915 |
=head2 rec1 |
=head2 rec1 |
916 |
|
|
917 |
Return all values in some field |
Return all values in some field |
928 |
return unless (defined($rec) && defined($rec->{$f})); |
return unless (defined($rec) && defined($rec->{$f})); |
929 |
warn "rec1($f) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
warn "rec1($f) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
930 |
if (ref($rec->{$f}) eq 'ARRAY') { |
if (ref($rec->{$f}) eq 'ARRAY') { |
931 |
return map { |
my @out; |
932 |
if (ref($_) eq 'HASH') { |
foreach my $h ( @{ $rec->{$f} } ) { |
933 |
values %{$_}; |
if (ref($h) eq 'HASH') { |
934 |
|
push @out, ( _pack_subfields_hash( $h ) ); |
935 |
} else { |
} else { |
936 |
$_; |
push @out, $h; |
937 |
} |
} |
938 |
} @{ $rec->{$f} }; |
} |
939 |
|
return @out; |
940 |
} elsif( defined($rec->{$f}) ) { |
} elsif( defined($rec->{$f}) ) { |
941 |
return $rec->{$f}; |
return $rec->{$f}; |
942 |
} |
} |
971 |
@v = rec('200') |
@v = rec('200') |
972 |
@v = rec('200','a') |
@v = rec('200','a') |
973 |
|
|
974 |
|
If rec() returns just single value, it will |
975 |
|
return scalar, not array. |
976 |
|
|
977 |
=cut |
=cut |
978 |
|
|
979 |
sub rec { |
sub rec { |
983 |
} elsif ($#_ == 1) { |
} elsif ($#_ == 1) { |
984 |
@out = rec2(@_); |
@out = rec2(@_); |
985 |
} |
} |
986 |
if (@out) { |
if ($#out == 0 && ! wantarray) { |
987 |
|
return $out[0]; |
988 |
|
} elsif (@out) { |
989 |
return @out; |
return @out; |
990 |
} else { |
} else { |
991 |
return ''; |
return ''; |
1069 |
|
|
1070 |
Consult lookup hashes for some value |
Consult lookup hashes for some value |
1071 |
|
|
1072 |
@v = lookup( $v ); |
@v = lookup( |
1073 |
@v = lookup( @v ); |
sub { |
1074 |
|
'ffkk/peri/mfn'.rec('000') |
1075 |
|
}, |
1076 |
|
'ffkk','peri','200-a-200-e', |
1077 |
|
sub { |
1078 |
|
first(rec(200,'a')).' '.first(rec('200','e')) |
1079 |
|
} |
1080 |
|
); |
1081 |
|
|
1082 |
|
Code like above will be B<automatically generated> using L<WebPAC::Parse> from |
1083 |
|
normal lookup definition in C<conf/lookup/something.pl> which looks like: |
1084 |
|
|
1085 |
|
lookup( |
1086 |
|
# which results to return from record recorded in lookup |
1087 |
|
sub { 'ffkk/peri/mfn' . rec('000') }, |
1088 |
|
# from which database and input |
1089 |
|
'ffkk','peri', |
1090 |
|
# such that following values match |
1091 |
|
sub { first(rec(200,'a')) . ' ' . first(rec('200','e')) }, |
1092 |
|
# if this part is missing, we will try to match same fields |
1093 |
|
# from lookup record and current one, or you can override |
1094 |
|
# which records to use from current record using |
1095 |
|
sub { rec('900','x') . ' ' . rec('900','y') }, |
1096 |
|
) |
1097 |
|
|
1098 |
|
You can think about this lookup as SQL (if that helps): |
1099 |
|
|
1100 |
|
select |
1101 |
|
sub { what } |
1102 |
|
from |
1103 |
|
database, input |
1104 |
|
where |
1105 |
|
sub { filter from lookuped record } |
1106 |
|
having |
1107 |
|
sub { optional filter on current record } |
1108 |
|
|
1109 |
|
Easy as pie, right? |
1110 |
|
|
1111 |
=cut |
=cut |
1112 |
|
|
1113 |
sub lookup { |
sub lookup { |
1114 |
my $k = shift or return; |
my ($what, $database, $input, $key, $having) = @_; |
1115 |
return unless (defined($lookup->{$k})); |
|
1116 |
if (ref($lookup->{$k}) eq 'ARRAY') { |
confess "lookup needs 5 arguments: what, database, input, key, having\n" unless ($#_ == 4); |
1117 |
return @{ $lookup->{$k} }; |
|
1118 |
|
warn "## lookup ($database, $input, $key)", $/ if ($debug > 1); |
1119 |
|
return unless (defined($lookup->{$database}->{$input}->{$key})); |
1120 |
|
|
1121 |
|
confess "lookup really need load_row_coderef added to data_structure\n" unless ($load_row_coderef); |
1122 |
|
|
1123 |
|
my $mfns; |
1124 |
|
my @having = $having->(); |
1125 |
|
|
1126 |
|
warn "## having = ", dump( @having ) if ($debug > 2); |
1127 |
|
|
1128 |
|
foreach my $h ( @having ) { |
1129 |
|
if (defined($lookup->{$database}->{$input}->{$key}->{$h})) { |
1130 |
|
warn "lookup for $database/$input/$key/$h return ",dump($lookup->{$database}->{$input}->{$key}->{$h}),"\n" if ($debug); |
1131 |
|
$mfns->{$_}++ foreach keys %{ $lookup->{$database}->{$input}->{$key}->{$h} }; |
1132 |
|
} |
1133 |
|
} |
1134 |
|
|
1135 |
|
return unless ($mfns); |
1136 |
|
|
1137 |
|
my @mfns = sort keys %$mfns; |
1138 |
|
|
1139 |
|
warn "# lookup loading $database/$input/$key mfn ", join(",",@mfns)," having ",dump(@having),"\n" if ($debug); |
1140 |
|
|
1141 |
|
my $old_rec = $rec; |
1142 |
|
my @out; |
1143 |
|
|
1144 |
|
foreach my $mfn (@mfns) { |
1145 |
|
$rec = $load_row_coderef->( $database, $input, $mfn ); |
1146 |
|
|
1147 |
|
warn "got $database/$input/$mfn = ", dump($rec), $/ if ($debug); |
1148 |
|
|
1149 |
|
my @vals = $what->(); |
1150 |
|
|
1151 |
|
push @out, ( @vals ); |
1152 |
|
|
1153 |
|
warn "lookup for mfn $mfn returned ", dump(@vals), $/ if ($debug); |
1154 |
|
} |
1155 |
|
|
1156 |
|
# if (ref($lookup->{$k}) eq 'ARRAY') { |
1157 |
|
# return @{ $lookup->{$k} }; |
1158 |
|
# } else { |
1159 |
|
# return $lookup->{$k}; |
1160 |
|
# } |
1161 |
|
|
1162 |
|
$rec = $old_rec; |
1163 |
|
|
1164 |
|
warn "## lookup returns = ", dump(@out), $/ if ($debug); |
1165 |
|
|
1166 |
|
if ($#out == 0) { |
1167 |
|
return $out[0]; |
1168 |
} else { |
} else { |
1169 |
return $lookup->{$k}; |
return @out; |
1170 |
|
} |
1171 |
|
} |
1172 |
|
|
1173 |
|
=head2 save_into_lookup |
1174 |
|
|
1175 |
|
Save value into lookup. It associates current database, input |
1176 |
|
and specific keys with one or more values which will be |
1177 |
|
associated over MFN. |
1178 |
|
|
1179 |
|
MFN will be extracted from first occurence current of field 000 |
1180 |
|
in current record, or if it doesn't exist from L<_set_config> C<_mfn>. |
1181 |
|
|
1182 |
|
my $nr = save_into_lookup($database,$input,$key,sub { |
1183 |
|
# code which produce one or more values |
1184 |
|
}); |
1185 |
|
|
1186 |
|
It returns number of items saved. |
1187 |
|
|
1188 |
|
This function shouldn't be called directly, it's called from code created by |
1189 |
|
L<WebPAC::Parser>. |
1190 |
|
|
1191 |
|
=cut |
1192 |
|
|
1193 |
|
sub save_into_lookup { |
1194 |
|
my ($database,$input,$key,$coderef) = @_; |
1195 |
|
die "save_into_lookup needs database" unless defined($database); |
1196 |
|
die "save_into_lookup needs input" unless defined($input); |
1197 |
|
die "save_into_lookup needs key" unless defined($key); |
1198 |
|
die "save_into_lookup needs CODE" unless ( defined($coderef) && ref($coderef) eq 'CODE' ); |
1199 |
|
|
1200 |
|
warn "## save_into_lookup rec = ", dump($rec), " config = ", dump($config), $/ if ($debug > 2); |
1201 |
|
|
1202 |
|
my $mfn = |
1203 |
|
defined($rec->{'000'}->[0]) ? $rec->{'000'}->[0] : |
1204 |
|
defined($config->{_mfn}) ? $config->{_mfn} : |
1205 |
|
die "mfn not defined or zero"; |
1206 |
|
|
1207 |
|
my $nr = 0; |
1208 |
|
|
1209 |
|
foreach my $v ( $coderef->() ) { |
1210 |
|
$lookup->{$database}->{$input}->{$key}->{$v}->{$mfn}++; |
1211 |
|
warn "# saved lookup $database/$input/$key [$v] $mfn\n" if ($debug > 1); |
1212 |
|
$nr++; |
1213 |
} |
} |
1214 |
|
|
1215 |
|
return $nr; |
1216 |
} |
} |
1217 |
|
|
1218 |
=head2 config |
=head2 config |
1339 |
} |
} |
1340 |
} |
} |
1341 |
|
|
1342 |
|
my $hash; |
1343 |
|
|
1344 |
|
=head2 set |
1345 |
|
|
1346 |
|
set( key => 'value' ); |
1347 |
|
|
1348 |
|
=cut |
1349 |
|
|
1350 |
|
sub set { |
1351 |
|
my ($k,$v) = @_; |
1352 |
|
warn "## set ( $k => ", dump($v), " )", $/ if ( $debug ); |
1353 |
|
$hash->{$k} = $v; |
1354 |
|
}; |
1355 |
|
|
1356 |
|
=head2 get |
1357 |
|
|
1358 |
|
get( 'key' ); |
1359 |
|
|
1360 |
|
=cut |
1361 |
|
|
1362 |
|
sub get { |
1363 |
|
my $k = shift || return; |
1364 |
|
my $v = $hash->{$k}; |
1365 |
|
warn "## get $k = ", dump( $v ), $/ if ( $debug ); |
1366 |
|
return $v; |
1367 |
|
} |
1368 |
|
|
1369 |
|
=head2 count |
1370 |
|
|
1371 |
|
if ( count( @result ) == 1 ) { |
1372 |
|
# do something if only 1 result is there |
1373 |
|
} |
1374 |
|
|
1375 |
|
=cut |
1376 |
|
|
1377 |
|
sub count { |
1378 |
|
warn "## count ",dump(@_),$/ if ( $debug ); |
1379 |
|
return @_ . ''; |
1380 |
|
} |
1381 |
|
|
1382 |
# END |
# END |
1383 |
1; |
1; |