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 0.10 |
Version 0.13 |
20 |
|
|
21 |
=cut |
=cut |
22 |
|
|
23 |
our $VERSION = '0.10'; |
our $VERSION = '0.13'; |
24 |
|
|
25 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
26 |
|
|
98 |
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_ref") if ($self->{lookup}); |
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_ref") if ($self->{lookup}); |
99 |
|
|
100 |
$log->logconfess("specify low-level file format module") unless ($self->{module}); |
$log->logconfess("specify low-level file format module") unless ($self->{module}); |
101 |
my $module = $self->{module}; |
my $module_path = $self->{module}; |
102 |
$module =~ s#::#/#g; |
$module_path =~ s#::#/#g; |
103 |
$module .= '.pm'; |
$module_path .= '.pm'; |
104 |
$log->debug("require low-level module $self->{module} from $module"); |
$log->debug("require low-level module $self->{module} from $module_path"); |
105 |
|
|
106 |
require $module; |
require $module_path; |
|
#eval $self->{module} .'->import'; |
|
107 |
|
|
108 |
# check if required subclasses are implemented |
# check if required subclasses are implemented |
109 |
foreach my $subclass (qw/open_db fetch_rec init/) { |
foreach my $subclass (qw/open_db fetch_rec init dump_rec/) { |
110 |
my $n = $self->{module} . '::' . $subclass; |
# FIXME |
|
if (! defined &{ $n }) { |
|
|
my $missing = "missing $subclass in $self->{module}"; |
|
|
$self->{$subclass} = sub { $log->logwarn($missing) }; |
|
|
} else { |
|
|
$self->{$subclass} = \&{ $n }; |
|
|
} |
|
|
} |
|
|
|
|
|
if ($self->{init}) { |
|
|
$log->debug("calling init"); |
|
|
$self->{init}->($self, @_); |
|
111 |
} |
} |
112 |
|
|
113 |
$self->{'encoding'} ||= 'ISO-8859-2'; |
$self->{'encoding'} ||= 'ISO-8859-2'; |
147 |
|
|
148 |
$input->open( |
$input->open( |
149 |
path => '/path/to/database/file', |
path => '/path/to/database/file', |
150 |
code_page => '852', |
code_page => 'cp852', |
151 |
limit => 500, |
limit => 500, |
152 |
offset => 6000, |
offset => 6000, |
|
lookup => $lookup_obj, |
|
153 |
stats => 1, |
stats => 1, |
154 |
lookup_ref => sub { |
lookup_coderef => sub { |
155 |
my ($k,$v) = @_; |
my $rec = shift; |
156 |
# store lookup $k => $v |
# store lookups |
157 |
}, |
}, |
158 |
modify_records => { |
modify_records => { |
159 |
900 => { '^a' => { ' : ' => '^b' } }, |
900 => { '^a' => { ' : ' => '^b' } }, |
160 |
901 => { '*' => { '^b' => ' ; ' } }, |
901 => { '*' => { '^b' => ' ; ' } }, |
161 |
}, |
}, |
162 |
|
modify_file => 'conf/modify/mapping.map', |
163 |
); |
); |
164 |
|
|
165 |
By default, C<code_page> is assumed to be C<852>. |
By default, C<code_page> is assumed to be C<cp852>. |
166 |
|
|
167 |
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. |
168 |
|
|
170 |
|
|
171 |
C<stats> create optional report about usage of fields and subfields |
C<stats> create optional report about usage of fields and subfields |
172 |
|
|
173 |
C<lookup_coderef> is closure to call when adding C<< key => 'value' >> combinations to |
C<lookup_coderef> is closure to called to save data into lookups |
|
lookup. |
|
174 |
|
|
175 |
C<modify_records> specify mapping from subfields to delimiters or from |
C<modify_records> specify mapping from subfields to delimiters or from |
176 |
delimiters to subfields, as well as oprations on fields (if subfield is |
delimiters to subfields, as well as oprations on fields (if subfield is |
177 |
defined as C<*>. |
defined as C<*>. |
178 |
|
|
179 |
|
C<modify_file> is alternative for C<modify_records> above which preserves order and offers |
180 |
|
(hopefully) simplier sintax than YAML or perl (see L</modify_file_regex>). This option |
181 |
|
overrides C<modify_records> if both exists for same input. |
182 |
|
|
183 |
Returns size of database, regardless of C<offset> and C<limit> |
Returns size of database, regardless of C<offset> and C<limit> |
184 |
parametars, see also C<size>. |
parametars, see also C<size>. |
185 |
|
|
195 |
$log->logconfess("lookup_coderef must be CODE, not ",ref($arg->{lookup_coderef})) |
$log->logconfess("lookup_coderef must be CODE, not ",ref($arg->{lookup_coderef})) |
196 |
if ($arg->{lookup_coderef} && ref($arg->{lookup_coderef}) ne 'CODE'); |
if ($arg->{lookup_coderef} && ref($arg->{lookup_coderef}) ne 'CODE'); |
197 |
|
|
198 |
|
$log->debug( $arg->{lookup_coderef} ? '' : 'not ', "using lookup_coderef"); |
199 |
|
|
200 |
$log->logcroak("need path") if (! $arg->{'path'}); |
$log->logcroak("need path") if (! $arg->{'path'}); |
201 |
my $code_page = $arg->{'code_page'} || '852'; |
my $code_page = $arg->{'code_page'} || 'cp852'; |
202 |
|
|
203 |
# store data in object |
# store data in object |
204 |
$self->{'input_code_page'} = $code_page; |
$self->{'input_code_page'} = $code_page; |
206 |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
207 |
} |
} |
208 |
|
|
|
# create Text::Iconv object |
|
|
$self->{iconv} = Text::Iconv->new($code_page,$self->{'encoding'}); ## FIXME remove! |
|
|
|
|
209 |
my $filter_ref; |
my $filter_ref; |
210 |
my $recode_regex; |
my $recode_regex; |
211 |
my $recode_map; |
my $recode_map; |
228 |
|
|
229 |
} |
} |
230 |
|
|
231 |
my $rec_regex = $self->modify_record_regexps(%{ $arg->{modify_records} }); |
my $rec_regex; |
232 |
$log->debug("rec_regex: ", Dumper($rec_regex)); |
if (my $p = $arg->{modify_file}) { |
233 |
|
$log->debug("using modify_file $p"); |
234 |
my ($db, $size) = $self->{open_db}->( $self, |
$rec_regex = $self->modify_file_regexps( $p ); |
235 |
path => $arg->{path}, |
} elsif (my $h = $arg->{modify_records}) { |
236 |
filter => sub { |
$log->debug("using modify_records ", Dumper( $h )); |
237 |
my ($l,$f_nr) = @_; |
$rec_regex = $self->modify_record_regexps(%{ $h }); |
238 |
return unless defined($l); |
} |
239 |
|
$log->debug("rec_regex: ", Dumper($rec_regex)) if ($rec_regex); |
|
## FIXME remove iconv! |
|
|
$l = $self->{iconv}->convert($l) if ($self->{iconv}); |
|
|
|
|
|
$l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
|
|
|
|
|
## FIXME remove this warning when we are sure that none of API is calling |
|
|
## this wrongly |
|
|
#warn "filter called without field number" unless ($f_nr); |
|
|
|
|
|
return $l unless ($rec_regex && $f_nr); |
|
|
|
|
|
# my $max_regex = 100; |
|
|
|
|
|
# apply regexps |
|
|
if ($rec_regex && defined($rec_regex->{$f_nr})) { |
|
|
$log->logconfess("regexps->{$f_nr} must be ARRAY") if (ref($rec_regex->{$f_nr}) ne 'ARRAY'); |
|
|
my $c = 0; |
|
|
foreach my $r (@{ $rec_regex->{$f_nr} }) { |
|
|
#$log->debug("\$l = $l\neval \$l =~ $r"); |
|
|
eval '$l =~ ' . $r; |
|
|
$log->error("error applying regex: $r") if ($@); |
|
|
|
|
|
# while ( $c < $max_regex && eval '$l =~ ' . $r ) { $c++ }; |
|
|
# $log->error("field $f_nr has more than $max_regex regex iterations\n\$l = $l\neval \$l =~ $r") if ($c == $max_regex); |
|
240 |
|
|
241 |
} |
my $class = $self->{module} || $log->logconfess("can't get low-level module name!"); |
|
} |
|
242 |
|
|
243 |
return $l; |
my $ll_db = $class->new( |
244 |
}, |
path => $arg->{path}, |
245 |
|
# filter => sub { |
246 |
|
# my ($l,$f_nr) = @_; |
247 |
|
# return unless defined($l); |
248 |
|
# from_to($l, $code_page, $self->{'encoding'}); |
249 |
|
# $l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
250 |
|
# return $l; |
251 |
|
# }, |
252 |
%{ $arg }, |
%{ $arg }, |
253 |
); |
); |
254 |
|
|
255 |
unless (defined($db)) { |
unless (defined($ll_db)) { |
256 |
$log->logwarn("can't open database $arg->{path}, skipping..."); |
$log->logwarn("can't open database $arg->{path}, skipping..."); |
257 |
return; |
return; |
258 |
} |
} |
259 |
|
|
260 |
|
my $size = $ll_db->size; |
261 |
|
|
262 |
unless ($size) { |
unless ($size) { |
263 |
$log->logwarn("no records in database $arg->{path}, skipping..."); |
$log->logwarn("no records in database $arg->{path}, skipping..."); |
264 |
return; |
return; |
290 |
|
|
291 |
$log->debug("position: $pos\n"); |
$log->debug("position: $pos\n"); |
292 |
|
|
293 |
my $rec = $self->{fetch_rec}->($self, $db, $pos ); |
my $rec = $ll_db->fetch_rec($pos, sub { |
294 |
|
my ($l,$f_nr) = @_; |
295 |
|
# return unless defined($l); |
296 |
|
# return $l unless ($rec_regex && $f_nr); |
297 |
|
|
298 |
|
$log->debug("-=> $f_nr ## $l"); |
299 |
|
|
300 |
|
# codepage conversion and recode_regex |
301 |
|
from_to($l, $code_page, $self->{'encoding'}); |
302 |
|
$l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
303 |
|
|
304 |
|
# apply regexps |
305 |
|
if ($rec_regex && defined($rec_regex->{$f_nr})) { |
306 |
|
$log->logconfess("regexps->{$f_nr} must be ARRAY") if (ref($rec_regex->{$f_nr}) ne 'ARRAY'); |
307 |
|
my $c = 0; |
308 |
|
foreach my $r (@{ $rec_regex->{$f_nr} }) { |
309 |
|
my $old_l = $l; |
310 |
|
eval '$l =~ ' . $r; |
311 |
|
if ($old_l ne $l) { |
312 |
|
$log->debug("REGEX on $f_nr eval \$l =~ $r\n## old l: [$old_l]\n## new l: [$l]"); |
313 |
|
} |
314 |
|
$log->error("error applying regex: $r") if ($@); |
315 |
|
} |
316 |
|
} |
317 |
|
|
318 |
|
$log->debug("<=- $f_nr ## $l"); |
319 |
|
return $l; |
320 |
|
}); |
321 |
|
|
322 |
$log->debug(sub { Dumper($rec) }); |
$log->debug(sub { Dumper($rec) }); |
323 |
|
|
339 |
# update counters for statistics |
# update counters for statistics |
340 |
if ($self->{stats}) { |
if ($self->{stats}) { |
341 |
|
|
342 |
|
# fetch clean record with regexpes applied for statistics |
343 |
|
my $rec = $ll_db->fetch_rec($pos); |
344 |
|
|
345 |
foreach my $fld (keys %{ $rec }) { |
foreach my $fld (keys %{ $rec }) { |
346 |
$self->{_stats}->{fld}->{ $fld }++; |
$self->{_stats}->{fld}->{ $fld }++; |
347 |
|
|
353 |
if (ref($row) eq 'HASH') { |
if (ref($row) eq 'HASH') { |
354 |
|
|
355 |
foreach my $sf (keys %{ $row }) { |
foreach my $sf (keys %{ $row }) { |
356 |
|
next if ($sf eq 'subfields'); |
357 |
$self->{_stats}->{sf}->{ $fld }->{ $sf }->{count}++; |
$self->{_stats}->{sf}->{ $fld }->{ $sf }->{count}++; |
358 |
$self->{_stats}->{sf}->{ $fld }->{ $sf }->{repeatable}++ |
$self->{_stats}->{sf}->{ $fld }->{ $sf }->{repeatable}++ |
359 |
if (ref($row->{$sf}) eq 'ARRAY'); |
if (ref($row->{$sf}) eq 'ARRAY'); |
538 |
return $out; |
return $out; |
539 |
} |
} |
540 |
|
|
541 |
|
=head2 dump |
542 |
|
|
543 |
|
Display humanly readable dump of record |
544 |
|
|
545 |
|
=cut |
546 |
|
|
547 |
|
sub dump { |
548 |
|
my $self = shift; |
549 |
|
|
550 |
|
return $self->{dump_rec}->($self, $self->{pos}); |
551 |
|
|
552 |
|
} |
553 |
|
|
554 |
=head2 modify_record_regexps |
=head2 modify_record_regexps |
555 |
|
|
556 |
Generate hash with regexpes to be applied using L<filter>. |
Generate hash with regexpes to be applied using l<filter>. |
557 |
|
|
558 |
my $regexpes = $input->modify_record_regexps( |
my $regexpes = $input->modify_record_regexps( |
559 |
900 => { '^a' => { ' : ' => '^b' } }, |
900 => { '^a' => { ' : ' => '^b' } }, |
562 |
|
|
563 |
=cut |
=cut |
564 |
|
|
565 |
|
sub _get_regex { |
566 |
|
my ($sf,$from,$to) = @_; |
567 |
|
if ($sf =~ /^\^/) { |
568 |
|
return |
569 |
|
's/\Q'. $sf .'\E([^\^]*?)\Q'. $from .'\E([^\^]*?)/'. $sf .'$1'. $to .'$2/'; |
570 |
|
} else { |
571 |
|
return |
572 |
|
's/\Q'. $from .'\E/'. $to .'/g'; |
573 |
|
} |
574 |
|
} |
575 |
|
|
576 |
sub modify_record_regexps { |
sub modify_record_regexps { |
577 |
my $self = shift; |
my $self = shift; |
578 |
my $modify_record = {@_}; |
my $modify_record = {@_}; |
579 |
|
|
580 |
my $regexpes; |
my $regexpes; |
581 |
|
|
582 |
|
my $log = $self->_get_logger(); |
583 |
|
|
584 |
foreach my $f (keys %$modify_record) { |
foreach my $f (keys %$modify_record) { |
585 |
warn "--- f: $f\n"; |
$log->debug("field: $f"); |
586 |
|
|
587 |
foreach my $sf (keys %{ $modify_record->{$f} }) { |
foreach my $sf (keys %{ $modify_record->{$f} }) { |
588 |
warn "---- sf: $sf\n"; |
$log->debug("subfield: $sf"); |
589 |
|
|
590 |
foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) { |
foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) { |
591 |
my $to = $modify_record->{$f}->{$sf}->{$from}; |
my $to = $modify_record->{$f}->{$sf}->{$from}; |
592 |
#die "no field?" unless defined($to); |
#die "no field?" unless defined($to); |
593 |
warn "----- transform: |$from| -> |$to|\n"; |
$log->debug("transform: |$from| -> |$to|"); |
|
|
|
|
if ($sf =~ /^\^/) { |
|
|
my $regex = |
|
|
's/\Q'. $sf .'\E([^\^]+)\Q'. $from .'\E([^\^]+)/'. $sf .'$1'. $to .'$2/g'; |
|
|
push @{ $regexpes->{$f} }, $regex; |
|
|
warn ">>>>> $regex [sf]\n"; |
|
|
} else { |
|
|
my $regex = |
|
|
's/\Q'. $from .'\E/'. $to .'/g'; |
|
|
push @{ $regexpes->{$f} }, $regex; |
|
|
warn ">>>>> $regex [global]\n"; |
|
|
} |
|
594 |
|
|
595 |
|
my $regex = _get_regex($sf,$from,$to); |
596 |
|
push @{ $regexpes->{$f} }, $regex; |
597 |
|
$log->debug("regex: $regex"); |
598 |
} |
} |
599 |
} |
} |
600 |
} |
} |
601 |
|
|
602 |
|
return $regexpes; |
603 |
|
} |
604 |
|
|
605 |
|
=head2 modify_file_regexps |
606 |
|
|
607 |
|
Generate hash with regexpes to be applied using l<filter> from |
608 |
|
pseudo hash/yaml format for regex mappings. |
609 |
|
|
610 |
|
It should be obvious: |
611 |
|
|
612 |
|
200 |
613 |
|
'^a' |
614 |
|
' : ' => '^e' |
615 |
|
' = ' => '^d' |
616 |
|
|
617 |
|
In field I<200> find C<'^a'> and then C<' : '>, and replace it with C<'^e'>. |
618 |
|
In field I<200> find C<'^a'> and then C<' = '>, and replace it with C<'^d'>. |
619 |
|
|
620 |
|
my $regexpes = $input->modify_file_regexps( 'conf/modify/common.pl' ); |
621 |
|
|
622 |
|
On undef path it will just return. |
623 |
|
|
624 |
|
=cut |
625 |
|
|
626 |
|
sub modify_file_regexps { |
627 |
|
my $self = shift; |
628 |
|
|
629 |
|
my $modify_path = shift || return; |
630 |
|
|
631 |
|
my $log = $self->_get_logger(); |
632 |
|
|
633 |
|
my $regexpes; |
634 |
|
|
635 |
|
CORE::open(my $fh, $modify_path) || $log->logdie("can't open modify file $modify_path: $!"); |
636 |
|
|
637 |
|
my ($f,$sf); |
638 |
|
|
639 |
|
while(<$fh>) { |
640 |
|
chomp; |
641 |
|
next if (/^#/ || /^\s*$/); |
642 |
|
|
643 |
|
if (/^\s*(\d+)\s*$/) { |
644 |
|
$f = $1; |
645 |
|
$log->debug("field: $f"); |
646 |
|
next; |
647 |
|
} elsif (/^\s*'([^']*)'\s*$/) { |
648 |
|
$sf = $1; |
649 |
|
$log->die("can't define subfiled before field in: $_") unless ($f); |
650 |
|
$log->debug("subfield: $sf"); |
651 |
|
} elsif (/^\s*'([^']*)'\s*=>\s*'([^']*)'\s*$/) { |
652 |
|
my ($from,$to) = ($1, $2); |
653 |
|
|
654 |
|
$log->debug("transform: |$from| -> |$to|"); |
655 |
|
|
656 |
|
my $regex = _get_regex($sf,$from,$to); |
657 |
|
push @{ $regexpes->{$f} }, $regex; |
658 |
|
$log->debug("regex: $regex"); |
659 |
|
} |
660 |
|
} |
661 |
|
|
662 |
return $regexpes; |
return $regexpes; |
663 |
} |
} |
664 |
|
|