2 |
|
|
3 |
use warnings; |
use warnings; |
4 |
use strict; |
use strict; |
5 |
|
use blib; |
6 |
|
use WebPAC::Common; |
7 |
|
use base 'WebPAC::Common'; |
8 |
use Data::Dumper; |
use Data::Dumper; |
9 |
|
|
10 |
=head1 NAME |
=head1 NAME |
13 |
|
|
14 |
=head1 VERSION |
=head1 VERSION |
15 |
|
|
16 |
Version 0.01 |
Version 0.09 |
17 |
|
|
18 |
=cut |
=cut |
19 |
|
|
20 |
our $VERSION = '0.01'; |
our $VERSION = '0.09'; |
21 |
|
|
22 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
23 |
|
|
49 |
code defined as code ref on format after field substitution to producing |
code defined as code ref on format after field substitution to producing |
50 |
output |
output |
51 |
|
|
52 |
|
There is one built-in filter called C<regex> which can be use like this: |
53 |
|
|
54 |
|
filter{regex(s/foo/bar/)} |
55 |
|
|
56 |
=item * |
=item * |
57 |
|
|
58 |
optional C<lookup{...}> will be then performed. See C<WebPAC::Lookups>. |
optional C<lookup{...}> will be then performed. See C<WebPAC::Lookups>. |
85 |
return length($_); |
return length($_); |
86 |
}, ... |
}, ... |
87 |
}, |
}, |
88 |
db => $webpac_db_obj, |
db => $db_obj, |
89 |
lookup_regex => $lookup->regex, |
lookup_regex => $lookup->regex, |
90 |
|
lookup => $lookup_obj, |
91 |
|
prefix => 'foobar', |
92 |
); |
); |
93 |
|
|
94 |
Parametar C<filter> defines user supplied snippets of perl code which can |
Parametar C<filter> defines user supplied snippets of perl code which can |
95 |
be use with C<filter{...}> notation. |
be use with C<filter{...}> notation. |
96 |
|
|
97 |
|
C<prefix> is used to form filename for database record (to support multiple |
98 |
|
source files which are joined in one database). |
99 |
|
|
100 |
Recommended parametar C<lookup_regex> is used to enable parsing of lookups |
Recommended parametar C<lookup_regex> is used to enable parsing of lookups |
101 |
in structures. |
in structures. If you pass this parametar, you must also pass C<lookup> |
102 |
|
which is C<WebPAC::Lookup> object. |
103 |
|
|
104 |
=cut |
=cut |
105 |
|
|
108 |
my $self = {@_}; |
my $self = {@_}; |
109 |
bless($self, $class); |
bless($self, $class); |
110 |
|
|
111 |
|
my $r = $self->{'lookup_regex'} ? 1 : 0; |
112 |
|
my $l = $self->{'lookup'} ? 1 : 0; |
113 |
|
|
114 |
|
my $log = $self->_get_logger(); |
115 |
|
|
116 |
|
# those two must be in pair |
117 |
|
if ( ($r & $l) != ($r || $l) ) { |
118 |
|
my $log = $self->_get_logger(); |
119 |
|
$log->logdie("lookup_regex and lookup must be in pair"); |
120 |
|
} |
121 |
|
|
122 |
|
$log->logdie("lookup must be WebPAC::Lookup object") if ($self->{'lookup'} && ! $self->{'lookup'}->isa('WebPAC::Lookup')); |
123 |
|
|
124 |
|
$log->warn("no prefix defined. please check that!") unless ($self->{'prefix'}); |
125 |
|
|
126 |
|
$log->debug("using lookup regex: ", $self->{lookup_regex}) if ($r && $l); |
127 |
|
|
128 |
|
if (! $self->{filter} || ! $self->{filter}->{regex}) { |
129 |
|
$log->debug("adding built-in filter regex"); |
130 |
|
$self->{filter}->{regex} = sub { |
131 |
|
my ($val, $regex) = @_; |
132 |
|
eval "\$val =~ $regex"; |
133 |
|
return $val; |
134 |
|
}; |
135 |
|
} |
136 |
|
|
137 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
138 |
} |
} |
139 |
|
|
140 |
|
=head2 all_tags |
141 |
|
|
142 |
|
Returns all tags in document in specified order |
143 |
|
|
144 |
|
my $sorted_tags = $self->all_tags(); |
145 |
|
|
146 |
|
=cut |
147 |
|
|
148 |
|
sub all_tags { |
149 |
|
my $self = shift; |
150 |
|
|
151 |
|
if (! $self->{_tags_by_order}) { |
152 |
|
|
153 |
|
my $log = $self->_get_logger; |
154 |
|
# sanity check |
155 |
|
$log->logdie("can't find self->{inport_xml}->{indexer}") unless ($self->{import_xml}->{indexer}); |
156 |
|
|
157 |
|
my @tags = keys %{ $self->{'import_xml'}->{'indexer'}}; |
158 |
|
$log->debug("unsorted tags: " . join(", ", @tags)); |
159 |
|
|
160 |
|
@tags = sort { $self->_sort_by_order } @tags; |
161 |
|
|
162 |
|
$log->debug("sorted tags: " . join(",", @tags) ); |
163 |
|
|
164 |
|
$self->{_tags_by_order} = \@tags; |
165 |
|
} |
166 |
|
|
167 |
|
return $self->{_tags_by_order}; |
168 |
|
} |
169 |
|
|
170 |
|
|
171 |
|
|
172 |
=head2 data_structure |
=head2 data_structure |
173 |
|
|
176 |
|
|
177 |
This structures are used to produce output. |
This structures are used to produce output. |
178 |
|
|
179 |
my @ds = $webpac->data_structure($rec); |
my $ds = $webpac->data_structure($rec); |
|
|
|
|
B<Note: historical oddity follows> |
|
|
|
|
|
This method will also set C<< $webpac->{'currnet_filename'} >> if there is |
|
|
C<< <filename> >> tag and C<< $webpac->{'headline'} >> if there is |
|
|
C<< <headline> >> tag. |
|
180 |
|
|
181 |
=cut |
=cut |
182 |
|
|
188 |
my $rec = shift; |
my $rec = shift; |
189 |
$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o); |
$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o); |
190 |
|
|
191 |
|
$log->debug("data_structure rec = ", sub { Dumper($rec) }); |
192 |
|
|
193 |
|
$log->logdie("need unique ID (mfn) in field 000 of record " . Dumper($rec) ) unless (defined($rec->{'000'})); |
194 |
|
|
195 |
|
my $id = $rec->{'000'}->[0] || $log->logdie("field 000 isn't array!"); |
196 |
|
|
197 |
my $cache_file; |
my $cache_file; |
198 |
|
|
199 |
if ($self->{'db'}) { |
if ($self->{'db'}) { |
200 |
my @ds = $self->{'db'}->get_ds($rec); |
my $ds = $self->{'db'}->load_ds( id => $id, prefix => $self->{prefix} ); |
201 |
return @ds if (@ds); |
$log->debug("load_ds( rec = ", sub { Dumper($rec) }, ") = ", sub { Dumper($ds) }); |
202 |
|
return $ds if ($ds); |
203 |
|
$log->debug("cache miss, creating"); |
204 |
} |
} |
205 |
|
|
206 |
undef $self->{'currnet_filename'}; |
my $tags = $self->all_tags(); |
|
undef $self->{'headline'}; |
|
|
|
|
|
my @sorted_tags; |
|
|
if ($self->{tags_by_order}) { |
|
|
@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; |
|
|
} |
|
207 |
|
|
208 |
my @ds; |
$log->debug("tags: ",sub { join(", ",@{ $tags }) }); |
209 |
|
|
210 |
$log->debug("tags: ",sub { join(", ",@sorted_tags) }); |
my $ds; |
211 |
|
|
212 |
foreach my $field (@sorted_tags) { |
foreach my $field (@{ $tags }) { |
213 |
|
|
214 |
my $row; |
my $row; |
215 |
|
|
216 |
#print "field $field [",$self->{'tag'},"] = ",Dumper($self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}}); |
#print "field $field [",$self->{'tag'},"] = ",Dumper($self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}}); |
217 |
|
|
218 |
foreach my $tag (@{$self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}}}) { |
foreach my $tag (@{$self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}}}) { |
219 |
my $format = $tag->{'value'} || $tag->{'content'}; |
my $format; |
220 |
|
|
221 |
$log->debug("format: $format"); |
$log->logdie("expected tag HASH and got $tag") unless (ref($tag) eq 'HASH'); |
222 |
|
$format = $tag->{'value'} || $tag->{'content'}; |
223 |
|
|
224 |
my @v; |
my @v; |
225 |
if ($self->{'lookup_regex'} && $format =~ $self->{'lookup_regex'}) { |
if ($self->{'lookup_regex'} && $format =~ $self->{'lookup_regex'}) { |
226 |
@v = $self->fill_in_to_arr($rec,$format); |
@v = $self->_rec_to_arr($rec,$format,'fill_in'); |
227 |
} else { |
} else { |
228 |
@v = $self->parse_to_arr($rec,$format); |
@v = $self->_rec_to_arr($rec,$format,'parse'); |
229 |
|
} |
230 |
|
if (! @v) { |
231 |
|
$log->debug("$field <",$self->{tag},"> format: $format no values"); |
232 |
|
next; |
233 |
|
} else { |
234 |
|
$log->debug("$field <",$self->{tag},"> format: $format values: ", join(",", @v)); |
235 |
} |
} |
|
next if (! @v); |
|
236 |
|
|
237 |
if ($tag->{'sort'}) { |
if ($tag->{'sort'}) { |
238 |
@v = $self->sort_arr(@v); |
@v = $self->sort_arr(@v); |
243 |
@v = map { $self->apply_format($tag->{'format_name'},$tag->{'format_delimiter'},$_) } @v; |
@v = map { $self->apply_format($tag->{'format_name'},$tag->{'format_delimiter'},$_) } @v; |
244 |
} |
} |
245 |
|
|
|
if ($field eq 'filename') { |
|
|
$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! |
|
|
} |
|
|
|
|
246 |
# delimiter will join repeatable fields |
# delimiter will join repeatable fields |
247 |
if ($tag->{'delimiter'}) { |
if ($tag->{'delimiter'}) { |
248 |
@v = ( join($tag->{'delimiter'}, @v) ); |
@v = ( join($tag->{'delimiter'}, @v) ); |
249 |
} |
} |
250 |
|
|
251 |
# default types |
# default types |
252 |
my @types = qw(display swish); |
my @types = qw(display search); |
253 |
# override by type attribute |
# override by type attribute |
254 |
@types = ( $tag->{'type'} ) if ($tag->{'type'}); |
@types = ( $tag->{'type'} ) if ($tag->{'type'}); |
255 |
|
|
256 |
foreach my $type (@types) { |
foreach my $type (@types) { |
257 |
# append to previous line? |
# append to previous line? |
258 |
$log->debug("type: $type ",sub { join(" ",@v) }, $row->{'append'} || 'no append'); |
$log->debug("tag $field / $type [",sub { join(",",@v) }, "] ", $row->{'append'} || 'no append'); |
259 |
if ($tag->{'append'}) { |
if ($tag->{'append'}) { |
260 |
|
|
261 |
# I will delimit appended part with |
# I will delimit appended part with |
282 |
|
|
283 |
# TODO: name_sigular, name_plural |
# TODO: name_sigular, name_plural |
284 |
my $name = $self->{'import_xml'}->{'indexer'}->{$field}->{'name'}; |
my $name = $self->{'import_xml'}->{'indexer'}->{$field}->{'name'}; |
285 |
$row->{'name'} = $name ? $self->_x($name) : $field; |
my $row_name = $name ? $self->_x($name) : $field; |
286 |
|
|
287 |
# post-sort all values in field |
# post-sort all values in field |
288 |
if ($self->{'import_xml'}->{'indexer'}->{$field}->{'sort'}) { |
if ($self->{'import_xml'}->{'indexer'}->{$field}->{'sort'}) { |
289 |
$log->warn("sort at field tag not implemented"); |
$log->warn("sort at field tag not implemented"); |
290 |
} |
} |
291 |
|
|
292 |
push @ds, $row; |
$ds->{$row_name} = $row; |
293 |
|
|
294 |
$log->debug("row $field: ",sub { Dumper($row) }); |
$log->debug("row $field: ",sub { Dumper($row) }); |
295 |
} |
} |
296 |
|
|
297 |
} |
} |
298 |
|
|
299 |
$self->{'db'}->put_gs( |
$self->{'db'}->save_ds( |
300 |
ds => \@ds, |
id => $id, |
301 |
current_filename => $self->{'current_filename'}, |
ds => $ds, |
302 |
headline => $self->{'headline'}, |
prefix => $self->{prefix}, |
303 |
) if ($self->{'db'}); |
) if ($self->{'db'}); |
304 |
|
|
305 |
return @ds; |
$log->debug("ds: ", sub { Dumper($ds) }); |
306 |
|
|
307 |
|
$log->logconfess("data structure returned is not array any more!") if wantarray; |
308 |
|
|
309 |
|
return $ds; |
310 |
|
|
311 |
} |
} |
312 |
|
|
318 |
|
|
319 |
my $text = $webpac->parse($rec,'eval{"v901^a" eq "Deskriptor"}descriptor: v250^a', $i); |
my $text = $webpac->parse($rec,'eval{"v901^a" eq "Deskriptor"}descriptor: v250^a', $i); |
320 |
|
|
321 |
|
Filters are implemented here. While simple form of filters looks like this: |
322 |
|
|
323 |
|
filter{name_of_filter} |
324 |
|
|
325 |
|
but, filters can also have variable number of parametars like this: |
326 |
|
|
327 |
|
filter{name_of_filter(param,param,param)} |
328 |
|
|
329 |
=cut |
=cut |
330 |
|
|
331 |
|
my $warn_once; |
332 |
|
|
333 |
sub parse { |
sub parse { |
334 |
my $self = shift; |
my $self = shift; |
335 |
|
|
336 |
my ($rec, $format_utf8, $i) = @_; |
my ($rec, $format_utf8, $i, $rec_size) = @_; |
337 |
|
|
338 |
return if (! $format_utf8); |
return if (! $format_utf8); |
339 |
|
|
347 |
|
|
348 |
my @out; |
my @out; |
349 |
|
|
350 |
$log->debug("format: $format"); |
$log->debug("format: $format [$i]"); |
351 |
|
|
352 |
my $eval_code; |
my $eval_code; |
353 |
# remove eval{...} from beginning |
# remove eval{...} from beginning |
357 |
# remove filter{...} from beginning |
# remove filter{...} from beginning |
358 |
$filter_name = $1 if ($format =~ s/^filter{([^}]+)}//s); |
$filter_name = $1 if ($format =~ s/^filter{([^}]+)}//s); |
359 |
|
|
360 |
|
# did we found any (att all) field from format in row? |
361 |
|
my $found_any; |
362 |
|
# prefix before first field which we preserve it $found_any |
363 |
my $prefix; |
my $prefix; |
364 |
my $all_found=0; |
|
365 |
|
my $f_step = 1; |
366 |
|
|
367 |
while ($format =~ s/^(.*?)(v|s)(\d+)(?:\^(\w))?//s) { |
while ($format =~ s/^(.*?)(v|s)(\d+)(?:\^(\w))?//s) { |
368 |
|
|
369 |
my $del = $1 || ''; |
my $del = $1 || ''; |
370 |
$prefix ||= $del if ($all_found == 0); |
$prefix = $del if ($f_step == 1); |
371 |
|
|
372 |
|
my $fld_type = lc($2); |
373 |
|
|
374 |
# repeatable index |
# repeatable index |
375 |
my $r = $i; |
my $r = $i; |
376 |
$r = 0 if (lc("$2") eq 's'); |
if ($fld_type eq 's') { |
377 |
|
if ($found_any->{'v'}) { |
378 |
|
$r = 0; |
379 |
|
} else { |
380 |
|
return; |
381 |
|
} |
382 |
|
} |
383 |
|
|
384 |
my $found = 0; |
my $found = 0; |
385 |
my $tmp = $self->get_data(\$rec,$3,$4,$r,\$found); |
my $tmp = $self->get_data(\$rec,$3,$4,$r,\$found,$rec_size); |
386 |
|
|
387 |
if ($found) { |
if ($found) { |
388 |
push @out, $del; |
$found_any->{$fld_type} += $found; |
389 |
push @out, $tmp; |
|
390 |
$all_found += $found; |
# we will skip delimiter before first occurence of field! |
391 |
|
push @out, $del unless($found_any->{$fld_type} == 1); |
392 |
|
push @out, $tmp if ($tmp); |
393 |
} |
} |
394 |
|
$f_step++; |
395 |
} |
} |
396 |
|
|
397 |
return if (! $all_found); |
# test if any fields found? |
398 |
|
return if (! $found_any->{'v'} && ! $found_any->{'s'}); |
399 |
|
|
400 |
my $out = join('',@out); |
my $out = join('',@out); |
401 |
|
|
415 |
return if (! $self->_eval($eval)); |
return if (! $self->_eval($eval)); |
416 |
} |
} |
417 |
|
|
418 |
if ($filter_name && $self->{'filter'}->{$filter_name}) { |
if ($filter_name) { |
419 |
$log->debug("about to filter{$filter_name} format: $out"); |
my @filter_args; |
420 |
$out = $self->{'filter'}->{$filter_name}->($out); |
if ($filter_name =~ s/(\w+)\((.*)\)/$1/) { |
421 |
return unless(defined($out)); |
@filter_args = split(/,/, $2); |
422 |
$log->debug("filter result: $out"); |
} |
423 |
|
if ($self->{'filter'}->{$filter_name}) { |
424 |
|
$log->debug("about to filter{$filter_name} format: $out with arguments: ", join(",", @filter_args)); |
425 |
|
unshift @filter_args, $out; |
426 |
|
$out = $self->{'filter'}->{$filter_name}->(@filter_args); |
427 |
|
return unless(defined($out)); |
428 |
|
$log->debug("filter result: $out"); |
429 |
|
} elsif (! $warn_once->{$filter_name}) { |
430 |
|
$log->warn("trying to use undefined filter $filter_name"); |
431 |
|
$warn_once->{$filter_name}++; |
432 |
|
} |
433 |
} |
} |
434 |
|
|
435 |
return $out; |
return $out; |
436 |
} |
} |
437 |
|
|
|
=head2 parse_to_arr |
|
|
|
|
|
Similar to C<parse>, but returns array of all repeatable fields |
|
|
|
|
|
my @arr = $webpac->parse_to_arr($rec,'v250^a'); |
|
|
|
|
|
=cut |
|
|
|
|
|
sub parse_to_arr { |
|
|
my $self = shift; |
|
|
|
|
|
my ($rec, $format_utf8) = @_; |
|
|
|
|
|
my $log = $self->_get_logger(); |
|
|
|
|
|
$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o); |
|
|
return if (! $format_utf8); |
|
|
|
|
|
my $i = 0; |
|
|
my @arr; |
|
|
|
|
|
while (my $v = $self->parse($rec,$format_utf8,$i++)) { |
|
|
push @arr, $v; |
|
|
} |
|
|
|
|
|
$log->debug("format '$format_utf8' returned ",--$i," elements: ", sub { join(" | ",@arr) }) if (@arr); |
|
|
|
|
|
return @arr; |
|
|
} |
|
|
|
|
|
|
|
438 |
=head2 fill_in |
=head2 fill_in |
439 |
|
|
440 |
Workhourse of all: takes record from in-memory structure of database and |
Workhourse of all: takes record from in-memory structure of database and |
456 |
This method will automatically decode UTF-8 string to local code page |
This method will automatically decode UTF-8 string to local code page |
457 |
if needed. |
if needed. |
458 |
|
|
459 |
|
There is optional parametar C<$record_size> which can be used to get sizes of |
460 |
|
all C<field^subfield> combinations in this format. |
461 |
|
|
462 |
|
my $text = $webpac->fill_in($rec,'got: v900^a v900^x',0,\$rec_size); |
463 |
|
|
464 |
=cut |
=cut |
465 |
|
|
466 |
sub fill_in { |
sub fill_in { |
468 |
|
|
469 |
my $log = $self->_get_logger(); |
my $log = $self->_get_logger(); |
470 |
|
|
471 |
my $rec = shift || $log->logconfess("need data record"); |
my ($rec,$format,$i,$rec_size) = @_; |
472 |
my $format = shift || $log->logconfess("need format to parse"); |
|
473 |
|
$log->logconfess("need data record") unless ($rec); |
474 |
|
$log->logconfess("need format to parse") unless($format); |
475 |
|
|
476 |
# iteration (for repeatable fields) |
# iteration (for repeatable fields) |
477 |
my $i = shift || 0; |
$i ||= 0; |
478 |
|
|
479 |
$log->logdie("infitite loop in format $format") if ($i > ($self->{'max_mfn'} || 9999)); |
$log->logdie("infitite loop in format $format") if ($i > ($self->{'max_mfn'} || 9999)); |
480 |
|
|
486 |
} |
} |
487 |
|
|
488 |
my $found = 0; |
my $found = 0; |
489 |
|
my $just_single = 1; |
490 |
|
|
491 |
my $eval_code; |
my $eval_code; |
492 |
# remove eval{...} from beginning |
# remove eval{...} from beginning |
496 |
# remove filter{...} from beginning |
# remove filter{...} from beginning |
497 |
$filter_name = $1 if ($format =~ s/^filter{([^}]+)}//s); |
$filter_name = $1 if ($format =~ s/^filter{([^}]+)}//s); |
498 |
|
|
499 |
# do actual replacement of placeholders |
{ |
500 |
# repeatable fields |
# fix warnings |
501 |
$format =~ s/v(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,$i,\$found)/ges; |
no warnings 'uninitialized'; |
502 |
# non-repeatable fields |
|
503 |
$format =~ s/s(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,0,\$found)/ges; |
# do actual replacement of placeholders |
504 |
|
# repeatable fields |
505 |
|
if ($format =~ s/v(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,$i,\$found,$rec_size)/ges) { |
506 |
|
$just_single = 0; |
507 |
|
} |
508 |
|
|
509 |
|
# non-repeatable fields |
510 |
|
if ($format =~ s/s(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,0,\$found,$rec_size)/ges) { |
511 |
|
return if ($i > 0 && $just_single); |
512 |
|
} |
513 |
|
} |
514 |
|
|
515 |
if ($found) { |
if ($found) { |
516 |
$log->debug("format: $format"); |
$log->debug("format: $format"); |
526 |
} |
} |
527 |
# do we have lookups? |
# do we have lookups? |
528 |
if ($self->{'lookup'}) { |
if ($self->{'lookup'}) { |
529 |
return $self->lookup($format); |
if ($self->{'lookup'}->can('lookup')) { |
530 |
|
my @lookup = $self->{lookup}->lookup($format); |
531 |
|
$log->debug("lookup $format", join(", ", @lookup)); |
532 |
|
return @lookup; |
533 |
|
} else { |
534 |
|
$log->warn("Have lookup object but can't invoke lookup method"); |
535 |
|
} |
536 |
} else { |
} else { |
537 |
return $format; |
return $format; |
538 |
} |
} |
542 |
} |
} |
543 |
|
|
544 |
|
|
545 |
=head2 fill_in_to_arr |
=head2 _rec_to_arr |
546 |
|
|
547 |
Similar to C<fill_in>, but returns array of all repeatable fields. Usable |
Similar to C<parse> and C<fill_in>, but returns array of all repeatable fields. Usable |
548 |
for fields which have lookups, so they shouldn't be parsed but rather |
for fields which have lookups, so they shouldn't be parsed but rather |
549 |
C<fill_id>ed. |
C<paste>d or C<fill_id>ed. Last argument is name of operation: C<paste> or C<fill_in>. |
550 |
|
|
551 |
my @arr = $webpac->fill_in_to_arr($rec,'[v900];;[v250^a]'); |
my @arr = $webpac->fill_in_to_arr($rec,'[v900];;[v250^a]','paste'); |
552 |
|
|
553 |
=cut |
=cut |
554 |
|
|
555 |
sub fill_in_to_arr { |
sub _rec_to_arr { |
556 |
my $self = shift; |
my $self = shift; |
557 |
|
|
558 |
my ($rec, $format_utf8) = @_; |
my ($rec, $format_utf8, $code) = @_; |
559 |
|
|
560 |
my $log = $self->_get_logger(); |
my $log = $self->_get_logger(); |
561 |
|
|
562 |
$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o); |
$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o); |
563 |
return if (! $format_utf8); |
return if (! $format_utf8); |
564 |
|
|
565 |
|
$log->debug("using $code on $format_utf8"); |
566 |
|
|
567 |
my $i = 0; |
my $i = 0; |
568 |
|
my $max = 0; |
569 |
my @arr; |
my @arr; |
570 |
|
my $rec_size = {}; |
571 |
|
|
572 |
while (my @v = $self->fill_in($rec,$format_utf8,$i++)) { |
while ($i <= $max) { |
573 |
push @arr, @v; |
my @v = $self->$code($rec,$format_utf8,$i++,\$rec_size); |
574 |
|
if ($rec_size) { |
575 |
|
foreach my $f (keys %{ $rec_size }) { |
576 |
|
$max = $rec_size->{$f} if ($rec_size->{$f} > $max); |
577 |
|
} |
578 |
|
$log->debug("max set to $max"); |
579 |
|
undef $rec_size; |
580 |
|
} |
581 |
|
if (@v) { |
582 |
|
push @arr, @v; |
583 |
|
} else { |
584 |
|
push @arr, '' if ($max > $i); |
585 |
|
} |
586 |
} |
} |
587 |
|
|
588 |
$log->debug("format '$format_utf8' returned ",--$i," elements: ", sub { join(" | ",@arr) }) if (@arr); |
$log->debug("format '$format_utf8' returned ",--$i," elements: ", sub { join(" | ",@arr) }) if (@arr); |
595 |
|
|
596 |
Returns value from record. |
Returns value from record. |
597 |
|
|
598 |
my $text = $self->get_data(\$rec,$f,$sf,$i,\$found); |
my $text = $self->get_data(\$rec,$f,$sf,$i,\$found,\$rec_size); |
599 |
|
|
600 |
|
Required arguments are: |
601 |
|
|
602 |
|
=over 8 |
603 |
|
|
604 |
|
=item C<$rec> |
605 |
|
|
606 |
|
record reference |
607 |
|
|
608 |
|
=item C<$f> |
609 |
|
|
610 |
|
field |
611 |
|
|
612 |
|
=item C<$sf> |
613 |
|
|
614 |
Arguments are: |
optional subfield |
|
record reference C<$rec>, |
|
|
field C<$f>, |
|
|
optional subfiled C<$sf>, |
|
|
index for repeatable values C<$i>. |
|
615 |
|
|
616 |
Optinal variable C<$found> will be incremeted if there |
=item C<$i> |
|
is field. |
|
617 |
|
|
618 |
Returns value or empty string. |
index offset for repeatable values ( 0 ... $rec_size->{'400^a'} ) |
619 |
|
|
620 |
|
=item C<$found> |
621 |
|
|
622 |
|
optional variable that will be incremeted if preset |
623 |
|
|
624 |
|
=item C<$rec_size> |
625 |
|
|
626 |
|
hash to hold maximum occurances of C<field^subfield> combinations |
627 |
|
(which can be accessed using keys in same format) |
628 |
|
|
629 |
|
=back |
630 |
|
|
631 |
|
Returns value or empty string, updates C<$found> and C<rec_size> |
632 |
|
if present. |
633 |
|
|
634 |
=cut |
=cut |
635 |
|
|
636 |
sub get_data { |
sub get_data { |
637 |
my $self = shift; |
my $self = shift; |
638 |
|
|
639 |
my ($rec,$f,$sf,$i,$found) = @_; |
my ($rec,$f,$sf,$i,$found,$cache) = @_; |
640 |
|
|
641 |
if ($$rec->{$f}) { |
return '' unless ($$rec->{$f} && ref($$rec->{$f}) eq 'ARRAY'); |
642 |
return '' if (! $$rec->{$f}->[$i]); |
|
643 |
|
if (defined($$cache)) { |
644 |
|
$$cache->{ $f . ( $sf ? '^' . $sf : '' ) } ||= scalar @{ $$rec->{$f} }; |
645 |
|
} |
646 |
|
|
647 |
|
return '' unless ($$rec->{$f}->[$i]); |
648 |
|
|
649 |
|
{ |
650 |
no strict 'refs'; |
no strict 'refs'; |
651 |
if ($sf && $$rec->{$f}->[$i]->{$sf}) { |
if (defined($sf)) { |
652 |
$$found++ if (defined($$found)); |
$$found++ if (defined($$found) && $$rec->{$f}->[$i]->{$sf}); |
653 |
return $$rec->{$f}->[$i]->{$sf}; |
return $$rec->{$f}->[$i]->{$sf}; |
654 |
} elsif ($$rec->{$f}->[$i]) { |
} else { |
655 |
$$found++ if (defined($$found)); |
$$found++ if (defined($$found)); |
656 |
# it still might have subfield, just |
# it still might have subfields, just |
657 |
# not specified, so we'll dump all |
# not specified, so we'll dump some debug info |
658 |
if ($$rec->{$f}->[$i] =~ /HASH/o) { |
if ($$rec->{$f}->[$i] =~ /HASH/o) { |
659 |
my $out; |
my $out; |
660 |
foreach my $k (keys %{$$rec->{$f}->[$i]}) { |
foreach my $k (keys %{$$rec->{$f}->[$i]}) { |
661 |
$out .= $$rec->{$f}->[$i]->{$k}." "; |
my $v = $$rec->{$f}->[$i]->{$k}; |
662 |
|
$out .= '$' . $k .':' . $v if ($v); |
663 |
} |
} |
664 |
return $out; |
return $out; |
665 |
} else { |
} else { |
666 |
return $$rec->{$f}->[$i]; |
return $$rec->{$f}->[$i]; |
667 |
} |
} |
668 |
} |
} |
|
} else { |
|
|
return ''; |
|
669 |
} |
} |
670 |
} |
} |
671 |
|
|
703 |
$log->debug("using format $name [$format] on $data to produce: $out"); |
$log->debug("using format $name [$format] on $data to produce: $out"); |
704 |
|
|
705 |
if ($self->{'lookup_regex'} && $out =~ $self->{'lookup_regex'}) { |
if ($self->{'lookup_regex'} && $out =~ $self->{'lookup_regex'}) { |
706 |
return $self->lookup($out); |
return $self->{'lookup'}->lookup($out); |
707 |
} else { |
} else { |
708 |
return $out; |
return $out; |
709 |
} |
} |
786 |
|
|
787 |
=cut |
=cut |
788 |
|
|
789 |
1; # End of WebPAC::DB |
1; # End of WebPAC::Normalize |