5 |
_get_ds _clean_ds |
_get_ds _clean_ds |
6 |
|
|
7 |
tag search display |
tag search display |
8 |
marc21 |
marc marc_indicators marc_repeatable_subfield |
9 |
|
|
10 |
rec1 rec2 rec |
rec1 rec2 rec |
11 |
regex prefix suffix surround |
regex prefix suffix surround |
16 |
use strict; |
use strict; |
17 |
|
|
18 |
#use base qw/WebPAC::Common/; |
#use base qw/WebPAC::Common/; |
19 |
use Data::Dumper; |
use Data::Dump qw/dump/; |
20 |
use Encode qw/from_to/; |
use Encode qw/from_to/; |
21 |
|
|
22 |
|
# debugging warn(s) |
23 |
|
my $debug = 0; |
24 |
|
|
25 |
|
|
26 |
=head1 NAME |
=head1 NAME |
27 |
|
|
28 |
WebPAC::Normalize - describe normalisaton rules using sets |
WebPAC::Normalize - describe normalisaton rules using sets |
47 |
|
|
48 |
Normalisation can generate multiple output normalized data. For now, supported output |
Normalisation can generate multiple output normalized data. For now, supported output |
49 |
types (on the left side of definition) are: C<tag>, C<display>, C<search> and |
types (on the left side of definition) are: C<tag>, C<display>, C<search> and |
50 |
C<marc21>. |
C<marc>. |
51 |
|
|
52 |
=head1 FUNCTIONS |
=head1 FUNCTIONS |
53 |
|
|
113 |
|
|
114 |
=cut |
=cut |
115 |
|
|
116 |
my $out; |
my ($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators); |
|
my $marc21; |
|
|
my $marc_encoding; |
|
117 |
|
|
118 |
sub _get_ds { |
sub _get_ds { |
119 |
return $out; |
return $out; |
129 |
|
|
130 |
sub _clean_ds { |
sub _clean_ds { |
131 |
my $a = {@_}; |
my $a = {@_}; |
132 |
$out = undef; |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators) = (); |
|
$marc21 = undef; |
|
133 |
$marc_encoding = $a->{marc_encoding}; |
$marc_encoding = $a->{marc_encoding}; |
134 |
} |
} |
135 |
|
|
147 |
$lookup = shift; |
$lookup = shift; |
148 |
} |
} |
149 |
|
|
150 |
=head2 _get_marc21_fields |
=head2 _get_marc_fields |
151 |
|
|
152 |
Get all fields defined by calls to C<marc21> |
Get all fields defined by calls to C<marc> |
153 |
|
|
154 |
$marc->add_fields( WebPAC::Normalize:_get_marc21_fields() ); |
$marc->add_fields( WebPAC::Normalize:_get_marc_fields() ); |
155 |
|
|
156 |
|
|
157 |
|
|
162 |
if any of indicators are different. This is sane for most cases except for |
if any of indicators are different. This is sane for most cases except for |
163 |
non-repeatable fields with repeatable subfields. |
non-repeatable fields with repeatable subfields. |
164 |
|
|
165 |
B<TODO>: implement exceptions to magic |
You can change behaviour of that using C<marc_repeatable_subfield>. |
166 |
|
|
167 |
=cut |
=cut |
168 |
|
|
169 |
sub _get_marc21_fields { |
sub _get_marc_fields { |
170 |
|
|
171 |
|
return if (! $marc_record || ref($marc_record) ne 'ARRAY' || $#{ $marc_record } < 0); |
172 |
|
|
173 |
|
# first, sort all existing fields |
174 |
|
# XXX might not be needed, but modern perl might randomize elements in hash |
175 |
|
my @sorted_marc_record = sort { |
176 |
|
$a->[0] . $a->[3] cmp $b->[0] . $b->[3] |
177 |
|
} @{ $marc_record }; |
178 |
|
|
179 |
|
# output marc fields |
180 |
my @m; |
my @m; |
181 |
my $last; |
|
182 |
foreach my $row (@{ $marc21 }) { |
# count unique field-subfields (used for offset when walking to next subfield) |
183 |
if ($last && |
my $u; |
184 |
$last->[0] eq $row->[0] && # check if field is same |
map { $u->{ $_->[0] . $_->[3] }++ } @sorted_marc_record; |
185 |
$last->[1] eq $row->[1] && # check for i1 |
|
186 |
$last->[2] eq $row->[2] && # and for i2 |
if ($debug) { |
187 |
$last->[3] ne $row->[3] # and subfield is different |
warn "## marc_repeatable_subfield ", dump( $marc_repeatable_subfield ), $/; |
188 |
) { |
warn "## marc_record ", dump( $marc_record ), $/; |
189 |
push @$last, ( $row->[3] , $row->[4] ); |
warn "## sorted_marc_record ", dump( \@sorted_marc_record ), $/; |
190 |
warn "## ++ added $row->[0] ^$row->[3] to $last->[0]\n"; |
warn "## subfield count ", dump( $u ), $/; |
191 |
next; |
} |
192 |
} elsif ($last) { |
|
193 |
push @m, $last; |
my $len = $#sorted_marc_record; |
194 |
|
my $visited; |
195 |
|
my $i = 0; |
196 |
|
my $field; |
197 |
|
|
198 |
|
foreach ( 0 .. $len ) { |
199 |
|
|
200 |
|
# find next element which isn't visited |
201 |
|
while ($visited->{$i}) { |
202 |
|
$i = ($i + 1) % ($len + 1); |
203 |
} |
} |
204 |
|
|
205 |
$last = $row; |
# mark it visited |
206 |
|
$visited->{$i}++; |
207 |
|
|
208 |
|
my $row = $sorted_marc_record[$i]; |
209 |
|
|
210 |
|
# field and subfield which is key for |
211 |
|
# marc_repeatable_subfield and u |
212 |
|
my $fsf = $row->[0] . $row->[3]; |
213 |
|
|
214 |
|
if ($debug > 1) { |
215 |
|
|
216 |
|
print "### field so far [", $#$field, "] : ", dump( $field ), " ", $field ? 'T' : 'F', $/; |
217 |
|
print "### this [$i]: ", dump( $row ),$/; |
218 |
|
print "### sf: ", $row->[3], " vs ", $field->[3], |
219 |
|
$marc_repeatable_subfield->{ $row->[0] . $row->[3] } ? ' (repeatable)' : '', $/, |
220 |
|
if ($#$field >= 0); |
221 |
|
|
222 |
|
} |
223 |
|
|
224 |
|
# if field exists |
225 |
|
if ( $#$field >= 0 ) { |
226 |
|
if ( |
227 |
|
$row->[0] ne $field->[0] || # field |
228 |
|
$row->[1] ne $field->[1] || # i1 |
229 |
|
$row->[2] ne $field->[2] # i2 |
230 |
|
) { |
231 |
|
push @m, $field; |
232 |
|
warn "## saved/1 ", dump( $field ),$/ if ($debug); |
233 |
|
$field = $row; |
234 |
|
|
235 |
|
} elsif ( |
236 |
|
( $row->[3] lt $field->[-2] ) # subfield which is not next (e.g. a after c) |
237 |
|
|| |
238 |
|
( $row->[3] eq $field->[-2] && # same subfield, but not repeatable |
239 |
|
! $marc_repeatable_subfield->{ $fsf } |
240 |
|
) |
241 |
|
) { |
242 |
|
push @m, $field; |
243 |
|
warn "## saved/2 ", dump( $field ),$/ if ($debug); |
244 |
|
$field = $row; |
245 |
|
|
246 |
|
} else { |
247 |
|
# append new subfields to existing field |
248 |
|
push @$field, ( $row->[3], $row->[4] ); |
249 |
|
} |
250 |
|
} else { |
251 |
|
# insert first field |
252 |
|
$field = $row; |
253 |
|
} |
254 |
|
|
255 |
|
if (! $marc_repeatable_subfield->{ $fsf }) { |
256 |
|
# make step to next subfield |
257 |
|
$i = ($i + $u->{ $fsf } ) % ($len + 1); |
258 |
|
} |
259 |
} |
} |
260 |
|
|
261 |
push @m, $last if ($last); |
if ($#$field >= 0) { |
262 |
|
push @m, $field; |
263 |
|
warn "## saved/3 ", dump( $field ),$/ if ($debug); |
264 |
|
} |
265 |
|
|
266 |
return @m; |
return @m; |
267 |
} |
} |
320 |
$out->{$name}->{search} = \@o; |
$out->{$name}->{search} = \@o; |
321 |
} |
} |
322 |
|
|
323 |
=head2 marc21 |
=head2 marc |
324 |
|
|
325 |
Save value for MARC field |
Save value for MARC field |
326 |
|
|
327 |
marc21('900','a', rec('200','a') ); |
marc('900','a', rec('200','a') ); |
328 |
|
|
329 |
=cut |
=cut |
330 |
|
|
331 |
sub marc21 { |
sub marc { |
332 |
my $f = shift or die "marc21 needs field"; |
my $f = shift or die "marc needs field"; |
333 |
die "marc21 field must be numer" unless ($f =~ /^\d+$/); |
die "marc field must be numer" unless ($f =~ /^\d+$/); |
334 |
|
|
335 |
my $sf = shift or die "marc21 needs subfield"; |
my $sf = shift or die "marc needs subfield"; |
336 |
|
|
337 |
foreach (@_) { |
foreach (@_) { |
338 |
my $v = $_; # make var read-write for Encode |
my $v = $_; # make var read-write for Encode |
339 |
next unless (defined($v) && $v !~ /^\s*$/); |
next unless (defined($v) && $v !~ /^\s*$/); |
340 |
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
341 |
push @{ $marc21 }, [ $f, ' ', ' ', $sf => $v ]; |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
342 |
|
push @{ $marc_record }, [ $f, $i1, $i2, $sf => $v ]; |
343 |
} |
} |
344 |
} |
} |
345 |
|
|
346 |
|
=head2 marc_repeatable_subfield |
347 |
|
|
348 |
|
Save values for MARC repetable subfield |
349 |
|
|
350 |
|
marc_repeatable_subfield('910', 'z', rec('909') ); |
351 |
|
|
352 |
|
=cut |
353 |
|
|
354 |
|
sub marc_repeatable_subfield { |
355 |
|
my ($f,$sf) = @_; |
356 |
|
die "marc_repeatable_subfield need field and subfield!\n" unless ($f && $sf); |
357 |
|
$marc_repeatable_subfield->{ $f . $sf }++; |
358 |
|
marc(@_); |
359 |
|
} |
360 |
|
|
361 |
|
=head2 marc_indicators |
362 |
|
|
363 |
|
Set both indicators for MARC field |
364 |
|
|
365 |
|
marc_indicators('900', ' ', 1); |
366 |
|
|
367 |
|
Any indicator value other than C<0-9> will be treated as undefined. |
368 |
|
|
369 |
|
=cut |
370 |
|
|
371 |
|
sub marc_indicators { |
372 |
|
my $f = shift || die "marc_indicators need field!\n"; |
373 |
|
my ($i1,$i2) = @_; |
374 |
|
die "marc_indicators($f, ...) need i1!\n" unless(defined($i1)); |
375 |
|
die "marc_indicators($f, $i1, ...) need i2!\n" unless(defined($i2)); |
376 |
|
|
377 |
|
$i1 = ' ' if ($i1 !~ /^\d$/); |
378 |
|
$i2 = ' ' if ($i2 !~ /^\d$/); |
379 |
|
@{ $marc_indicators->{$f} } = ($i1,$i2); |
380 |
|
} |
381 |
|
|
382 |
|
|
383 |
=head1 Functions to extract data from input |
=head1 Functions to extract data from input |
384 |
|
|
385 |
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 |
454 |
sub regex { |
sub regex { |
455 |
my $r = shift; |
my $r = shift; |
456 |
my @out; |
my @out; |
457 |
#warn "r: $r\n",Dumper(\@_); |
#warn "r: $r\n", dump(\@_); |
458 |
foreach my $t (@_) { |
foreach my $t (@_) { |
459 |
next unless ($t); |
next unless ($t); |
460 |
eval "\$t =~ $r"; |
eval "\$t =~ $r"; |