2 |
|
|
3 |
use warnings; |
use warnings; |
4 |
use strict; |
use strict; |
5 |
|
use base 'WebPAC::Common'; |
6 |
use Data::Dumper; |
use Data::Dumper; |
7 |
|
|
8 |
=head1 NAME |
=head1 NAME |
11 |
|
|
12 |
=head1 VERSION |
=head1 VERSION |
13 |
|
|
14 |
Version 0.01 |
Version 0.07 |
15 |
|
|
16 |
=cut |
=cut |
17 |
|
|
18 |
our $VERSION = '0.01'; |
our $VERSION = '0.07'; |
19 |
|
|
20 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
21 |
|
|
47 |
code defined as code ref on format after field substitution to producing |
code defined as code ref on format after field substitution to producing |
48 |
output |
output |
49 |
|
|
50 |
|
There is one built-in filter called C<regex> which can be use like this: |
51 |
|
|
52 |
|
filter{regex(s/foo/bar/)} |
53 |
|
|
54 |
=item * |
=item * |
55 |
|
|
56 |
optional C<lookup{...}> will be then performed. See C<WebPAC::Lookups>. |
optional C<lookup{...}> will be then performed. See C<WebPAC::Lookups>. |
83 |
return length($_); |
return length($_); |
84 |
}, ... |
}, ... |
85 |
}, |
}, |
86 |
db => $webpac_db_obj, |
db => $db_obj, |
87 |
lookup_regex => $lookup->regex, |
lookup_regex => $lookup->regex, |
88 |
|
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. |
in structures. If you pass this parametar, you must also pass C<lookup> |
100 |
|
which is C<WebPAC::Lookup> object. |
101 |
|
|
102 |
=cut |
=cut |
103 |
|
|
106 |
my $self = {@_}; |
my $self = {@_}; |
107 |
bless($self, $class); |
bless($self, $class); |
108 |
|
|
109 |
|
my $r = $self->{'lookup_regex'} ? 1 : 0; |
110 |
|
my $l = $self->{'lookup'} ? 1 : 0; |
111 |
|
|
112 |
|
my $log = $self->_get_logger(); |
113 |
|
|
114 |
|
# those two must be in pair |
115 |
|
if ( ($r & $l) != ($r || $l) ) { |
116 |
|
my $log = $self->_get_logger(); |
117 |
|
$log->logdie("lookup_regex and lookup must be in pair"); |
118 |
|
} |
119 |
|
|
120 |
|
$log->logdie("lookup must be WebPAC::Lookup object") if ($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 = $self->{'db'}->get_ds($rec); |
my $ds = $self->{'db'}->load_ds( id => $id, prefix => $self->{prefix} ); |
168 |
return @ds if (@ds); |
$log->debug("load_ds( rec = ", sub { Dumper($rec) }, ") = ", sub { Dumper($ds) }); |
169 |
|
return $ds if ($ds); |
170 |
|
$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 |
|
|
189 |
#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'}}); |
190 |
|
|
191 |
foreach my $tag (@{$self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}}}) { |
foreach my $tag (@{$self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}}}) { |
192 |
my $format = $tag->{'value'} || $tag->{'content'}; |
my $format; |
193 |
|
|
194 |
|
$log->logdie("expected tag HASH and got $tag") unless (ref($tag) eq 'HASH'); |
195 |
|
$format = $tag->{'value'} || $tag->{'content'}; |
196 |
|
|
197 |
$log->debug("format: $format"); |
$log->debug("format: $format"); |
198 |
|
|
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 |
} |
} |
266 |
|
|
267 |
} |
} |
268 |
|
|
269 |
$self->{'db'}->put_gs( |
$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 |
return @ds; |
$log->debug("ds: ", sub { Dumper($ds) }); |
276 |
|
|
277 |
|
$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 |
|
|
369 |
return if (! $self->_eval($eval)); |
return if (! $self->_eval($eval)); |
370 |
} |
} |
371 |
|
|
372 |
if ($filter_name && $self->{'filter'}->{$filter_name}) { |
if ($filter_name) { |
373 |
$log->debug("about to filter{$filter_name} format: $out"); |
my @filter_args; |
374 |
$out = $self->{'filter'}->{$filter_name}->($out); |
if ($filter_name =~ s/(\w+)\((.*)\)/$1/) { |
375 |
return unless(defined($out)); |
@filter_args = split(/,/, $2); |
376 |
$log->debug("filter result: $out"); |
} |
377 |
|
if ($self->{'filter'}->{$filter_name}) { |
378 |
|
$log->debug("about to filter{$filter_name} format: $out with arguments: ", join(",", @filter_args)); |
379 |
|
unshift @filter_args, $out; |
380 |
|
$out = $self->{'filter'}->{$filter_name}->(@filter_args); |
381 |
|
return unless(defined($out)); |
382 |
|
$log->debug("filter result: $out"); |
383 |
|
} elsif (! $warn_once->{$filter_name}) { |
384 |
|
$log->warn("trying to use undefined filter $filter_name"); |
385 |
|
$warn_once->{$filter_name}++; |
386 |
|
} |
387 |
} |
} |
388 |
|
|
389 |
return $out; |
return $out; |
492 |
} |
} |
493 |
# do we have lookups? |
# do we have lookups? |
494 |
if ($self->{'lookup'}) { |
if ($self->{'lookup'}) { |
495 |
return $self->lookup($format); |
if ($self->{'lookup'}->can('lookup')) { |
496 |
|
my @lookup = $self->{lookup}->lookup($format); |
497 |
|
$log->debug("lookup $format", join(", ", @lookup)); |
498 |
|
return @lookup; |
499 |
|
} else { |
500 |
|
$log->warn("Have lookup object but can't invoke lookup method"); |
501 |
|
} |
502 |
} else { |
} else { |
503 |
return $format; |
return $format; |
504 |
} |
} |
571 |
if ($sf && $$rec->{$f}->[$i]->{$sf}) { |
if ($sf && $$rec->{$f}->[$i]->{$sf}) { |
572 |
$$found++ if (defined($$found)); |
$$found++ if (defined($$found)); |
573 |
return $$rec->{$f}->[$i]->{$sf}; |
return $$rec->{$f}->[$i]->{$sf}; |
574 |
} elsif ($$rec->{$f}->[$i]) { |
} elsif (! $sf && $$rec->{$f}->[$i]) { |
575 |
$$found++ if (defined($$found)); |
$$found++ if (defined($$found)); |
576 |
# it still might have subfield, just |
# it still might have subfield, just |
577 |
# not specified, so we'll dump all |
# not specified, so we'll dump all |
584 |
} else { |
} else { |
585 |
return $$rec->{$f}->[$i]; |
return $$rec->{$f}->[$i]; |
586 |
} |
} |
587 |
|
} else { |
588 |
|
return ''; |
589 |
} |
} |
590 |
} else { |
} else { |
591 |
return ''; |
return ''; |
626 |
$log->debug("using format $name [$format] on $data to produce: $out"); |
$log->debug("using format $name [$format] on $data to produce: $out"); |
627 |
|
|
628 |
if ($self->{'lookup_regex'} && $out =~ $self->{'lookup_regex'}) { |
if ($self->{'lookup_regex'} && $out =~ $self->{'lookup_regex'}) { |
629 |
return $self->lookup($out); |
return $self->{'lookup'}->lookup($out); |
630 |
} else { |
} else { |
631 |
return $out; |
return $out; |
632 |
} |
} |
709 |
|
|
710 |
=cut |
=cut |
711 |
|
|
712 |
1; # End of WebPAC::DB |
1; # End of WebPAC::Normalize |