7 |
_debug |
_debug |
8 |
_pack_subfields_hash |
_pack_subfields_hash |
9 |
|
|
10 |
|
to |
11 |
search_display search display sorted |
search_display search display sorted |
12 |
|
|
|
marc marc_indicators marc_repeatable_subfield |
|
|
marc_compose marc_leader marc_fixed |
|
|
marc_duplicate marc_remove marc_count |
|
|
marc_original_order |
|
|
|
|
13 |
rec1 rec2 rec |
rec1 rec2 rec |
14 |
frec |
frec frec_eq frec_ne |
15 |
regex prefix suffix surround |
regex prefix suffix surround |
16 |
first lookup join_with |
first lookup join_with |
17 |
save_into_lookup |
save_into_lookup |
21 |
get set |
get set |
22 |
count |
count |
23 |
|
|
24 |
|
row |
25 |
|
rec_array |
26 |
|
|
27 |
/; |
/; |
28 |
|
|
29 |
use warnings; |
use warnings; |
31 |
|
|
32 |
#use base qw/WebPAC::Common/; |
#use base qw/WebPAC::Common/; |
33 |
use Data::Dump qw/dump/; |
use Data::Dump qw/dump/; |
|
use Storable qw/dclone/; |
|
34 |
use Carp qw/confess/; |
use Carp qw/confess/; |
35 |
|
|
36 |
# debugging warn(s) |
# debugging warn(s) |
37 |
my $debug = 0; |
my $debug = 0; |
38 |
|
_debug( $debug ); |
39 |
|
|
40 |
|
# FIXME |
41 |
use WebPAC::Normalize::ISBN; |
use WebPAC::Normalize::ISBN; |
42 |
push @EXPORT, ( 'isbn_10', 'isbn_13' ); |
push @EXPORT, ( 'isbn_10', 'isbn_13' ); |
43 |
|
|
44 |
|
use WebPAC::Normalize::MARC; |
45 |
|
push @EXPORT, ( qw/ |
46 |
|
marc marc_indicators marc_repeatable_subfield |
47 |
|
marc_compose marc_leader marc_fixed |
48 |
|
marc_duplicate marc_remove marc_count |
49 |
|
marc_original_order |
50 |
|
marc_template |
51 |
|
/); |
52 |
|
|
53 |
|
use Storable qw/dclone/; |
54 |
|
|
55 |
=head1 NAME |
=head1 NAME |
56 |
|
|
57 |
WebPAC::Normalize - describe normalisaton rules using sets |
WebPAC::Normalize - describe normalisaton rules using sets |
58 |
|
|
59 |
=cut |
=cut |
60 |
|
|
61 |
our $VERSION = '0.32'; |
our $VERSION = '0.36'; |
62 |
|
|
63 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
64 |
|
|
142 |
|
|
143 |
sub _set_ds { |
sub _set_ds { |
144 |
$rec = shift or die "no record hash"; |
$rec = shift or die "no record hash"; |
145 |
|
$WebPAC::Normalize::MARC::rec = $rec; |
146 |
|
} |
147 |
|
|
148 |
|
=head2 |
149 |
|
|
150 |
|
my $rec = _get_rec(); |
151 |
|
|
152 |
|
=cut |
153 |
|
|
154 |
|
sub _get_rec { $rec }; |
155 |
|
|
156 |
|
sub rec_array { |
157 |
|
my $d = $rec->{ $_[0] }; |
158 |
|
return @$d if ref($d) eq 'ARRAY'; |
159 |
|
die "field $_[0] not array: ",dump( $d ); |
160 |
} |
} |
161 |
|
|
162 |
=head2 _set_config |
=head2 _set_config |
195 |
|
|
196 |
=cut |
=cut |
197 |
|
|
198 |
my ($out, $marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $marc_leader); |
my $out; |
|
my ($marc_record_offset, $marc_fetch_offset) = (0, 0); |
|
199 |
|
|
200 |
sub _get_ds { |
sub _get_ds { |
201 |
#warn "## out = ",dump($out); |
#warn "## out = ",dump($out); |
212 |
|
|
213 |
sub _clean_ds { |
sub _clean_ds { |
214 |
my $a = {@_}; |
my $a = {@_}; |
215 |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $marc_leader) = (); |
$out = undef; |
216 |
($marc_record_offset, $marc_fetch_offset) = (0,0); |
WebPAC::Normalize::MARC::_clean(); |
|
$marc_encoding = $a->{marc_encoding}; |
|
217 |
} |
} |
218 |
|
|
219 |
=head2 _set_lookup |
=head2 _set_lookup |
261 |
$load_row_coderef = $coderef; |
$load_row_coderef = $coderef; |
262 |
} |
} |
263 |
|
|
|
=head2 _get_marc_fields |
|
|
|
|
|
Get all fields defined by calls to C<marc> |
|
|
|
|
|
$marc->add_fields( WebPAC::Normalize:_get_marc_fields() ); |
|
|
|
|
|
We are using I<magic> which detect repeatable fields only from |
|
|
sequence of field/subfield data generated by normalization. |
|
|
|
|
|
Repeatable field is created when there is second occurence of same subfield or |
|
|
if any of indicators are different. |
|
|
|
|
|
This is sane for most cases. Something like: |
|
|
|
|
|
900a-1 900b-1 900c-1 |
|
|
900a-2 900b-2 |
|
|
900a-3 |
|
|
|
|
|
will be created from any combination of: |
|
|
|
|
|
900a-1 900a-2 900a-3 900b-1 900b-2 900c-1 |
|
|
|
|
|
and following rules: |
|
|
|
|
|
marc('900','a', rec('200','a') ); |
|
|
marc('900','b', rec('200','b') ); |
|
|
marc('900','c', rec('200','c') ); |
|
|
|
|
|
which might not be what you have in mind. If you need repeatable subfield, |
|
|
define it using C<marc_repeatable_subfield> like this: |
|
|
|
|
|
marc_repeatable_subfield('900','a'); |
|
|
marc('900','a', rec('200','a') ); |
|
|
marc('900','b', rec('200','b') ); |
|
|
marc('900','c', rec('200','c') ); |
|
|
|
|
|
will create: |
|
|
|
|
|
900a-1 900a-2 900a-3 900b-1 900c-1 |
|
|
900b-2 |
|
|
|
|
|
There is also support for returning next or specific using: |
|
|
|
|
|
while (my $mf = WebPAC::Normalize:_get_marc_fields( fetch_next => 1 ) ) { |
|
|
# do something with $mf |
|
|
} |
|
|
|
|
|
will always return fields from next MARC record or |
|
|
|
|
|
my $mf = WebPAC::Normalize::_get_marc_fields( offset => 42 ); |
|
|
|
|
|
will return 42th copy record (if it exists). |
|
|
|
|
|
=cut |
|
|
|
|
|
my $fetch_pos; |
|
|
|
|
|
sub _get_marc_fields { |
|
|
|
|
|
my $arg = {@_}; |
|
|
warn "### _get_marc_fields arg: ", dump($arg), $/ if ($debug > 2); |
|
|
$fetch_pos = $marc_fetch_offset; |
|
|
if ($arg->{offset}) { |
|
|
$fetch_pos = $arg->{offset}; |
|
|
} elsif($arg->{fetch_next}) { |
|
|
$marc_fetch_offset++; |
|
|
} |
|
|
|
|
|
return if (! $marc_record || ref($marc_record) ne 'ARRAY'); |
|
|
|
|
|
warn "### full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 2); |
|
|
|
|
|
my $marc_rec = $marc_record->[ $fetch_pos ]; |
|
|
|
|
|
warn "## _get_marc_fields (at offset: $fetch_pos) -- marc_record = ", dump( @$marc_rec ), $/ if ($debug > 1); |
|
|
|
|
|
return if (! $marc_rec || ref($marc_rec) ne 'ARRAY' || $#{ $marc_rec } < 0); |
|
|
|
|
|
# first, sort all existing fields |
|
|
# XXX might not be needed, but modern perl might randomize elements in hash |
|
|
my @sorted_marc_record = sort { |
|
|
$a->[0] . ( $a->[3] || '' ) cmp $b->[0] . ( $b->[3] || '') |
|
|
} @{ $marc_rec }; |
|
|
|
|
|
@sorted_marc_record = @{ $marc_rec }; ### FIXME disable sorting |
|
|
|
|
|
# output marc fields |
|
|
my @m; |
|
|
|
|
|
# count unique field-subfields (used for offset when walking to next subfield) |
|
|
my $u; |
|
|
map { $u->{ $_->[0] . ( $_->[3] || '') }++ } @sorted_marc_record; |
|
|
|
|
|
if ($debug) { |
|
|
warn "## marc_repeatable_subfield = ", dump( $marc_repeatable_subfield ), $/ if ( $marc_repeatable_subfield ); |
|
|
warn "## marc_record[$fetch_pos] = ", dump( $marc_rec ), $/; |
|
|
warn "## sorted_marc_record = ", dump( \@sorted_marc_record ), $/; |
|
|
warn "## subfield count = ", dump( $u ), $/; |
|
|
} |
|
|
|
|
|
my $len = $#sorted_marc_record; |
|
|
my $visited; |
|
|
my $i = 0; |
|
|
my $field; |
|
|
|
|
|
foreach ( 0 .. $len ) { |
|
|
|
|
|
# find next element which isn't visited |
|
|
while ($visited->{$i}) { |
|
|
$i = ($i + 1) % ($len + 1); |
|
|
} |
|
|
|
|
|
# mark it visited |
|
|
$visited->{$i}++; |
|
|
|
|
|
my $row = dclone( $sorted_marc_record[$i] ); |
|
|
|
|
|
# field and subfield which is key for |
|
|
# marc_repeatable_subfield and u |
|
|
my $fsf = $row->[0] . ( $row->[3] || '' ); |
|
|
|
|
|
if ($debug > 1) { |
|
|
|
|
|
print "### field so far [", $#$field, "] : ", dump( $field ), " ", $field ? 'T' : 'F', $/; |
|
|
print "### this [$i]: ", dump( $row ),$/; |
|
|
print "### sf: ", $row->[3], " vs ", $field->[3], |
|
|
$marc_repeatable_subfield->{ $row->[0] . $row->[3] } ? ' (repeatable)' : '', $/, |
|
|
if ($#$field >= 0); |
|
|
|
|
|
} |
|
|
|
|
|
# if field exists |
|
|
if ( $#$field >= 0 ) { |
|
|
if ( |
|
|
$row->[0] ne $field->[0] || # field |
|
|
$row->[1] ne $field->[1] || # i1 |
|
|
$row->[2] ne $field->[2] # i2 |
|
|
) { |
|
|
push @m, $field; |
|
|
warn "## saved/1 ", dump( $field ),$/ if ($debug); |
|
|
$field = $row; |
|
|
|
|
|
} elsif ( |
|
|
( $row->[3] lt $field->[-2] ) # subfield which is not next (e.g. a after c) |
|
|
|| |
|
|
( $row->[3] eq $field->[-2] && # same subfield, but not repeatable |
|
|
! $marc_repeatable_subfield->{ $fsf } |
|
|
) |
|
|
) { |
|
|
push @m, $field; |
|
|
warn "## saved/2 ", dump( $field ),$/ if ($debug); |
|
|
$field = $row; |
|
|
|
|
|
} else { |
|
|
# append new subfields to existing field |
|
|
push @$field, ( $row->[3], $row->[4] ); |
|
|
} |
|
|
} else { |
|
|
# insert first field |
|
|
$field = $row; |
|
|
} |
|
|
|
|
|
if (! $marc_repeatable_subfield->{ $fsf }) { |
|
|
# make step to next subfield |
|
|
$i = ($i + $u->{ $fsf } ) % ($len + 1); |
|
|
} |
|
|
} |
|
|
|
|
|
if ($#$field >= 0) { |
|
|
push @m, $field; |
|
|
warn "## saved/3 ", dump( $field ),$/ if ($debug); |
|
|
} |
|
|
|
|
|
return \@m; |
|
|
} |
|
|
|
|
|
=head2 _get_marc_leader |
|
|
|
|
|
Return leader from currently fetched record by L</_get_marc_fields> |
|
|
|
|
|
print WebPAC::Normalize::_get_marc_leader(); |
|
|
|
|
|
=cut |
|
|
|
|
|
sub _get_marc_leader { |
|
|
die "no fetch_pos, did you called _get_marc_fields first?" unless ( defined( $fetch_pos ) ); |
|
|
return $marc_leader->[ $fetch_pos ]; |
|
|
} |
|
|
|
|
264 |
=head2 _debug |
=head2 _debug |
265 |
|
|
266 |
Change level of debug warnings |
Change level of debug warnings |
274 |
return $debug unless defined($l); |
return $debug unless defined($l); |
275 |
warn "debug level $l",$/ if ($l > 0); |
warn "debug level $l",$/ if ($l > 0); |
276 |
$debug = $l; |
$debug = $l; |
277 |
|
$WebPAC::Normalize::MARC::debug = $debug; |
278 |
} |
} |
279 |
|
|
280 |
=head1 Functions to create C<data_structure> |
=head1 Functions to create C<data_structure> |
281 |
|
|
282 |
Those functions generally have to first in your normalization file. |
Those functions generally have to first in your normalization file. |
283 |
|
|
284 |
|
=head2 to |
285 |
|
|
286 |
|
Generic way to set values for some name |
287 |
|
|
288 |
|
to('field-name', 'name-value' => rec('200','a') ); |
289 |
|
|
290 |
|
There are many helpers defined below which might be easier to use. |
291 |
|
|
292 |
|
=cut |
293 |
|
|
294 |
|
sub to { |
295 |
|
my $type = shift or confess "need type -- BUG?"; |
296 |
|
my $name = shift or confess "needs name as first argument"; |
297 |
|
my @o = grep { defined($_) && $_ ne '' } @_; |
298 |
|
return unless (@o); |
299 |
|
$out->{$name}->{$type} = \@o; |
300 |
|
} |
301 |
|
|
302 |
=head2 search_display |
=head2 search_display |
303 |
|
|
304 |
Define output for L<search> and L<display> at the same time |
Define output for L<search> and L<display> at the same time |
305 |
|
|
306 |
search_display('Title', rec('200','a') ); |
search_display('Title', rec('200','a') ); |
307 |
|
|
|
|
|
308 |
=cut |
=cut |
309 |
|
|
310 |
sub search_display { |
sub search_display { |
317 |
|
|
318 |
=head2 tag |
=head2 tag |
319 |
|
|
320 |
Old name for L<search_display>, but supported |
Old name for L<search_display>, it will probably be removed at one point. |
321 |
|
|
322 |
=cut |
=cut |
323 |
|
|
333 |
|
|
334 |
=cut |
=cut |
335 |
|
|
336 |
sub _field { |
sub display { to( 'display', @_ ) } |
|
my $type = shift or confess "need type -- BUG?"; |
|
|
my $name = shift or confess "needs name as first argument"; |
|
|
my @o = grep { defined($_) && $_ ne '' } @_; |
|
|
return unless (@o); |
|
|
$out->{$name}->{$type} = \@o; |
|
|
} |
|
|
|
|
|
sub display { _field( 'display', @_ ) } |
|
337 |
|
|
338 |
=head2 search |
=head2 search |
339 |
|
|
343 |
|
|
344 |
=cut |
=cut |
345 |
|
|
346 |
sub search { _field( 'search', @_ ) } |
sub search { to( 'search', @_ ) } |
347 |
|
|
348 |
=head2 sorted |
=head2 sorted |
349 |
|
|
353 |
|
|
354 |
=cut |
=cut |
355 |
|
|
356 |
sub sorted { _field( 'sorted', @_ ) } |
sub sorted { to( 'sorted', @_ ) } |
|
|
|
357 |
|
|
358 |
=head2 marc_leader |
=head2 row |
359 |
|
|
360 |
Setup fields within MARC leader or get leader |
Insert new row of data into output module |
361 |
|
|
362 |
marc_leader('05','c'); |
row( column => 'foo', column2 => 'bar' ); |
|
my $leader = marc_leader(); |
|
363 |
|
|
364 |
=cut |
=cut |
365 |
|
|
366 |
sub marc_leader { |
use Data::Dump qw/dump/; |
|
my ($offset,$value) = @_; |
|
|
|
|
|
if ($offset) { |
|
|
$marc_leader->[ $marc_record_offset ]->{ $offset } = $value; |
|
|
} else { |
|
|
|
|
|
if (defined($marc_leader)) { |
|
|
die "marc_leader not array = ", dump( $marc_leader ) unless (ref($marc_leader) eq 'ARRAY'); |
|
|
return $marc_leader->[ $marc_record_offset ]; |
|
|
} else { |
|
|
return; |
|
|
} |
|
|
} |
|
|
} |
|
|
|
|
|
=head2 marc_fixed |
|
|
|
|
|
Create control/indentifier fields with values in fixed positions |
|
|
|
|
|
marc_fixed('008', 00, '070402'); |
|
|
marc_fixed('008', 39, '|'); |
|
|
|
|
|
Positions not specified will be filled with spaces (C<0x20>). |
|
|
|
|
|
There will be no effort to extend last specified value to full length of |
|
|
field in standard. |
|
|
|
|
|
=cut |
|
|
|
|
|
sub marc_fixed { |
|
|
my ($f, $pos, $val) = @_; |
|
|
die "need marc(field, position, value)" unless defined($f) && defined($pos); |
|
|
|
|
|
confess "need val" unless defined $val; |
|
|
|
|
|
my $update = 0; |
|
|
|
|
|
map { |
|
|
if ($_->[0] eq $f) { |
|
|
my $old = $_->[1]; |
|
|
if (length($old) <= $pos) { |
|
|
$_->[1] .= ' ' x ( $pos - length($old) ) . $val; |
|
|
warn "## marc_fixed($f,$pos,'$val') append '$old' -> '$_->[1]'\n" if ($debug > 1); |
|
|
} else { |
|
|
$_->[1] = substr($old, 0, $pos) . $val . substr($old, $pos + length($val)); |
|
|
warn "## marc_fixed($f,$pos,'$val') update '$old' -> '$_->[1]'\n" if ($debug > 1); |
|
|
} |
|
|
$update++; |
|
|
} |
|
|
} @{ $marc_record->[ $marc_record_offset ] }; |
|
|
|
|
|
if (! $update) { |
|
|
my $v = ' ' x $pos . $val; |
|
|
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $v ]; |
|
|
warn "## marc_fixed($f,$pos,'val') created '$v'\n" if ($debug > 1); |
|
|
} |
|
|
} |
|
|
|
|
|
=head2 marc |
|
|
|
|
|
Save value for MARC field |
|
|
|
|
|
marc('900','a', rec('200','a') ); |
|
|
marc('001', rec('000') ); |
|
|
|
|
|
=cut |
|
|
|
|
|
sub marc { |
|
|
my $f = shift or die "marc needs field"; |
|
|
die "marc field must be numer" unless ($f =~ /^\d+$/); |
|
|
|
|
|
my $sf; |
|
|
if ($f >= 10) { |
|
|
$sf = shift or die "marc needs subfield"; |
|
|
} |
|
|
|
|
|
foreach (@_) { |
|
|
my $v = $_; # make var read-write for Encode |
|
|
next unless (defined($v) && $v !~ /^\s*$/); |
|
|
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
|
|
if (defined $sf) { |
|
|
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $i1, $i2, $sf => $v ]; |
|
|
} else { |
|
|
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $v ]; |
|
|
} |
|
|
} |
|
|
} |
|
|
|
|
|
=head2 marc_repeatable_subfield |
|
|
|
|
|
Save values for MARC repetable subfield |
|
|
|
|
|
marc_repeatable_subfield('910', 'z', rec('909') ); |
|
|
|
|
|
=cut |
|
|
|
|
|
sub marc_repeatable_subfield { |
|
|
my ($f,$sf) = @_; |
|
|
die "marc_repeatable_subfield need field and subfield!\n" unless ($f && $sf); |
|
|
$marc_repeatable_subfield->{ $f . $sf }++; |
|
|
marc(@_); |
|
|
} |
|
|
|
|
|
=head2 marc_indicators |
|
|
|
|
|
Set both indicators for MARC field |
|
|
|
|
|
marc_indicators('900', ' ', 1); |
|
|
|
|
|
Any indicator value other than C<0-9> will be treated as undefined. |
|
|
|
|
|
=cut |
|
|
|
|
|
sub marc_indicators { |
|
|
my $f = shift || die "marc_indicators need field!\n"; |
|
|
my ($i1,$i2) = @_; |
|
|
die "marc_indicators($f, ...) need i1!\n" unless(defined($i1)); |
|
|
die "marc_indicators($f, $i1, ...) need i2!\n" unless(defined($i2)); |
|
|
|
|
|
$i1 = ' ' if ($i1 !~ /^\d$/); |
|
|
$i2 = ' ' if ($i2 !~ /^\d$/); |
|
|
@{ $marc_indicators->{$f} } = ($i1,$i2); |
|
|
} |
|
|
|
|
|
=head2 marc_compose |
|
|
|
|
|
Save values for each MARC subfield explicitly |
|
|
|
|
|
marc_compose('900', |
|
|
'a', rec('200','a') |
|
|
'b', rec('201','a') |
|
|
'a', rec('200','b') |
|
|
'c', rec('200','c') |
|
|
); |
|
|
|
|
|
If you specify C<+> for subfield, value will be appended |
|
|
to previous defined subfield. |
|
|
|
|
|
=cut |
|
|
|
|
|
sub marc_compose { |
|
|
my $f = shift or die "marc_compose needs field"; |
|
|
die "marc_compose field must be numer" unless ($f =~ /^\d+$/); |
|
|
|
|
|
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
|
|
my $m = [ $f, $i1, $i2 ]; |
|
|
|
|
|
warn "### marc_compose input subfields = ", dump(@_),$/ if ($debug > 2); |
|
|
|
|
|
if ($#_ % 2 != 1) { |
|
|
die "ERROR: marc_compose",dump($f,@_)," not valid (must be even).\nDo you need to add first() or join() around some argument?\n"; |
|
|
} |
|
|
|
|
|
while (@_) { |
|
|
my $sf = shift; |
|
|
my $v = shift; |
|
|
|
|
|
next unless (defined($v) && $v !~ /^\s*$/); |
|
|
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
|
|
if ($sf ne '+') { |
|
|
push @$m, ( $sf, $v ); |
|
|
} else { |
|
|
$m->[ $#$m ] .= $v; |
|
|
} |
|
|
} |
|
|
|
|
|
warn "## marc_compose current marc = ", dump( $m ),$/ if ($debug > 1); |
|
|
|
|
|
push @{ $marc_record->[ $marc_record_offset ] }, $m if ($#{$m} > 2); |
|
|
} |
|
|
|
|
|
=head2 marc_duplicate |
|
|
|
|
|
Generate copy of current MARC record and continue working on copy |
|
|
|
|
|
marc_duplicate(); |
|
|
|
|
|
Copies can be accessed using C<< _get_marc_fields( fetch_next => 1 ) >> or |
|
|
C<< _get_marc_fields( offset => 42 ) >>. |
|
|
|
|
|
=cut |
|
|
|
|
|
sub marc_duplicate { |
|
|
my $m = $marc_record->[ -1 ]; |
|
|
die "can't duplicate record which isn't defined" unless ($m); |
|
|
push @{ $marc_record }, dclone( $m ); |
|
|
push @{ $marc_leader }, dclone( marc_leader() ); |
|
|
warn "## marc_duplicate = ", dump(@$marc_leader, @$marc_record), $/ if ($debug > 1); |
|
|
$marc_record_offset = $#{ $marc_record }; |
|
|
warn "## marc_record_offset = $marc_record_offset", $/ if ($debug > 1); |
|
|
|
|
|
} |
|
|
|
|
|
=head2 marc_remove |
|
|
|
|
|
Remove some field or subfield from MARC record. |
|
|
|
|
|
marc_remove('200'); |
|
|
marc_remove('200','a'); |
|
|
|
|
|
This will erase field C<200> or C<200^a> from current MARC record. |
|
|
|
|
|
marc_remove('*'); |
|
|
|
|
|
Will remove all fields in current MARC record. |
|
|
|
|
|
This is useful after calling C<marc_duplicate> or on it's own (but, you |
|
|
should probably just remove that subfield definition if you are not |
|
|
using C<marc_duplicate>). |
|
|
|
|
|
FIXME: support fields < 10. |
|
|
|
|
|
=cut |
|
|
|
|
|
sub marc_remove { |
|
|
my ($f, $sf) = @_; |
|
|
|
|
|
die "marc_remove needs record number" unless defined($f); |
|
|
|
|
|
my $marc = $marc_record->[ $marc_record_offset ]; |
|
|
|
|
|
warn "### marc_remove before = ", dump( $marc ), $/ if ($debug > 2); |
|
|
|
|
|
if ($f eq '*') { |
|
|
|
|
|
delete( $marc_record->[ $marc_record_offset ] ); |
|
|
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
|
|
|
|
|
} else { |
|
|
|
|
|
my $i = 0; |
|
|
foreach ( 0 .. $#{ $marc } ) { |
|
|
last unless (defined $marc->[$i]); |
|
|
warn "#### working on ",dump( @{ $marc->[$i] }), $/ if ($debug > 3); |
|
|
if ($marc->[$i]->[0] eq $f) { |
|
|
if (! defined $sf) { |
|
|
# remove whole field |
|
|
splice @$marc, $i, 1; |
|
|
warn "#### slice \@\$marc, $i, 1 = ",dump( @{ $marc }), $/ if ($debug > 3); |
|
|
$i--; |
|
|
} else { |
|
|
foreach my $j ( 0 .. (( $#{ $marc->[$i] } - 3 ) / 2) ) { |
|
|
my $o = ($j * 2) + 3; |
|
|
if ($marc->[$i]->[$o] eq $sf) { |
|
|
# remove subfield |
|
|
splice @{$marc->[$i]}, $o, 2; |
|
|
warn "#### slice \@{\$marc->[$i]}, $o, 2 = ", dump( @{ $marc }), $/ if ($debug > 3); |
|
|
# is record now empty? |
|
|
if ($#{ $marc->[$i] } == 2) { |
|
|
splice @$marc, $i, 1; |
|
|
warn "#### slice \@\$marc, $i, 1 = ", dump( @{ $marc }), $/ if ($debug > 3); |
|
|
$i--; |
|
|
}; |
|
|
} |
|
|
} |
|
|
} |
|
|
} |
|
|
$i++; |
|
|
} |
|
|
|
|
|
warn "### marc_remove($f", $sf ? ",$sf" : "", ") after = ", dump( $marc ), $/ if ($debug > 2); |
|
|
|
|
|
$marc_record->[ $marc_record_offset ] = $marc; |
|
|
} |
|
|
|
|
|
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
|
|
} |
|
|
|
|
|
=head2 marc_original_order |
|
|
|
|
|
Copy all subfields preserving original order to marc field. |
|
|
|
|
|
marc_original_order( marc_field_number, original_input_field_number ); |
|
|
|
|
|
Please note that field numbers are consistent with other commands (marc |
|
|
field number first), but somewhat counter-intuitive (destination and then |
|
|
source). |
|
|
|
|
|
You might want to use this command if you are just renaming subfields or |
|
|
using pre-processing modify_record in C<config.yml> and don't need any |
|
|
post-processing or want to preserve order of original subfields. |
|
|
|
|
|
|
|
|
=cut |
|
|
|
|
|
sub marc_original_order { |
|
|
|
|
|
my ($to, $from) = @_; |
|
|
die "marc_original_order needs from and to fields\n" unless ($from && $to); |
|
|
|
|
|
return unless defined($rec->{$from}); |
|
|
|
|
|
my $r = $rec->{$from}; |
|
|
die "record field $from isn't array\n" unless (ref($r) eq 'ARRAY'); |
|
|
|
|
|
my ($i1,$i2) = defined($marc_indicators->{$to}) ? @{ $marc_indicators->{$to} } : (' ',' '); |
|
|
warn "## marc_original_order($to,$from) source = ", dump( $r ),$/ if ($debug > 1); |
|
|
|
|
|
foreach my $d (@$r) { |
|
|
|
|
|
if (! defined($d->{subfields}) && ref($d->{subfields}) ne 'ARRAY') { |
|
|
warn "# marc_original_order($to,$from): field $from doesn't have subfields specification\n"; |
|
|
next; |
|
|
} |
|
|
|
|
|
my @sfs = @{ $d->{subfields} }; |
|
|
|
|
|
die "field $from doesn't have even number of subfields specifications\n" unless($#sfs % 2 == 1); |
|
|
|
|
|
warn "#--> d: ",dump($d), "\n#--> sfs: ",dump(@sfs),$/ if ($debug > 2); |
|
|
|
|
|
my $m = [ $to, $i1, $i2 ]; |
|
|
|
|
|
while (my $sf = shift @sfs) { |
|
|
|
|
|
warn "#--> sf: ",dump($sf), $/ if ($debug > 2); |
|
|
my $offset = shift @sfs; |
|
|
die "corrupted sufields specification for field $from\n" unless defined($offset); |
|
|
|
|
|
my $v; |
|
|
if (ref($d->{$sf}) eq 'ARRAY') { |
|
|
$v = $d->{$sf}->[$offset] if (defined($d->{$sf}->[$offset])); |
|
|
} elsif ($offset == 0) { |
|
|
$v = $d->{$sf}; |
|
|
} else { |
|
|
die "field $from subfield '$sf' need occurence $offset which doesn't exist", dump($d->{$sf}); |
|
|
} |
|
|
push @$m, ( $sf, $v ) if (defined($v)); |
|
|
} |
|
|
|
|
|
if ($#{$m} > 2) { |
|
|
push @{ $marc_record->[ $marc_record_offset ] }, $m; |
|
|
} |
|
|
} |
|
|
|
|
|
warn "## marc_record = ", dump( $marc_record ),$/ if ($debug > 1); |
|
|
} |
|
|
|
|
|
=head2 marc_count |
|
|
|
|
|
Return number of MARC records created using L</marc_duplicate>. |
|
|
|
|
|
print "created ", marc_count(), " records"; |
|
367 |
|
|
368 |
=cut |
sub row { |
369 |
|
die "array doesn't have even number of elements but $#_: ",dump( @_ ) if $#_ % 2 != 1; |
370 |
|
|
371 |
sub marc_count { |
push @{ $out->{'_rows'} }, {@_}; |
|
return $#{ $marc_record }; |
|
372 |
} |
} |
373 |
|
|
374 |
|
|
391 |
|
|
392 |
warn "## _pack_subfields_hash( ",dump(@_), " )\n" if ($debug > 1); |
warn "## _pack_subfields_hash( ",dump(@_), " )\n" if ($debug > 1); |
393 |
|
|
394 |
my ($h,$include_subfields) = @_; |
my ($hash,$include_subfields) = @_; |
395 |
|
|
396 |
# sanity and ease of use |
# sanity and ease of use |
397 |
return $h if (ref($h) ne 'HASH'); |
return $hash if (ref($hash) ne 'HASH'); |
398 |
|
|
399 |
|
my $h = dclone( $hash ); |
400 |
|
|
401 |
if ( defined($h->{subfields}) ) { |
if ( defined($h->{subfields}) ) { |
402 |
my $sfs = delete $h->{subfields} || die "no subfields?"; |
my $sfs = delete $h->{subfields} || die "no subfields?"; |
487 |
} else { |
} else { |
488 |
$_->{$sf}; |
$_->{$sf}; |
489 |
} |
} |
490 |
} grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} }; |
} grep { ref($_) eq 'HASH' && defined $_->{$sf} } @{ $rec->{$f} }; |
491 |
} |
} |
492 |
|
|
493 |
=head2 rec |
=head2 rec |
533 |
return shift @out; |
return shift @out; |
534 |
} |
} |
535 |
|
|
536 |
|
=head2 frec_eq |
537 |
|
|
538 |
|
=head2 frec_ne |
539 |
|
|
540 |
|
Check if first values from two fields are same or different |
541 |
|
|
542 |
|
if ( frec_eq( 900 => 'a', 910 => 'c' ) ) { |
543 |
|
# values are same |
544 |
|
} else { |
545 |
|
# values are different |
546 |
|
} |
547 |
|
|
548 |
|
Strictly speaking C<frec_eq> and C<frec_ne> wouldn't be needed if you |
549 |
|
could write something like: |
550 |
|
|
551 |
|
if ( frec( '900','a' ) eq frec( '910','c' ) ) { |
552 |
|
# yada tada |
553 |
|
} |
554 |
|
|
555 |
|
but you can't since our parser L<WebPAC::Parser> will remove all whitespaces |
556 |
|
in order to parse text and create invalid function C<eqfrec>. |
557 |
|
|
558 |
|
=cut |
559 |
|
|
560 |
|
sub frec_eq { |
561 |
|
my ( $f1,$sf1, $f2, $sf2 ) = @_; |
562 |
|
return (rec( $f1, $sf1 ))[0] eq (rec( $f2, $sf2 ))[0]; |
563 |
|
} |
564 |
|
|
565 |
|
sub frec_ne { |
566 |
|
return ! frec_eq( @_ ); |
567 |
|
} |
568 |
|
|
569 |
=head2 regex |
=head2 regex |
570 |
|
|
571 |
Apply regex to some or all values |
Apply regex to some or all values |
861 |
|
|
862 |
sub id { |
sub id { |
863 |
my $mfn = $config->{_mfn} || die "no _mfn in config data"; |
my $mfn = $config->{_mfn} || die "no _mfn in config data"; |
864 |
return $mfn . $#{$marc_record} ? $#{$marc_record} + 1 : ''; |
return $mfn . ( WebPAC::Normalize::MARC::_created_marc_records() || '' ); |
865 |
} |
} |
866 |
|
|
867 |
=head2 join_with |
=head2 join_with |