1 |
package WebPAC::Normalize; |
package WebPAC::Normalize; |
2 |
|
use Exporter 'import'; |
3 |
|
@EXPORT = qw/ |
4 |
|
_set_rec _set_lookup |
5 |
|
_set_load_row |
6 |
|
_get_ds _clean_ds |
7 |
|
_debug |
8 |
|
_pack_subfields_hash |
9 |
|
|
10 |
|
tag search display |
11 |
|
marc marc_indicators marc_repeatable_subfield |
12 |
|
marc_compose marc_leader marc_fixed |
13 |
|
marc_duplicate marc_remove marc_count |
14 |
|
marc_original_order |
15 |
|
|
16 |
|
rec1 rec2 rec |
17 |
|
regex prefix suffix surround |
18 |
|
first lookup join_with |
19 |
|
save_into_lookup |
20 |
|
|
21 |
|
split_rec_on |
22 |
|
|
23 |
|
get set |
24 |
|
count |
25 |
|
/; |
26 |
|
|
27 |
use warnings; |
use warnings; |
28 |
use strict; |
use strict; |
29 |
use Data::Dumper; |
|
30 |
|
#use base qw/WebPAC::Common/; |
31 |
|
use Data::Dump qw/dump/; |
32 |
|
use Storable qw/dclone/; |
33 |
|
use Carp qw/confess/; |
34 |
|
|
35 |
|
# debugging warn(s) |
36 |
|
my $debug = 0; |
37 |
|
|
38 |
|
|
39 |
=head1 NAME |
=head1 NAME |
40 |
|
|
41 |
WebPAC::Normalize - data mungling for normalisation |
WebPAC::Normalize - describe normalisaton rules using sets |
42 |
|
|
43 |
=head1 VERSION |
=head1 VERSION |
44 |
|
|
45 |
Version 0.01 |
Version 0.29 |
46 |
|
|
47 |
=cut |
=cut |
48 |
|
|
49 |
our $VERSION = '0.01'; |
our $VERSION = '0.29'; |
50 |
|
|
51 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
52 |
|
|
53 |
This package contains code that mungle data to produce normalized format. |
This module uses C<conf/normalize/*.pl> files to perform normalisation |
54 |
|
from input records using perl functions which are specialized for set |
55 |
|
processing. |
56 |
|
|
57 |
|
Sets are implemented as arrays, and normalisation file is valid perl, which |
58 |
|
means that you check it's validity before running WebPAC using |
59 |
|
C<perl -c normalize.pl>. |
60 |
|
|
61 |
|
Normalisation can generate multiple output normalized data. For now, supported output |
62 |
|
types (on the left side of definition) are: C<tag>, C<display>, C<search> and |
63 |
|
C<marc>. |
64 |
|
|
65 |
|
=head1 FUNCTIONS |
66 |
|
|
67 |
|
Functions which start with C<_> are private and used by WebPAC internally. |
68 |
|
All other functions are available for use within normalisation rules. |
69 |
|
|
70 |
|
=head2 data_structure |
71 |
|
|
72 |
|
Return data structure |
73 |
|
|
74 |
|
my $ds = WebPAC::Normalize::data_structure( |
75 |
|
lookup => $lookup_hash, |
76 |
|
row => $row, |
77 |
|
rules => $normalize_pl_config, |
78 |
|
marc_encoding => 'utf-8', |
79 |
|
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<row>, C<rules> and C<log> are mandatory while all |
87 |
|
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 |
It contains several assumptions: |
This function will B<die> if normalizastion can't be evaled. |
93 |
|
|
94 |
=over |
Since this function isn't exported you have to call it with |
95 |
|
C<WebPAC::Normalize::data_structure>. |
96 |
|
|
97 |
=item * |
=cut |
98 |
|
|
99 |
format of fields is defined using C<v123^a> notation for repeatable fields |
my $load_row_coderef; |
|
or C<s123^a> for single (or first) value, where C<123> is field number and |
|
|
C<a> is subfield. |
|
100 |
|
|
101 |
=item * |
sub data_structure { |
102 |
|
my $arg = {@_}; |
103 |
|
|
104 |
source data records (C<$rec>) have unique identifiers in field C<000> |
die "need row argument" unless ($arg->{row}); |
105 |
|
die "need normalisation argument" unless ($arg->{rules}); |
106 |
|
|
107 |
=item * |
no strict 'subs'; |
108 |
|
_set_lookup( $arg->{lookup} ) if defined($arg->{lookup}); |
109 |
|
_set_rec( $arg->{row} ); |
110 |
|
_set_config( $arg->{config} ) if defined($arg->{config}); |
111 |
|
_clean_ds( %{ $arg } ); |
112 |
|
$load_row_coderef = $arg->{load_row_coderef}; |
113 |
|
|
114 |
optional C<eval{length('v123^a') == 3}> tag at B<beginning of format> will be |
eval "$arg->{rules}"; |
115 |
perl code that is evaluated before producing output (value of field will be |
die "error evaling $arg->{rules}: $@\n" if ($@); |
|
interpolated before that) |
|
116 |
|
|
117 |
=item * |
return _get_ds(); |
118 |
|
} |
119 |
|
|
120 |
optional C<filter{filter_name}> at B<begining of format> will apply perl |
=head2 _set_rec |
|
code defined as code ref on format after field substitution to producing |
|
|
output |
|
121 |
|
|
122 |
=item * |
Set current record hash |
123 |
|
|
124 |
optional C<lookup{...}> will be then performed. See C<WebPAC::Lookups>. |
_set_rec( $rec ); |
125 |
|
|
126 |
|
=cut |
127 |
|
|
128 |
|
my $rec; |
129 |
|
|
130 |
|
sub _set_rec { |
131 |
|
$rec = shift or die "no record hash"; |
132 |
|
} |
133 |
|
|
134 |
=item * |
=head2 _set_config |
135 |
|
|
136 |
at end, optional C<format>s rules are resolved. Format rules are similar to |
Set current config hash |
137 |
C<sprintf> and can also contain C<lookup{...}> which is performed after |
|
138 |
values are inserted in format. |
_set_config( $config ); |
139 |
|
|
140 |
|
Magic keys are: |
141 |
|
|
142 |
|
=over 4 |
143 |
|
|
144 |
|
=item _ |
145 |
|
|
146 |
|
Code of current database |
147 |
|
|
148 |
|
=item _mfn |
149 |
|
|
150 |
|
Current MFN |
151 |
|
|
152 |
=back |
=back |
153 |
|
|
154 |
This also describes order in which transformations are applied (eval, |
=cut |
|
filter, lookup, format) which is important to undestand when deciding how to |
|
|
solve your data mungling and normalisation process. |
|
155 |
|
|
156 |
|
my $config; |
157 |
|
|
158 |
|
sub _set_config { |
159 |
|
$config = shift; |
160 |
|
} |
161 |
|
|
162 |
|
=head2 _get_ds |
163 |
|
|
164 |
=head1 FUNCTIONS |
Return hash formatted as data structure |
165 |
|
|
166 |
=head2 new |
my $ds = _get_ds(); |
167 |
|
|
168 |
Create new normalisation object |
=cut |
169 |
|
|
170 |
my $n = new WebPAC::Normalize::Something( |
my ($out, $marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $marc_leader); |
171 |
filter => { |
my ($marc_record_offset, $marc_fetch_offset) = (0, 0); |
172 |
'filter_name_1' => sub { |
|
173 |
# filter code |
sub _get_ds { |
174 |
return length($_); |
return $out; |
175 |
}, ... |
} |
|
}, |
|
|
db => $webpac_db_obj, |
|
|
lookup_regex => $lookup->regex, |
|
|
); |
|
176 |
|
|
177 |
Parametar C<filter> defines user supplied snippets of perl code which can |
=head2 _clean_ds |
|
be use with C<filter{...}> notation. |
|
178 |
|
|
179 |
Recommended parametar C<lookup_regex> is used to enable parsing of lookups |
Clean data structure hash for next record |
180 |
in structures. |
|
181 |
|
_clean_ds(); |
182 |
|
|
183 |
=cut |
=cut |
184 |
|
|
185 |
sub new { |
sub _clean_ds { |
186 |
my $class = shift; |
my $a = {@_}; |
187 |
my $self = {@_}; |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $marc_leader) = (); |
188 |
bless($self, $class); |
($marc_record_offset, $marc_fetch_offset) = (0,0); |
189 |
|
$marc_encoding = $a->{marc_encoding}; |
190 |
|
} |
191 |
|
|
192 |
|
=head2 _set_lookup |
193 |
|
|
194 |
|
Set current lookup hash |
195 |
|
|
196 |
|
_set_lookup( $lookup ); |
197 |
|
|
198 |
$self ? return $self : return undef; |
=cut |
199 |
|
|
200 |
|
my $lookup; |
201 |
|
|
202 |
|
sub _set_lookup { |
203 |
|
$lookup = shift; |
204 |
} |
} |
205 |
|
|
206 |
|
=head2 _get_lookup |
207 |
|
|
208 |
=head2 data_structure |
Get current lookup hash |
209 |
|
|
210 |
Create in-memory data structure which represents normalized layout from |
my $lookup = _get_lookup(); |
|
C<conf/normalize/*.xml>. |
|
211 |
|
|
212 |
This structures are used to produce output. |
=cut |
213 |
|
|
214 |
my @ds = $webpac->data_structure($rec); |
sub _get_lookup { |
215 |
|
return $lookup; |
216 |
|
} |
217 |
|
|
218 |
B<Note: historical oddity follows> |
=head2 _set_load_row |
219 |
|
|
220 |
This method will also set C<< $webpac->{'currnet_filename'} >> if there is |
Setup code reference which will return L<data_structure> from |
221 |
C<< <filename> >> tag and C<< $webpac->{'headline'} >> if there is |
L<WebPAC::Store> |
222 |
C<< <headline> >> tag. |
|
223 |
|
_set_load_row(sub { |
224 |
|
my ($database,$input,$mfn) = @_; |
225 |
|
$store->load_row( database => $database, input => $input, id => $mfn ); |
226 |
|
}); |
227 |
|
|
228 |
=cut |
=cut |
229 |
|
|
230 |
sub data_structure { |
sub _set_load_row { |
231 |
my $self = shift; |
my $coderef = shift; |
232 |
|
confess "argument isn't CODE" unless ref($coderef) eq 'CODE'; |
233 |
|
|
234 |
my $log = $self->_get_logger(); |
$load_row_coderef = $coderef; |
235 |
|
} |
236 |
|
|
237 |
my $rec = shift; |
=head2 _get_marc_fields |
|
$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o); |
|
238 |
|
|
239 |
my $cache_file; |
Get all fields defined by calls to C<marc> |
240 |
|
|
241 |
if ($self->{'db'}) { |
$marc->add_fields( WebPAC::Normalize:_get_marc_fields() ); |
|
my @ds = $self->{'db'}->get_ds($rec); |
|
|
return @ds if (@ds); |
|
|
} |
|
242 |
|
|
243 |
undef $self->{'currnet_filename'}; |
We are using I<magic> which detect repeatable fields only from |
244 |
undef $self->{'headline'}; |
sequence of field/subfield data generated by normalization. |
245 |
|
|
246 |
my @sorted_tags; |
Repeatable field is created when there is second occurence of same subfield or |
247 |
if ($self->{tags_by_order}) { |
if any of indicators are different. |
|
@sorted_tags = @{$self->{tags_by_order}}; |
|
|
} else { |
|
|
@sorted_tags = sort { $self->_sort_by_order } keys %{$self->{'import_xml'}->{'indexer'}}; |
|
|
$self->{tags_by_order} = \@sorted_tags; |
|
|
} |
|
248 |
|
|
249 |
my @ds; |
This is sane for most cases. Something like: |
250 |
|
|
251 |
$log->debug("tags: ",sub { join(", ",@sorted_tags) }); |
900a-1 900b-1 900c-1 |
252 |
|
900a-2 900b-2 |
253 |
|
900a-3 |
254 |
|
|
255 |
foreach my $field (@sorted_tags) { |
will be created from any combination of: |
256 |
|
|
257 |
my $row; |
900a-1 900a-2 900a-3 900b-1 900b-2 900c-1 |
258 |
|
|
259 |
#print "field $field [",$self->{'tag'},"] = ",Dumper($self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}}); |
and following rules: |
260 |
|
|
261 |
foreach my $tag (@{$self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}}}) { |
marc('900','a', rec('200','a') ); |
262 |
my $format = $tag->{'value'} || $tag->{'content'}; |
marc('900','b', rec('200','b') ); |
263 |
|
marc('900','c', rec('200','c') ); |
264 |
|
|
265 |
$log->debug("format: $format"); |
which might not be what you have in mind. If you need repeatable subfield, |
266 |
|
define it using C<marc_repeatable_subfield> like this: |
267 |
|
|
268 |
my @v; |
marc_repeatable_subfield('900','a'); |
269 |
if ($self->{'lookup_regex'} && $format =~ $self->{'lookup_regex'}) { |
marc('900','a', rec('200','a') ); |
270 |
@v = $self->fill_in_to_arr($rec,$format); |
marc('900','b', rec('200','b') ); |
271 |
} else { |
marc('900','c', rec('200','c') ); |
|
@v = $self->parse_to_arr($rec,$format); |
|
|
} |
|
|
next if (! @v); |
|
272 |
|
|
273 |
if ($tag->{'sort'}) { |
will create: |
|
@v = $self->sort_arr(@v); |
|
|
} |
|
274 |
|
|
275 |
# use format? |
900a-1 900a-2 900a-3 900b-1 900c-1 |
276 |
if ($tag->{'format_name'}) { |
900b-2 |
|
@v = map { $self->apply_format($tag->{'format_name'},$tag->{'format_delimiter'},$_) } @v; |
|
|
} |
|
277 |
|
|
278 |
if ($field eq 'filename') { |
There is also support for returning next or specific using: |
|
$self->{'current_filename'} = join('',@v); |
|
|
$log->debug("filename: ",$self->{'current_filename'}); |
|
|
} elsif ($field eq 'headline') { |
|
|
$self->{'headline'} .= join('',@v); |
|
|
$log->debug("headline: ",$self->{'headline'}); |
|
|
next; # don't return headline in data_structure! |
|
|
} |
|
279 |
|
|
280 |
# delimiter will join repeatable fields |
while (my $mf = WebPAC::Normalize:_get_marc_fields( fetch_next => 1 ) ) { |
281 |
if ($tag->{'delimiter'}) { |
# do something with $mf |
282 |
@v = ( join($tag->{'delimiter'}, @v) ); |
} |
|
} |
|
283 |
|
|
284 |
# default types |
will always return fields from next MARC record or |
|
my @types = qw(display swish); |
|
|
# override by type attribute |
|
|
@types = ( $tag->{'type'} ) if ($tag->{'type'}); |
|
|
|
|
|
foreach my $type (@types) { |
|
|
# append to previous line? |
|
|
$log->debug("type: $type ",sub { join(" ",@v) }, $row->{'append'} || 'no append'); |
|
|
if ($tag->{'append'}) { |
|
|
|
|
|
# I will delimit appended part with |
|
|
# delimiter (or ,) |
|
|
my $d = $tag->{'delimiter'}; |
|
|
# default delimiter |
|
|
$d ||= " "; |
|
|
|
|
|
my $last = pop @{$row->{$type}}; |
|
|
$d = "" if (! $last); |
|
|
$last .= $d . join($d, @v); |
|
|
push @{$row->{$type}}, $last; |
|
285 |
|
|
286 |
} else { |
my $mf = WebPAC::Normalize::_get_marc_fields( offset => 42 ); |
|
push @{$row->{$type}}, @v; |
|
|
} |
|
|
} |
|
287 |
|
|
288 |
|
will return 42th copy record (if it exists). |
289 |
|
|
290 |
|
=cut |
291 |
|
|
292 |
|
my $fetch_pos; |
293 |
|
|
294 |
|
sub _get_marc_fields { |
295 |
|
|
296 |
|
my $arg = {@_}; |
297 |
|
warn "### _get_marc_fields arg: ", dump($arg), $/ if ($debug > 2); |
298 |
|
$fetch_pos = $marc_fetch_offset; |
299 |
|
if ($arg->{offset}) { |
300 |
|
$fetch_pos = $arg->{offset}; |
301 |
|
} elsif($arg->{fetch_next}) { |
302 |
|
$marc_fetch_offset++; |
303 |
|
} |
304 |
|
|
305 |
|
return if (! $marc_record || ref($marc_record) ne 'ARRAY'); |
306 |
|
|
307 |
|
warn "### full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 2); |
308 |
|
|
309 |
|
my $marc_rec = $marc_record->[ $fetch_pos ]; |
310 |
|
|
311 |
|
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); |
314 |
|
|
315 |
|
# first, sort all existing fields |
316 |
|
# XXX might not be needed, but modern perl might randomize elements in hash |
317 |
|
my @sorted_marc_record = sort { |
318 |
|
$a->[0] . ( $a->[3] || '' ) cmp $b->[0] . ( $b->[3] || '') |
319 |
|
} @{ $marc_rec }; |
320 |
|
|
321 |
|
@sorted_marc_record = @{ $marc_rec }; ### FIXME disable sorting |
322 |
|
|
323 |
|
# output marc fields |
324 |
|
my @m; |
325 |
|
|
326 |
|
# count unique field-subfields (used for offset when walking to next subfield) |
327 |
|
my $u; |
328 |
|
map { $u->{ $_->[0] . ( $_->[3] || '') }++ } @sorted_marc_record; |
329 |
|
|
330 |
|
if ($debug) { |
331 |
|
warn "## marc_repeatable_subfield = ", dump( $marc_repeatable_subfield ), $/ if ( $marc_repeatable_subfield ); |
332 |
|
warn "## marc_record[$fetch_pos] = ", dump( $marc_rec ), $/; |
333 |
|
warn "## sorted_marc_record = ", dump( \@sorted_marc_record ), $/; |
334 |
|
warn "## subfield count = ", dump( $u ), $/; |
335 |
|
} |
336 |
|
|
337 |
|
my $len = $#sorted_marc_record; |
338 |
|
my $visited; |
339 |
|
my $i = 0; |
340 |
|
my $field; |
341 |
|
|
342 |
|
foreach ( 0 .. $len ) { |
343 |
|
|
344 |
|
# find next element which isn't visited |
345 |
|
while ($visited->{$i}) { |
346 |
|
$i = ($i + 1) % ($len + 1); |
347 |
} |
} |
348 |
|
|
349 |
if ($row) { |
# mark it visited |
350 |
$row->{'tag'} = $field; |
$visited->{$i}++; |
351 |
|
|
352 |
# TODO: name_sigular, name_plural |
my $row = dclone( $sorted_marc_record[$i] ); |
353 |
my $name = $self->{'import_xml'}->{'indexer'}->{$field}->{'name'}; |
|
354 |
$row->{'name'} = $name ? $self->_x($name) : $field; |
# field and subfield which is key for |
355 |
|
# marc_repeatable_subfield and u |
356 |
# post-sort all values in field |
my $fsf = $row->[0] . ( $row->[3] || '' ); |
357 |
if ($self->{'import_xml'}->{'indexer'}->{$field}->{'sort'}) { |
|
358 |
$log->warn("sort at field tag not implemented"); |
if ($debug > 1) { |
359 |
} |
|
360 |
|
print "### field so far [", $#$field, "] : ", dump( $field ), " ", $field ? 'T' : 'F', $/; |
361 |
|
print "### this [$i]: ", dump( $row ),$/; |
362 |
|
print "### sf: ", $row->[3], " vs ", $field->[3], |
363 |
|
$marc_repeatable_subfield->{ $row->[0] . $row->[3] } ? ' (repeatable)' : '', $/, |
364 |
|
if ($#$field >= 0); |
365 |
|
|
366 |
push @ds, $row; |
} |
367 |
|
|
368 |
|
# if field exists |
369 |
|
if ( $#$field >= 0 ) { |
370 |
|
if ( |
371 |
|
$row->[0] ne $field->[0] || # field |
372 |
|
$row->[1] ne $field->[1] || # i1 |
373 |
|
$row->[2] ne $field->[2] # i2 |
374 |
|
) { |
375 |
|
push @m, $field; |
376 |
|
warn "## saved/1 ", dump( $field ),$/ if ($debug); |
377 |
|
$field = $row; |
378 |
|
|
379 |
|
} elsif ( |
380 |
|
( $row->[3] lt $field->[-2] ) # subfield which is not next (e.g. a after c) |
381 |
|
|| |
382 |
|
( $row->[3] eq $field->[-2] && # same subfield, but not repeatable |
383 |
|
! $marc_repeatable_subfield->{ $fsf } |
384 |
|
) |
385 |
|
) { |
386 |
|
push @m, $field; |
387 |
|
warn "## saved/2 ", dump( $field ),$/ if ($debug); |
388 |
|
$field = $row; |
389 |
|
|
390 |
$log->debug("row $field: ",sub { Dumper($row) }); |
} else { |
391 |
|
# append new subfields to existing field |
392 |
|
push @$field, ( $row->[3], $row->[4] ); |
393 |
|
} |
394 |
|
} else { |
395 |
|
# insert first field |
396 |
|
$field = $row; |
397 |
} |
} |
398 |
|
|
399 |
|
if (! $marc_repeatable_subfield->{ $fsf }) { |
400 |
|
# make step to next subfield |
401 |
|
$i = ($i + $u->{ $fsf } ) % ($len + 1); |
402 |
|
} |
403 |
} |
} |
404 |
|
|
405 |
$self->{'db'}->put_gs( |
if ($#$field >= 0) { |
406 |
ds => \@ds, |
push @m, $field; |
407 |
current_filename => $self->{'current_filename'}, |
warn "## saved/3 ", dump( $field ),$/ if ($debug); |
408 |
headline => $self->{'headline'}, |
} |
409 |
) if ($self->{'db'}); |
|
410 |
|
return \@m; |
411 |
|
} |
412 |
|
|
413 |
|
=head2 _get_marc_leader |
414 |
|
|
415 |
|
Return leader from currently fetched record by L</_get_marc_fields> |
416 |
|
|
417 |
return @ds; |
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 parse |
=head2 _debug |
427 |
|
|
428 |
Perform smart parsing of string, skipping delimiters for fields which aren't |
Change level of debug warnings |
|
defined. It can also eval code in format starting with C<eval{...}> and |
|
|
return output or nothing depending on eval code. |
|
429 |
|
|
430 |
my $text = $webpac->parse($rec,'eval{"v901^a" eq "Deskriptor"}descriptor: v250^a', $i); |
_debug( 2 ); |
431 |
|
|
432 |
=cut |
=cut |
433 |
|
|
434 |
sub parse { |
sub _debug { |
435 |
my $self = shift; |
my $l = shift; |
436 |
|
return $debug unless defined($l); |
437 |
|
warn "debug level $l",$/ if ($l > 0); |
438 |
|
$debug = $l; |
439 |
|
} |
440 |
|
|
441 |
my ($rec, $format_utf8, $i) = @_; |
=head1 Functions to create C<data_structure> |
442 |
|
|
443 |
return if (! $format_utf8); |
Those functions generally have to first in your normalization file. |
444 |
|
|
445 |
my $log = $self->_get_logger(); |
=head2 tag |
446 |
|
|
447 |
$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o); |
Define new tag for I<search> and I<display>. |
448 |
|
|
449 |
$i = 0 if (! $i); |
tag('Title', rec('200','a') ); |
450 |
|
|
|
my $format = $self->_x($format_utf8) || $log->logconfess("can't convert '$format_utf8' from UTF-8 to ",$self->{'code_page'}); |
|
451 |
|
|
452 |
my @out; |
=cut |
453 |
|
|
454 |
$log->debug("format: $format"); |
sub tag { |
455 |
|
my $name = shift or die "tag needs name as first argument"; |
456 |
|
my @o = grep { defined($_) && $_ ne '' } @_; |
457 |
|
return unless (@o); |
458 |
|
$out->{$name}->{tag} = $name; |
459 |
|
$out->{$name}->{search} = \@o; |
460 |
|
$out->{$name}->{display} = \@o; |
461 |
|
} |
462 |
|
|
463 |
my $eval_code; |
=head2 display |
|
# remove eval{...} from beginning |
|
|
$eval_code = $1 if ($format =~ s/^eval{([^}]+)}//s); |
|
464 |
|
|
465 |
my $filter_name; |
Define tag just for I<display> |
|
# remove filter{...} from beginning |
|
|
$filter_name = $1 if ($format =~ s/^filter{([^}]+)}//s); |
|
466 |
|
|
467 |
my $prefix; |
@v = display('Title', rec('200','a') ); |
|
my $all_found=0; |
|
468 |
|
|
469 |
while ($format =~ s/^(.*?)(v|s)(\d+)(?:\^(\w))?//s) { |
=cut |
470 |
|
|
471 |
|
sub display { |
472 |
|
my $name = shift or die "display needs name as first argument"; |
473 |
|
my @o = grep { defined($_) && $_ ne '' } @_; |
474 |
|
return unless (@o); |
475 |
|
$out->{$name}->{tag} = $name; |
476 |
|
$out->{$name}->{display} = \@o; |
477 |
|
} |
478 |
|
|
479 |
my $del = $1 || ''; |
=head2 search |
|
$prefix ||= $del if ($all_found == 0); |
|
480 |
|
|
481 |
# repeatable index |
Prepare values just for I<search> |
|
my $r = $i; |
|
|
$r = 0 if (lc("$2") eq 's'); |
|
482 |
|
|
483 |
my $found = 0; |
@v = search('Title', rec('200','a') ); |
484 |
my $tmp = $self->get_data(\$rec,$3,$4,$r,\$found); |
|
485 |
|
=cut |
486 |
|
|
487 |
if ($found) { |
sub search { |
488 |
push @out, $del; |
my $name = shift or die "search needs name as first argument"; |
489 |
push @out, $tmp; |
my @o = grep { defined($_) && $_ ne '' } @_; |
490 |
$all_found += $found; |
return unless (@o); |
491 |
|
$out->{$name}->{tag} = $name; |
492 |
|
$out->{$name}->{search} = \@o; |
493 |
|
} |
494 |
|
|
495 |
|
=head2 marc_leader |
496 |
|
|
497 |
|
Setup fields within MARC leader or get leader |
498 |
|
|
499 |
|
marc_leader('05','c'); |
500 |
|
my $leader = marc_leader(); |
501 |
|
|
502 |
|
=cut |
503 |
|
|
504 |
|
sub marc_leader { |
505 |
|
my ($offset,$value) = @_; |
506 |
|
|
507 |
|
if ($offset) { |
508 |
|
$marc_leader->[ $marc_record_offset ]->{ $offset } = $value; |
509 |
|
} else { |
510 |
|
|
511 |
|
if (defined($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 |
return if (! $all_found); |
=head2 marc_fixed |
521 |
|
|
522 |
my $out = join('',@out); |
Create control/indentifier fields with values in fixed positions |
523 |
|
|
524 |
if ($out) { |
marc_fixed('008', 00, '070402'); |
525 |
# add rest of format (suffix) |
marc_fixed('008', 39, '|'); |
|
$out .= $format; |
|
526 |
|
|
527 |
# add prefix if not there |
Positions not specified will be filled with spaces (C<0x20>). |
|
$out = $prefix . $out if ($out !~ m/^\Q$prefix\E/); |
|
528 |
|
|
529 |
$log->debug("result: $out"); |
There will be no effort to extend last specified value to full length of |
530 |
|
field in standard. |
531 |
|
|
532 |
|
=cut |
533 |
|
|
534 |
|
sub marc_fixed { |
535 |
|
my ($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 ($debug > 1); |
558 |
} |
} |
559 |
|
} |
560 |
|
|
561 |
|
=head2 marc |
562 |
|
|
563 |
|
Save value for MARC field |
564 |
|
|
565 |
if ($eval_code) { |
marc('900','a', rec('200','a') ); |
566 |
my $eval = $self->fill_in($rec,$eval_code,$i) || return; |
marc('001', rec('000') ); |
567 |
$log->debug("about to eval{$eval} format: $out"); |
|
568 |
return if (! $self->_eval($eval)); |
=cut |
569 |
|
|
570 |
|
sub marc { |
571 |
|
my $f = shift or die "marc needs field"; |
572 |
|
die "marc field must be numer" unless ($f =~ /^\d+$/); |
573 |
|
|
574 |
|
my $sf; |
575 |
|
if ($f >= 10) { |
576 |
|
$sf = shift or die "marc needs subfield"; |
577 |
} |
} |
578 |
|
|
579 |
if ($filter_name && $self->{'filter'}->{$filter_name}) { |
foreach (@_) { |
580 |
$log->debug("about to filter{$filter_name} format: $out"); |
my $v = $_; # make var read-write for Encode |
581 |
$out = $self->{'filter'}->{$filter_name}->($out); |
next unless (defined($v) && $v !~ /^\s*$/); |
582 |
return unless(defined($out)); |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
583 |
$log->debug("filter result: $out"); |
if (defined $sf) { |
584 |
|
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $i1, $i2, $sf => $v ]; |
585 |
|
} else { |
586 |
|
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $v ]; |
587 |
|
} |
588 |
} |
} |
589 |
|
} |
590 |
|
|
591 |
return $out; |
=head2 marc_repeatable_subfield |
592 |
|
|
593 |
|
Save values for MARC repetable subfield |
594 |
|
|
595 |
|
marc_repeatable_subfield('910', 'z', rec('909') ); |
596 |
|
|
597 |
|
=cut |
598 |
|
|
599 |
|
sub marc_repeatable_subfield { |
600 |
|
my ($f,$sf) = @_; |
601 |
|
die "marc_repeatable_subfield need field and subfield!\n" unless ($f && $sf); |
602 |
|
$marc_repeatable_subfield->{ $f . $sf }++; |
603 |
|
marc(@_); |
604 |
} |
} |
605 |
|
|
606 |
=head2 parse_to_arr |
=head2 marc_indicators |
607 |
|
|
608 |
Similar to C<parse>, but returns array of all repeatable fields |
Set both indicators for MARC field |
609 |
|
|
610 |
my @arr = $webpac->parse_to_arr($rec,'v250^a'); |
marc_indicators('900', ' ', 1); |
611 |
|
|
612 |
|
Any indicator value other than C<0-9> will be treated as undefined. |
613 |
|
|
614 |
=cut |
=cut |
615 |
|
|
616 |
sub parse_to_arr { |
sub marc_indicators { |
617 |
my $self = shift; |
my $f = shift || die "marc_indicators need field!\n"; |
618 |
|
my ($i1,$i2) = @_; |
619 |
|
die "marc_indicators($f, ...) need i1!\n" unless(defined($i1)); |
620 |
|
die "marc_indicators($f, $i1, ...) need i2!\n" unless(defined($i2)); |
621 |
|
|
622 |
|
$i1 = ' ' if ($i1 !~ /^\d$/); |
623 |
|
$i2 = ' ' if ($i2 !~ /^\d$/); |
624 |
|
@{ $marc_indicators->{$f} } = ($i1,$i2); |
625 |
|
} |
626 |
|
|
627 |
|
=head2 marc_compose |
628 |
|
|
629 |
my ($rec, $format_utf8) = @_; |
Save values for each MARC subfield explicitly |
630 |
|
|
631 |
my $log = $self->_get_logger(); |
marc_compose('900', |
632 |
|
'a', rec('200','a') |
633 |
|
'b', rec('201','a') |
634 |
|
'a', rec('200','b') |
635 |
|
'c', rec('200','c') |
636 |
|
); |
637 |
|
|
638 |
$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o); |
If you specify C<+> for subfield, value will be appended |
639 |
return if (! $format_utf8); |
to previous defined subfield. |
640 |
|
|
641 |
my $i = 0; |
=cut |
642 |
my @arr; |
|
643 |
|
sub marc_compose { |
644 |
|
my $f = shift or die "marc_compose needs field"; |
645 |
|
die "marc_compose field must be numer" unless ($f =~ /^\d+$/); |
646 |
|
|
647 |
while (my $v = $self->parse($rec,$format_utf8,$i++)) { |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
648 |
push @arr, $v; |
my $m = [ $f, $i1, $i2 ]; |
649 |
|
|
650 |
|
warn "### marc_compose input subfields = ", dump(@_),$/ 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 (@_) { |
657 |
|
my $sf = shift; |
658 |
|
my $v = shift; |
659 |
|
|
660 |
|
next unless (defined($v) && $v !~ /^\s*$/); |
661 |
|
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
662 |
|
if ($sf ne '+') { |
663 |
|
push @$m, ( $sf, $v ); |
664 |
|
} else { |
665 |
|
$m->[ $#$m ] .= $v; |
666 |
|
} |
667 |
} |
} |
668 |
|
|
669 |
$log->debug("format '$format_utf8' returned ",--$i," elements: ", sub { join(" | ",@arr) }) if (@arr); |
warn "## marc_compose current marc = ", dump( $m ),$/ if ($debug > 1); |
670 |
|
|
671 |
return @arr; |
push @{ $marc_record->[ $marc_record_offset ] }, $m if ($#{$m} > 2); |
672 |
} |
} |
673 |
|
|
674 |
|
=head2 marc_duplicate |
675 |
|
|
676 |
=head2 fill_in |
Generate copy of current MARC record and continue working on copy |
677 |
|
|
678 |
Workhourse of all: takes record from in-memory structure of database and |
marc_duplicate(); |
|
strings with placeholders and returns string or array of with substituted |
|
|
values from record. |
|
679 |
|
|
680 |
my $text = $webpac->fill_in($rec,'v250^a'); |
Copies can be accessed using C<< _get_marc_fields( fetch_next => 1 ) >> or |
681 |
|
C<< _get_marc_fields( offset => 42 ) >>. |
682 |
|
|
683 |
Optional argument is ordinal number for repeatable fields. By default, |
=cut |
|
it's assume to be first repeatable field (fields are perl array, so first |
|
|
element is 0). |
|
|
Following example will read second value from repeatable field. |
|
684 |
|
|
685 |
my $text = $webpac->fill_in($rec,'Title: v250^a',1); |
sub marc_duplicate { |
686 |
|
my $m = $marc_record->[ -1 ]; |
687 |
|
die "can't duplicate record which isn't defined" unless ($m); |
688 |
|
push @{ $marc_record }, dclone( $m ); |
689 |
|
push @{ $marc_leader }, dclone( marc_leader() ); |
690 |
|
warn "## marc_duplicate = ", dump(@$marc_leader, @$marc_record), $/ if ($debug > 1); |
691 |
|
$marc_record_offset = $#{ $marc_record }; |
692 |
|
warn "## marc_record_offset = $marc_record_offset", $/ if ($debug > 1); |
693 |
|
|
694 |
This function B<does not> perform parsing of format to inteligenty skip |
} |
|
delimiters before fields which aren't used. |
|
695 |
|
|
696 |
This method will automatically decode UTF-8 string to local code page |
=head2 marc_remove |
|
if needed. |
|
697 |
|
|
698 |
=cut |
Remove some field or subfield from MARC record. |
699 |
|
|
700 |
sub fill_in { |
marc_remove('200'); |
701 |
my $self = shift; |
marc_remove('200','a'); |
702 |
|
|
703 |
my $log = $self->_get_logger(); |
This will erase field C<200> or C<200^a> from current MARC record. |
704 |
|
|
705 |
my $rec = shift || $log->logconfess("need data record"); |
marc_remove('*'); |
|
my $format = shift || $log->logconfess("need format to parse"); |
|
|
# iteration (for repeatable fields) |
|
|
my $i = shift || 0; |
|
706 |
|
|
707 |
$log->logdie("infitite loop in format $format") if ($i > ($self->{'max_mfn'} || 9999)); |
Will remove all fields in current MARC record. |
708 |
|
|
709 |
# FIXME remove for speedup? |
This is useful after calling C<marc_duplicate> or on it's own (but, you |
710 |
$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o); |
should probably just remove that subfield definition if you are not |
711 |
|
using C<marc_duplicate>). |
712 |
|
|
713 |
if (utf8::is_utf8($format)) { |
FIXME: support fields < 10. |
|
$format = $self->_x($format); |
|
|
} |
|
714 |
|
|
715 |
my $found = 0; |
=cut |
716 |
|
|
717 |
my $eval_code; |
sub marc_remove { |
718 |
# remove eval{...} from beginning |
my ($f, $sf) = @_; |
|
$eval_code = $1 if ($format =~ s/^eval{([^}]+)}//s); |
|
719 |
|
|
720 |
my $filter_name; |
die "marc_remove needs record number" unless defined($f); |
|
# remove filter{...} from beginning |
|
|
$filter_name = $1 if ($format =~ s/^filter{([^}]+)}//s); |
|
721 |
|
|
722 |
# do actual replacement of placeholders |
my $marc = $marc_record->[ $marc_record_offset ]; |
723 |
# repeatable fields |
|
724 |
$format =~ s/v(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,$i,\$found)/ges; |
warn "### marc_remove before = ", dump( $marc ), $/ if ($debug > 2); |
725 |
# non-repeatable fields |
|
726 |
$format =~ s/s(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,0,\$found)/ges; |
if ($f eq '*') { |
727 |
|
|
728 |
|
delete( $marc_record->[ $marc_record_offset ] ); |
729 |
|
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
730 |
|
|
|
if ($found) { |
|
|
$log->debug("format: $format"); |
|
|
if ($eval_code) { |
|
|
my $eval = $self->fill_in($rec,$eval_code,$i); |
|
|
return if (! $self->_eval($eval)); |
|
|
} |
|
|
if ($filter_name && $self->{'filter'}->{$filter_name}) { |
|
|
$log->debug("filter '$filter_name' for $format"); |
|
|
$format = $self->{'filter'}->{$filter_name}->($format); |
|
|
return unless(defined($format)); |
|
|
$log->debug("filter result: $format"); |
|
|
} |
|
|
# do we have lookups? |
|
|
if ($self->{'lookup'}) { |
|
|
return $self->lookup($format); |
|
|
} else { |
|
|
return $format; |
|
|
} |
|
731 |
} else { |
} else { |
732 |
return; |
|
733 |
|
my $i = 0; |
734 |
|
foreach ( 0 .. $#{ $marc } ) { |
735 |
|
last unless (defined $marc->[$i]); |
736 |
|
warn "#### working on ",dump( @{ $marc->[$i] }), $/ if ($debug > 3); |
737 |
|
if ($marc->[$i]->[0] eq $f) { |
738 |
|
if (! defined $sf) { |
739 |
|
# remove whole field |
740 |
|
splice @$marc, $i, 1; |
741 |
|
warn "#### slice \@\$marc, $i, 1 = ",dump( @{ $marc }), $/ if ($debug > 3); |
742 |
|
$i--; |
743 |
|
} else { |
744 |
|
foreach my $j ( 0 .. (( $#{ $marc->[$i] } - 3 ) / 2) ) { |
745 |
|
my $o = ($j * 2) + 3; |
746 |
|
if ($marc->[$i]->[$o] eq $sf) { |
747 |
|
# remove subfield |
748 |
|
splice @{$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 |
|
$i++; |
761 |
|
} |
762 |
|
|
763 |
|
warn "### marc_remove($f", $sf ? ",$sf" : "", ") after = ", dump( $marc ), $/ if ($debug > 2); |
764 |
|
|
765 |
|
$marc_record->[ $marc_record_offset ] = $marc; |
766 |
} |
} |
767 |
|
|
768 |
|
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
769 |
} |
} |
770 |
|
|
771 |
|
=head2 marc_original_order |
772 |
|
|
773 |
|
Copy all subfields preserving original order to marc field. |
774 |
|
|
775 |
=head2 fill_in_to_arr |
marc_original_order( marc_field_number, original_input_field_number ); |
776 |
|
|
777 |
Similar to C<fill_in>, but returns array of all repeatable fields. Usable |
Please note that field numbers are consistent with other commands (marc |
778 |
for fields which have lookups, so they shouldn't be parsed but rather |
field number first), but somewhat counter-intuitive (destination and then |
779 |
C<fill_id>ed. |
source). |
780 |
|
|
781 |
|
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 |
783 |
|
post-processing or want to preserve order of original subfields. |
784 |
|
|
|
my @arr = $webpac->fill_in_to_arr($rec,'[v900];;[v250^a]'); |
|
785 |
|
|
786 |
=cut |
=cut |
787 |
|
|
788 |
sub fill_in_to_arr { |
sub marc_original_order { |
|
my $self = shift; |
|
789 |
|
|
790 |
my ($rec, $format_utf8) = @_; |
my ($to, $from) = @_; |
791 |
|
die "marc_original_order needs from and to fields\n" unless ($from && $to); |
792 |
|
|
793 |
my $log = $self->_get_logger(); |
return unless defined($rec->{$from}); |
794 |
|
|
795 |
$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o); |
my $r = $rec->{$from}; |
796 |
return if (! $format_utf8); |
die "record field $from isn't array\n" unless (ref($r) eq 'ARRAY'); |
797 |
|
|
798 |
my $i = 0; |
my ($i1,$i2) = defined($marc_indicators->{$to}) ? @{ $marc_indicators->{$to} } : (' ',' '); |
799 |
my @arr; |
warn "## marc_original_order($to,$from) source = ", dump( $r ),$/ if ($debug > 1); |
800 |
|
|
801 |
while (my @v = $self->fill_in($rec,$format_utf8,$i++)) { |
foreach my $d (@$r) { |
802 |
push @arr, @v; |
|
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} }; |
809 |
|
|
810 |
|
die "field $from doesn't have even number of subfields specifications\n" unless($#sfs % 2 == 1); |
811 |
|
|
812 |
|
warn "#--> d: ",dump($d), "\n#--> sfs: ",dump(@sfs),$/ if ($debug > 2); |
813 |
|
|
814 |
$log->debug("format '$format_utf8' returned ",--$i," elements: ", sub { join(" | ",@arr) }) if (@arr); |
my $m = [ $to, $i1, $i2 ]; |
815 |
|
|
816 |
return @arr; |
while (my $sf = shift @sfs) { |
817 |
|
|
818 |
|
warn "#--> sf: ",dump($sf), $/ if ($debug > 2); |
819 |
|
my $offset = shift @sfs; |
820 |
|
die "corrupted sufields specification for field $from\n" unless defined($offset); |
821 |
|
|
822 |
|
my $v; |
823 |
|
if (ref($d->{$sf}) eq 'ARRAY') { |
824 |
|
$v = $d->{$sf}->[$offset] if (defined($d->{$sf}->[$offset])); |
825 |
|
} elsif ($offset == 0) { |
826 |
|
$v = $d->{$sf}; |
827 |
|
} else { |
828 |
|
die "field $from subfield '$sf' need occurence $offset which doesn't exist", dump($d->{$sf}); |
829 |
|
} |
830 |
|
push @$m, ( $sf, $v ) if (defined($v)); |
831 |
|
} |
832 |
|
|
833 |
|
if ($#{$m} > 2) { |
834 |
|
push @{ $marc_record->[ $marc_record_offset ] }, $m; |
835 |
|
} |
836 |
|
} |
837 |
|
|
838 |
|
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 |
=head2 get_data |
=cut |
848 |
|
|
849 |
|
sub marc_count { |
850 |
|
return $#{ $marc_record }; |
851 |
|
} |
852 |
|
|
|
Returns value from record. |
|
853 |
|
|
854 |
my $text = $self->get_data(\$rec,$f,$sf,$i,\$found); |
=head1 Functions to extract data from input |
855 |
|
|
856 |
Arguments are: |
This function should be used inside functions to create C<data_structure> described |
857 |
record reference C<$rec>, |
above. |
|
field C<$f>, |
|
|
optional subfiled C<$sf>, |
|
|
index for repeatable values C<$i>. |
|
858 |
|
|
859 |
Optinal variable C<$found> will be incremeted if there |
=head2 _pack_subfields_hash |
|
is field. |
|
860 |
|
|
861 |
Returns value or empty string. |
@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 |
=cut |
868 |
|
|
869 |
sub get_data { |
sub _pack_subfields_hash { |
870 |
my $self = shift; |
|
871 |
|
warn "## _pack_subfields_hash( ",dump(@_), " )\n" if ($debug > 1); |
872 |
|
|
873 |
|
my ($h,$include_subfields) = @_; |
874 |
|
|
875 |
my ($rec,$f,$sf,$i,$found) = @_; |
if ( defined($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 |
if ($$rec->{$f}) { |
push @out, $h->{$sf}; |
887 |
return '' if (! $$rec->{$f}->[$i]); |
} else { |
888 |
no strict 'refs'; |
#warn "====> $sf $o / $#$sfs ", dump( $sfs, $h->{$sf} ), "\n"; |
889 |
if ($sf && $$rec->{$f}->[$i]->{$sf}) { |
push @out, $h->{$sf}->[$o]; |
890 |
$$found++ if (defined($$found)); |
} |
891 |
return $$rec->{$f}->[$i]->{$sf}; |
} |
892 |
} elsif ($$rec->{$f}->[$i]) { |
if ($include_subfields) { |
893 |
$$found++ if (defined($$found)); |
return join('', @out); |
894 |
# it still might have subfield, just |
} else { |
895 |
# not specified, so we'll dump all |
return @out; |
896 |
if ($$rec->{$f}->[$i] =~ /HASH/o) { |
} |
897 |
my $out; |
} else { |
898 |
foreach my $k (keys %{$$rec->{$f}->[$i]}) { |
if ($include_subfields) { |
899 |
$out .= $$rec->{$f}->[$i]->{$k}." "; |
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 |
return $out; |
} |
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 |
916 |
|
|
917 |
|
Return all values in some field |
918 |
|
|
919 |
|
@v = rec1('200') |
920 |
|
|
921 |
|
TODO: order of values is probably same as in source data, need to investigate that |
922 |
|
|
923 |
|
=cut |
924 |
|
|
925 |
|
sub rec1 { |
926 |
|
my $f = shift; |
927 |
|
warn "rec1($f) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
928 |
|
return unless (defined($rec) && defined($rec->{$f})); |
929 |
|
warn "rec1($f) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
930 |
|
if (ref($rec->{$f}) eq 'ARRAY') { |
931 |
|
my @out; |
932 |
|
foreach my $h ( @{ $rec->{$f} } ) { |
933 |
|
if (ref($h) eq 'HASH') { |
934 |
|
push @out, ( _pack_subfields_hash( $h ) ); |
935 |
} else { |
} else { |
936 |
return $$rec->{$f}->[$i]; |
push @out, $h; |
937 |
} |
} |
938 |
} |
} |
939 |
|
return @out; |
940 |
|
} elsif( defined($rec->{$f}) ) { |
941 |
|
return $rec->{$f}; |
942 |
|
} |
943 |
|
} |
944 |
|
|
945 |
|
=head2 rec2 |
946 |
|
|
947 |
|
Return all values in specific field and subfield |
948 |
|
|
949 |
|
@v = rec2('200','a') |
950 |
|
|
951 |
|
=cut |
952 |
|
|
953 |
|
sub rec2 { |
954 |
|
my $f = shift; |
955 |
|
return unless (defined($rec && $rec->{$f})); |
956 |
|
my $sf = shift; |
957 |
|
warn "rec2($f,$sf) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
958 |
|
return map { |
959 |
|
if (ref($_->{$sf}) eq 'ARRAY') { |
960 |
|
@{ $_->{$sf} }; |
961 |
|
} else { |
962 |
|
$_->{$sf}; |
963 |
|
} |
964 |
|
} grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} }; |
965 |
|
} |
966 |
|
|
967 |
|
=head2 rec |
968 |
|
|
969 |
|
syntaxtic sugar for |
970 |
|
|
971 |
|
@v = rec('200') |
972 |
|
@v = rec('200','a') |
973 |
|
|
974 |
|
If rec() returns just single value, it will |
975 |
|
return scalar, not array. |
976 |
|
|
977 |
|
=cut |
978 |
|
|
979 |
|
sub rec { |
980 |
|
my @out; |
981 |
|
if ($#_ == 0) { |
982 |
|
@out = rec1(@_); |
983 |
|
} elsif ($#_ == 1) { |
984 |
|
@out = rec2(@_); |
985 |
|
} |
986 |
|
if ($#out == 0 && ! wantarray) { |
987 |
|
return $out[0]; |
988 |
|
} elsif (@out) { |
989 |
|
return @out; |
990 |
} else { |
} else { |
991 |
return ''; |
return ''; |
992 |
} |
} |
993 |
} |
} |
994 |
|
|
995 |
|
=head2 regex |
996 |
|
|
997 |
|
Apply regex to some or all values |
998 |
|
|
999 |
|
@v = regex( 's/foo/bar/g', @v ); |
1000 |
|
|
1001 |
|
=cut |
1002 |
|
|
1003 |
|
sub regex { |
1004 |
|
my $r = shift; |
1005 |
|
my @out; |
1006 |
|
#warn "r: $r\n", dump(\@_); |
1007 |
|
foreach my $t (@_) { |
1008 |
|
next unless ($t); |
1009 |
|
eval "\$t =~ $r"; |
1010 |
|
push @out, $t if ($t && $t ne ''); |
1011 |
|
} |
1012 |
|
return @out; |
1013 |
|
} |
1014 |
|
|
1015 |
|
=head2 prefix |
1016 |
|
|
1017 |
|
Prefix all values with a string |
1018 |
|
|
1019 |
|
@v = prefix( 'my_', @v ); |
1020 |
|
|
1021 |
|
=cut |
1022 |
|
|
1023 |
|
sub prefix { |
1024 |
|
my $p = shift; |
1025 |
|
return @_ unless defined( $p ); |
1026 |
|
return map { $p . $_ } grep { defined($_) } @_; |
1027 |
|
} |
1028 |
|
|
1029 |
|
=head2 suffix |
1030 |
|
|
1031 |
|
suffix all values with a string |
1032 |
|
|
1033 |
|
@v = suffix( '_my', @v ); |
1034 |
|
|
1035 |
|
=cut |
1036 |
|
|
1037 |
|
sub suffix { |
1038 |
|
my $s = shift; |
1039 |
|
return @_ unless defined( $s ); |
1040 |
|
return map { $_ . $s } grep { defined($_) } @_; |
1041 |
|
} |
1042 |
|
|
1043 |
|
=head2 surround |
1044 |
|
|
1045 |
|
surround all values with a two strings |
1046 |
|
|
1047 |
|
@v = surround( 'prefix_', '_suffix', @v ); |
1048 |
|
|
1049 |
|
=cut |
1050 |
|
|
1051 |
=head2 apply_format |
sub surround { |
1052 |
|
my $p = shift; |
1053 |
|
my $s = shift; |
1054 |
|
$p = '' unless defined( $p ); |
1055 |
|
$s = '' unless defined( $s ); |
1056 |
|
return map { $p . $_ . $s } grep { defined($_) } @_; |
1057 |
|
} |
1058 |
|
|
1059 |
Apply format specified in tag with C<format_name="name"> and |
=head2 first |
|
C<format_delimiter=";;">. |
|
1060 |
|
|
1061 |
my $text = $webpac->apply_format($format_name,$format_delimiter,$data); |
Return first element |
1062 |
|
|
1063 |
Formats can contain C<lookup{...}> if you need them. |
$v = first( @v ); |
1064 |
|
|
1065 |
=cut |
=cut |
1066 |
|
|
1067 |
sub apply_format { |
sub first { |
1068 |
my $self = shift; |
my $r = shift; |
1069 |
|
return $r; |
1070 |
|
} |
1071 |
|
|
1072 |
my ($name,$delimiter,$data) = @_; |
=head2 lookup |
1073 |
|
|
1074 |
my $log = $self->_get_logger(); |
Consult lookup hashes for some value |
1075 |
|
|
1076 |
if (! $self->{'import_xml'}->{'format'}->{$name}) { |
@v = lookup( |
1077 |
$log->warn("<format name=\"$name\"> is not defined in ",$self->{'import_xml_file'}); |
sub { |
1078 |
return $data; |
'ffkk/peri/mfn'.rec('000') |
1079 |
|
}, |
1080 |
|
'ffkk','peri','200-a-200-e', |
1081 |
|
sub { |
1082 |
|
first(rec(200,'a')).' '.first(rec('200','e')) |
1083 |
} |
} |
1084 |
|
); |
1085 |
|
|
1086 |
|
Code like above will be B<automatically generated> using L<WebPAC::Parse> from |
1087 |
|
normal lookup definition in C<conf/lookup/something.pl> which looks like: |
1088 |
|
|
1089 |
|
lookup( |
1090 |
|
# which results to return from record recorded in lookup |
1091 |
|
sub { 'ffkk/peri/mfn' . rec('000') }, |
1092 |
|
# from which database and input |
1093 |
|
'ffkk','peri', |
1094 |
|
# such that following values match |
1095 |
|
sub { first(rec(200,'a')) . ' ' . first(rec('200','e')) }, |
1096 |
|
# if this part is missing, we will try to match same fields |
1097 |
|
# from lookup record and current one, or you can override |
1098 |
|
# which records to use from current record using |
1099 |
|
sub { rec('900','x') . ' ' . rec('900','y') }, |
1100 |
|
) |
1101 |
|
|
1102 |
|
You can think about this lookup as SQL (if that helps): |
1103 |
|
|
1104 |
|
select |
1105 |
|
sub { what } |
1106 |
|
from |
1107 |
|
database, input |
1108 |
|
where |
1109 |
|
sub { filter from lookuped record } |
1110 |
|
having |
1111 |
|
sub { optional filter on current record } |
1112 |
|
|
1113 |
|
Easy as pie, right? |
1114 |
|
|
1115 |
|
=cut |
1116 |
|
|
1117 |
|
sub lookup { |
1118 |
|
my ($what, $database, $input, $key, $having) = @_; |
1119 |
|
|
1120 |
$log->warn("no delimiter for format $name") if (! $delimiter); |
confess "lookup needs 5 arguments: what, database, input, key, having\n" unless ($#_ == 4); |
1121 |
|
|
1122 |
my $format = $self->_x($self->{'import_xml'}->{'format'}->{$name}->{'content'}) || $log->logdie("can't find format '$name'"); |
warn "## lookup ($database, $input, $key)", $/ if ($debug > 1); |
1123 |
|
return unless (defined($lookup->{$database}->{$input}->{$key})); |
1124 |
|
|
1125 |
my @data = split(/\Q$delimiter\E/, $data); |
confess "lookup really need load_row_coderef added to data_structure\n" unless ($load_row_coderef); |
1126 |
|
|
1127 |
my $out = sprintf($format, @data); |
my $mfns; |
1128 |
$log->debug("using format $name [$format] on $data to produce: $out"); |
my @having = $having->(); |
1129 |
|
|
1130 |
if ($self->{'lookup_regex'} && $out =~ $self->{'lookup_regex'}) { |
warn "## having = ", dump( @having ) if ($debug > 2); |
1131 |
return $self->lookup($out); |
|
1132 |
|
foreach my $h ( @having ) { |
1133 |
|
if (defined($lookup->{$database}->{$input}->{$key}->{$h})) { |
1134 |
|
warn "lookup for $database/$input/$key/$h return ",dump($lookup->{$database}->{$input}->{$key}->{$h}),"\n" if ($debug); |
1135 |
|
$mfns->{$_}++ foreach keys %{ $lookup->{$database}->{$input}->{$key}->{$h} }; |
1136 |
|
} |
1137 |
|
} |
1138 |
|
|
1139 |
|
return unless ($mfns); |
1140 |
|
|
1141 |
|
my @mfns = sort keys %$mfns; |
1142 |
|
|
1143 |
|
warn "# lookup loading $database/$input/$key mfn ", join(",",@mfns)," having ",dump(@having),"\n" if ($debug); |
1144 |
|
|
1145 |
|
my $old_rec = $rec; |
1146 |
|
my @out; |
1147 |
|
|
1148 |
|
foreach my $mfn (@mfns) { |
1149 |
|
$rec = $load_row_coderef->( $database, $input, $mfn ); |
1150 |
|
|
1151 |
|
warn "got $database/$input/$mfn = ", dump($rec), $/ if ($debug); |
1152 |
|
|
1153 |
|
my @vals = $what->(); |
1154 |
|
|
1155 |
|
push @out, ( @vals ); |
1156 |
|
|
1157 |
|
warn "lookup for mfn $mfn returned ", dump(@vals), $/ if ($debug); |
1158 |
|
} |
1159 |
|
|
1160 |
|
# if (ref($lookup->{$k}) eq 'ARRAY') { |
1161 |
|
# return @{ $lookup->{$k} }; |
1162 |
|
# } else { |
1163 |
|
# return $lookup->{$k}; |
1164 |
|
# } |
1165 |
|
|
1166 |
|
$rec = $old_rec; |
1167 |
|
|
1168 |
|
warn "## lookup returns = ", dump(@out), $/ if ($debug); |
1169 |
|
|
1170 |
|
if ($#out == 0) { |
1171 |
|
return $out[0]; |
1172 |
} else { |
} else { |
1173 |
return $out; |
return @out; |
1174 |
} |
} |
1175 |
|
} |
1176 |
|
|
1177 |
|
=head2 save_into_lookup |
1178 |
|
|
1179 |
|
Save value into lookup. It associates current database, input |
1180 |
|
and specific keys with one or more values which will be |
1181 |
|
associated over MFN. |
1182 |
|
|
1183 |
|
MFN will be extracted from first occurence current of field 000 |
1184 |
|
in current record, or if it doesn't exist from L<_set_config> C<_mfn>. |
1185 |
|
|
1186 |
|
my $nr = save_into_lookup($database,$input,$key,sub { |
1187 |
|
# code which produce one or more values |
1188 |
|
}); |
1189 |
|
|
1190 |
|
It returns number of items saved. |
1191 |
|
|
1192 |
|
This function shouldn't be called directly, it's called from code created by |
1193 |
|
L<WebPAC::Parser>. |
1194 |
|
|
1195 |
|
=cut |
1196 |
|
|
1197 |
|
sub save_into_lookup { |
1198 |
|
my ($database,$input,$key,$coderef) = @_; |
1199 |
|
die "save_into_lookup needs database" unless defined($database); |
1200 |
|
die "save_into_lookup needs input" unless defined($input); |
1201 |
|
die "save_into_lookup needs key" unless defined($key); |
1202 |
|
die "save_into_lookup needs CODE" unless ( defined($coderef) && ref($coderef) eq 'CODE' ); |
1203 |
|
|
1204 |
|
warn "## save_into_lookup rec = ", dump($rec), " config = ", dump($config), $/ if ($debug > 2); |
1205 |
|
|
1206 |
|
my $mfn = |
1207 |
|
defined($rec->{'000'}->[0]) ? $rec->{'000'}->[0] : |
1208 |
|
defined($config->{_mfn}) ? $config->{_mfn} : |
1209 |
|
die "mfn not defined or zero"; |
1210 |
|
|
1211 |
|
my $nr = 0; |
1212 |
|
|
1213 |
|
foreach my $v ( $coderef->() ) { |
1214 |
|
$lookup->{$database}->{$input}->{$key}->{$v}->{$mfn}++; |
1215 |
|
warn "# saved lookup $database/$input/$key [$v] $mfn\n" if ($debug > 1); |
1216 |
|
$nr++; |
1217 |
|
} |
1218 |
|
|
1219 |
|
return $nr; |
1220 |
} |
} |
1221 |
|
|
1222 |
=head2 sort_arr |
=head2 config |
1223 |
|
|
1224 |
|
Consult config values stored in C<config.yml> |
1225 |
|
|
1226 |
Sort array ignoring case and html in data |
# return database code (key under databases in yaml) |
1227 |
|
$database_code = config(); # use _ from hash |
1228 |
|
$database_name = config('name'); |
1229 |
|
$database_input_name = config('input name'); |
1230 |
|
$tag = config('input normalize tag'); |
1231 |
|
|
1232 |
my @sorted = $webpac->sort_arr(@unsorted); |
Up to three levels are supported. |
1233 |
|
|
1234 |
=cut |
=cut |
1235 |
|
|
1236 |
sub sort_arr { |
sub config { |
1237 |
my $self = shift; |
return unless ($config); |
1238 |
|
|
1239 |
|
my $p = shift; |
1240 |
|
|
1241 |
|
$p ||= ''; |
1242 |
|
|
1243 |
my $log = $self->_get_logger(); |
my $v; |
1244 |
|
|
1245 |
|
warn "### getting config($p)\n" if ($debug > 1); |
1246 |
|
|
1247 |
|
my @p = split(/\s+/,$p); |
1248 |
|
if ($#p < 0) { |
1249 |
|
$v = $config->{ '_' }; # special, database code |
1250 |
|
} else { |
1251 |
|
|
1252 |
|
my $c = dclone( $config ); |
1253 |
|
|
1254 |
|
foreach my $k (@p) { |
1255 |
|
warn "### k: $k c = ",dump($c),$/ if ($debug > 1); |
1256 |
|
if (ref($c) eq 'ARRAY') { |
1257 |
|
$c = shift @$c; |
1258 |
|
warn "config($p) taking first occurence of '$k', probably not what you wanted!\n"; |
1259 |
|
last; |
1260 |
|
} |
1261 |
|
|
1262 |
|
if (! defined($c->{$k}) ) { |
1263 |
|
$c = undef; |
1264 |
|
last; |
1265 |
|
} else { |
1266 |
|
$c = $c->{$k}; |
1267 |
|
} |
1268 |
|
} |
1269 |
|
$v = $c if ($c); |
1270 |
|
|
1271 |
# FIXME add Schwartzian Transformation? |
} |
1272 |
|
|
1273 |
my @sorted = sort { |
warn "## config( '$p' ) = ",dump( $v ),$/ if ($v && $debug); |
1274 |
$a =~ s#<[^>]+/*>##; |
warn "config( '$p' ) is empty\n" if (! $v); |
|
$b =~ s#<[^>]+/*>##; |
|
|
lc($b) cmp lc($a) |
|
|
} @_; |
|
|
$log->debug("sorted values: ",sub { join(", ",@sorted) }); |
|
1275 |
|
|
1276 |
return @sorted; |
return $v; |
1277 |
} |
} |
1278 |
|
|
1279 |
|
=head2 id |
1280 |
|
|
1281 |
=head1 INTERNAL METHODS |
Returns unique id of this record |
1282 |
|
|
1283 |
=head2 _sort_by_order |
$id = id(); |
1284 |
|
|
1285 |
Sort xml tags data structure accoding to C<order=""> attribute. |
Returns C<42/2> for 2nd occurence of MFN 42. |
1286 |
|
|
1287 |
=cut |
=cut |
1288 |
|
|
1289 |
sub _sort_by_order { |
sub id { |
1290 |
my $self = shift; |
my $mfn = $config->{_mfn} || die "no _mfn in config data"; |
1291 |
|
return $mfn . $#{$marc_record} ? $#{$marc_record} + 1 : ''; |
1292 |
|
} |
1293 |
|
|
1294 |
|
=head2 join_with |
1295 |
|
|
1296 |
my $va = $self->{'import_xml'}->{'indexer'}->{$a}->{'order'} || |
Joins walues with some delimiter |
1297 |
$self->{'import_xml'}->{'indexer'}->{$a}; |
|
1298 |
my $vb = $self->{'import_xml'}->{'indexer'}->{$b}->{'order'} || |
$v = join_with(", ", @v); |
1299 |
$self->{'import_xml'}->{'indexer'}->{$b}; |
|
1300 |
|
=cut |
1301 |
|
|
1302 |
return $va <=> $vb; |
sub join_with { |
1303 |
|
my $d = shift; |
1304 |
|
warn "### join_with('$d',",dump(@_),")\n" if ($debug > 2); |
1305 |
|
my $v = join($d, grep { defined($_) && $_ ne '' } @_); |
1306 |
|
return '' unless defined($v); |
1307 |
|
return $v; |
1308 |
} |
} |
1309 |
|
|
1310 |
=head2 _x |
=head2 split_rec_on |
1311 |
|
|
1312 |
Convert strings from C<conf/normalize/*.xml> encoding into application |
Split record subfield on some regex and take one of parts out |
|
specific encoding (optinally specified using C<code_page> to C<new> |
|
|
constructor). |
|
1313 |
|
|
1314 |
my $text = $n->_x('normalize text string'); |
$a_before_semi_column = |
1315 |
|
split_rec_on('200','a', /\s*;\s*/, $part); |
1316 |
|
|
1317 |
This is a stub so that other modules doesn't have to implement it. |
C<$part> is optional number of element. First element is |
1318 |
|
B<1>, not 0! |
1319 |
|
|
1320 |
|
If there is no C<$part> parameter or C<$part> is 0, this function will |
1321 |
|
return all values produced by splitting. |
1322 |
|
|
1323 |
=cut |
=cut |
1324 |
|
|
1325 |
sub _x { |
sub split_rec_on { |
1326 |
my $self = shift; |
die "split_rec_on need (fld,sf,regex[,part]" if ($#_ < 2); |
1327 |
return shift; |
|
1328 |
|
my ($fld, $sf, $regex, $part) = @_; |
1329 |
|
warn "### regex ", ref($regex), $regex, $/ if ($debug > 2); |
1330 |
|
|
1331 |
|
my @r = rec( $fld, $sf ); |
1332 |
|
my $v = shift @r; |
1333 |
|
warn "### first rec($fld,$sf) = ",dump($v),$/ if ($debug > 2); |
1334 |
|
|
1335 |
|
return '' if ( ! defined($v) || $v =~ /^\s*$/); |
1336 |
|
|
1337 |
|
my @s = split( $regex, $v ); |
1338 |
|
warn "## split_rec_on($fld,$sf,$regex,$part) = ",dump(@s),$/ if ($debug > 1); |
1339 |
|
if ($part && $part > 0) { |
1340 |
|
return $s[ $part - 1 ]; |
1341 |
|
} else { |
1342 |
|
return @s; |
1343 |
|
} |
1344 |
} |
} |
1345 |
|
|
1346 |
|
my $hash; |
1347 |
|
|
1348 |
=head1 AUTHOR |
=head2 set |
1349 |
|
|
1350 |
Dobrica Pavlinusic, C<< <dpavlin@rot13.org> >> |
set( key => 'value' ); |
1351 |
|
|
1352 |
=head1 COPYRIGHT & LICENSE |
=cut |
1353 |
|
|
1354 |
|
sub set { |
1355 |
|
my ($k,$v) = @_; |
1356 |
|
warn "## set ( $k => ", dump($v), " )", $/ if ( $debug ); |
1357 |
|
$hash->{$k} = $v; |
1358 |
|
}; |
1359 |
|
|
1360 |
|
=head2 get |
1361 |
|
|
1362 |
|
get( 'key' ); |
1363 |
|
|
1364 |
|
=cut |
1365 |
|
|
1366 |
|
sub get { |
1367 |
|
my $k = shift || return; |
1368 |
|
my $v = $hash->{$k}; |
1369 |
|
warn "## get $k = ", dump( $v ), $/ if ( $debug ); |
1370 |
|
return $v; |
1371 |
|
} |
1372 |
|
|
1373 |
Copyright 2005 Dobrica Pavlinusic, All Rights Reserved. |
=head2 count |
1374 |
|
|
1375 |
This program is free software; you can redistribute it and/or modify it |
if ( count( @result ) == 1 ) { |
1376 |
under the same terms as Perl itself. |
# do something if only 1 result is there |
1377 |
|
} |
1378 |
|
|
1379 |
=cut |
=cut |
1380 |
|
|
1381 |
1; # End of WebPAC::DB |
sub count { |
1382 |
|
warn "## count ",dump(@_),$/ if ( $debug ); |
1383 |
|
return @_ . ''; |
1384 |
|
} |
1385 |
|
|
1386 |
|
# END |
1387 |
|
1; |