5 |
_get_ds _clean_ds |
_get_ds _clean_ds |
6 |
|
|
7 |
tag search display |
tag search display |
8 |
|
marc marc_indicators marc_repeatable_subfield |
9 |
|
|
10 |
rec1 rec2 rec |
rec1 rec2 rec |
11 |
regex prefix suffix surround |
regex prefix suffix surround |
12 |
first lookup join_with |
first lookup join_with |
17 |
|
|
18 |
#use base qw/WebPAC::Common/; |
#use base qw/WebPAC::Common/; |
19 |
use Data::Dumper; |
use Data::Dumper; |
20 |
|
use Encode qw/from_to/; |
21 |
|
|
22 |
=head1 NAME |
=head1 NAME |
23 |
|
|
25 |
|
|
26 |
=head1 VERSION |
=head1 VERSION |
27 |
|
|
28 |
Version 0.05 |
Version 0.06 |
29 |
|
|
30 |
=cut |
=cut |
31 |
|
|
32 |
our $VERSION = '0.05'; |
our $VERSION = '0.06'; |
33 |
|
|
34 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
35 |
|
|
42 |
C<perl -c normalize.pl>. |
C<perl -c normalize.pl>. |
43 |
|
|
44 |
Normalisation can generate multiple output normalized data. For now, supported output |
Normalisation can generate multiple output normalized data. For now, supported output |
45 |
types (on the left side of definition) are: C<tag>, C<display> and C<search>. |
types (on the left side of definition) are: C<tag>, C<display>, C<search> and |
46 |
|
C<marc>. |
47 |
|
|
48 |
=head1 FUNCTIONS |
=head1 FUNCTIONS |
49 |
|
|
58 |
lookup => $lookup->lookup_hash, |
lookup => $lookup->lookup_hash, |
59 |
row => $row, |
row => $row, |
60 |
rules => $normalize_pl_config, |
rules => $normalize_pl_config, |
61 |
|
marc_encoding => 'utf-8', |
62 |
); |
); |
63 |
|
|
64 |
|
Options C<lookup>, C<row>, C<rules> and C<log> are mandatory while all |
65 |
|
other are optional. |
66 |
|
|
67 |
This function will B<die> if normalizastion can't be evaled. |
This function will B<die> if normalizastion can't be evaled. |
68 |
|
|
69 |
Since this function isn't exported you have to call it with |
Since this function isn't exported you have to call it with |
80 |
no strict 'subs'; |
no strict 'subs'; |
81 |
_set_lookup( $arg->{lookup} ); |
_set_lookup( $arg->{lookup} ); |
82 |
_set_rec( $arg->{row} ); |
_set_rec( $arg->{row} ); |
83 |
_clean_ds(); |
_clean_ds( %{ $arg } ); |
84 |
eval "$arg->{rules}"; |
eval "$arg->{rules}"; |
85 |
die "error evaling $arg->{rules}: $@\n" if ($@); |
die "error evaling $arg->{rules}: $@\n" if ($@); |
86 |
|
|
87 |
return _get_ds(); |
return _get_ds(); |
88 |
} |
} |
89 |
|
|
109 |
|
|
110 |
=cut |
=cut |
111 |
|
|
112 |
my $out; |
my ($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators); |
113 |
|
|
114 |
sub _get_ds { |
sub _get_ds { |
115 |
return $out; |
return $out; |
124 |
=cut |
=cut |
125 |
|
|
126 |
sub _clean_ds { |
sub _clean_ds { |
127 |
$out = undef; |
my $a = {@_}; |
128 |
|
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators) = (undef); |
129 |
|
$marc_encoding = $a->{marc_encoding}; |
130 |
} |
} |
131 |
|
|
132 |
=head2 _set_lookup |
=head2 _set_lookup |
143 |
$lookup = shift; |
$lookup = shift; |
144 |
} |
} |
145 |
|
|
146 |
|
=head2 _get_marc_fields |
147 |
|
|
148 |
|
Get all fields defined by calls to C<marc> |
149 |
|
|
150 |
|
$marc->add_fields( WebPAC::Normalize:_get_marc_fields() ); |
151 |
|
|
152 |
|
|
153 |
|
|
154 |
|
We are using I<magic> which detect repeatable fields only from |
155 |
|
sequence of field/subfield data generated by normalization. |
156 |
|
|
157 |
|
Repeatable field is created if there is second occurence of same subfield or |
158 |
|
if any of indicators are different. This is sane for most cases except for |
159 |
|
non-repeatable fields with repeatable subfields. |
160 |
|
|
161 |
|
You can change behaviour of that using C<marc_repeatable_subfield>. |
162 |
|
|
163 |
|
=cut |
164 |
|
|
165 |
|
sub _get_marc_fields { |
166 |
|
my @m; |
167 |
|
my $last; |
168 |
|
foreach my $row (@{ $marc_record }) { |
169 |
|
if ($last && |
170 |
|
$last->[0] eq $row->[0] && # check if field is same |
171 |
|
$last->[1] eq $row->[1] && # check for i1 |
172 |
|
$last->[2] eq $row->[2] && # and for i2 |
173 |
|
( $last->[3] ne $row->[3] || # and subfield is different |
174 |
|
$last->[3] eq $row->[3] && # or subfield is same, |
175 |
|
$marc_repeatable_subfield->{ $row->[3] } # but is repeatable |
176 |
|
) |
177 |
|
) { |
178 |
|
push @$last, ( $row->[3] , $row->[4] ); |
179 |
|
next; |
180 |
|
} elsif ($last) { |
181 |
|
push @m, $last; |
182 |
|
} |
183 |
|
|
184 |
|
$last = $row; |
185 |
|
} |
186 |
|
|
187 |
|
push @m, $last if ($last); |
188 |
|
|
189 |
|
return @m; |
190 |
|
} |
191 |
|
|
192 |
|
=head1 Functions to create C<data_structure> |
193 |
|
|
194 |
|
Those functions generally have to first in your normalization file. |
195 |
|
|
196 |
=head2 tag |
=head2 tag |
197 |
|
|
198 |
Define new tag for I<search> and I<display>. |
Define new tag for I<search> and I<display>. |
243 |
$out->{$name}->{search} = \@o; |
$out->{$name}->{search} = \@o; |
244 |
} |
} |
245 |
|
|
246 |
|
=head2 marc |
247 |
|
|
248 |
|
Save value for MARC field |
249 |
|
|
250 |
|
marc('900','a', rec('200','a') ); |
251 |
|
|
252 |
|
=cut |
253 |
|
|
254 |
|
sub marc { |
255 |
|
my $f = shift or die "marc needs field"; |
256 |
|
die "marc field must be numer" unless ($f =~ /^\d+$/); |
257 |
|
|
258 |
|
my $sf = shift or die "marc needs subfield"; |
259 |
|
|
260 |
|
foreach (@_) { |
261 |
|
my $v = $_; # make var read-write for Encode |
262 |
|
next unless (defined($v) && $v !~ /^\s*$/); |
263 |
|
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
264 |
|
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
265 |
|
push @{ $marc_record }, [ $f, $i1, $i2, $sf => $v ]; |
266 |
|
} |
267 |
|
} |
268 |
|
|
269 |
|
=head2 marc_repeatable_subfield |
270 |
|
|
271 |
|
Save values for MARC repetable subfield |
272 |
|
|
273 |
|
marc_repeatable_subfield('910', 'z', rec('909') ); |
274 |
|
|
275 |
|
=cut |
276 |
|
|
277 |
|
sub marc_repeatable_subfield { |
278 |
|
die "marc_repeatable_subfield need subfield!\n" unless (defined($_[1])); |
279 |
|
$marc_repeatable_subfield->{ $_[1] }++; |
280 |
|
marc(@_); |
281 |
|
} |
282 |
|
|
283 |
|
=head2 marc_indicators |
284 |
|
|
285 |
|
Set both indicators for MARC field |
286 |
|
|
287 |
|
marc_indicators('900', ' ', 1); |
288 |
|
|
289 |
|
Any indicator value other than C<0-9> will be treated as undefined. |
290 |
|
|
291 |
|
=cut |
292 |
|
|
293 |
|
sub marc_indicators { |
294 |
|
my $f = shift || die "marc_indicators need field!\n"; |
295 |
|
my ($i1,$i2) = @_; |
296 |
|
die "marc_indicators($f, ...) need i1!\n" unless(defined($i1)); |
297 |
|
die "marc_indicators($f, $i1, ...) need i2!\n" unless(defined($i2)); |
298 |
|
|
299 |
|
$i1 = ' ' if ($i1 !~ /^\d$/); |
300 |
|
$i2 = ' ' if ($i2 !~ /^\d$/); |
301 |
|
@{ $marc_indicators->{$f} } = ($i1,$i2); |
302 |
|
} |
303 |
|
|
304 |
|
|
305 |
|
=head1 Functions to extract data from input |
306 |
|
|
307 |
|
This function should be used inside functions to create C<data_structure> described |
308 |
|
above. |
309 |
|
|
310 |
=head2 rec1 |
=head2 rec1 |
311 |
|
|
312 |
Return all values in some field |
Return all values in some field |