1 |
package WebPAC::Normalize; |
2 |
use Exporter 'import'; |
3 |
@EXPORT = qw/ |
4 |
_set_rec _set_lookup |
5 |
_get_ds _clean_ds |
6 |
|
7 |
tag search display |
8 |
marc marc_indicators marc_repeatable_subfield |
9 |
|
10 |
rec1 rec2 rec |
11 |
regex prefix suffix surround |
12 |
first lookup join_with |
13 |
/; |
14 |
|
15 |
use warnings; |
16 |
use strict; |
17 |
|
18 |
#use base qw/WebPAC::Common/; |
19 |
use Data::Dump qw/dump/; |
20 |
use Encode qw/from_to/; |
21 |
|
22 |
# debugging warn(s) |
23 |
my $debug = 0; |
24 |
|
25 |
|
26 |
=head1 NAME |
27 |
|
28 |
WebPAC::Normalize - describe normalisaton rules using sets |
29 |
|
30 |
=head1 VERSION |
31 |
|
32 |
Version 0.06 |
33 |
|
34 |
=cut |
35 |
|
36 |
our $VERSION = '0.06'; |
37 |
|
38 |
=head1 SYNOPSIS |
39 |
|
40 |
This module uses C<conf/normalize/*.pl> files to perform normalisation |
41 |
from input records using perl functions which are specialized for set |
42 |
processing. |
43 |
|
44 |
Sets are implemented as arrays, and normalisation file is valid perl, which |
45 |
means that you check it's validity before running WebPAC using |
46 |
C<perl -c normalize.pl>. |
47 |
|
48 |
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 |
50 |
C<marc>. |
51 |
|
52 |
=head1 FUNCTIONS |
53 |
|
54 |
Functions which start with C<_> are private and used by WebPAC internally. |
55 |
All other functions are available for use within normalisation rules. |
56 |
|
57 |
=head2 data_structure |
58 |
|
59 |
Return data structure |
60 |
|
61 |
my $ds = WebPAC::Normalize::data_structure( |
62 |
lookup => $lookup->lookup_hash, |
63 |
row => $row, |
64 |
rules => $normalize_pl_config, |
65 |
marc_encoding => 'utf-8', |
66 |
); |
67 |
|
68 |
Options C<lookup>, C<row>, C<rules> and C<log> are mandatory while all |
69 |
other are optional. |
70 |
|
71 |
This function will B<die> if normalizastion can't be evaled. |
72 |
|
73 |
Since this function isn't exported you have to call it with |
74 |
C<WebPAC::Normalize::data_structure>. |
75 |
|
76 |
=cut |
77 |
|
78 |
sub data_structure { |
79 |
my $arg = {@_}; |
80 |
|
81 |
die "need row argument" unless ($arg->{row}); |
82 |
die "need normalisation argument" unless ($arg->{rules}); |
83 |
|
84 |
no strict 'subs'; |
85 |
_set_lookup( $arg->{lookup} ); |
86 |
_set_rec( $arg->{row} ); |
87 |
_clean_ds( %{ $arg } ); |
88 |
eval "$arg->{rules}"; |
89 |
die "error evaling $arg->{rules}: $@\n" if ($@); |
90 |
|
91 |
return _get_ds(); |
92 |
} |
93 |
|
94 |
=head2 _set_rec |
95 |
|
96 |
Set current record hash |
97 |
|
98 |
_set_rec( $rec ); |
99 |
|
100 |
=cut |
101 |
|
102 |
my $rec; |
103 |
|
104 |
sub _set_rec { |
105 |
$rec = shift or die "no record hash"; |
106 |
} |
107 |
|
108 |
=head2 _get_ds |
109 |
|
110 |
Return hash formatted as data structure |
111 |
|
112 |
my $ds = _get_ds(); |
113 |
|
114 |
=cut |
115 |
|
116 |
my ($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators); |
117 |
|
118 |
sub _get_ds { |
119 |
return $out; |
120 |
} |
121 |
|
122 |
=head2 _clean_ds |
123 |
|
124 |
Clean data structure hash for next record |
125 |
|
126 |
_clean_ds(); |
127 |
|
128 |
=cut |
129 |
|
130 |
sub _clean_ds { |
131 |
my $a = {@_}; |
132 |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators) = (); |
133 |
$marc_encoding = $a->{marc_encoding}; |
134 |
} |
135 |
|
136 |
=head2 _set_lookup |
137 |
|
138 |
Set current lookup hash |
139 |
|
140 |
_set_lookup( $lookup ); |
141 |
|
142 |
=cut |
143 |
|
144 |
my $lookup; |
145 |
|
146 |
sub _set_lookup { |
147 |
$lookup = shift; |
148 |
} |
149 |
|
150 |
=head2 _get_marc_fields |
151 |
|
152 |
Get all fields defined by calls to C<marc> |
153 |
|
154 |
$marc->add_fields( WebPAC::Normalize:_get_marc_fields() ); |
155 |
|
156 |
|
157 |
|
158 |
We are using I<magic> which detect repeatable fields only from |
159 |
sequence of field/subfield data generated by normalization. |
160 |
|
161 |
Repeatable field is created if there is second occurence of same subfield or |
162 |
if any of indicators are different. This is sane for most cases except for |
163 |
non-repeatable fields with repeatable subfields. |
164 |
|
165 |
You can change behaviour of that using C<marc_repeatable_subfield>. |
166 |
|
167 |
=cut |
168 |
|
169 |
sub _get_marc_fields { |
170 |
|
171 |
|
172 |
# first, sort all existing fields |
173 |
# XXX might not be needed, but modern perl might randomize elements in hash |
174 |
my @sorted_marc_record = sort { |
175 |
$a->[0] . $a->[3] cmp $b->[0] . $b->[3] |
176 |
} @{ $marc_record }; |
177 |
|
178 |
# output marc fields |
179 |
my @m; |
180 |
|
181 |
# count unique field-subfields (used for offset when walking to next subfield) |
182 |
my $u; |
183 |
map { $u->{ $_->[0] . $_->[3] }++ } @sorted_marc_record; |
184 |
|
185 |
if ($debug) { |
186 |
warn "## marc_repeatable_subfield ", dump( $marc_repeatable_subfield ), $/; |
187 |
warn "## marc_record ", dump( $marc_record ), $/; |
188 |
warn "## sorted_marc_record ", dump( \@sorted_marc_record ), $/; |
189 |
warn "## subfield count ", dump( $u ), $/; |
190 |
} |
191 |
|
192 |
my $len = $#sorted_marc_record; |
193 |
my $visited; |
194 |
my $i = 0; |
195 |
my $field; |
196 |
|
197 |
foreach ( 0 .. $len ) { |
198 |
|
199 |
# find next element which isn't visited |
200 |
while ($visited->{$i}) { |
201 |
$i = ($i + 1) % ($len + 1); |
202 |
} |
203 |
|
204 |
# mark it visited |
205 |
$visited->{$i}++; |
206 |
|
207 |
my $row = $sorted_marc_record[$i]; |
208 |
|
209 |
# field and subfield which is key for |
210 |
# marc_repeatable_subfield and u |
211 |
my $fsf = $row->[0] . $row->[3]; |
212 |
|
213 |
if ($debug > 1) { |
214 |
|
215 |
print "### field so far [", $#$field, "] : ", dump( $field ), " ", $field ? 'T' : 'F', $/; |
216 |
print "### this [$i]: ", dump( $row ),$/; |
217 |
print "### sf: ", $row->[3], " vs ", $field->[3], |
218 |
$marc_repeatable_subfield->{ $row->[0] . $row->[3] } ? ' (repeatable)' : '', $/, |
219 |
if ($#$field >= 0); |
220 |
|
221 |
} |
222 |
|
223 |
# if field exists |
224 |
if ( $#$field >= 0 ) { |
225 |
if ( |
226 |
$row->[0] ne $field->[0] || # field |
227 |
$row->[1] ne $field->[1] || # i1 |
228 |
$row->[2] ne $field->[2] # i2 |
229 |
) { |
230 |
push @m, $field; |
231 |
warn "## saved/1 ", dump( $field ),$/ if ($debug); |
232 |
$field = $row; |
233 |
|
234 |
} elsif ( |
235 |
( $row->[3] lt $field->[-2] ) # subfield which is not next (e.g. a after c) |
236 |
|| |
237 |
( $row->[3] eq $field->[-2] && # same subfield, but not repeatable |
238 |
! $marc_repeatable_subfield->{ $fsf } |
239 |
) |
240 |
) { |
241 |
push @m, $field; |
242 |
warn "## saved/2 ", dump( $field ),$/ if ($debug); |
243 |
$field = $row; |
244 |
|
245 |
} else { |
246 |
# append new subfields to existing field |
247 |
push @$field, ( $row->[3], $row->[4] ); |
248 |
} |
249 |
} else { |
250 |
# insert first field |
251 |
$field = $row; |
252 |
} |
253 |
|
254 |
if (! $marc_repeatable_subfield->{ $fsf }) { |
255 |
# make step to next subfield |
256 |
$i = ($i + $u->{ $fsf } ) % ($len + 1); |
257 |
} |
258 |
} |
259 |
|
260 |
if ($#$field >= 0) { |
261 |
push @m, $field; |
262 |
warn "## saved/3 ", dump( $field ),$/ if ($debug); |
263 |
} |
264 |
|
265 |
return @m; |
266 |
} |
267 |
|
268 |
=head1 Functions to create C<data_structure> |
269 |
|
270 |
Those functions generally have to first in your normalization file. |
271 |
|
272 |
=head2 tag |
273 |
|
274 |
Define new tag for I<search> and I<display>. |
275 |
|
276 |
tag('Title', rec('200','a') ); |
277 |
|
278 |
|
279 |
=cut |
280 |
|
281 |
sub tag { |
282 |
my $name = shift or die "tag needs name as first argument"; |
283 |
my @o = grep { defined($_) && $_ ne '' } @_; |
284 |
return unless (@o); |
285 |
$out->{$name}->{tag} = $name; |
286 |
$out->{$name}->{search} = \@o; |
287 |
$out->{$name}->{display} = \@o; |
288 |
} |
289 |
|
290 |
=head2 display |
291 |
|
292 |
Define tag just for I<display> |
293 |
|
294 |
@v = display('Title', rec('200','a') ); |
295 |
|
296 |
=cut |
297 |
|
298 |
sub display { |
299 |
my $name = shift or die "display needs name as first argument"; |
300 |
my @o = grep { defined($_) && $_ ne '' } @_; |
301 |
return unless (@o); |
302 |
$out->{$name}->{tag} = $name; |
303 |
$out->{$name}->{display} = \@o; |
304 |
} |
305 |
|
306 |
=head2 search |
307 |
|
308 |
Prepare values just for I<search> |
309 |
|
310 |
@v = search('Title', rec('200','a') ); |
311 |
|
312 |
=cut |
313 |
|
314 |
sub search { |
315 |
my $name = shift or die "search needs name as first argument"; |
316 |
my @o = grep { defined($_) && $_ ne '' } @_; |
317 |
return unless (@o); |
318 |
$out->{$name}->{tag} = $name; |
319 |
$out->{$name}->{search} = \@o; |
320 |
} |
321 |
|
322 |
=head2 marc |
323 |
|
324 |
Save value for MARC field |
325 |
|
326 |
marc('900','a', rec('200','a') ); |
327 |
|
328 |
=cut |
329 |
|
330 |
sub marc { |
331 |
my $f = shift or die "marc needs field"; |
332 |
die "marc field must be numer" unless ($f =~ /^\d+$/); |
333 |
|
334 |
my $sf = shift or die "marc needs subfield"; |
335 |
|
336 |
foreach (@_) { |
337 |
my $v = $_; # make var read-write for Encode |
338 |
next unless (defined($v) && $v !~ /^\s*$/); |
339 |
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
340 |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
341 |
push @{ $marc_record }, [ $f, $i1, $i2, $sf => $v ]; |
342 |
} |
343 |
} |
344 |
|
345 |
=head2 marc_repeatable_subfield |
346 |
|
347 |
Save values for MARC repetable subfield |
348 |
|
349 |
marc_repeatable_subfield('910', 'z', rec('909') ); |
350 |
|
351 |
=cut |
352 |
|
353 |
sub marc_repeatable_subfield { |
354 |
my ($f,$sf) = @_; |
355 |
die "marc_repeatable_subfield need field and subfield!\n" unless ($f && $sf); |
356 |
$marc_repeatable_subfield->{ $f . $sf }++; |
357 |
marc(@_); |
358 |
} |
359 |
|
360 |
=head2 marc_indicators |
361 |
|
362 |
Set both indicators for MARC field |
363 |
|
364 |
marc_indicators('900', ' ', 1); |
365 |
|
366 |
Any indicator value other than C<0-9> will be treated as undefined. |
367 |
|
368 |
=cut |
369 |
|
370 |
sub marc_indicators { |
371 |
my $f = shift || die "marc_indicators need field!\n"; |
372 |
my ($i1,$i2) = @_; |
373 |
die "marc_indicators($f, ...) need i1!\n" unless(defined($i1)); |
374 |
die "marc_indicators($f, $i1, ...) need i2!\n" unless(defined($i2)); |
375 |
|
376 |
$i1 = ' ' if ($i1 !~ /^\d$/); |
377 |
$i2 = ' ' if ($i2 !~ /^\d$/); |
378 |
@{ $marc_indicators->{$f} } = ($i1,$i2); |
379 |
} |
380 |
|
381 |
|
382 |
=head1 Functions to extract data from input |
383 |
|
384 |
This function should be used inside functions to create C<data_structure> described |
385 |
above. |
386 |
|
387 |
=head2 rec1 |
388 |
|
389 |
Return all values in some field |
390 |
|
391 |
@v = rec1('200') |
392 |
|
393 |
TODO: order of values is probably same as in source data, need to investigate that |
394 |
|
395 |
=cut |
396 |
|
397 |
sub rec1 { |
398 |
my $f = shift; |
399 |
return unless (defined($rec) && defined($rec->{$f})); |
400 |
if (ref($rec->{$f}) eq 'ARRAY') { |
401 |
return map { |
402 |
if (ref($_) eq 'HASH') { |
403 |
values %{$_}; |
404 |
} else { |
405 |
$_; |
406 |
} |
407 |
} @{ $rec->{$f} }; |
408 |
} elsif( defined($rec->{$f}) ) { |
409 |
return $rec->{$f}; |
410 |
} |
411 |
} |
412 |
|
413 |
=head2 rec2 |
414 |
|
415 |
Return all values in specific field and subfield |
416 |
|
417 |
@v = rec2('200','a') |
418 |
|
419 |
=cut |
420 |
|
421 |
sub rec2 { |
422 |
my $f = shift; |
423 |
return unless (defined($rec && $rec->{$f})); |
424 |
my $sf = shift; |
425 |
return map { $_->{$sf} } grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} }; |
426 |
} |
427 |
|
428 |
=head2 rec |
429 |
|
430 |
syntaxtic sugar for |
431 |
|
432 |
@v = rec('200') |
433 |
@v = rec('200','a') |
434 |
|
435 |
=cut |
436 |
|
437 |
sub rec { |
438 |
if ($#_ == 0) { |
439 |
return rec1(@_); |
440 |
} elsif ($#_ == 1) { |
441 |
return rec2(@_); |
442 |
} |
443 |
} |
444 |
|
445 |
=head2 regex |
446 |
|
447 |
Apply regex to some or all values |
448 |
|
449 |
@v = regex( 's/foo/bar/g', @v ); |
450 |
|
451 |
=cut |
452 |
|
453 |
sub regex { |
454 |
my $r = shift; |
455 |
my @out; |
456 |
#warn "r: $r\n", dump(\@_); |
457 |
foreach my $t (@_) { |
458 |
next unless ($t); |
459 |
eval "\$t =~ $r"; |
460 |
push @out, $t if ($t && $t ne ''); |
461 |
} |
462 |
return @out; |
463 |
} |
464 |
|
465 |
=head2 prefix |
466 |
|
467 |
Prefix all values with a string |
468 |
|
469 |
@v = prefix( 'my_', @v ); |
470 |
|
471 |
=cut |
472 |
|
473 |
sub prefix { |
474 |
my $p = shift or die "prefix needs string as first argument"; |
475 |
return map { $p . $_ } grep { defined($_) } @_; |
476 |
} |
477 |
|
478 |
=head2 suffix |
479 |
|
480 |
suffix all values with a string |
481 |
|
482 |
@v = suffix( '_my', @v ); |
483 |
|
484 |
=cut |
485 |
|
486 |
sub suffix { |
487 |
my $s = shift or die "suffix needs string as first argument"; |
488 |
return map { $_ . $s } grep { defined($_) } @_; |
489 |
} |
490 |
|
491 |
=head2 surround |
492 |
|
493 |
surround all values with a two strings |
494 |
|
495 |
@v = surround( 'prefix_', '_suffix', @v ); |
496 |
|
497 |
=cut |
498 |
|
499 |
sub surround { |
500 |
my $p = shift or die "surround need prefix as first argument"; |
501 |
my $s = shift or die "surround needs suffix as second argument"; |
502 |
return map { $p . $_ . $s } grep { defined($_) } @_; |
503 |
} |
504 |
|
505 |
=head2 first |
506 |
|
507 |
Return first element |
508 |
|
509 |
$v = first( @v ); |
510 |
|
511 |
=cut |
512 |
|
513 |
sub first { |
514 |
my $r = shift; |
515 |
return $r; |
516 |
} |
517 |
|
518 |
=head2 lookup |
519 |
|
520 |
Consult lookup hashes for some value |
521 |
|
522 |
@v = lookup( $v ); |
523 |
@v = lookup( @v ); |
524 |
|
525 |
=cut |
526 |
|
527 |
sub lookup { |
528 |
my $k = shift or return; |
529 |
return unless (defined($lookup->{$k})); |
530 |
if (ref($lookup->{$k}) eq 'ARRAY') { |
531 |
return @{ $lookup->{$k} }; |
532 |
} else { |
533 |
return $lookup->{$k}; |
534 |
} |
535 |
} |
536 |
|
537 |
=head2 join_with |
538 |
|
539 |
Joins walues with some delimiter |
540 |
|
541 |
$v = join_with(", ", @v); |
542 |
|
543 |
=cut |
544 |
|
545 |
sub join_with { |
546 |
my $d = shift; |
547 |
return join($d, grep { defined($_) && $_ ne '' } @_); |
548 |
} |
549 |
|
550 |
# END |
551 |
1; |