7 |
|
|
8 |
use WebPAC::Common; |
use WebPAC::Common; |
9 |
use base qw/WebPAC::Common/; |
use base qw/WebPAC::Common/; |
|
use Text::Iconv; |
|
10 |
use Data::Dumper; |
use Data::Dumper; |
11 |
|
use Encode qw/from_to/; |
12 |
|
|
13 |
=head1 NAME |
=head1 NAME |
14 |
|
|
16 |
|
|
17 |
=head1 VERSION |
=head1 VERSION |
18 |
|
|
19 |
Version 0.08 |
Version 0.11 |
20 |
|
|
21 |
=cut |
=cut |
22 |
|
|
23 |
our $VERSION = '0.08'; |
our $VERSION = '0.11'; |
24 |
|
|
25 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
26 |
|
|
39 |
|
|
40 |
Perhaps a little code snippet. |
Perhaps a little code snippet. |
41 |
|
|
42 |
use WebPAC::Input; |
use WebPAC::Input; |
43 |
|
|
44 |
my $db = WebPAC::Input->new( |
my $db = WebPAC::Input->new( |
45 |
module => 'WebPAC::Input::ISIS', |
module => 'WebPAC::Input::ISIS', |
|
config => $config, |
|
46 |
low_mem => 1, |
low_mem => 1, |
47 |
); |
); |
48 |
|
|
49 |
$db->open('/path/to/database'); |
$db->open( path => '/path/to/database' ); |
50 |
print "database size: ",$db->size,"\n"; |
print "database size: ",$db->size,"\n"; |
51 |
while (my $rec = $db->fetch) { |
while (my $rec = $db->fetch) { |
52 |
# do something with $rec |
# do something with $rec |
68 |
no_progress_bar => 1, |
no_progress_bar => 1, |
69 |
); |
); |
70 |
|
|
71 |
C<module> is low-level file format module. See L<WebPAC::Input::Isis> and |
C<module> is low-level file format module. See L<WebPAC::Input::ISIS> and |
72 |
L<WebPAC::Input::MARC>. |
L<WebPAC::Input::MARC>. |
73 |
|
|
74 |
Optional parametar C<encoding> specify application code page (which will be |
Optional parametar C<encoding> specify application code page (which will be |
159 |
|
|
160 |
$input->open( |
$input->open( |
161 |
path => '/path/to/database/file', |
path => '/path/to/database/file', |
162 |
code_page => '852', |
code_page => 'cp852', |
163 |
limit => 500, |
limit => 500, |
164 |
offset => 6000, |
offset => 6000, |
165 |
lookup => $lookup_obj, |
lookup => $lookup_obj, |
168 |
my ($k,$v) = @_; |
my ($k,$v) = @_; |
169 |
# store lookup $k => $v |
# store lookup $k => $v |
170 |
}, |
}, |
171 |
|
modify_records => { |
172 |
|
900 => { '^a' => { ' : ' => '^b' } }, |
173 |
|
901 => { '*' => { '^b' => ' ; ' } }, |
174 |
|
}, |
175 |
); |
); |
176 |
|
|
177 |
By default, C<code_page> is assumed to be C<852>. |
By default, C<code_page> is assumed to be C<cp852>. |
178 |
|
|
179 |
C<offset> is optional parametar to position at some offset before reading from database. |
C<offset> is optional parametar to position at some offset before reading from database. |
180 |
|
|
182 |
|
|
183 |
C<stats> create optional report about usage of fields and subfields |
C<stats> create optional report about usage of fields and subfields |
184 |
|
|
185 |
C<lookup_coderef> is closure to call when adding key => $value combinations to |
C<lookup_coderef> is closure to call when adding C<< key => 'value' >> combinations to |
186 |
lookup. |
lookup. |
187 |
|
|
188 |
|
C<modify_records> specify mapping from subfields to delimiters or from |
189 |
|
delimiters to subfields, as well as oprations on fields (if subfield is |
190 |
|
defined as C<*>. |
191 |
|
|
192 |
Returns size of database, regardless of C<offset> and C<limit> |
Returns size of database, regardless of C<offset> and C<limit> |
193 |
parametars, see also C<size>. |
parametars, see also C<size>. |
194 |
|
|
205 |
if ($arg->{lookup_coderef} && ref($arg->{lookup_coderef}) ne 'CODE'); |
if ($arg->{lookup_coderef} && ref($arg->{lookup_coderef}) ne 'CODE'); |
206 |
|
|
207 |
$log->logcroak("need path") if (! $arg->{'path'}); |
$log->logcroak("need path") if (! $arg->{'path'}); |
208 |
my $code_page = $arg->{'code_page'} || '852'; |
my $code_page = $arg->{'code_page'} || 'cp852'; |
209 |
|
|
210 |
# store data in object |
# store data in object |
211 |
$self->{'input_code_page'} = $code_page; |
$self->{'input_code_page'} = $code_page; |
213 |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
214 |
} |
} |
215 |
|
|
|
# create Text::Iconv object |
|
|
$self->{iconv} = Text::Iconv->new($code_page,$self->{'encoding'}); ## FIXME remove! |
|
|
|
|
216 |
my $filter_ref; |
my $filter_ref; |
217 |
|
my $recode_regex; |
218 |
|
my $recode_map; |
219 |
|
|
220 |
if ($self->{recode}) { |
if ($self->{recode}) { |
221 |
my @r = split(/\s/, $self->{recode}); |
my @r = split(/\s/, $self->{recode}); |
222 |
if ($#r % 2 != 1) { |
if ($#r % 2 != 1) { |
223 |
$log->logwarn("recode needs even number of elements (some number of valid pairs)"); |
$log->logwarn("recode needs even number of elements (some number of valid pairs)"); |
224 |
} else { |
} else { |
|
my $recode; |
|
225 |
while (@r) { |
while (@r) { |
226 |
my $from = shift @r; |
my $from = shift @r; |
227 |
my $to = shift @r; |
my $to = shift @r; |
228 |
$recode->{$from} = $to; |
$recode_map->{$from} = $to; |
229 |
} |
} |
230 |
|
|
231 |
my $regex = join '|' => keys %{ $recode }; |
$recode_regex = join '|' => keys %{ $recode_map }; |
|
|
|
|
$log->debug("using recode regex: $regex"); |
|
|
|
|
|
$filter_ref = sub { |
|
|
my $t = shift; |
|
|
$t =~ s/($regex)/$recode->{$1}/g; |
|
|
return $t; |
|
|
}; |
|
232 |
|
|
233 |
|
$log->debug("using recode regex: $recode_regex"); |
234 |
} |
} |
235 |
|
|
236 |
} |
} |
237 |
|
|
238 |
|
my $rec_regex = $self->modify_record_regexps(%{ $arg->{modify_records} }); |
239 |
|
$log->debug("rec_regex: ", Dumper($rec_regex)); |
240 |
|
|
241 |
my ($db, $size) = $self->{open_db}->( $self, |
my ($db, $size) = $self->{open_db}->( $self, |
242 |
path => $arg->{path}, |
path => $arg->{path}, |
243 |
filter => $filter_ref, |
# filter => sub { |
244 |
|
# my ($l,$f_nr) = @_; |
245 |
|
# return unless defined($l); |
246 |
|
# from_to($l, $code_page, $self->{'encoding'}); |
247 |
|
# $l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
248 |
|
# return $l; |
249 |
|
# }, |
250 |
%{ $arg }, |
%{ $arg }, |
251 |
); |
); |
252 |
|
|
286 |
|
|
287 |
$log->debug("position: $pos\n"); |
$log->debug("position: $pos\n"); |
288 |
|
|
289 |
my $rec = $self->{fetch_rec}->($self, $db, $pos ); |
my $rec = $self->{fetch_rec}->($self, $db, $pos, sub { |
290 |
|
my ($l,$f_nr) = @_; |
291 |
|
# return unless defined($l); |
292 |
|
# return $l unless ($rec_regex && $f_nr); |
293 |
|
|
294 |
|
$log->debug("-=> $f_nr ## $l"); |
295 |
|
|
296 |
|
# codepage conversion and recode_regex |
297 |
|
# from_to($l, $code_page, $self->{'encoding'}); |
298 |
|
from_to($l, $code_page, 'utf-8'); |
299 |
|
$l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
300 |
|
|
301 |
|
# apply regexps |
302 |
|
if ($rec_regex && defined($rec_regex->{$f_nr})) { |
303 |
|
$log->logconfess("regexps->{$f_nr} must be ARRAY") if (ref($rec_regex->{$f_nr}) ne 'ARRAY'); |
304 |
|
my $c = 0; |
305 |
|
foreach my $r (@{ $rec_regex->{$f_nr} }) { |
306 |
|
#$log->debug("\$l = $l\neval \$l =~ $r"); |
307 |
|
eval '$l =~ ' . $r; |
308 |
|
$log->error("error applying regex: $r") if ($@); |
309 |
|
} |
310 |
|
} |
311 |
|
|
312 |
|
$log->debug("<=- $f_nr ## $l"); |
313 |
|
return $l; |
314 |
|
}); |
315 |
|
|
316 |
$log->debug(sub { Dumper($rec) }); |
$log->debug(sub { Dumper($rec) }); |
317 |
|
|
333 |
# update counters for statistics |
# update counters for statistics |
334 |
if ($self->{stats}) { |
if ($self->{stats}) { |
335 |
|
|
336 |
|
# fetch clean record with regexpes applied for statistics |
337 |
|
my $rec = $self->{fetch_rec}->($self, $db, $pos); |
338 |
|
|
339 |
foreach my $fld (keys %{ $rec }) { |
foreach my $fld (keys %{ $rec }) { |
340 |
$self->{_stats}->{fld}->{ $fld }++; |
$self->{_stats}->{fld}->{ $fld }++; |
341 |
|
|
347 |
if (ref($row) eq 'HASH') { |
if (ref($row) eq 'HASH') { |
348 |
|
|
349 |
foreach my $sf (keys %{ $row }) { |
foreach my $sf (keys %{ $row }) { |
350 |
|
next if ($sf eq 'subfields'); |
351 |
$self->{_stats}->{sf}->{ $fld }->{ $sf }->{count}++; |
$self->{_stats}->{sf}->{ $fld }->{ $sf }->{count}++; |
352 |
$self->{_stats}->{sf}->{ $fld }->{ $sf }->{repeatable}++ |
$self->{_stats}->{sf}->{ $fld }->{ $sf }->{repeatable}++ |
353 |
if (ref($row->{$sf}) eq 'ARRAY'); |
if (ref($row->{$sf}) eq 'ARRAY'); |
532 |
return $out; |
return $out; |
533 |
} |
} |
534 |
|
|
535 |
|
=head2 modify_record_regexps |
536 |
|
|
537 |
|
Generate hash with regexpes to be applied using L<filter>. |
538 |
|
|
539 |
|
my $regexpes = $input->modify_record_regexps( |
540 |
|
900 => { '^a' => { ' : ' => '^b' } }, |
541 |
|
901 => { '*' => { '^b' => ' ; ' } }, |
542 |
|
); |
543 |
|
|
544 |
|
=cut |
545 |
|
|
546 |
|
sub modify_record_regexps { |
547 |
|
my $self = shift; |
548 |
|
my $modify_record = {@_}; |
549 |
|
|
550 |
|
my $regexpes; |
551 |
|
|
552 |
|
foreach my $f (keys %$modify_record) { |
553 |
|
warn "--- f: $f\n"; |
554 |
|
foreach my $sf (keys %{ $modify_record->{$f} }) { |
555 |
|
warn "---- sf: $sf\n"; |
556 |
|
foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) { |
557 |
|
my $to = $modify_record->{$f}->{$sf}->{$from}; |
558 |
|
#die "no field?" unless defined($to); |
559 |
|
warn "----- transform: |$from| -> |$to|\n"; |
560 |
|
|
561 |
|
if ($sf =~ /^\^/) { |
562 |
|
my $regex = |
563 |
|
's/\Q'. $sf .'\E([^\^]+)\Q'. $from .'\E([^\^]+)/'. $sf .'$1'. $to .'$2/g'; |
564 |
|
push @{ $regexpes->{$f} }, $regex; |
565 |
|
warn ">>>>> $regex [sf]\n"; |
566 |
|
} else { |
567 |
|
my $regex = |
568 |
|
's/\Q'. $from .'\E/'. $to .'/g'; |
569 |
|
push @{ $regexpes->{$f} }, $regex; |
570 |
|
warn ">>>>> $regex [global]\n"; |
571 |
|
} |
572 |
|
|
573 |
|
} |
574 |
|
} |
575 |
|
} |
576 |
|
|
577 |
|
return $regexpes; |
578 |
|
} |
579 |
|
|
580 |
=head1 MEMORY USAGE |
=head1 MEMORY USAGE |
581 |
|
|
582 |
C<low_mem> options is double-edged sword. If enabled, WebPAC |
C<low_mem> options is double-edged sword. If enabled, WebPAC |
614 |
|
|
615 |
=head1 COPYRIGHT & LICENSE |
=head1 COPYRIGHT & LICENSE |
616 |
|
|
617 |
Copyright 2005 Dobrica Pavlinusic, All Rights Reserved. |
Copyright 2005-2006 Dobrica Pavlinusic, All Rights Reserved. |
618 |
|
|
619 |
This program is free software; you can redistribute it and/or modify it |
This program is free software; you can redistribute it and/or modify it |
620 |
under the same terms as Perl itself. |
under the same terms as Perl itself. |