11 |
|
|
12 |
=head1 VERSION |
=head1 VERSION |
13 |
|
|
14 |
Version 0.01 |
Version 0.08 |
15 |
|
|
16 |
=cut |
=cut |
17 |
|
|
18 |
our $VERSION = '0.01'; |
our $VERSION = '0.08'; |
19 |
|
|
20 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
21 |
|
|
47 |
code defined as code ref on format after field substitution to producing |
code defined as code ref on format after field substitution to producing |
48 |
output |
output |
49 |
|
|
50 |
|
There is one built-in filter called C<regex> which can be use like this: |
51 |
|
|
52 |
|
filter{regex(s/foo/bar/)} |
53 |
|
|
54 |
=item * |
=item * |
55 |
|
|
56 |
optional C<lookup{...}> will be then performed. See C<WebPAC::Lookups>. |
optional C<lookup{...}> will be then performed. See C<WebPAC::Lookups>. |
86 |
db => $db_obj, |
db => $db_obj, |
87 |
lookup_regex => $lookup->regex, |
lookup_regex => $lookup->regex, |
88 |
lookup => $lookup_obj, |
lookup => $lookup_obj, |
89 |
|
prefix => 'foobar', |
90 |
); |
); |
91 |
|
|
92 |
Parametar C<filter> defines user supplied snippets of perl code which can |
Parametar C<filter> defines user supplied snippets of perl code which can |
93 |
be use with C<filter{...}> notation. |
be use with C<filter{...}> notation. |
94 |
|
|
95 |
|
C<prefix> is used to form filename for database record (to support multiple |
96 |
|
source files which are joined in one database). |
97 |
|
|
98 |
Recommended parametar C<lookup_regex> is used to enable parsing of lookups |
Recommended parametar C<lookup_regex> is used to enable parsing of lookups |
99 |
in structures. If you pass this parametar, you must also pass C<lookup> |
in structures. If you pass this parametar, you must also pass C<lookup> |
100 |
which is C<WebPAC::Lookup> object. |
which is C<WebPAC::Lookup> object. |
119 |
|
|
120 |
$log->logdie("lookup must be WebPAC::Lookup object") if ($self->{'lookup'} && ! $self->{'lookup'}->isa('WebPAC::Lookup')); |
$log->logdie("lookup must be WebPAC::Lookup object") if ($self->{'lookup'} && ! $self->{'lookup'}->isa('WebPAC::Lookup')); |
121 |
|
|
122 |
|
$log->warn("no prefix defined. please check that!") unless ($self->{'prefix'}); |
123 |
|
|
124 |
|
$log->debug("using lookup regex: ", $self->{lookup_regex}) if ($r && $l); |
125 |
|
|
126 |
|
if (! $self->{filter} || ! $self->{filter}->{regex}) { |
127 |
|
$log->debug("adding built-in filter regex"); |
128 |
|
$self->{filter}->{regex} = sub { |
129 |
|
my ($val, $regex) = @_; |
130 |
|
eval "\$val =~ $regex"; |
131 |
|
return $val; |
132 |
|
}; |
133 |
|
} |
134 |
|
|
135 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
136 |
} |
} |
137 |
|
|
143 |
|
|
144 |
This structures are used to produce output. |
This structures are used to produce output. |
145 |
|
|
146 |
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. |
|
147 |
|
|
148 |
=cut |
=cut |
149 |
|
|
155 |
my $rec = shift; |
my $rec = shift; |
156 |
$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o); |
$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o); |
157 |
|
|
158 |
|
$log->debug("data_structure rec = ", sub { Dumper($rec) }); |
159 |
|
|
160 |
|
$log->logdie("need unique ID (mfn) in field 000 of record " . Dumper($rec) ) unless (defined($rec->{'000'})); |
161 |
|
|
162 |
|
my $id = $rec->{'000'}->[0] || $log->logdie("field 000 isn't array!"); |
163 |
|
|
164 |
my $cache_file; |
my $cache_file; |
165 |
|
|
166 |
if ($self->{'db'}) { |
if ($self->{'db'}) { |
167 |
my @ds = $self->{'db'}->load_ds($rec); |
my $ds = $self->{'db'}->load_ds( id => $id, prefix => $self->{prefix} ); |
168 |
$log->debug("load_ds( rec = ", sub { Dumper($rec) }, ") = ", sub { Dumper(@ds) }); |
$log->debug("load_ds( rec = ", sub { Dumper($rec) }, ") = ", sub { Dumper($ds) }); |
169 |
return @ds if ($#ds > 0); |
return $ds if ($ds); |
170 |
$log->debug("cache miss, creating"); |
$log->debug("cache miss, creating"); |
171 |
} |
} |
172 |
|
|
|
undef $self->{'currnet_filename'}; |
|
|
undef $self->{'headline'}; |
|
|
|
|
173 |
my @sorted_tags; |
my @sorted_tags; |
174 |
if ($self->{tags_by_order}) { |
if ($self->{tags_by_order}) { |
175 |
@sorted_tags = @{$self->{tags_by_order}}; |
@sorted_tags = @{$self->{tags_by_order}}; |
178 |
$self->{tags_by_order} = \@sorted_tags; |
$self->{tags_by_order} = \@sorted_tags; |
179 |
} |
} |
180 |
|
|
181 |
my @ds; |
my $ds; |
182 |
|
|
183 |
$log->debug("tags: ",sub { join(", ",@sorted_tags) }); |
$log->debug("tags: ",sub { join(", ",@sorted_tags) }); |
184 |
|
|
213 |
@v = map { $self->apply_format($tag->{'format_name'},$tag->{'format_delimiter'},$_) } @v; |
@v = map { $self->apply_format($tag->{'format_name'},$tag->{'format_delimiter'},$_) } @v; |
214 |
} |
} |
215 |
|
|
|
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! |
|
|
} |
|
|
|
|
216 |
# delimiter will join repeatable fields |
# delimiter will join repeatable fields |
217 |
if ($tag->{'delimiter'}) { |
if ($tag->{'delimiter'}) { |
218 |
@v = ( join($tag->{'delimiter'}, @v) ); |
@v = ( join($tag->{'delimiter'}, @v) ); |
219 |
} |
} |
220 |
|
|
221 |
# default types |
# default types |
222 |
my @types = qw(display swish); |
my @types = qw(display search); |
223 |
# override by type attribute |
# override by type attribute |
224 |
@types = ( $tag->{'type'} ) if ($tag->{'type'}); |
@types = ( $tag->{'type'} ) if ($tag->{'type'}); |
225 |
|
|
226 |
foreach my $type (@types) { |
foreach my $type (@types) { |
227 |
# append to previous line? |
# append to previous line? |
228 |
$log->debug("type: $type ",sub { join(" ",@v) }, $row->{'append'} || 'no append'); |
$log->debug("type: $type ",sub { join(" ",@v) }, " ", $row->{'append'} || 'no append'); |
229 |
if ($tag->{'append'}) { |
if ($tag->{'append'}) { |
230 |
|
|
231 |
# I will delimit appended part with |
# I will delimit appended part with |
252 |
|
|
253 |
# TODO: name_sigular, name_plural |
# TODO: name_sigular, name_plural |
254 |
my $name = $self->{'import_xml'}->{'indexer'}->{$field}->{'name'}; |
my $name = $self->{'import_xml'}->{'indexer'}->{$field}->{'name'}; |
255 |
$row->{'name'} = $name ? $self->_x($name) : $field; |
my $row_name = $name ? $self->_x($name) : $field; |
256 |
|
|
257 |
# post-sort all values in field |
# post-sort all values in field |
258 |
if ($self->{'import_xml'}->{'indexer'}->{$field}->{'sort'}) { |
if ($self->{'import_xml'}->{'indexer'}->{$field}->{'sort'}) { |
259 |
$log->warn("sort at field tag not implemented"); |
$log->warn("sort at field tag not implemented"); |
260 |
} |
} |
261 |
|
|
262 |
push @ds, $row; |
$ds->{$row_name} = $row; |
263 |
|
|
264 |
$log->debug("row $field: ",sub { Dumper($row) }); |
$log->debug("row $field: ",sub { Dumper($row) }); |
265 |
} |
} |
267 |
} |
} |
268 |
|
|
269 |
$self->{'db'}->save_ds( |
$self->{'db'}->save_ds( |
270 |
ds => \@ds, |
id => $id, |
271 |
current_filename => $self->{'current_filename'}, |
ds => $ds, |
272 |
headline => $self->{'headline'}, |
prefix => $self->{prefix}, |
273 |
) if ($self->{'db'}); |
) if ($self->{'db'}); |
274 |
|
|
275 |
$log->debug("ds: ", sub { Dumper(@ds) }); |
$log->debug("ds: ", sub { Dumper($ds) }); |
276 |
|
|
277 |
return @ds; |
$log->logconfess("data structure returned is not array any more!") if wantarray; |
278 |
|
|
279 |
|
return $ds; |
280 |
|
|
281 |
} |
} |
282 |
|
|
288 |
|
|
289 |
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); |
290 |
|
|
291 |
|
Filters are implemented here. While simple form of filters looks like this: |
292 |
|
|
293 |
|
filter{name_of_filter} |
294 |
|
|
295 |
|
but, filters can also have variable number of parametars like this: |
296 |
|
|
297 |
|
filter{name_of_filter(param,param,param)} |
298 |
|
|
299 |
=cut |
=cut |
300 |
|
|
301 |
|
my $warn_once; |
302 |
|
|
303 |
sub parse { |
sub parse { |
304 |
my $self = shift; |
my $self = shift; |
305 |
|
|
317 |
|
|
318 |
my @out; |
my @out; |
319 |
|
|
320 |
$log->debug("format: $format"); |
$log->debug("format: $format [$i]"); |
321 |
|
|
322 |
my $eval_code; |
my $eval_code; |
323 |
# remove eval{...} from beginning |
# remove eval{...} from beginning |
327 |
# remove filter{...} from beginning |
# remove filter{...} from beginning |
328 |
$filter_name = $1 if ($format =~ s/^filter{([^}]+)}//s); |
$filter_name = $1 if ($format =~ s/^filter{([^}]+)}//s); |
329 |
|
|
330 |
|
# did we found any (att all) field from format in row? |
331 |
|
my $found_any; |
332 |
|
# prefix before first field which we preserve it $found_any |
333 |
my $prefix; |
my $prefix; |
334 |
my $all_found=0; |
|
335 |
|
my $f_step = 1; |
336 |
|
|
337 |
while ($format =~ s/^(.*?)(v|s)(\d+)(?:\^(\w))?//s) { |
while ($format =~ s/^(.*?)(v|s)(\d+)(?:\^(\w))?//s) { |
338 |
|
|
339 |
my $del = $1 || ''; |
my $del = $1 || ''; |
340 |
$prefix ||= $del if ($all_found == 0); |
$prefix = $del if ($f_step == 1); |
341 |
|
|
342 |
|
my $fld_type = lc($2); |
343 |
|
|
344 |
# repeatable index |
# repeatable index |
345 |
my $r = $i; |
my $r = $i; |
346 |
$r = 0 if (lc("$2") eq 's'); |
if ($fld_type eq 's') { |
347 |
|
if ($found_any->{'v'}) { |
348 |
|
$r = 0; |
349 |
|
} else { |
350 |
|
return; |
351 |
|
} |
352 |
|
} |
353 |
|
|
354 |
my $found = 0; |
my $found = 0; |
355 |
my $tmp = $self->get_data(\$rec,$3,$4,$r,\$found); |
my $tmp = $self->get_data(\$rec,$3,$4,$r,\$found); |
356 |
|
|
357 |
if ($found) { |
if ($found) { |
358 |
push @out, $del; |
$found_any->{$fld_type} += $found; |
359 |
|
|
360 |
|
# we will skip delimiter before first occurence of field! |
361 |
|
push @out, $del unless($found_any == 1); |
362 |
push @out, $tmp; |
push @out, $tmp; |
|
$all_found += $found; |
|
363 |
} |
} |
364 |
|
$f_step++; |
365 |
} |
} |
366 |
|
|
367 |
return if (! $all_found); |
# test if any fields found? |
368 |
|
return if (! $found_any->{'v'} && ! $found_any->{'s'}); |
369 |
|
|
370 |
my $out = join('',@out); |
my $out = join('',@out); |
371 |
|
|
385 |
return if (! $self->_eval($eval)); |
return if (! $self->_eval($eval)); |
386 |
} |
} |
387 |
|
|
388 |
if ($filter_name && $self->{'filter'}->{$filter_name}) { |
if ($filter_name) { |
389 |
$log->debug("about to filter{$filter_name} format: $out"); |
my @filter_args; |
390 |
$out = $self->{'filter'}->{$filter_name}->($out); |
if ($filter_name =~ s/(\w+)\((.*)\)/$1/) { |
391 |
return unless(defined($out)); |
@filter_args = split(/,/, $2); |
392 |
$log->debug("filter result: $out"); |
} |
393 |
|
if ($self->{'filter'}->{$filter_name}) { |
394 |
|
$log->debug("about to filter{$filter_name} format: $out with arguments: ", join(",", @filter_args)); |
395 |
|
unshift @filter_args, $out; |
396 |
|
$out = $self->{'filter'}->{$filter_name}->(@filter_args); |
397 |
|
return unless(defined($out)); |
398 |
|
$log->debug("filter result: $out"); |
399 |
|
} elsif (! $warn_once->{$filter_name}) { |
400 |
|
$log->warn("trying to use undefined filter $filter_name"); |
401 |
|
$warn_once->{$filter_name}++; |
402 |
|
} |
403 |
} |
} |
404 |
|
|
405 |
return $out; |
return $out; |
479 |
} |
} |
480 |
|
|
481 |
my $found = 0; |
my $found = 0; |
482 |
|
my $just_single = 1; |
483 |
|
|
484 |
my $eval_code; |
my $eval_code; |
485 |
# remove eval{...} from beginning |
# remove eval{...} from beginning |
491 |
|
|
492 |
# do actual replacement of placeholders |
# do actual replacement of placeholders |
493 |
# repeatable fields |
# repeatable fields |
494 |
$format =~ s/v(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,$i,\$found)/ges; |
if ($format =~ s/v(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,$i,\$found)/ges) { |
495 |
|
$just_single = 0; |
496 |
|
} |
497 |
|
|
498 |
# non-repeatable fields |
# non-repeatable fields |
499 |
$format =~ s/s(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,0,\$found)/ges; |
if ($format =~ s/s(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,0,\$found)/ges) { |
500 |
|
return if ($i > 0 && $just_single); |
501 |
|
} |
502 |
|
|
503 |
if ($found) { |
if ($found) { |
504 |
$log->debug("format: $format"); |
$log->debug("format: $format"); |
515 |
# do we have lookups? |
# do we have lookups? |
516 |
if ($self->{'lookup'}) { |
if ($self->{'lookup'}) { |
517 |
if ($self->{'lookup'}->can('lookup')) { |
if ($self->{'lookup'}->can('lookup')) { |
518 |
return $self->{'lookup'}->lookup($format); |
my @lookup = $self->{lookup}->lookup($format); |
519 |
|
$log->debug("lookup $format", join(", ", @lookup)); |
520 |
|
return @lookup; |
521 |
} else { |
} else { |
522 |
$log->warn("Have lookup object but can't invoke lookup method"); |
$log->warn("Have lookup object but can't invoke lookup method"); |
523 |
} |
} |
593 |
if ($sf && $$rec->{$f}->[$i]->{$sf}) { |
if ($sf && $$rec->{$f}->[$i]->{$sf}) { |
594 |
$$found++ if (defined($$found)); |
$$found++ if (defined($$found)); |
595 |
return $$rec->{$f}->[$i]->{$sf}; |
return $$rec->{$f}->[$i]->{$sf}; |
596 |
} elsif ($$rec->{$f}->[$i]) { |
} elsif (! $sf && $$rec->{$f}->[$i]) { |
597 |
$$found++ if (defined($$found)); |
$$found++ if (defined($$found)); |
598 |
# it still might have subfield, just |
# it still might have subfield, just |
599 |
# not specified, so we'll dump all |
# not specified, so we'll dump all |
606 |
} else { |
} else { |
607 |
return $$rec->{$f}->[$i]; |
return $$rec->{$f}->[$i]; |
608 |
} |
} |
609 |
|
} else { |
610 |
|
return ''; |
611 |
} |
} |
612 |
} else { |
} else { |
613 |
return ''; |
return ''; |
731 |
|
|
732 |
=cut |
=cut |
733 |
|
|
734 |
1; # End of WebPAC::DB |
1; # End of WebPAC::Normalize |