7 |
|
|
8 |
use WebPAC::Common; |
use WebPAC::Common; |
9 |
use base qw/WebPAC::Common/; |
use base qw/WebPAC::Common/; |
10 |
use Data::Dumper; |
use Data::Dump qw/dump/; |
11 |
use Encode qw/from_to/; |
use Encode qw/from_to/; |
12 |
|
|
13 |
=head1 NAME |
=head1 NAME |
16 |
|
|
17 |
=head1 VERSION |
=head1 VERSION |
18 |
|
|
19 |
Version 0.14 |
Version 0.18 |
20 |
|
|
21 |
=cut |
=cut |
22 |
|
|
23 |
our $VERSION = '0.14'; |
our $VERSION = '0.18'; |
24 |
|
|
25 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
26 |
|
|
64 |
encoding => 'ISO-8859-2', |
encoding => 'ISO-8859-2', |
65 |
recode => 'char pairs', |
recode => 'char pairs', |
66 |
no_progress_bar => 1, |
no_progress_bar => 1, |
67 |
|
input_config => { |
68 |
|
mapping => [ 'foo', 'bar', 'baz' ], |
69 |
|
}, |
70 |
); |
); |
71 |
|
|
72 |
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 |
105 |
|
|
106 |
require $module_path; |
require $module_path; |
107 |
|
|
|
# check if required subclasses are implemented |
|
|
foreach my $subclass (qw/open_db fetch_rec init dump_rec/) { |
|
|
# FIXME |
|
|
} |
|
|
|
|
108 |
$self->{'encoding'} ||= 'ISO-8859-2'; |
$self->{'encoding'} ||= 'ISO-8859-2'; |
109 |
|
|
110 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
227 |
$log->debug("using modify_file $p"); |
$log->debug("using modify_file $p"); |
228 |
$rec_regex = $self->modify_file_regexps( $p ); |
$rec_regex = $self->modify_file_regexps( $p ); |
229 |
} elsif (my $h = $arg->{modify_records}) { |
} elsif (my $h = $arg->{modify_records}) { |
230 |
$log->debug("using modify_records ", Dumper( $h )); |
$log->debug("using modify_records ", sub { dump( $h ) }); |
231 |
$rec_regex = $self->modify_record_regexps(%{ $h }); |
$rec_regex = $self->modify_record_regexps(%{ $h }); |
232 |
} |
} |
233 |
$log->debug("rec_regex: ", Dumper($rec_regex)) if ($rec_regex); |
$log->debug("rec_regex: ", sub { dump($rec_regex) }) if ($rec_regex); |
234 |
|
|
235 |
my $class = $self->{module} || $log->logconfess("can't get low-level module name!"); |
my $class = $self->{module} || $log->logconfess("can't get low-level module name!"); |
236 |
|
|
237 |
my $ll_db = $class->new( |
my $ll_db = $class->new( |
238 |
path => $arg->{path}, |
path => $arg->{path}, |
239 |
|
input_config => $arg->{input_config} || $self->{input_config}, |
240 |
# filter => sub { |
# filter => sub { |
241 |
# my ($l,$f_nr) = @_; |
# my ($l,$f_nr) = @_; |
242 |
# return unless defined($l); |
# return unless defined($l); |
286 |
$log->debug("position: $pos\n"); |
$log->debug("position: $pos\n"); |
287 |
|
|
288 |
my $rec = $ll_db->fetch_rec($pos, sub { |
my $rec = $ll_db->fetch_rec($pos, sub { |
289 |
my ($l,$f_nr) = @_; |
my ($l,$f_nr,$debug) = @_; |
290 |
# return unless defined($l); |
# return unless defined($l); |
291 |
# return $l unless ($rec_regex && $f_nr); |
# return $l unless ($rec_regex && $f_nr); |
292 |
|
|
293 |
|
return unless ( defined($l) && defined($f_nr) ); |
294 |
|
|
295 |
|
warn "-=> $f_nr ## |$l|\n" if ($debug); |
296 |
$log->debug("-=> $f_nr ## $l"); |
$log->debug("-=> $f_nr ## $l"); |
297 |
|
|
298 |
# codepage conversion and recode_regex |
# codepage conversion and recode_regex |
305 |
my $c = 0; |
my $c = 0; |
306 |
foreach my $r (@{ $rec_regex->{$f_nr} }) { |
foreach my $r (@{ $rec_regex->{$f_nr} }) { |
307 |
my $old_l = $l; |
my $old_l = $l; |
308 |
eval '$l =~ ' . $r; |
$log->logconfess("expected regex in ", dump( $r )) unless defined($r->{regex}); |
309 |
|
eval '$l =~ ' . $r->{regex}; |
310 |
if ($old_l ne $l) { |
if ($old_l ne $l) { |
311 |
$log->debug("REGEX on $f_nr eval \$l =~ $r\n## old l: [$old_l]\n## new l: [$l]"); |
my $d = "|$old_l| -> |$l| "; # . $r->{regex}; |
312 |
|
$d .= ' +' . $r->{line} . ' ' . $r->{file} if defined($r->{line}); |
313 |
|
$d .= ' ' . $r->{debug} if defined($r->{debug}); |
314 |
|
$log->debug("MODIFY $d"); |
315 |
|
warn "*** $d\n" if ($debug); |
316 |
|
|
317 |
} |
} |
318 |
$log->error("error applying regex: $r") if ($@); |
$log->error("error applying regex: $r") if ($@); |
319 |
} |
} |
320 |
} |
} |
321 |
|
|
322 |
$log->debug("<=- $f_nr ## $l"); |
$log->debug("<=- $f_nr ## |$l|"); |
323 |
|
warn "<=- $f_nr ## $l\n" if ($debug); |
324 |
return $l; |
return $l; |
325 |
}); |
}); |
326 |
|
|
327 |
$log->debug(sub { Dumper($rec) }); |
$log->debug(sub { dump($rec) }); |
328 |
|
|
329 |
if (! $rec) { |
if (! $rec) { |
330 |
$log->warn("record $pos empty? skipping..."); |
$log->warn("record $pos empty? skipping..."); |
353 |
foreach my $fld (keys %{ $rec }) { |
foreach my $fld (keys %{ $rec }) { |
354 |
$self->{_stats}->{fld}->{ $fld }++; |
$self->{_stats}->{fld}->{ $fld }++; |
355 |
|
|
356 |
$log->logdie("invalid record fild $fld, not ARRAY") |
#$log->logdie("invalid record fild $fld, not ARRAY") |
357 |
unless (ref($rec->{ $fld }) eq 'ARRAY'); |
next unless (ref($rec->{ $fld }) eq 'ARRAY'); |
358 |
|
|
359 |
foreach my $row (@{ $rec->{$fld} }) { |
foreach my $row (@{ $rec->{$fld} }) { |
360 |
|
|
483 |
|
|
484 |
sub seek { |
sub seek { |
485 |
my $self = shift; |
my $self = shift; |
486 |
my $pos = shift || return; |
my $pos = shift; |
487 |
|
|
488 |
my $log = $self->_get_logger(); |
my $log = $self->_get_logger(); |
489 |
|
|
490 |
|
$log->logconfess("called without pos") unless defined($pos); |
491 |
|
|
492 |
if ($pos < 1) { |
if ($pos < 1) { |
493 |
$log->warn("seek before first record"); |
$log->warn("seek before first record"); |
494 |
$pos = 1; |
$pos = 1; |
523 |
|
|
524 |
my $out = join("\n", |
my $out = join("\n", |
525 |
map { |
map { |
526 |
my $f = $_ || die "no field"; |
my $f = $_; |
527 |
|
die "no field in ", dump( $s->{fld} ) unless defined( $f ); |
528 |
my $v = $s->{fld}->{$f} || die "no s->{fld}->{$f}"; |
my $v = $s->{fld}->{$f} || die "no s->{fld}->{$f}"; |
529 |
$max_fld = $v if ($v > $max_fld); |
$max_fld = $v if ($v > $max_fld); |
530 |
|
|
531 |
my $o = sprintf("%4s %d ~", $f, $v); |
my $o = sprintf("%4s %d ~", $f, $v); |
532 |
|
|
533 |
if (defined($s->{sf}->{$f})) { |
if (defined($s->{sf}->{$f})) { |
534 |
|
my @subfields = keys %{ $s->{sf}->{$f} }; |
535 |
map { |
map { |
536 |
$o .= sprintf(" %s:%d%s", $_, |
$o .= sprintf(" %s:%d%s", $_, |
537 |
$s->{sf}->{$f}->{$_}->{count}, |
$s->{sf}->{$f}->{$_}->{count}, |
538 |
$s->{sf}->{$f}->{$_}->{repeatable} ? '*' : '', |
$s->{sf}->{$f}->{$_}->{repeatable} ? '*' : '', |
539 |
); |
); |
540 |
} sort keys %{ $s->{sf}->{$f} }; |
} ( |
541 |
|
# first indicators and other special subfields |
542 |
|
sort( grep { length($_) > 1 } @subfields ), |
543 |
|
# then subfileds (single char) |
544 |
|
sort( grep { length($_) == 1 } @subfields ), |
545 |
|
); |
546 |
} |
} |
547 |
|
|
548 |
if (my $v_r = $s->{repeatable}->{$f}) { |
if (my $v_r = $s->{repeatable}->{$f}) { |
550 |
} |
} |
551 |
|
|
552 |
$o; |
$o; |
553 |
} sort { $a cmp $b } keys %{ $s->{fld} } |
} sort { |
554 |
|
if ( $a =~ m/^\d+$/ && $b =~ m/^\d+$/ ) { |
555 |
|
$a <=> $b |
556 |
|
} else { |
557 |
|
$a cmp $b |
558 |
|
} |
559 |
|
} keys %{ $s->{fld} } |
560 |
); |
); |
561 |
|
|
562 |
$log->debug( sub { Dumper($s) } ); |
$log->debug( sub { dump($s) } ); |
563 |
|
|
564 |
return $out; |
return $out; |
565 |
} |
} |
566 |
|
|
567 |
=head2 dump |
=head2 dump_ascii |
568 |
|
|
569 |
Display humanly readable dump of record |
Display humanly readable dump of record |
570 |
|
|
571 |
=cut |
=cut |
572 |
|
|
573 |
sub dump { |
sub dump_ascii { |
574 |
my $self = shift; |
my $self = shift; |
575 |
|
|
576 |
return $self->{ll_db}->dump_rec( $self->{pos} ); |
return unless $self->{ll_db}; |
577 |
|
|
578 |
|
if ($self->{ll_db}->can('dump_ascii')) { |
579 |
|
return $self->{ll_db}->dump_ascii( $self->{pos} ); |
580 |
|
} else { |
581 |
|
return dump( $self->{ll_db}->fetch_rec( $self->{pos} ) ); |
582 |
|
} |
583 |
} |
} |
584 |
|
|
585 |
=head2 modify_record_regexps |
=head2 _get_regex |
586 |
|
|
587 |
Generate hash with regexpes to be applied using l<filter>. |
Helper function called which create regexps to be execute on code. |
588 |
|
|
589 |
my $regexpes = $input->modify_record_regexps( |
_get_regex( 900, 'regex:[0-9]+' ,'numbers' ); |
590 |
900 => { '^a' => { ' : ' => '^b' } }, |
_get_regex( 900, '^b', ' : ^b' ); |
591 |
901 => { '*' => { '^b' => ' ; ' } }, |
|
592 |
); |
It supports perl regexps with C<regex:> prefix to from value and has |
593 |
|
additional logic to skip empty subfields. |
594 |
|
|
595 |
=cut |
=cut |
596 |
|
|
597 |
sub _get_regex { |
sub _get_regex { |
598 |
my ($sf,$from,$to) = @_; |
my ($sf,$from,$to) = @_; |
599 |
|
|
600 |
|
# protect / |
601 |
|
$from =~ s!/!\\/!gs; |
602 |
|
$to =~ s!/!\\/!gs; |
603 |
|
|
604 |
|
if ($from =~ m/^regex:(.+)$/) { |
605 |
|
$from = $1; |
606 |
|
} else { |
607 |
|
$from = '\Q' . $from . '\E'; |
608 |
|
} |
609 |
if ($sf =~ /^\^/) { |
if ($sf =~ /^\^/) { |
610 |
|
my $need_subfield_data = '*'; # no |
611 |
|
# if from is also subfield, require some data in between |
612 |
|
# to correctly skip empty subfields |
613 |
|
$need_subfield_data = '+' if ($from =~ m/^\\Q\^/); |
614 |
return |
return |
615 |
's/\Q'. $sf .'\E([^\^]*?)\Q'. $from .'\E([^\^]*?)/'. $sf .'$1'. $to .'$2/'; |
's/\Q'. $sf .'\E([^\^]' . $need_subfield_data . '?)'. $from .'([^\^]*?)/'. $sf .'$1'. $to .'$2/'; |
616 |
} else { |
} else { |
617 |
return |
return |
618 |
's/\Q'. $from .'\E/'. $to .'/g'; |
's/'. $from .'/'. $to .'/g'; |
619 |
} |
} |
620 |
} |
} |
621 |
|
|
622 |
|
|
623 |
|
=head2 modify_record_regexps |
624 |
|
|
625 |
|
Generate hash with regexpes to be applied using L<filter>. |
626 |
|
|
627 |
|
my $regexpes = $input->modify_record_regexps( |
628 |
|
900 => { '^a' => { ' : ' => '^b' } }, |
629 |
|
901 => { '*' => { '^b' => ' ; ' } }, |
630 |
|
); |
631 |
|
|
632 |
|
=cut |
633 |
|
|
634 |
sub modify_record_regexps { |
sub modify_record_regexps { |
635 |
my $self = shift; |
my $self = shift; |
636 |
my $modify_record = {@_}; |
my $modify_record = {@_}; |
648 |
foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) { |
foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) { |
649 |
my $to = $modify_record->{$f}->{$sf}->{$from}; |
my $to = $modify_record->{$f}->{$sf}->{$from}; |
650 |
#die "no field?" unless defined($to); |
#die "no field?" unless defined($to); |
651 |
$log->debug("transform: |$from| -> |$to|"); |
my $d = "|$from| -> |$to|"; |
652 |
|
$log->debug("transform: $d"); |
653 |
|
|
654 |
my $regex = _get_regex($sf,$from,$to); |
my $regex = _get_regex($sf,$from,$to); |
655 |
push @{ $regexpes->{$f} }, $regex; |
push @{ $regexpes->{$f} }, { regex => $regex, debug => $d }; |
656 |
$log->debug("regex: $regex"); |
$log->debug("regex: $regex"); |
657 |
} |
} |
658 |
} |
} |
663 |
|
|
664 |
=head2 modify_file_regexps |
=head2 modify_file_regexps |
665 |
|
|
666 |
Generate hash with regexpes to be applied using l<filter> from |
Generate hash with regexpes to be applied using L<filter> from |
667 |
pseudo hash/yaml format for regex mappings. |
pseudo hash/yaml format for regex mappings. |
668 |
|
|
669 |
It should be obvious: |
It should be obvious: |
713 |
$log->debug("transform: |$from| -> |$to|"); |
$log->debug("transform: |$from| -> |$to|"); |
714 |
|
|
715 |
my $regex = _get_regex($sf,$from,$to); |
my $regex = _get_regex($sf,$from,$to); |
716 |
push @{ $regexpes->{$f} }, $regex; |
push @{ $regexpes->{$f} }, { |
717 |
|
regex => $regex, |
718 |
|
file => $modify_path, |
719 |
|
line => $., |
720 |
|
}; |
721 |
$log->debug("regex: $regex"); |
$log->debug("regex: $regex"); |
722 |
} |
} |
723 |
} |
} |