3 |
@EXPORT = qw/ |
@EXPORT = qw/ |
4 |
_set_rec _set_lookup |
_set_rec _set_lookup |
5 |
_get_ds _clean_ds |
_get_ds _clean_ds |
6 |
|
_debug |
7 |
|
|
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 |
11 |
|
marc_duplicate marc_remove |
12 |
|
marc_original_order |
13 |
|
|
14 |
rec1 rec2 rec |
rec1 rec2 rec |
15 |
regex prefix suffix surround |
regex prefix suffix surround |
16 |
first lookup join_with |
first lookup join_with |
17 |
|
|
18 |
|
split_rec_on |
19 |
/; |
/; |
20 |
|
|
21 |
use warnings; |
use warnings; |
24 |
#use base qw/WebPAC::Common/; |
#use base qw/WebPAC::Common/; |
25 |
use Data::Dump qw/dump/; |
use Data::Dump qw/dump/; |
26 |
use Encode qw/from_to/; |
use Encode qw/from_to/; |
27 |
|
use Storable qw/dclone/; |
28 |
|
|
29 |
# debugging warn(s) |
# debugging warn(s) |
30 |
my $debug = 0; |
my $debug = 0; |
36 |
|
|
37 |
=head1 VERSION |
=head1 VERSION |
38 |
|
|
39 |
Version 0.06 |
Version 0.15 |
40 |
|
|
41 |
=cut |
=cut |
42 |
|
|
43 |
our $VERSION = '0.06'; |
our $VERSION = '0.15'; |
44 |
|
|
45 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
46 |
|
|
70 |
row => $row, |
row => $row, |
71 |
rules => $normalize_pl_config, |
rules => $normalize_pl_config, |
72 |
marc_encoding => 'utf-8', |
marc_encoding => 'utf-8', |
73 |
|
config => $config, |
74 |
); |
); |
75 |
|
|
76 |
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 |
92 |
no strict 'subs'; |
no strict 'subs'; |
93 |
_set_lookup( $arg->{lookup} ); |
_set_lookup( $arg->{lookup} ); |
94 |
_set_rec( $arg->{row} ); |
_set_rec( $arg->{row} ); |
95 |
|
_set_config( $arg->{config} ); |
96 |
_clean_ds( %{ $arg } ); |
_clean_ds( %{ $arg } ); |
97 |
eval "$arg->{rules}"; |
eval "$arg->{rules}"; |
98 |
die "error evaling $arg->{rules}: $@\n" if ($@); |
die "error evaling $arg->{rules}: $@\n" if ($@); |
114 |
$rec = shift or die "no record hash"; |
$rec = shift or die "no record hash"; |
115 |
} |
} |
116 |
|
|
117 |
|
=head2 _set_config |
118 |
|
|
119 |
|
Set current config hash |
120 |
|
|
121 |
|
_set_config( $config ); |
122 |
|
|
123 |
|
Magic keys are: |
124 |
|
|
125 |
|
=over 4 |
126 |
|
|
127 |
|
=item _ |
128 |
|
|
129 |
|
Code of current database |
130 |
|
|
131 |
|
=item _mfn |
132 |
|
|
133 |
|
Current MFN |
134 |
|
|
135 |
|
=back |
136 |
|
|
137 |
|
=cut |
138 |
|
|
139 |
|
my $config; |
140 |
|
|
141 |
|
sub _set_config { |
142 |
|
$config = shift; |
143 |
|
} |
144 |
|
|
145 |
=head2 _get_ds |
=head2 _get_ds |
146 |
|
|
147 |
Return hash formatted as data structure |
Return hash formatted as data structure |
150 |
|
|
151 |
=cut |
=cut |
152 |
|
|
153 |
my ($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators); |
my ($out, $marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators); |
154 |
|
my ($marc_record_offset, $marc_fetch_offset) = (0, 0); |
155 |
|
|
156 |
sub _get_ds { |
sub _get_ds { |
157 |
return $out; |
return $out; |
168 |
sub _clean_ds { |
sub _clean_ds { |
169 |
my $a = {@_}; |
my $a = {@_}; |
170 |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators) = (); |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators) = (); |
171 |
|
($marc_record_offset, $marc_fetch_offset) = (0,0); |
172 |
$marc_encoding = $a->{marc_encoding}; |
$marc_encoding = $a->{marc_encoding}; |
173 |
} |
} |
174 |
|
|
192 |
|
|
193 |
$marc->add_fields( WebPAC::Normalize:_get_marc_fields() ); |
$marc->add_fields( WebPAC::Normalize:_get_marc_fields() ); |
194 |
|
|
|
|
|
|
|
|
195 |
We are using I<magic> which detect repeatable fields only from |
We are using I<magic> which detect repeatable fields only from |
196 |
sequence of field/subfield data generated by normalization. |
sequence of field/subfield data generated by normalization. |
197 |
|
|
198 |
Repeatable field is created if there is second occurence of same subfield or |
Repeatable field is created when there is second occurence of same subfield or |
199 |
if any of indicators are different. This is sane for most cases except for |
if any of indicators are different. |
200 |
non-repeatable fields with repeatable subfields. |
|
201 |
|
This is sane for most cases. Something like: |
202 |
|
|
203 |
|
900a-1 900b-1 900c-1 |
204 |
|
900a-2 900b-2 |
205 |
|
900a-3 |
206 |
|
|
207 |
|
will be created from any combination of: |
208 |
|
|
209 |
|
900a-1 900a-2 900a-3 900b-1 900b-2 900c-1 |
210 |
|
|
211 |
|
and following rules: |
212 |
|
|
213 |
|
marc('900','a', rec('200','a') ); |
214 |
|
marc('900','b', rec('200','b') ); |
215 |
|
marc('900','c', rec('200','c') ); |
216 |
|
|
217 |
|
which might not be what you have in mind. If you need repeatable subfield, |
218 |
|
define it using C<marc_repeatable_subfield> like this: |
219 |
|
|
220 |
|
marc_repeatable_subfield('900','a'); |
221 |
|
marc('900','a', rec('200','a') ); |
222 |
|
marc('900','b', rec('200','b') ); |
223 |
|
marc('900','c', rec('200','c') ); |
224 |
|
|
225 |
|
will create: |
226 |
|
|
227 |
|
900a-1 900a-2 900a-3 900b-1 900c-1 |
228 |
|
900b-2 |
229 |
|
|
230 |
You can change behaviour of that using C<marc_repeatable_subfield>. |
There is also support for returning next or specific using: |
231 |
|
|
232 |
|
while (my $mf = WebPAC::Normalize:_get_marc_fields( fetch_next => 1 ) ) { |
233 |
|
# do something with $mf |
234 |
|
} |
235 |
|
|
236 |
|
will always return fields from next MARC record or |
237 |
|
|
238 |
|
my $mf = WebPAC::Normalize::_get_marc_fields( offset => 42 ); |
239 |
|
|
240 |
|
will return 42th copy record (if it exists). |
241 |
|
|
242 |
=cut |
=cut |
243 |
|
|
244 |
sub _get_marc_fields { |
sub _get_marc_fields { |
245 |
|
|
246 |
|
my $arg = {@_}; |
247 |
|
warn "### _get_marc_fields arg: ", dump($arg), $/ if ($debug > 2); |
248 |
|
my $offset = $marc_fetch_offset; |
249 |
|
if ($arg->{offset}) { |
250 |
|
$offset = $arg->{offset}; |
251 |
|
} elsif($arg->{fetch_next}) { |
252 |
|
$marc_fetch_offset++; |
253 |
|
} |
254 |
|
|
255 |
|
return if (! $marc_record || ref($marc_record) ne 'ARRAY'); |
256 |
|
|
257 |
|
warn "### full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 2); |
258 |
|
|
259 |
|
my $marc_rec = $marc_record->[ $offset ]; |
260 |
|
|
261 |
|
warn "## _get_marc_fields (at offset: $offset) -- marc_record = ", dump( @$marc_rec ), $/ if ($debug > 1); |
262 |
|
|
263 |
|
return if (! $marc_rec || ref($marc_rec) ne 'ARRAY' || $#{ $marc_rec } < 0); |
264 |
|
|
265 |
# first, sort all existing fields |
# first, sort all existing fields |
266 |
# 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 |
267 |
my @sorted_marc_record = sort { |
my @sorted_marc_record = sort { |
268 |
$a->[0] . $a->[3] cmp $b->[0] . $b->[3] |
$a->[0] . ( $a->[3] || '' ) cmp $b->[0] . ( $b->[3] || '') |
269 |
} @{ $marc_record }; |
} @{ $marc_rec }; |
270 |
|
|
271 |
|
@sorted_marc_record = @{ $marc_rec }; ### FIXME disable sorting |
272 |
|
|
273 |
# output marc fields |
# output marc fields |
274 |
my @m; |
my @m; |
275 |
|
|
276 |
# count unique field-subfields (used for offset when walking to next subfield) |
# count unique field-subfields (used for offset when walking to next subfield) |
277 |
my $u; |
my $u; |
278 |
map { $u->{ $_->[0] . $_->[3] }++ } @sorted_marc_record; |
map { $u->{ $_->[0] . ( $_->[3] || '') }++ } @sorted_marc_record; |
279 |
|
|
280 |
if ($debug) { |
if ($debug) { |
281 |
warn "## marc_repeatable_subfield ", dump( $marc_repeatable_subfield ), $/; |
warn "## marc_repeatable_subfield = ", dump( $marc_repeatable_subfield ), $/ if ( $marc_repeatable_subfield ); |
282 |
warn "## marc_record ", dump( $marc_record ), $/; |
warn "## marc_record[$offset] = ", dump( $marc_rec ), $/; |
283 |
warn "## sorted_marc_record ", dump( \@sorted_marc_record ), $/; |
warn "## sorted_marc_record = ", dump( \@sorted_marc_record ), $/; |
284 |
warn "## subfield count ", dump( $u ), $/; |
warn "## subfield count = ", dump( $u ), $/; |
285 |
} |
} |
286 |
|
|
287 |
my $len = $#sorted_marc_record; |
my $len = $#sorted_marc_record; |
299 |
# mark it visited |
# mark it visited |
300 |
$visited->{$i}++; |
$visited->{$i}++; |
301 |
|
|
302 |
my $row = $sorted_marc_record[$i]; |
my $row = dclone( $sorted_marc_record[$i] ); |
303 |
|
|
304 |
# field and subfield which is key for |
# field and subfield which is key for |
305 |
# marc_repeatable_subfield and u |
# marc_repeatable_subfield and u |
306 |
my $fsf = $row->[0] . $row->[3]; |
my $fsf = $row->[0] . ( $row->[3] || '' ); |
307 |
|
|
308 |
if ($debug > 1) { |
if ($debug > 1) { |
309 |
|
|
357 |
warn "## saved/3 ", dump( $field ),$/ if ($debug); |
warn "## saved/3 ", dump( $field ),$/ if ($debug); |
358 |
} |
} |
359 |
|
|
360 |
return @m; |
return \@m; |
361 |
|
} |
362 |
|
|
363 |
|
=head2 _debug |
364 |
|
|
365 |
|
Change level of debug warnings |
366 |
|
|
367 |
|
_debug( 2 ); |
368 |
|
|
369 |
|
=cut |
370 |
|
|
371 |
|
sub _debug { |
372 |
|
my $l = shift; |
373 |
|
return $debug unless defined($l); |
374 |
|
warn "debug level $l",$/ if ($l > 0); |
375 |
|
$debug = $l; |
376 |
} |
} |
377 |
|
|
378 |
=head1 Functions to create C<data_structure> |
=head1 Functions to create C<data_structure> |
429 |
$out->{$name}->{search} = \@o; |
$out->{$name}->{search} = \@o; |
430 |
} |
} |
431 |
|
|
432 |
|
=head2 marc_leader |
433 |
|
|
434 |
|
Setup fields within MARC leader or get leader |
435 |
|
|
436 |
|
marc_leader('05','c'); |
437 |
|
my $leader = marc_leader(); |
438 |
|
|
439 |
|
=cut |
440 |
|
|
441 |
|
sub marc_leader { |
442 |
|
my ($offset,$value) = @_; |
443 |
|
|
444 |
|
if ($offset) { |
445 |
|
$out->{' leader'}->{ $offset } = $value; |
446 |
|
} else { |
447 |
|
return $out->{' leader'}; |
448 |
|
} |
449 |
|
} |
450 |
|
|
451 |
=head2 marc |
=head2 marc |
452 |
|
|
453 |
Save value for MARC field |
Save value for MARC field |
454 |
|
|
455 |
marc('900','a', rec('200','a') ); |
marc('900','a', rec('200','a') ); |
456 |
|
marc('001', rec('000') ); |
457 |
|
|
458 |
=cut |
=cut |
459 |
|
|
461 |
my $f = shift or die "marc needs field"; |
my $f = shift or die "marc needs field"; |
462 |
die "marc field must be numer" unless ($f =~ /^\d+$/); |
die "marc field must be numer" unless ($f =~ /^\d+$/); |
463 |
|
|
464 |
my $sf = shift or die "marc needs subfield"; |
my $sf; |
465 |
|
if ($f >= 10) { |
466 |
|
$sf = shift or die "marc needs subfield"; |
467 |
|
} |
468 |
|
|
469 |
foreach (@_) { |
foreach (@_) { |
470 |
my $v = $_; # make var read-write for Encode |
my $v = $_; # make var read-write for Encode |
471 |
next unless (defined($v) && $v !~ /^\s*$/); |
next unless (defined($v) && $v !~ /^\s*$/); |
472 |
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
473 |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
474 |
push @{ $marc_record }, [ $f, $i1, $i2, $sf => $v ]; |
if (defined $sf) { |
475 |
|
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $i1, $i2, $sf => $v ]; |
476 |
|
} else { |
477 |
|
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $v ]; |
478 |
|
} |
479 |
} |
} |
480 |
} |
} |
481 |
|
|
515 |
@{ $marc_indicators->{$f} } = ($i1,$i2); |
@{ $marc_indicators->{$f} } = ($i1,$i2); |
516 |
} |
} |
517 |
|
|
518 |
|
=head2 marc_compose |
519 |
|
|
520 |
|
Save values for each MARC subfield explicitly |
521 |
|
|
522 |
|
marc_compose('900', |
523 |
|
'a', rec('200','a') |
524 |
|
'b', rec('201','a') |
525 |
|
'a', rec('200','b') |
526 |
|
'c', rec('200','c') |
527 |
|
); |
528 |
|
|
529 |
|
If you specify C<+> for subfield, value will be appended |
530 |
|
to previous defined subfield. |
531 |
|
|
532 |
|
=cut |
533 |
|
|
534 |
|
sub marc_compose { |
535 |
|
my $f = shift or die "marc_compose needs field"; |
536 |
|
die "marc_compose field must be numer" unless ($f =~ /^\d+$/); |
537 |
|
|
538 |
|
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
539 |
|
my $m = [ $f, $i1, $i2 ]; |
540 |
|
|
541 |
|
warn "### marc_compose input subfields = ", dump(@_),$/ if ($debug > 2); |
542 |
|
|
543 |
|
while (@_) { |
544 |
|
my $sf = shift or die "marc_compose $f needs subfield"; |
545 |
|
my $v = shift; |
546 |
|
|
547 |
|
next unless (defined($v) && $v !~ /^\s*$/); |
548 |
|
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
549 |
|
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
550 |
|
if ($sf ne '+') { |
551 |
|
push @$m, ( $sf, $v ); |
552 |
|
} else { |
553 |
|
$m->[ $#$m ] .= $v; |
554 |
|
} |
555 |
|
} |
556 |
|
|
557 |
|
warn "## marc_compose current marc = ", dump( $m ),$/ if ($debug > 1); |
558 |
|
|
559 |
|
push @{ $marc_record->[ $marc_record_offset ] }, $m if ($#{$m} > 2); |
560 |
|
} |
561 |
|
|
562 |
|
=head2 marc_duplicate |
563 |
|
|
564 |
|
Generate copy of current MARC record and continue working on copy |
565 |
|
|
566 |
|
marc_duplicate(); |
567 |
|
|
568 |
|
Copies can be accessed using C<< _get_marc_fields( fetch_next => 1 ) >> or |
569 |
|
C<< _get_marc_fields( offset => 42 ) >>. |
570 |
|
|
571 |
|
=cut |
572 |
|
|
573 |
|
sub marc_duplicate { |
574 |
|
my $m = $marc_record->[ -1 ]; |
575 |
|
die "can't duplicate record which isn't defined" unless ($m); |
576 |
|
push @{ $marc_record }, dclone( $m ); |
577 |
|
warn "## marc_duplicate = ", dump(@$marc_record), $/ if ($debug > 1); |
578 |
|
$marc_record_offset = $#{ $marc_record }; |
579 |
|
warn "## marc_record_offset = $marc_record_offset", $/ if ($debug > 1); |
580 |
|
} |
581 |
|
|
582 |
|
=head2 marc_remove |
583 |
|
|
584 |
|
Remove some field or subfield from MARC record. |
585 |
|
|
586 |
|
marc_remove('200'); |
587 |
|
marc_remove('200','a'); |
588 |
|
|
589 |
|
This will erase field C<200> or C<200^a> from current MARC record. |
590 |
|
|
591 |
|
This is useful after calling C<marc_duplicate> or on it's own (but, you |
592 |
|
should probably just remove that subfield definition if you are not |
593 |
|
using C<marc_duplicate>). |
594 |
|
|
595 |
|
FIXME: support fields < 10. |
596 |
|
|
597 |
|
=cut |
598 |
|
|
599 |
|
sub marc_remove { |
600 |
|
my ($f, $sf) = @_; |
601 |
|
|
602 |
|
die "marc_remove needs record number" unless defined($f); |
603 |
|
|
604 |
|
my $marc = $marc_record->[ $marc_record_offset ]; |
605 |
|
|
606 |
|
warn "### marc_remove before = ", dump( $marc ), $/ if ($debug > 2); |
607 |
|
|
608 |
|
my $i = 0; |
609 |
|
foreach ( 0 .. $#{ $marc } ) { |
610 |
|
last unless (defined $marc->[$i]); |
611 |
|
warn "#### working on ",dump( @{ $marc->[$i] }), $/ if ($debug > 3); |
612 |
|
if ($marc->[$i]->[0] eq $f) { |
613 |
|
if (! defined $sf) { |
614 |
|
# remove whole field |
615 |
|
splice @$marc, $i, 1; |
616 |
|
warn "#### slice \@\$marc, $i, 1 = ",dump( @{ $marc }), $/ if ($debug > 3); |
617 |
|
$i--; |
618 |
|
} else { |
619 |
|
foreach my $j ( 0 .. (( $#{ $marc->[$i] } - 3 ) / 2) ) { |
620 |
|
my $o = ($j * 2) + 3; |
621 |
|
if ($marc->[$i]->[$o] eq $sf) { |
622 |
|
# remove subfield |
623 |
|
splice @{$marc->[$i]}, $o, 2; |
624 |
|
warn "#### slice \@{\$marc->[$i]}, $o, 2 = ", dump( @{ $marc }), $/ if ($debug > 3); |
625 |
|
# is record now empty? |
626 |
|
if ($#{ $marc->[$i] } == 2) { |
627 |
|
splice @$marc, $i, 1; |
628 |
|
warn "#### slice \@\$marc, $i, 1 = ", dump( @{ $marc }), $/ if ($debug > 3); |
629 |
|
$i--; |
630 |
|
}; |
631 |
|
} |
632 |
|
} |
633 |
|
} |
634 |
|
} |
635 |
|
$i++; |
636 |
|
} |
637 |
|
|
638 |
|
warn "### marc_remove($f", $sf ? ",$sf" : "", ") after = ", dump( $marc ), $/ if ($debug > 2); |
639 |
|
|
640 |
|
$marc_record->[ $marc_record_offset ] = $marc; |
641 |
|
|
642 |
|
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
643 |
|
} |
644 |
|
|
645 |
|
=head2 marc_original_order |
646 |
|
|
647 |
|
Copy all subfields preserving original order to marc field. |
648 |
|
|
649 |
|
marc_original_order(210, 260); |
650 |
|
|
651 |
|
You might want to use this command if you are just renaming subfields or |
652 |
|
using pre-processing modify_record in C<config.yml> and don't need any |
653 |
|
post-processing or want to preserve order of original subfields. |
654 |
|
|
655 |
|
=cut |
656 |
|
|
657 |
|
sub marc_original_order { |
658 |
|
|
659 |
|
my ($from, $to) = @_; |
660 |
|
die "marc_original_order needs from and to fields\n" unless ($from && $to); |
661 |
|
|
662 |
|
my $r = $rec->{$from} || return; |
663 |
|
die "record field $from isn't array\n" unless (ref($r) eq 'ARRAY'); |
664 |
|
|
665 |
|
my ($i1,$i2) = defined($marc_indicators->{$to}) ? @{ $marc_indicators->{$to} } : (' ',' '); |
666 |
|
warn "## marc_original_order($from,$to) source = ", dump( $r ),$/ if ($debug > 1); |
667 |
|
|
668 |
|
foreach my $d (@$r) { |
669 |
|
|
670 |
|
if (! defined($d->{subfields}) && ref($d->{subfields}) ne 'ARRAY') { |
671 |
|
warn "# marc_original_order($from,$to): field $from doesn't have subfields specification\n"; |
672 |
|
next; |
673 |
|
} |
674 |
|
|
675 |
|
my @sfs = @{ $d->{subfields} }; |
676 |
|
|
677 |
|
die "field $from doesn't have even number of subfields specifications\n" unless($#sfs % 2 == 1); |
678 |
|
|
679 |
|
warn "#--> d: ",dump($d), "\n#--> sfs: ",dump(@sfs),$/; |
680 |
|
|
681 |
|
my $m = [ $to, $i1, $i2 ]; |
682 |
|
|
683 |
|
while (my $sf = shift @sfs) { |
684 |
|
warn "#--> sf: ",dump($sf), $/; |
685 |
|
my $offset = shift @sfs; |
686 |
|
die "corrupted sufields specification for field $from\n" unless defined($offset); |
687 |
|
|
688 |
|
my $v; |
689 |
|
if (ref($d->{$sf}) eq 'ARRAY') { |
690 |
|
$v = $d->{$sf}->[$offset] if (defined($d->{$sf}->[$offset])); |
691 |
|
} elsif ($offset == 0) { |
692 |
|
$v = $d->{$sf}; |
693 |
|
} else { |
694 |
|
die "field $from subfield '$sf' need occurence $offset which doesn't exist", dump($d->{$sf}); |
695 |
|
} |
696 |
|
push @$m, ( $sf, $v ) if (defined($v)); |
697 |
|
} |
698 |
|
|
699 |
|
if ($#{$m} > 2) { |
700 |
|
push @{ $marc_record->[ $marc_record_offset ] }, $m; |
701 |
|
} |
702 |
|
} |
703 |
|
|
704 |
|
warn "## marc_record = ", dump( $marc_record ),$/ if ($debug > 1); |
705 |
|
|
706 |
|
warn "# marc_original_order is partly implemented"; |
707 |
|
} |
708 |
|
|
709 |
|
|
710 |
=head1 Functions to extract data from input |
=head1 Functions to extract data from input |
711 |
|
|
724 |
|
|
725 |
sub rec1 { |
sub rec1 { |
726 |
my $f = shift; |
my $f = shift; |
727 |
|
warn "rec1($f) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
728 |
return unless (defined($rec) && defined($rec->{$f})); |
return unless (defined($rec) && defined($rec->{$f})); |
729 |
|
warn "rec1($f) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
730 |
if (ref($rec->{$f}) eq 'ARRAY') { |
if (ref($rec->{$f}) eq 'ARRAY') { |
731 |
return map { |
return map { |
732 |
if (ref($_) eq 'HASH') { |
if (ref($_) eq 'HASH') { |
752 |
my $f = shift; |
my $f = shift; |
753 |
return unless (defined($rec && $rec->{$f})); |
return unless (defined($rec && $rec->{$f})); |
754 |
my $sf = shift; |
my $sf = shift; |
755 |
return map { $_->{$sf} } grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} }; |
warn "rec2($f,$sf) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
756 |
|
return map { |
757 |
|
if (ref($_->{$sf}) eq 'ARRAY') { |
758 |
|
@{ $_->{$sf} }; |
759 |
|
} else { |
760 |
|
$_->{$sf}; |
761 |
|
} |
762 |
|
} grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} }; |
763 |
} |
} |
764 |
|
|
765 |
=head2 rec |
=head2 rec |
772 |
=cut |
=cut |
773 |
|
|
774 |
sub rec { |
sub rec { |
775 |
|
my @out; |
776 |
if ($#_ == 0) { |
if ($#_ == 0) { |
777 |
return rec1(@_); |
@out = rec1(@_); |
778 |
} elsif ($#_ == 1) { |
} elsif ($#_ == 1) { |
779 |
return rec2(@_); |
@out = rec2(@_); |
780 |
|
} |
781 |
|
if (@out) { |
782 |
|
return @out; |
783 |
|
} else { |
784 |
|
return ''; |
785 |
} |
} |
786 |
} |
} |
787 |
|
|
814 |
=cut |
=cut |
815 |
|
|
816 |
sub prefix { |
sub prefix { |
817 |
my $p = shift or die "prefix needs string as first argument"; |
my $p = shift or return; |
818 |
return map { $p . $_ } grep { defined($_) } @_; |
return map { $p . $_ } grep { defined($_) } @_; |
819 |
} |
} |
820 |
|
|
877 |
} |
} |
878 |
} |
} |
879 |
|
|
880 |
|
=head2 config |
881 |
|
|
882 |
|
Consult config values stored in C<config.yml> |
883 |
|
|
884 |
|
# return database code (key under databases in yaml) |
885 |
|
$database_code = config(); # use _ from hash |
886 |
|
$database_name = config('name'); |
887 |
|
$database_input_name = config('input name'); |
888 |
|
$tag = config('input normalize tag'); |
889 |
|
|
890 |
|
Up to three levels are supported. |
891 |
|
|
892 |
|
=cut |
893 |
|
|
894 |
|
sub config { |
895 |
|
return unless ($config); |
896 |
|
|
897 |
|
my $p = shift; |
898 |
|
|
899 |
|
$p ||= ''; |
900 |
|
|
901 |
|
my $v; |
902 |
|
|
903 |
|
warn "### getting config($p)\n" if ($debug > 1); |
904 |
|
|
905 |
|
my @p = split(/\s+/,$p); |
906 |
|
if ($#p < 0) { |
907 |
|
$v = $config->{ '_' }; # special, database code |
908 |
|
} else { |
909 |
|
|
910 |
|
my $c = dclone( $config ); |
911 |
|
|
912 |
|
foreach my $k (@p) { |
913 |
|
warn "### k: $k c = ",dump($c),$/ if ($debug > 1); |
914 |
|
if (ref($c) eq 'ARRAY') { |
915 |
|
$c = shift @$c; |
916 |
|
warn "config($p) taking first occurence of '$k', probably not what you wanted!\n"; |
917 |
|
last; |
918 |
|
} |
919 |
|
|
920 |
|
if (! defined($c->{$k}) ) { |
921 |
|
$c = undef; |
922 |
|
last; |
923 |
|
} else { |
924 |
|
$c = $c->{$k}; |
925 |
|
} |
926 |
|
} |
927 |
|
$v = $c if ($c); |
928 |
|
|
929 |
|
} |
930 |
|
|
931 |
|
warn "## config( '$p' ) = ",dump( $v ),$/ if ($v && $debug); |
932 |
|
warn "config( '$p' ) is empty\n" if (! $v); |
933 |
|
|
934 |
|
return $v; |
935 |
|
} |
936 |
|
|
937 |
|
=head2 id |
938 |
|
|
939 |
|
Returns unique id of this record |
940 |
|
|
941 |
|
$id = id(); |
942 |
|
|
943 |
|
Returns C<42/2> for 2nd occurence of MFN 42. |
944 |
|
|
945 |
|
=cut |
946 |
|
|
947 |
|
sub id { |
948 |
|
my $mfn = $config->{_mfn} || die "no _mfn in config data"; |
949 |
|
return $mfn . $#{$marc_record} ? $#{$marc_record} + 1 : ''; |
950 |
|
} |
951 |
|
|
952 |
=head2 join_with |
=head2 join_with |
953 |
|
|
954 |
Joins walues with some delimiter |
Joins walues with some delimiter |
959 |
|
|
960 |
sub join_with { |
sub join_with { |
961 |
my $d = shift; |
my $d = shift; |
962 |
return join($d, grep { defined($_) && $_ ne '' } @_); |
warn "### join_with('$d',",dump(@_),")\n" if ($debug > 2); |
963 |
|
my $v = join($d, grep { defined($_) && $_ ne '' } @_); |
964 |
|
return '' unless defined($v); |
965 |
|
return $v; |
966 |
|
} |
967 |
|
|
968 |
|
=head2 split_rec_on |
969 |
|
|
970 |
|
Split record subfield on some regex and take one of parts out |
971 |
|
|
972 |
|
$a_before_semi_column = |
973 |
|
split_rec_on('200','a', /\s*;\s*/, $part); |
974 |
|
|
975 |
|
C<$part> is optional number of element. First element is |
976 |
|
B<1>, not 0! |
977 |
|
|
978 |
|
If there is no C<$part> parameter or C<$part> is 0, this function will |
979 |
|
return all values produced by splitting. |
980 |
|
|
981 |
|
=cut |
982 |
|
|
983 |
|
sub split_rec_on { |
984 |
|
die "split_rec_on need (fld,sf,regex[,part]" if ($#_ < 2); |
985 |
|
|
986 |
|
my ($fld, $sf, $regex, $part) = @_; |
987 |
|
warn "### regex ", ref($regex), $regex, $/ if ($debug > 2); |
988 |
|
|
989 |
|
my @r = rec( $fld, $sf ); |
990 |
|
my $v = shift @r; |
991 |
|
warn "### first rec($fld,$sf) = ",dump($v),$/ if ($debug > 2); |
992 |
|
|
993 |
|
return '' if ( ! defined($v) || $v =~ /^\s*$/); |
994 |
|
|
995 |
|
my @s = split( $regex, $v ); |
996 |
|
warn "## split_rec_on($fld,$sf,$regex,$part) = ",dump(@s),$/ if ($debug > 1); |
997 |
|
if ($part && $part > 0) { |
998 |
|
return $s[ $part - 1 ]; |
999 |
|
} else { |
1000 |
|
return @s; |
1001 |
|
} |
1002 |
} |
} |
1003 |
|
|
1004 |
# END |
# END |