1 |
package WebPAC::Normalize; |
package WebPAC::Normalize; |
2 |
use Exporter 'import'; |
use Exporter 'import'; |
3 |
@EXPORT = qw/ |
our @EXPORT = qw/ |
4 |
_set_rec _set_lookup |
_set_ds _set_lookup |
5 |
|
_set_load_row |
6 |
_get_ds _clean_ds |
_get_ds _clean_ds |
7 |
_debug |
_debug |
8 |
|
_pack_subfields_hash |
9 |
|
|
10 |
tag search display |
to |
11 |
marc marc_indicators marc_repeatable_subfield |
search_display search display sorted |
|
marc_compose marc_leader |
|
|
marc_duplicate marc_remove |
|
12 |
|
|
13 |
rec1 rec2 rec |
rec1 rec2 rec |
14 |
|
frec frec_eq frec_ne |
15 |
regex prefix suffix surround |
regex prefix suffix surround |
16 |
first lookup join_with |
first lookup join_with |
17 |
|
save_into_lookup |
18 |
|
|
19 |
split_rec_on |
split_rec_on |
20 |
|
|
21 |
|
get set |
22 |
|
count |
23 |
|
|
24 |
|
row |
25 |
|
rec_array |
26 |
|
|
27 |
/; |
/; |
28 |
|
|
29 |
use warnings; |
use warnings; |
31 |
|
|
32 |
#use base qw/WebPAC::Common/; |
#use base qw/WebPAC::Common/; |
33 |
use Data::Dump qw/dump/; |
use Data::Dump qw/dump/; |
34 |
use Encode qw/from_to/; |
use Carp qw/confess/; |
|
use Storable qw/dclone/; |
|
35 |
|
|
36 |
# debugging warn(s) |
# debugging warn(s) |
37 |
my $debug = 0; |
my $debug = 0; |
38 |
|
_debug( $debug ); |
39 |
|
|
40 |
|
# FIXME |
41 |
|
use WebPAC::Normalize::ISBN; |
42 |
|
push @EXPORT, ( 'isbn_10', 'isbn_13' ); |
43 |
|
|
44 |
=head1 NAME |
use WebPAC::Normalize::MARC; |
45 |
|
push @EXPORT, ( qw/ |
46 |
|
marc marc_indicators marc_repeatable_subfield |
47 |
|
marc_compose marc_leader marc_fixed |
48 |
|
marc_duplicate marc_remove marc_count |
49 |
|
marc_original_order |
50 |
|
marc_template |
51 |
|
/); |
52 |
|
|
53 |
WebPAC::Normalize - describe normalisaton rules using sets |
use Storable qw/dclone/; |
54 |
|
|
55 |
=head1 VERSION |
=head1 NAME |
56 |
|
|
57 |
Version 0.10 |
WebPAC::Normalize - describe normalisaton rules using sets |
58 |
|
|
59 |
=cut |
=cut |
60 |
|
|
61 |
our $VERSION = '0.10'; |
our $VERSION = '0.36'; |
62 |
|
|
63 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
64 |
|
|
71 |
C<perl -c normalize.pl>. |
C<perl -c normalize.pl>. |
72 |
|
|
73 |
Normalisation can generate multiple output normalized data. For now, supported output |
Normalisation can generate multiple output normalized data. For now, supported output |
74 |
types (on the left side of definition) are: C<tag>, C<display>, C<search> and |
types (on the left side of definition) are: C<search_display>, C<display>, C<search> and |
75 |
C<marc>. |
C<marc>. |
76 |
|
|
77 |
=head1 FUNCTIONS |
=head1 FUNCTIONS |
84 |
Return data structure |
Return data structure |
85 |
|
|
86 |
my $ds = WebPAC::Normalize::data_structure( |
my $ds = WebPAC::Normalize::data_structure( |
87 |
lookup => $lookup->lookup_hash, |
lookup => $lookup_hash, |
88 |
row => $row, |
row => $row, |
89 |
rules => $normalize_pl_config, |
rules => $normalize_pl_config, |
90 |
marc_encoding => 'utf-8', |
marc_encoding => 'utf-8', |
91 |
|
config => $config, |
92 |
|
load_row_coderef => sub { |
93 |
|
my ($database,$input,$mfn) = @_; |
94 |
|
$store->load_row( database => $database, input => $input, id => $mfn ); |
95 |
|
}, |
96 |
); |
); |
97 |
|
|
98 |
Options C<lookup>, C<row>, C<rules> and C<log> are mandatory while all |
Options C<row>, C<rules> and C<log> are mandatory while all |
99 |
other are optional. |
other are optional. |
100 |
|
|
101 |
|
C<load_row_coderef> is closure only used when executing lookups, so they will |
102 |
|
die if it's not defined. |
103 |
|
|
104 |
This function will B<die> if normalizastion can't be evaled. |
This function will B<die> if normalizastion can't be evaled. |
105 |
|
|
106 |
Since this function isn't exported you have to call it with |
Since this function isn't exported you have to call it with |
108 |
|
|
109 |
=cut |
=cut |
110 |
|
|
111 |
|
my $load_row_coderef; |
112 |
|
|
113 |
sub data_structure { |
sub data_structure { |
114 |
my $arg = {@_}; |
my $arg = {@_}; |
115 |
|
|
116 |
die "need row argument" unless ($arg->{row}); |
die "need row argument" unless ($arg->{row}); |
117 |
die "need normalisation argument" unless ($arg->{rules}); |
die "need normalisation argument" unless ($arg->{rules}); |
118 |
|
|
119 |
no strict 'subs'; |
_set_lookup( $arg->{lookup} ) if defined($arg->{lookup}); |
120 |
_set_lookup( $arg->{lookup} ); |
_set_ds( $arg->{row} ); |
121 |
_set_rec( $arg->{row} ); |
_set_config( $arg->{config} ) if defined($arg->{config}); |
122 |
_clean_ds( %{ $arg } ); |
_clean_ds( %{ $arg } ); |
123 |
eval "$arg->{rules}"; |
$load_row_coderef = $arg->{load_row_coderef}; |
124 |
|
|
125 |
|
no strict 'subs'; |
126 |
|
no warnings 'redefine'; |
127 |
|
eval "$arg->{rules};"; |
128 |
die "error evaling $arg->{rules}: $@\n" if ($@); |
die "error evaling $arg->{rules}: $@\n" if ($@); |
129 |
|
|
130 |
return _get_ds(); |
return _get_ds(); |
131 |
} |
} |
132 |
|
|
133 |
=head2 _set_rec |
=head2 _set_ds |
134 |
|
|
135 |
Set current record hash |
Set current record hash |
136 |
|
|
137 |
_set_rec( $rec ); |
_set_ds( $rec ); |
138 |
|
|
139 |
=cut |
=cut |
140 |
|
|
141 |
my $rec; |
my $rec; |
142 |
|
|
143 |
sub _set_rec { |
sub _set_ds { |
144 |
$rec = shift or die "no record hash"; |
$rec = shift or die "no record hash"; |
145 |
|
$WebPAC::Normalize::MARC::rec = $rec; |
146 |
|
} |
147 |
|
|
148 |
|
=head2 |
149 |
|
|
150 |
|
my $rec = _get_rec(); |
151 |
|
|
152 |
|
=cut |
153 |
|
|
154 |
|
sub _get_rec { $rec }; |
155 |
|
|
156 |
|
sub rec_array { |
157 |
|
my $d = $rec->{ $_[0] }; |
158 |
|
return @$d if ref($d) eq 'ARRAY'; |
159 |
|
die "field $_[0] not array: ",dump( $d ); |
160 |
|
} |
161 |
|
|
162 |
|
=head2 _set_config |
163 |
|
|
164 |
|
Set current config hash |
165 |
|
|
166 |
|
_set_config( $config ); |
167 |
|
|
168 |
|
Magic keys are: |
169 |
|
|
170 |
|
=over 4 |
171 |
|
|
172 |
|
=item _ |
173 |
|
|
174 |
|
Code of current database |
175 |
|
|
176 |
|
=item _mfn |
177 |
|
|
178 |
|
Current MFN |
179 |
|
|
180 |
|
=back |
181 |
|
|
182 |
|
=cut |
183 |
|
|
184 |
|
my $config; |
185 |
|
|
186 |
|
sub _set_config { |
187 |
|
$config = shift; |
188 |
} |
} |
189 |
|
|
190 |
=head2 _get_ds |
=head2 _get_ds |
195 |
|
|
196 |
=cut |
=cut |
197 |
|
|
198 |
my ($out, $marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators); |
my $out; |
|
my ($marc_record_offset, $marc_fetch_offset) = (0, 0); |
|
199 |
|
|
200 |
sub _get_ds { |
sub _get_ds { |
201 |
|
#warn "## out = ",dump($out); |
202 |
return $out; |
return $out; |
203 |
} |
} |
204 |
|
|
212 |
|
|
213 |
sub _clean_ds { |
sub _clean_ds { |
214 |
my $a = {@_}; |
my $a = {@_}; |
215 |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators) = (); |
$out = undef; |
216 |
($marc_record_offset, $marc_fetch_offset) = (0,0); |
WebPAC::Normalize::MARC::_clean(); |
|
$marc_encoding = $a->{marc_encoding}; |
|
217 |
} |
} |
218 |
|
|
219 |
=head2 _set_lookup |
=head2 _set_lookup |
230 |
$lookup = shift; |
$lookup = shift; |
231 |
} |
} |
232 |
|
|
233 |
=head2 _get_marc_fields |
=head2 _get_lookup |
|
|
|
|
Get all fields defined by calls to C<marc> |
|
|
|
|
|
$marc->add_fields( WebPAC::Normalize:_get_marc_fields() ); |
|
|
|
|
|
We are using I<magic> which detect repeatable fields only from |
|
|
sequence of field/subfield data generated by normalization. |
|
|
|
|
|
Repeatable field is created when there is second occurence of same subfield or |
|
|
if any of indicators are different. |
|
|
|
|
|
This is sane for most cases. Something like: |
|
|
|
|
|
900a-1 900b-1 900c-1 |
|
|
900a-2 900b-2 |
|
|
900a-3 |
|
|
|
|
|
will be created from any combination of: |
|
234 |
|
|
235 |
900a-1 900a-2 900a-3 900b-1 900b-2 900c-1 |
Get current lookup hash |
|
|
|
|
and following rules: |
|
|
|
|
|
marc('900','a', rec('200','a') ); |
|
|
marc('900','b', rec('200','b') ); |
|
|
marc('900','c', rec('200','c') ); |
|
|
|
|
|
which might not be what you have in mind. If you need repeatable subfield, |
|
|
define it using C<marc_repeatable_subfield> like this: |
|
|
|
|
|
marc_repeatable_subfield('900','a'); |
|
|
marc('900','a', rec('200','a') ); |
|
|
marc('900','b', rec('200','b') ); |
|
|
marc('900','c', rec('200','c') ); |
|
|
|
|
|
will create: |
|
|
|
|
|
900a-1 900a-2 900a-3 900b-1 900c-1 |
|
|
900b-2 |
|
|
|
|
|
There is also support for returning next or specific using: |
|
|
|
|
|
while (my $mf = WebPAC::Normalize:_get_marc_fields( fetch_next => 1 ) ) { |
|
|
# do something with $mf |
|
|
} |
|
236 |
|
|
237 |
will always return fields from next MARC record or |
my $lookup = _get_lookup(); |
|
|
|
|
my $mf = WebPAC::Normalize::_get_marc_fields( offset => 42 ); |
|
|
|
|
|
will return 42th copy record (if it exists). |
|
238 |
|
|
239 |
=cut |
=cut |
240 |
|
|
241 |
sub _get_marc_fields { |
sub _get_lookup { |
242 |
|
return $lookup; |
243 |
my $arg = {@_}; |
} |
|
warn "### _get_marc_fields arg: ", dump($arg), $/ if ($debug > 2); |
|
|
my $offset = $marc_fetch_offset; |
|
|
if ($arg->{offset}) { |
|
|
$offset = $arg->{offset}; |
|
|
} elsif($arg->{fetch_next}) { |
|
|
$marc_fetch_offset++; |
|
|
} |
|
|
|
|
|
return if (! $marc_record || ref($marc_record) ne 'ARRAY'); |
|
|
|
|
|
warn "### full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 2); |
|
|
|
|
|
my $marc_rec = $marc_record->[ $offset ]; |
|
|
|
|
|
warn "## _get_marc_fields (at offset: $offset) -- marc_record = ", dump( @$marc_rec ), $/ if ($debug > 1); |
|
|
|
|
|
return if (! $marc_rec || ref($marc_rec) ne 'ARRAY' || $#{ $marc_rec } < 0); |
|
|
|
|
|
# first, sort all existing fields |
|
|
# XXX might not be needed, but modern perl might randomize elements in hash |
|
|
my @sorted_marc_record = sort { |
|
|
$a->[0] . ( $a->[3] || '' ) cmp $b->[0] . ( $b->[3] || '') |
|
|
} @{ $marc_rec }; |
|
|
|
|
|
@sorted_marc_record = @{ $marc_rec }; ### FIXME disable sorting |
|
|
|
|
|
# output marc fields |
|
|
my @m; |
|
|
|
|
|
# count unique field-subfields (used for offset when walking to next subfield) |
|
|
my $u; |
|
|
map { $u->{ $_->[0] . ( $_->[3] || '') }++ } @sorted_marc_record; |
|
|
|
|
|
if ($debug) { |
|
|
warn "## marc_repeatable_subfield = ", dump( $marc_repeatable_subfield ), $/ if ( $marc_repeatable_subfield ); |
|
|
warn "## marc_record[$offset] = ", dump( $marc_rec ), $/; |
|
|
warn "## sorted_marc_record = ", dump( \@sorted_marc_record ), $/; |
|
|
warn "## subfield count = ", dump( $u ), $/; |
|
|
} |
|
|
|
|
|
my $len = $#sorted_marc_record; |
|
|
my $visited; |
|
|
my $i = 0; |
|
|
my $field; |
|
|
|
|
|
foreach ( 0 .. $len ) { |
|
|
|
|
|
# find next element which isn't visited |
|
|
while ($visited->{$i}) { |
|
|
$i = ($i + 1) % ($len + 1); |
|
|
} |
|
|
|
|
|
# mark it visited |
|
|
$visited->{$i}++; |
|
|
|
|
|
my $row = dclone( $sorted_marc_record[$i] ); |
|
|
|
|
|
# field and subfield which is key for |
|
|
# marc_repeatable_subfield and u |
|
|
my $fsf = $row->[0] . ( $row->[3] || '' ); |
|
|
|
|
|
if ($debug > 1) { |
|
|
|
|
|
print "### field so far [", $#$field, "] : ", dump( $field ), " ", $field ? 'T' : 'F', $/; |
|
|
print "### this [$i]: ", dump( $row ),$/; |
|
|
print "### sf: ", $row->[3], " vs ", $field->[3], |
|
|
$marc_repeatable_subfield->{ $row->[0] . $row->[3] } ? ' (repeatable)' : '', $/, |
|
|
if ($#$field >= 0); |
|
244 |
|
|
245 |
} |
=head2 _set_load_row |
246 |
|
|
247 |
# if field exists |
Setup code reference which will return L<data_structure> from |
248 |
if ( $#$field >= 0 ) { |
L<WebPAC::Store> |
|
if ( |
|
|
$row->[0] ne $field->[0] || # field |
|
|
$row->[1] ne $field->[1] || # i1 |
|
|
$row->[2] ne $field->[2] # i2 |
|
|
) { |
|
|
push @m, $field; |
|
|
warn "## saved/1 ", dump( $field ),$/ if ($debug); |
|
|
$field = $row; |
|
|
|
|
|
} elsif ( |
|
|
( $row->[3] lt $field->[-2] ) # subfield which is not next (e.g. a after c) |
|
|
|| |
|
|
( $row->[3] eq $field->[-2] && # same subfield, but not repeatable |
|
|
! $marc_repeatable_subfield->{ $fsf } |
|
|
) |
|
|
) { |
|
|
push @m, $field; |
|
|
warn "## saved/2 ", dump( $field ),$/ if ($debug); |
|
|
$field = $row; |
|
249 |
|
|
250 |
} else { |
_set_load_row(sub { |
251 |
# append new subfields to existing field |
my ($database,$input,$mfn) = @_; |
252 |
push @$field, ( $row->[3], $row->[4] ); |
$store->load_row( database => $database, input => $input, id => $mfn ); |
253 |
} |
}); |
|
} else { |
|
|
# insert first field |
|
|
$field = $row; |
|
|
} |
|
254 |
|
|
255 |
if (! $marc_repeatable_subfield->{ $fsf }) { |
=cut |
|
# make step to next subfield |
|
|
$i = ($i + $u->{ $fsf } ) % ($len + 1); |
|
|
} |
|
|
} |
|
256 |
|
|
257 |
if ($#$field >= 0) { |
sub _set_load_row { |
258 |
push @m, $field; |
my $coderef = shift; |
259 |
warn "## saved/3 ", dump( $field ),$/ if ($debug); |
confess "argument isn't CODE" unless ref($coderef) eq 'CODE'; |
|
} |
|
260 |
|
|
261 |
return @m; |
$load_row_coderef = $coderef; |
262 |
} |
} |
263 |
|
|
264 |
=head2 _debug |
=head2 _debug |
274 |
return $debug unless defined($l); |
return $debug unless defined($l); |
275 |
warn "debug level $l",$/ if ($l > 0); |
warn "debug level $l",$/ if ($l > 0); |
276 |
$debug = $l; |
$debug = $l; |
277 |
|
$WebPAC::Normalize::MARC::debug = $debug; |
278 |
} |
} |
279 |
|
|
280 |
=head1 Functions to create C<data_structure> |
=head1 Functions to create C<data_structure> |
281 |
|
|
282 |
Those functions generally have to first in your normalization file. |
Those functions generally have to first in your normalization file. |
283 |
|
|
284 |
=head2 tag |
=head2 to |
285 |
|
|
286 |
Define new tag for I<search> and I<display>. |
Generic way to set values for some name |
287 |
|
|
288 |
tag('Title', rec('200','a') ); |
to('field-name', 'name-value' => rec('200','a') ); |
289 |
|
|
290 |
|
There are many helpers defined below which might be easier to use. |
291 |
|
|
292 |
=cut |
=cut |
293 |
|
|
294 |
sub tag { |
sub to { |
295 |
my $name = shift or die "tag needs name as first argument"; |
my $type = shift or confess "need type -- BUG?"; |
296 |
|
my $name = shift or confess "needs name as first argument"; |
297 |
my @o = grep { defined($_) && $_ ne '' } @_; |
my @o = grep { defined($_) && $_ ne '' } @_; |
298 |
return unless (@o); |
return unless (@o); |
299 |
$out->{$name}->{tag} = $name; |
$out->{$name}->{$type} = \@o; |
|
$out->{$name}->{search} = \@o; |
|
|
$out->{$name}->{display} = \@o; |
|
300 |
} |
} |
301 |
|
|
302 |
=head2 display |
=head2 search_display |
303 |
|
|
304 |
Define tag just for I<display> |
Define output for L<search> and L<display> at the same time |
305 |
|
|
306 |
@v = display('Title', rec('200','a') ); |
search_display('Title', rec('200','a') ); |
307 |
|
|
308 |
=cut |
=cut |
309 |
|
|
310 |
sub display { |
sub search_display { |
311 |
my $name = shift or die "display needs name as first argument"; |
my $name = shift or die "search_display needs name as first argument"; |
312 |
my @o = grep { defined($_) && $_ ne '' } @_; |
my @o = grep { defined($_) && $_ ne '' } @_; |
313 |
return unless (@o); |
return unless (@o); |
314 |
$out->{$name}->{tag} = $name; |
$out->{$name}->{search} = \@o; |
315 |
$out->{$name}->{display} = \@o; |
$out->{$name}->{display} = \@o; |
316 |
} |
} |
317 |
|
|
318 |
=head2 search |
=head2 tag |
|
|
|
|
Prepare values just for I<search> |
|
319 |
|
|
320 |
@v = search('Title', rec('200','a') ); |
Old name for L<search_display>, it will probably be removed at one point. |
321 |
|
|
322 |
=cut |
=cut |
323 |
|
|
324 |
sub search { |
sub tag { |
325 |
my $name = shift or die "search needs name as first argument"; |
search_display( @_ ); |
|
my @o = grep { defined($_) && $_ ne '' } @_; |
|
|
return unless (@o); |
|
|
$out->{$name}->{tag} = $name; |
|
|
$out->{$name}->{search} = \@o; |
|
326 |
} |
} |
327 |
|
|
328 |
=head2 marc_leader |
=head2 display |
329 |
|
|
330 |
Setup fields within MARC leader or get leader |
Define output just for I<display> |
331 |
|
|
332 |
marc_leader('05','c'); |
@v = display('Title', rec('200','a') ); |
|
my $leader = marc_leader(); |
|
333 |
|
|
334 |
=cut |
=cut |
335 |
|
|
336 |
sub marc_leader { |
sub display { to( 'display', @_ ) } |
|
my ($offset,$value) = @_; |
|
337 |
|
|
338 |
if ($offset) { |
=head2 search |
|
$out->{' leader'}->{ $offset } = $value; |
|
|
} else { |
|
|
return $out->{' leader'}; |
|
|
} |
|
|
} |
|
|
|
|
|
=head2 marc |
|
339 |
|
|
340 |
Save value for MARC field |
Prepare values just for I<search> |
341 |
|
|
342 |
marc('900','a', rec('200','a') ); |
@v = search('Title', rec('200','a') ); |
|
marc('001', rec('000') ); |
|
343 |
|
|
344 |
=cut |
=cut |
345 |
|
|
346 |
sub marc { |
sub search { to( 'search', @_ ) } |
|
my $f = shift or die "marc needs field"; |
|
|
die "marc field must be numer" unless ($f =~ /^\d+$/); |
|
|
|
|
|
my $sf; |
|
|
if ($f >= 10) { |
|
|
$sf = shift or die "marc needs subfield"; |
|
|
} |
|
|
|
|
|
foreach (@_) { |
|
|
my $v = $_; # make var read-write for Encode |
|
|
next unless (defined($v) && $v !~ /^\s*$/); |
|
|
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
|
|
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
|
|
if (defined $sf) { |
|
|
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $i1, $i2, $sf => $v ]; |
|
|
} else { |
|
|
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $v ]; |
|
|
} |
|
|
} |
|
|
} |
|
347 |
|
|
348 |
=head2 marc_repeatable_subfield |
=head2 sorted |
349 |
|
|
350 |
Save values for MARC repetable subfield |
Insert into lists which will be automatically sorted |
351 |
|
|
352 |
marc_repeatable_subfield('910', 'z', rec('909') ); |
sorted('Title', rec('200','a') ); |
353 |
|
|
354 |
=cut |
=cut |
355 |
|
|
356 |
sub marc_repeatable_subfield { |
sub sorted { to( 'sorted', @_ ) } |
|
my ($f,$sf) = @_; |
|
|
die "marc_repeatable_subfield need field and subfield!\n" unless ($f && $sf); |
|
|
$marc_repeatable_subfield->{ $f . $sf }++; |
|
|
marc(@_); |
|
|
} |
|
357 |
|
|
358 |
=head2 marc_indicators |
=head2 row |
359 |
|
|
360 |
Set both indicators for MARC field |
Insert new row of data into output module |
361 |
|
|
362 |
marc_indicators('900', ' ', 1); |
row( column => 'foo', column2 => 'bar' ); |
|
|
|
|
Any indicator value other than C<0-9> will be treated as undefined. |
|
363 |
|
|
364 |
=cut |
=cut |
365 |
|
|
366 |
sub marc_indicators { |
use Data::Dump qw/dump/; |
|
my $f = shift || die "marc_indicators need field!\n"; |
|
|
my ($i1,$i2) = @_; |
|
|
die "marc_indicators($f, ...) need i1!\n" unless(defined($i1)); |
|
|
die "marc_indicators($f, $i1, ...) need i2!\n" unless(defined($i2)); |
|
367 |
|
|
368 |
$i1 = ' ' if ($i1 !~ /^\d$/); |
sub row { |
369 |
$i2 = ' ' if ($i2 !~ /^\d$/); |
die "array doesn't have odd number of elements but $#_: ",dump( @_ ) if $#_ % 2 == 1; |
370 |
@{ $marc_indicators->{$f} } = ($i1,$i2); |
my $table = shift @_; |
371 |
|
push @{ $out->{'_rows'}->{$table} }, {@_}; |
372 |
} |
} |
373 |
|
|
|
=head2 marc_compose |
|
374 |
|
|
375 |
Save values for each MARC subfield explicitly |
=head1 Functions to extract data from input |
|
|
|
|
marc_compose('900', |
|
|
'a', rec('200','a') |
|
|
'b', rec('201','a') |
|
|
'a', rec('200','b') |
|
|
'c', rec('200','c') |
|
|
); |
|
|
|
|
|
=cut |
|
|
|
|
|
sub marc_compose { |
|
|
my $f = shift or die "marc_compose needs field"; |
|
|
die "marc_compose field must be numer" unless ($f =~ /^\d+$/); |
|
|
|
|
|
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
|
|
my $m = [ $f, $i1, $i2 ]; |
|
|
|
|
|
while (@_) { |
|
|
my $sf = shift or die "marc_compose $f needs subfield"; |
|
|
my $v = shift; |
|
|
|
|
|
next unless (defined($v) && $v !~ /^\s*$/); |
|
|
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
|
|
push @$m, ( $sf, $v ); |
|
|
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
|
|
} |
|
|
|
|
|
warn "## marc_compose(d) ", dump( $m ),$/ if ($debug > 1); |
|
|
|
|
|
push @{ $marc_record->[ $marc_record_offset ] }, $m if ($#{$m} > 2); |
|
|
} |
|
376 |
|
|
377 |
=head2 marc_duplicate |
This function should be used inside functions to create C<data_structure> described |
378 |
|
above. |
379 |
|
|
380 |
Generate copy of current MARC record and continue working on copy |
=head2 _pack_subfields_hash |
381 |
|
|
382 |
marc_duplicate(); |
@subfields = _pack_subfields_hash( $h ); |
383 |
|
$subfields = _pack_subfields_hash( $h, 1 ); |
384 |
|
|
385 |
Copies can be accessed using C<< _get_marc_fields( fetch_next => 1 ) >> or |
Return each subfield value in array or pack them all together and return scalar |
386 |
C<< _get_marc_fields( offset => 42 ) >>. |
with subfields (denoted by C<^>) and values. |
387 |
|
|
388 |
=cut |
=cut |
389 |
|
|
390 |
sub marc_duplicate { |
sub _pack_subfields_hash { |
|
my $m = $marc_record->[ -1 ]; |
|
|
die "can't duplicate record which isn't defined" unless ($m); |
|
|
push @{ $marc_record }, dclone( $m ); |
|
|
warn "## marc_duplicate = ", dump(@$marc_record), $/ if ($debug > 1); |
|
|
$marc_record_offset = $#{ $marc_record }; |
|
|
warn "## marc_record_offset = $marc_record_offset", $/ if ($debug > 1); |
|
|
} |
|
|
|
|
|
=head2 marc_remove |
|
391 |
|
|
392 |
Remove some field or subfield from MARC record. |
warn "## _pack_subfields_hash( ",dump(@_), " )\n" if ($debug > 1); |
393 |
|
|
394 |
marc_remove('200'); |
my ($hash,$include_subfields) = @_; |
|
marc_remove('200','a'); |
|
395 |
|
|
396 |
This will erase field C<200> or C<200^a> from current MARC record. |
# sanity and ease of use |
397 |
|
return $hash if (ref($hash) ne 'HASH'); |
|
This is useful after calling C<marc_duplicate> or on it's own (but, you |
|
|
should probably just remove that subfield definition if you are not |
|
|
using C<marc_duplicate>). |
|
|
|
|
|
FIXME: support fields < 10. |
|
|
|
|
|
=cut |
|
398 |
|
|
399 |
sub marc_remove { |
my $h = dclone( $hash ); |
|
my ($f, $sf) = @_; |
|
400 |
|
|
401 |
die "marc_remove needs record number" unless defined($f); |
if ( defined($h->{subfields}) ) { |
402 |
|
my $sfs = delete $h->{subfields} || die "no subfields?"; |
403 |
|
my @out; |
404 |
|
while (@$sfs) { |
405 |
|
my $sf = shift @$sfs; |
406 |
|
push @out, '^' . $sf if ($include_subfields); |
407 |
|
my $o = shift @$sfs; |
408 |
|
if ($o == 0 && ref( $h->{$sf} ) ne 'ARRAY' ) { |
409 |
|
# single element subfields are not arrays |
410 |
|
#warn "====> $sf $o / $#$sfs ", dump( $sfs, $h->{$sf} ), "\n"; |
411 |
|
|
412 |
my $marc = $marc_record->[ $marc_record_offset ]; |
push @out, $h->{$sf}; |
|
|
|
|
warn "### marc_remove before = ", dump( $marc ), $/ if ($debug > 2); |
|
|
|
|
|
foreach my $i ( 0 .. $#{ $marc } ) { |
|
|
last unless (defined $marc->[$i]); |
|
|
warn "#### working on ",dump( @{ $marc->[$i] }), $/ if ($debug > 3); |
|
|
if ($marc->[$i]->[0] eq $f) { |
|
|
if (! defined $sf) { |
|
|
# remove whole field |
|
|
splice @$marc, $i, 1; |
|
|
warn "#### slice \@\$marc, $i, 1 = ",dump( @{ $marc }), $/ if ($debug > 3); |
|
|
$i--; |
|
413 |
} else { |
} else { |
414 |
foreach my $j ( 0 .. (( $#{ $marc->[$i] } - 3 ) / 2) ) { |
#warn "====> $sf $o / $#$sfs ", dump( $sfs, $h->{$sf} ), "\n"; |
415 |
my $o = ($j * 2) + 3; |
push @out, $h->{$sf}->[$o]; |
416 |
if ($marc->[$i]->[$o] eq $sf) { |
} |
417 |
# remove subfield |
} |
418 |
splice @{$marc->[$i]}, $o, 2; |
if ($include_subfields) { |
419 |
warn "#### slice \@{\$marc->[$i]}, $o, 2 = ", dump( @{ $marc }), $/ if ($debug > 3); |
return join('', @out); |
420 |
# is record now empty? |
} else { |
421 |
if ($#{ $marc->[$i] } == 2) { |
return @out; |
422 |
splice @$marc, $i, 1; |
} |
423 |
warn "#### slice \@\$marc, $i, 1 = ", dump( @{ $marc }), $/ if ($debug > 3); |
} else { |
424 |
$i--; |
if ($include_subfields) { |
425 |
}; |
my $out = ''; |
426 |
} |
foreach my $sf (sort keys %$h) { |
427 |
|
if (ref($h->{$sf}) eq 'ARRAY') { |
428 |
|
$out .= '^' . $sf . join('^' . $sf, @{ $h->{$sf} }); |
429 |
|
} else { |
430 |
|
$out .= '^' . $sf . $h->{$sf}; |
431 |
} |
} |
432 |
} |
} |
433 |
|
return $out; |
434 |
|
} else { |
435 |
|
# FIXME this should probably be in alphabetical order instead of hash order |
436 |
|
values %{$h}; |
437 |
} |
} |
438 |
} |
} |
|
|
|
|
warn "### marc_remove($f", $sf ? ",$sf" : "", ") after = ", dump( $marc ), $/ if ($debug > 2); |
|
|
|
|
|
$marc_record->[ $marc_record_offset ] = $marc; |
|
|
|
|
|
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
|
439 |
} |
} |
440 |
|
|
|
=head1 Functions to extract data from input |
|
|
|
|
|
This function should be used inside functions to create C<data_structure> described |
|
|
above. |
|
|
|
|
441 |
=head2 rec1 |
=head2 rec1 |
442 |
|
|
443 |
Return all values in some field |
Return all values in some field |
454 |
return unless (defined($rec) && defined($rec->{$f})); |
return unless (defined($rec) && defined($rec->{$f})); |
455 |
warn "rec1($f) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
warn "rec1($f) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
456 |
if (ref($rec->{$f}) eq 'ARRAY') { |
if (ref($rec->{$f}) eq 'ARRAY') { |
457 |
return map { |
my @out; |
458 |
if (ref($_) eq 'HASH') { |
foreach my $h ( @{ $rec->{$f} } ) { |
459 |
values %{$_}; |
if (ref($h) eq 'HASH') { |
460 |
|
push @out, ( _pack_subfields_hash( $h ) ); |
461 |
} else { |
} else { |
462 |
$_; |
push @out, $h; |
463 |
} |
} |
464 |
} @{ $rec->{$f} }; |
} |
465 |
|
return @out; |
466 |
} elsif( defined($rec->{$f}) ) { |
} elsif( defined($rec->{$f}) ) { |
467 |
return $rec->{$f}; |
return $rec->{$f}; |
468 |
} |
} |
480 |
my $f = shift; |
my $f = shift; |
481 |
return unless (defined($rec && $rec->{$f})); |
return unless (defined($rec && $rec->{$f})); |
482 |
my $sf = shift; |
my $sf = shift; |
483 |
return map { $_->{$sf} } grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} }; |
warn "rec2($f,$sf) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
484 |
|
return map { |
485 |
|
if (ref($_->{$sf}) eq 'ARRAY') { |
486 |
|
@{ $_->{$sf} }; |
487 |
|
} else { |
488 |
|
$_->{$sf}; |
489 |
|
} |
490 |
|
} grep { ref($_) eq 'HASH' && defined $_->{$sf} } @{ $rec->{$f} }; |
491 |
} |
} |
492 |
|
|
493 |
=head2 rec |
=head2 rec |
497 |
@v = rec('200') |
@v = rec('200') |
498 |
@v = rec('200','a') |
@v = rec('200','a') |
499 |
|
|
500 |
|
If rec() returns just single value, it will |
501 |
|
return scalar, not array. |
502 |
|
|
503 |
=cut |
=cut |
504 |
|
|
505 |
sub rec { |
sub rec { |
506 |
|
my @out; |
507 |
if ($#_ == 0) { |
if ($#_ == 0) { |
508 |
return rec1(@_); |
@out = rec1(@_); |
509 |
} elsif ($#_ == 1) { |
} elsif ($#_ == 1) { |
510 |
return rec2(@_); |
@out = rec2(@_); |
511 |
|
} |
512 |
|
if ($#out == 0 && ! wantarray) { |
513 |
|
return $out[0]; |
514 |
|
} elsif (@out) { |
515 |
|
return @out; |
516 |
|
} else { |
517 |
|
return ''; |
518 |
} |
} |
519 |
} |
} |
520 |
|
|
521 |
|
=head2 frec |
522 |
|
|
523 |
|
Returns first value from field |
524 |
|
|
525 |
|
$v = frec('200'); |
526 |
|
$v = frec('200','a'); |
527 |
|
|
528 |
|
=cut |
529 |
|
|
530 |
|
sub frec { |
531 |
|
my @out = rec(@_); |
532 |
|
warn "rec(",dump(@_),") has more than one return value, ignoring\n" if $#out > 0; |
533 |
|
return shift @out; |
534 |
|
} |
535 |
|
|
536 |
|
=head2 frec_eq |
537 |
|
|
538 |
|
=head2 frec_ne |
539 |
|
|
540 |
|
Check if first values from two fields are same or different |
541 |
|
|
542 |
|
if ( frec_eq( 900 => 'a', 910 => 'c' ) ) { |
543 |
|
# values are same |
544 |
|
} else { |
545 |
|
# values are different |
546 |
|
} |
547 |
|
|
548 |
|
Strictly speaking C<frec_eq> and C<frec_ne> wouldn't be needed if you |
549 |
|
could write something like: |
550 |
|
|
551 |
|
if ( frec( '900','a' ) eq frec( '910','c' ) ) { |
552 |
|
# yada tada |
553 |
|
} |
554 |
|
|
555 |
|
but you can't since our parser L<WebPAC::Parser> will remove all whitespaces |
556 |
|
in order to parse text and create invalid function C<eqfrec>. |
557 |
|
|
558 |
|
=cut |
559 |
|
|
560 |
|
sub frec_eq { |
561 |
|
my ( $f1,$sf1, $f2, $sf2 ) = @_; |
562 |
|
return (rec( $f1, $sf1 ))[0] eq (rec( $f2, $sf2 ))[0]; |
563 |
|
} |
564 |
|
|
565 |
|
sub frec_ne { |
566 |
|
return ! frec_eq( @_ ); |
567 |
|
} |
568 |
|
|
569 |
=head2 regex |
=head2 regex |
570 |
|
|
571 |
Apply regex to some or all values |
Apply regex to some or all values |
595 |
=cut |
=cut |
596 |
|
|
597 |
sub prefix { |
sub prefix { |
598 |
my $p = shift or die "prefix needs string as first argument"; |
my $p = shift; |
599 |
|
return @_ unless defined( $p ); |
600 |
return map { $p . $_ } grep { defined($_) } @_; |
return map { $p . $_ } grep { defined($_) } @_; |
601 |
} |
} |
602 |
|
|
609 |
=cut |
=cut |
610 |
|
|
611 |
sub suffix { |
sub suffix { |
612 |
my $s = shift or die "suffix needs string as first argument"; |
my $s = shift; |
613 |
|
return @_ unless defined( $s ); |
614 |
return map { $_ . $s } grep { defined($_) } @_; |
return map { $_ . $s } grep { defined($_) } @_; |
615 |
} |
} |
616 |
|
|
623 |
=cut |
=cut |
624 |
|
|
625 |
sub surround { |
sub surround { |
626 |
my $p = shift or die "surround need prefix as first argument"; |
my $p = shift; |
627 |
my $s = shift or die "surround needs suffix as second argument"; |
my $s = shift; |
628 |
|
$p = '' unless defined( $p ); |
629 |
|
$s = '' unless defined( $s ); |
630 |
return map { $p . $_ . $s } grep { defined($_) } @_; |
return map { $p . $_ . $s } grep { defined($_) } @_; |
631 |
} |
} |
632 |
|
|
647 |
|
|
648 |
Consult lookup hashes for some value |
Consult lookup hashes for some value |
649 |
|
|
650 |
@v = lookup( $v ); |
@v = lookup( |
651 |
@v = lookup( @v ); |
sub { |
652 |
|
'ffkk/peri/mfn'.rec('000') |
653 |
|
}, |
654 |
|
'ffkk','peri','200-a-200-e', |
655 |
|
sub { |
656 |
|
first(rec(200,'a')).' '.first(rec('200','e')) |
657 |
|
} |
658 |
|
); |
659 |
|
|
660 |
|
Code like above will be B<automatically generated> using L<WebPAC::Parse> from |
661 |
|
normal lookup definition in C<conf/lookup/something.pl> which looks like: |
662 |
|
|
663 |
|
lookup( |
664 |
|
# which results to return from record recorded in lookup |
665 |
|
sub { 'ffkk/peri/mfn' . rec('000') }, |
666 |
|
# from which database and input |
667 |
|
'ffkk','peri', |
668 |
|
# such that following values match |
669 |
|
sub { first(rec(200,'a')) . ' ' . first(rec('200','e')) }, |
670 |
|
# if this part is missing, we will try to match same fields |
671 |
|
# from lookup record and current one, or you can override |
672 |
|
# which records to use from current record using |
673 |
|
sub { rec('900','x') . ' ' . rec('900','y') }, |
674 |
|
) |
675 |
|
|
676 |
|
You can think about this lookup as SQL (if that helps): |
677 |
|
|
678 |
|
select |
679 |
|
sub { what } |
680 |
|
from |
681 |
|
database, input |
682 |
|
where |
683 |
|
sub { filter from lookuped record } |
684 |
|
having |
685 |
|
sub { optional filter on current record } |
686 |
|
|
687 |
|
Easy as pie, right? |
688 |
|
|
689 |
=cut |
=cut |
690 |
|
|
691 |
sub lookup { |
sub lookup { |
692 |
my $k = shift or return; |
my ($what, $database, $input, $key, $having) = @_; |
693 |
return unless (defined($lookup->{$k})); |
|
694 |
if (ref($lookup->{$k}) eq 'ARRAY') { |
confess "lookup needs 5 arguments: what, database, input, key, having\n" unless ($#_ == 4); |
695 |
return @{ $lookup->{$k} }; |
|
696 |
|
warn "## lookup ($database, $input, $key)", $/ if ($debug > 1); |
697 |
|
return unless (defined($lookup->{$database}->{$input}->{$key})); |
698 |
|
|
699 |
|
confess "lookup really need load_row_coderef added to data_structure\n" unless ($load_row_coderef); |
700 |
|
|
701 |
|
my $mfns; |
702 |
|
my @having = $having->(); |
703 |
|
|
704 |
|
warn "## having = ", dump( @having ) if ($debug > 2); |
705 |
|
|
706 |
|
foreach my $h ( @having ) { |
707 |
|
if (defined($lookup->{$database}->{$input}->{$key}->{$h})) { |
708 |
|
warn "lookup for $database/$input/$key/$h return ",dump($lookup->{$database}->{$input}->{$key}->{$h}),"\n" if ($debug); |
709 |
|
$mfns->{$_}++ foreach keys %{ $lookup->{$database}->{$input}->{$key}->{$h} }; |
710 |
|
} |
711 |
|
} |
712 |
|
|
713 |
|
return unless ($mfns); |
714 |
|
|
715 |
|
my @mfns = sort keys %$mfns; |
716 |
|
|
717 |
|
warn "# lookup loading $database/$input/$key mfn ", join(",",@mfns)," having ",dump(@having),"\n" if ($debug); |
718 |
|
|
719 |
|
my $old_rec = $rec; |
720 |
|
my @out; |
721 |
|
|
722 |
|
foreach my $mfn (@mfns) { |
723 |
|
$rec = $load_row_coderef->( $database, $input, $mfn ); |
724 |
|
|
725 |
|
warn "got $database/$input/$mfn = ", dump($rec), $/ if ($debug); |
726 |
|
|
727 |
|
my @vals = $what->(); |
728 |
|
|
729 |
|
push @out, ( @vals ); |
730 |
|
|
731 |
|
warn "lookup for mfn $mfn returned ", dump(@vals), $/ if ($debug); |
732 |
|
} |
733 |
|
|
734 |
|
# if (ref($lookup->{$k}) eq 'ARRAY') { |
735 |
|
# return @{ $lookup->{$k} }; |
736 |
|
# } else { |
737 |
|
# return $lookup->{$k}; |
738 |
|
# } |
739 |
|
|
740 |
|
$rec = $old_rec; |
741 |
|
|
742 |
|
warn "## lookup returns = ", dump(@out), $/ if ($debug); |
743 |
|
|
744 |
|
if ($#out == 0) { |
745 |
|
return $out[0]; |
746 |
} else { |
} else { |
747 |
return $lookup->{$k}; |
return @out; |
748 |
|
} |
749 |
|
} |
750 |
|
|
751 |
|
=head2 save_into_lookup |
752 |
|
|
753 |
|
Save value into lookup. It associates current database, input |
754 |
|
and specific keys with one or more values which will be |
755 |
|
associated over MFN. |
756 |
|
|
757 |
|
MFN will be extracted from first occurence current of field 000 |
758 |
|
in current record, or if it doesn't exist from L<_set_config> C<_mfn>. |
759 |
|
|
760 |
|
my $nr = save_into_lookup($database,$input,$key,sub { |
761 |
|
# code which produce one or more values |
762 |
|
}); |
763 |
|
|
764 |
|
It returns number of items saved. |
765 |
|
|
766 |
|
This function shouldn't be called directly, it's called from code created by |
767 |
|
L<WebPAC::Parser>. |
768 |
|
|
769 |
|
=cut |
770 |
|
|
771 |
|
sub save_into_lookup { |
772 |
|
my ($database,$input,$key,$coderef) = @_; |
773 |
|
die "save_into_lookup needs database" unless defined($database); |
774 |
|
die "save_into_lookup needs input" unless defined($input); |
775 |
|
die "save_into_lookup needs key" unless defined($key); |
776 |
|
die "save_into_lookup needs CODE" unless ( defined($coderef) && ref($coderef) eq 'CODE' ); |
777 |
|
|
778 |
|
warn "## save_into_lookup rec = ", dump($rec), " config = ", dump($config), $/ if ($debug > 2); |
779 |
|
|
780 |
|
my $mfn = |
781 |
|
defined($rec->{'000'}->[0]) ? $rec->{'000'}->[0] : |
782 |
|
defined($config->{_mfn}) ? $config->{_mfn} : |
783 |
|
die "mfn not defined or zero"; |
784 |
|
|
785 |
|
my $nr = 0; |
786 |
|
|
787 |
|
foreach my $v ( $coderef->() ) { |
788 |
|
$lookup->{$database}->{$input}->{$key}->{$v}->{$mfn}++; |
789 |
|
warn "# saved lookup $database/$input/$key [$v] $mfn\n" if ($debug > 1); |
790 |
|
$nr++; |
791 |
|
} |
792 |
|
|
793 |
|
return $nr; |
794 |
|
} |
795 |
|
|
796 |
|
=head2 config |
797 |
|
|
798 |
|
Consult config values stored in C<config.yml> |
799 |
|
|
800 |
|
# return database code (key under databases in yaml) |
801 |
|
$database_code = config(); # use _ from hash |
802 |
|
$database_name = config('name'); |
803 |
|
$database_input_name = config('input name'); |
804 |
|
|
805 |
|
Up to three levels are supported. |
806 |
|
|
807 |
|
=cut |
808 |
|
|
809 |
|
sub config { |
810 |
|
return unless ($config); |
811 |
|
|
812 |
|
my $p = shift; |
813 |
|
|
814 |
|
$p ||= ''; |
815 |
|
|
816 |
|
my $v; |
817 |
|
|
818 |
|
warn "### getting config($p)\n" if ($debug > 1); |
819 |
|
|
820 |
|
my @p = split(/\s+/,$p); |
821 |
|
if ($#p < 0) { |
822 |
|
$v = $config->{ '_' }; # special, database code |
823 |
|
} else { |
824 |
|
|
825 |
|
my $c = dclone( $config ); |
826 |
|
|
827 |
|
foreach my $k (@p) { |
828 |
|
warn "### k: $k c = ",dump($c),$/ if ($debug > 1); |
829 |
|
if (ref($c) eq 'ARRAY') { |
830 |
|
$c = shift @$c; |
831 |
|
warn "config($p) taking first occurence of '$k', probably not what you wanted!\n"; |
832 |
|
last; |
833 |
|
} |
834 |
|
|
835 |
|
if (! defined($c->{$k}) ) { |
836 |
|
$c = undef; |
837 |
|
last; |
838 |
|
} else { |
839 |
|
$c = $c->{$k}; |
840 |
|
} |
841 |
|
} |
842 |
|
$v = $c if ($c); |
843 |
|
|
844 |
} |
} |
845 |
|
|
846 |
|
warn "## config( '$p' ) = ",dump( $v ),$/ if ($v && $debug); |
847 |
|
warn "config( '$p' ) is empty\n" if (! $v); |
848 |
|
|
849 |
|
return $v; |
850 |
|
} |
851 |
|
|
852 |
|
=head2 id |
853 |
|
|
854 |
|
Returns unique id of this record |
855 |
|
|
856 |
|
$id = id(); |
857 |
|
|
858 |
|
Returns C<42/2> for 2nd occurence of MFN 42. |
859 |
|
|
860 |
|
=cut |
861 |
|
|
862 |
|
sub id { |
863 |
|
my $mfn = $config->{_mfn} || die "no _mfn in config data"; |
864 |
|
return $mfn . ( WebPAC::Normalize::MARC::_created_marc_records() || '' ); |
865 |
} |
} |
866 |
|
|
867 |
=head2 join_with |
=head2 join_with |
874 |
|
|
875 |
sub join_with { |
sub join_with { |
876 |
my $d = shift; |
my $d = shift; |
877 |
return join($d, grep { defined($_) && $_ ne '' } @_); |
warn "### join_with('$d',",dump(@_),")\n" if ($debug > 2); |
878 |
|
my $v = join($d, grep { defined($_) && $_ ne '' } @_); |
879 |
|
return '' unless defined($v); |
880 |
|
return $v; |
881 |
} |
} |
882 |
|
|
883 |
=head2 split_rec_on |
=head2 split_rec_on |
905 |
my $v = shift @r; |
my $v = shift @r; |
906 |
warn "### first rec($fld,$sf) = ",dump($v),$/ if ($debug > 2); |
warn "### first rec($fld,$sf) = ",dump($v),$/ if ($debug > 2); |
907 |
|
|
908 |
return '' if( ! defined($v) || $v =~ /^\s*$/); |
return '' if ( ! defined($v) || $v =~ /^\s*$/); |
909 |
|
|
910 |
my @s = split( $regex, $v ); |
my @s = split( $regex, $v ); |
911 |
warn "## split_rec_on($fld,$sf,$regex,$part) = ",dump(@s),$/ if ($debug > 1); |
warn "## split_rec_on($fld,$sf,$regex,$part) = ",dump(@s),$/ if ($debug > 1); |
916 |
} |
} |
917 |
} |
} |
918 |
|
|
919 |
|
my $hash; |
920 |
|
|
921 |
|
=head2 set |
922 |
|
|
923 |
|
set( key => 'value' ); |
924 |
|
|
925 |
|
=cut |
926 |
|
|
927 |
|
sub set { |
928 |
|
my ($k,$v) = @_; |
929 |
|
warn "## set ( $k => ", dump($v), " )", $/ if ( $debug ); |
930 |
|
$hash->{$k} = $v; |
931 |
|
}; |
932 |
|
|
933 |
|
=head2 get |
934 |
|
|
935 |
|
get( 'key' ); |
936 |
|
|
937 |
|
=cut |
938 |
|
|
939 |
|
sub get { |
940 |
|
my $k = shift || return; |
941 |
|
my $v = $hash->{$k}; |
942 |
|
warn "## get $k = ", dump( $v ), $/ if ( $debug ); |
943 |
|
return $v; |
944 |
|
} |
945 |
|
|
946 |
|
=head2 count |
947 |
|
|
948 |
|
if ( count( @result ) == 1 ) { |
949 |
|
# do something if only 1 result is there |
950 |
|
} |
951 |
|
|
952 |
|
=cut |
953 |
|
|
954 |
|
sub count { |
955 |
|
warn "## count ",dump(@_),$/ if ( $debug ); |
956 |
|
return @_ . ''; |
957 |
|
} |
958 |
|
|
959 |
# END |
# END |
960 |
1; |
1; |