3 |
use warnings; |
use warnings; |
4 |
use strict; |
use strict; |
5 |
|
|
6 |
use blib; |
use lib 'lib'; |
7 |
|
|
8 |
use WebPAC::Common; |
use WebPAC::Common; |
9 |
use base qw/WebPAC::Common/; |
use base qw/WebPAC::Common/; |
10 |
use Data::Dump qw/dump/; |
use Data::Dump qw/dump/; |
11 |
use Encode qw/from_to/; |
use Encode qw/decode from_to/; |
12 |
|
use YAML; |
13 |
|
|
14 |
=head1 NAME |
=head1 NAME |
15 |
|
|
16 |
WebPAC::Input - read different file formats into WebPAC |
WebPAC::Input - read different file formats into WebPAC |
17 |
|
|
|
=head1 VERSION |
|
|
|
|
|
Version 0.17 |
|
|
|
|
18 |
=cut |
=cut |
19 |
|
|
20 |
our $VERSION = '0.17'; |
our $VERSION = '0.19'; |
21 |
|
|
22 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
23 |
|
|
58 |
|
|
59 |
my $db = new WebPAC::Input( |
my $db = new WebPAC::Input( |
60 |
module => 'WebPAC::Input::MARC', |
module => 'WebPAC::Input::MARC', |
|
encoding => 'ISO-8859-2', |
|
61 |
recode => 'char pairs', |
recode => 'char pairs', |
62 |
no_progress_bar => 1, |
no_progress_bar => 1, |
63 |
|
input_config => { |
64 |
|
mapping => [ 'foo', 'bar', 'baz' ], |
65 |
|
}, |
66 |
); |
); |
67 |
|
|
68 |
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 |
69 |
L<WebPAC::Input::MARC>. |
L<WebPAC::Input::MARC>. |
70 |
|
|
|
Optional parametar C<encoding> specify application code page (which will be |
|
|
used internally). This should probably be your terminal encoding, and by |
|
|
default, it C<ISO-8859-2>. |
|
|
|
|
71 |
C<recode> is optional string constisting of character or words pairs that |
C<recode> is optional string constisting of character or words pairs that |
72 |
should be replaced in input stream. |
should be replaced in input stream. |
73 |
|
|
85 |
|
|
86 |
my $log = $self->_get_logger; |
my $log = $self->_get_logger; |
87 |
|
|
88 |
$log->logconfess("code_page argument is not suppored any more. change it to encoding") if ($self->{lookup}); |
$log->logconfess("code_page argument is not suppored any more.") if $self->{code_page}; |
89 |
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_ref") if ($self->{lookup}); |
$log->logconfess("encoding argument is not suppored any more.") if $self->{encoding}; |
90 |
$log->logconfess("low_mem argument is not suppored any more. rewrite it to load_row and save_row") if ($self->{low_mem}); |
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_ref") if $self->{lookup}; |
91 |
|
$log->logconfess("low_mem argument is not suppored any more. rewrite it to load_row and save_row") if $self->{low_mem}; |
92 |
|
|
93 |
$log->logconfess("specify low-level file format module") unless ($self->{module}); |
$log->logconfess("specify low-level file format module") unless ($self->{module}); |
94 |
my $module_path = $self->{module}; |
my $module_path = $self->{module}; |
98 |
|
|
99 |
require $module_path; |
require $module_path; |
100 |
|
|
|
$self->{'encoding'} ||= 'ISO-8859-2'; |
|
|
|
|
101 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
102 |
} |
} |
103 |
|
|
109 |
|
|
110 |
$input->open( |
$input->open( |
111 |
path => '/path/to/database/file', |
path => '/path/to/database/file', |
112 |
code_page => 'cp852', |
input_encoding => 'cp852', |
113 |
|
strict_encoding => 0, |
114 |
limit => 500, |
limit => 500, |
115 |
offset => 6000, |
offset => 6000, |
116 |
stats => 1, |
stats => 1, |
135 |
|
|
136 |
); |
); |
137 |
|
|
138 |
By default, C<code_page> is assumed to be C<cp852>. |
By default, C<input_encoding> is assumed to be C<cp852>. |
139 |
|
|
140 |
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. |
141 |
|
|
156 |
C<save_row> and C<load_row> are low-level implementation of store engine. Calling convention |
C<save_row> and C<load_row> are low-level implementation of store engine. Calling convention |
157 |
is documented in example above. |
is documented in example above. |
158 |
|
|
159 |
|
C<strict_encoding> should really default to 1, but it doesn't for now. |
160 |
|
|
161 |
Returns size of database, regardless of C<offset> and C<limit> |
Returns size of database, regardless of C<offset> and C<limit> |
162 |
parametars, see also C<size>. |
parametars, see also C<size>. |
163 |
|
|
168 |
my $arg = {@_}; |
my $arg = {@_}; |
169 |
|
|
170 |
my $log = $self->_get_logger(); |
my $log = $self->_get_logger(); |
171 |
|
$log->debug( "arguments: ",dump( $arg )); |
172 |
|
|
173 |
|
$log->logconfess("encoding argument is not suppored any more.") if $self->{encoding}; |
174 |
|
$log->logconfess("code_page argument is not suppored any more.") if $self->{code_page}; |
175 |
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_coderef") if ($arg->{lookup}); |
$log->logconfess("lookup argument is not suppored any more. rewrite call to lookup_coderef") if ($arg->{lookup}); |
176 |
$log->logconfess("lookup_coderef must be CODE, not ",ref($arg->{lookup_coderef})) |
$log->logconfess("lookup_coderef must be CODE, not ",ref($arg->{lookup_coderef})) |
177 |
if ($arg->{lookup_coderef} && ref($arg->{lookup_coderef}) ne 'CODE'); |
if ($arg->{lookup_coderef} && ref($arg->{lookup_coderef}) ne 'CODE'); |
179 |
$log->debug( $arg->{lookup_coderef} ? '' : 'not ', "using lookup_coderef"); |
$log->debug( $arg->{lookup_coderef} ? '' : 'not ', "using lookup_coderef"); |
180 |
|
|
181 |
$log->logcroak("need path") if (! $arg->{'path'}); |
$log->logcroak("need path") if (! $arg->{'path'}); |
182 |
my $code_page = $arg->{'code_page'} || 'cp852'; |
my $input_encoding = $arg->{'input_encoding'} || $self->{'input_encoding'} || 'cp852'; |
183 |
|
|
184 |
# store data in object |
# store data in object |
|
$self->{'input_code_page'} = $code_page; |
|
185 |
foreach my $v (qw/path offset limit/) { |
foreach my $v (qw/path offset limit/) { |
186 |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
$self->{$v} = $arg->{$v} if ($arg->{$v}); |
187 |
} |
} |
232 |
|
|
233 |
my $ll_db = $class->new( |
my $ll_db = $class->new( |
234 |
path => $arg->{path}, |
path => $arg->{path}, |
235 |
|
input_config => $arg->{input_config} || $self->{input_config}, |
236 |
# filter => sub { |
# filter => sub { |
237 |
# my ($l,$f_nr) = @_; |
# my ($l,$f_nr) = @_; |
238 |
# return unless defined($l); |
# return unless defined($l); |
239 |
# from_to($l, $code_page, $self->{'encoding'}); |
# $l = decode($input_encoding, $l); |
240 |
# $l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
# $l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
241 |
# return $l; |
# return $l; |
242 |
# }, |
# }, |
274 |
# store size for later |
# store size for later |
275 |
$self->{size} = ($to_rec - $from_rec) ? ($to_rec - $from_rec + 1) : 0; |
$self->{size} = ($to_rec - $from_rec) ? ($to_rec - $from_rec + 1) : 0; |
276 |
|
|
277 |
$log->info("processing $self->{size}/$size records [$from_rec-$to_rec] convert $code_page -> $self->{encoding}", $self->{stats} ? ' [stats]' : ''); |
my $strict_encoding = $arg->{strict_encoding} || $self->{strict_encoding}; ## FIXME should be 1 really |
278 |
|
|
279 |
|
$log->info("processing $self->{size}/$size records [$from_rec-$to_rec]", |
280 |
|
" encoding $input_encoding ", $strict_encoding ? ' [strict]' : '', |
281 |
|
$self->{stats} ? ' [stats]' : '', |
282 |
|
); |
283 |
|
|
284 |
# read database |
# read database |
285 |
for (my $pos = $from_rec; $pos <= $to_rec; $pos++) { |
for (my $pos = $from_rec; $pos <= $to_rec; $pos++) { |
287 |
$log->debug("position: $pos\n"); |
$log->debug("position: $pos\n"); |
288 |
|
|
289 |
my $rec = $ll_db->fetch_rec($pos, sub { |
my $rec = $ll_db->fetch_rec($pos, sub { |
290 |
my ($l,$f_nr) = @_; |
my ($l,$f_nr,$debug) = @_; |
291 |
# return unless defined($l); |
# return unless defined($l); |
292 |
# return $l unless ($rec_regex && $f_nr); |
# return $l unless ($rec_regex && $f_nr); |
293 |
|
|
294 |
|
return unless ( defined($l) && defined($f_nr) ); |
295 |
|
|
296 |
|
warn "-=> $f_nr ## |$l|\n" if ($debug); |
297 |
$log->debug("-=> $f_nr ## $l"); |
$log->debug("-=> $f_nr ## $l"); |
298 |
|
|
299 |
# codepage conversion and recode_regex |
# codepage conversion and recode_regex |
300 |
from_to($l, $code_page, $self->{'encoding'}); |
$l = decode($input_encoding, $l, 1); |
301 |
$l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
$l =~ s/($recode_regex)/$recode_map->{$1}/g if ($recode_regex && $recode_map); |
302 |
|
|
303 |
# apply regexps |
# apply regexps |
306 |
my $c = 0; |
my $c = 0; |
307 |
foreach my $r (@{ $rec_regex->{$f_nr} }) { |
foreach my $r (@{ $rec_regex->{$f_nr} }) { |
308 |
my $old_l = $l; |
my $old_l = $l; |
309 |
eval '$l =~ ' . $r; |
$log->logconfess("expected regex in ", dump( $r )) unless defined($r->{regex}); |
310 |
|
eval '$l =~ ' . $r->{regex}; |
311 |
if ($old_l ne $l) { |
if ($old_l ne $l) { |
312 |
$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}; |
313 |
|
$d .= ' +' . $r->{line} . ' ' . $r->{file} if defined($r->{line}); |
314 |
|
$d .= ' ' . $r->{debug} if defined($r->{debug}); |
315 |
|
$log->debug("MODIFY $d"); |
316 |
|
warn "*** $d\n" if ($debug); |
317 |
|
|
318 |
} |
} |
319 |
$log->error("error applying regex: $r") if ($@); |
$log->error("error applying regex: ",dump($r), $@) if $@; |
320 |
} |
} |
321 |
} |
} |
322 |
|
|
323 |
$log->debug("<=- $f_nr ## $l"); |
$log->debug("<=- $f_nr ## |$l|"); |
324 |
|
warn "<=- $f_nr ## $l\n" if ($debug); |
325 |
return $l; |
return $l; |
326 |
}); |
}); |
327 |
|
|
354 |
foreach my $fld (keys %{ $rec }) { |
foreach my $fld (keys %{ $rec }) { |
355 |
$self->{_stats}->{fld}->{ $fld }++; |
$self->{_stats}->{fld}->{ $fld }++; |
356 |
|
|
357 |
$log->logdie("invalid record fild $fld, not ARRAY") |
#$log->logdie("invalid record fild $fld, not ARRAY") |
358 |
unless (ref($rec->{ $fld }) eq 'ARRAY'); |
next unless (ref($rec->{ $fld }) eq 'ARRAY'); |
359 |
|
|
360 |
foreach my $row (@{ $rec->{$fld} }) { |
foreach my $row (@{ $rec->{$fld} }) { |
361 |
|
|
488 |
|
|
489 |
my $log = $self->_get_logger(); |
my $log = $self->_get_logger(); |
490 |
|
|
491 |
$log->confess("called without pos") unless defined($pos); |
$log->logconfess("called without pos") unless defined($pos); |
492 |
|
|
493 |
if ($pos < 1) { |
if ($pos < 1) { |
494 |
$log->warn("seek before first record"); |
$log->warn("seek before first record"); |
524 |
|
|
525 |
my $out = join("\n", |
my $out = join("\n", |
526 |
map { |
map { |
527 |
my $f = $_ || die "no field"; |
my $f = $_; |
528 |
|
die "no field in ", dump( $s->{fld} ) unless defined( $f ); |
529 |
my $v = $s->{fld}->{$f} || die "no s->{fld}->{$f}"; |
my $v = $s->{fld}->{$f} || die "no s->{fld}->{$f}"; |
530 |
$max_fld = $v if ($v > $max_fld); |
$max_fld = $v if ($v > $max_fld); |
531 |
|
|
532 |
my $o = sprintf("%4s %d ~", $f, $v); |
my $o = sprintf("%4s %d ~", $f, $v); |
533 |
|
|
534 |
if (defined($s->{sf}->{$f})) { |
if (defined($s->{sf}->{$f})) { |
535 |
|
my @subfields = keys %{ $s->{sf}->{$f} }; |
536 |
map { |
map { |
537 |
$o .= sprintf(" %s:%d%s", $_, |
$o .= sprintf(" %s:%d%s", $_, |
538 |
$s->{sf}->{$f}->{$_}->{count}, |
$s->{sf}->{$f}->{$_}->{count}, |
539 |
$s->{sf}->{$f}->{$_}->{repeatable} ? '*' : '', |
$s->{sf}->{$f}->{$_}->{repeatable} ? '*' : '', |
540 |
); |
); |
541 |
} sort keys %{ $s->{sf}->{$f} }; |
} ( |
542 |
|
# first indicators and other special subfields |
543 |
|
sort( grep { length($_) > 1 } @subfields ), |
544 |
|
# then subfileds (single char) |
545 |
|
sort( grep { length($_) == 1 } @subfields ), |
546 |
|
); |
547 |
} |
} |
548 |
|
|
549 |
if (my $v_r = $s->{repeatable}->{$f}) { |
if (my $v_r = $s->{repeatable}->{$f}) { |
551 |
} |
} |
552 |
|
|
553 |
$o; |
$o; |
554 |
} sort { $a cmp $b } keys %{ $s->{fld} } |
} sort { |
555 |
|
if ( $a =~ m/^\d+$/ && $b =~ m/^\d+$/ ) { |
556 |
|
$a <=> $b |
557 |
|
} else { |
558 |
|
$a cmp $b |
559 |
|
} |
560 |
|
} keys %{ $s->{fld} } |
561 |
); |
); |
562 |
|
|
563 |
$log->debug( sub { dump($s) } ); |
$log->debug( sub { dump($s) } ); |
564 |
|
|
565 |
|
my $path = 'var/stats.yml'; |
566 |
|
YAML::DumpFile( $path, $s ); |
567 |
|
$log->info( 'created ', $path, ' with ', -s $path, ' bytes' ); |
568 |
|
|
569 |
return $out; |
return $out; |
570 |
} |
} |
571 |
|
|
580 |
|
|
581 |
return unless $self->{ll_db}; |
return unless $self->{ll_db}; |
582 |
|
|
583 |
if ($self->{ll_db}->can('dump_rec')) { |
if ($self->{ll_db}->can('dump_ascii')) { |
584 |
return $self->{ll_db}->dump_ascii( $self->{pos} ); |
return $self->{ll_db}->dump_ascii( $self->{pos} ); |
585 |
} else { |
} else { |
586 |
return dump( $self->{ll_db}->fetch_rec( $self->{pos} ) ); |
return dump( $self->{ll_db}->fetch_rec( $self->{pos} ) ); |
587 |
} |
} |
588 |
} |
} |
589 |
|
|
590 |
=head2 modify_record_regexps |
=head2 _get_regex |
591 |
|
|
592 |
Generate hash with regexpes to be applied using l<filter>. |
Helper function called which create regexps to be execute on code. |
593 |
|
|
594 |
my $regexpes = $input->modify_record_regexps( |
_get_regex( 900, 'regex:[0-9]+' ,'numbers' ); |
595 |
900 => { '^a' => { ' : ' => '^b' } }, |
_get_regex( 900, '^b', ' : ^b' ); |
596 |
901 => { '*' => { '^b' => ' ; ' } }, |
|
597 |
); |
It supports perl regexps with C<regex:> prefix to from value and has |
598 |
|
additional logic to skip empty subfields. |
599 |
|
|
600 |
=cut |
=cut |
601 |
|
|
602 |
sub _get_regex { |
sub _get_regex { |
603 |
my ($sf,$from,$to) = @_; |
my ($sf,$from,$to) = @_; |
604 |
|
|
605 |
|
# protect / |
606 |
|
$from =~ s!/!\\/!gs; |
607 |
|
$to =~ s!/!\\/!gs; |
608 |
|
|
609 |
if ($from =~ m/^regex:(.+)$/) { |
if ($from =~ m/^regex:(.+)$/) { |
610 |
$from = $1; |
$from = $1; |
611 |
} else { |
} else { |
612 |
$from = '\Q' . $from . '\E'; |
$from = '\Q' . $from . '\E'; |
613 |
} |
} |
614 |
if ($sf =~ /^\^/) { |
if ($sf =~ /^\^/) { |
615 |
|
my $need_subfield_data = '*'; # no |
616 |
|
# if from is also subfield, require some data in between |
617 |
|
# to correctly skip empty subfields |
618 |
|
$need_subfield_data = '+' if ($from =~ m/^\\Q\^/); |
619 |
return |
return |
620 |
's/\Q'. $sf .'\E([^\^]*?)'. $from .'([^\^]*?)/'. $sf .'$1'. $to .'$2/'; |
's/\Q'. $sf .'\E([^\^]' . $need_subfield_data . '?)'. $from .'([^\^]*?)/'. $sf .'$1'. $to .'$2/'; |
621 |
} else { |
} else { |
622 |
return |
return |
623 |
's/'. $from .'/'. $to .'/g'; |
's/'. $from .'/'. $to .'/g'; |
624 |
} |
} |
625 |
} |
} |
626 |
|
|
627 |
|
|
628 |
|
=head2 modify_record_regexps |
629 |
|
|
630 |
|
Generate hash with regexpes to be applied using L<filter>. |
631 |
|
|
632 |
|
my $regexpes = $input->modify_record_regexps( |
633 |
|
900 => { '^a' => { ' : ' => '^b' } }, |
634 |
|
901 => { '*' => { '^b' => ' ; ' } }, |
635 |
|
); |
636 |
|
|
637 |
|
=cut |
638 |
|
|
639 |
sub modify_record_regexps { |
sub modify_record_regexps { |
640 |
my $self = shift; |
my $self = shift; |
641 |
my $modify_record = {@_}; |
my $modify_record = {@_}; |
653 |
foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) { |
foreach my $from (keys %{ $modify_record->{$f}->{$sf} }) { |
654 |
my $to = $modify_record->{$f}->{$sf}->{$from}; |
my $to = $modify_record->{$f}->{$sf}->{$from}; |
655 |
#die "no field?" unless defined($to); |
#die "no field?" unless defined($to); |
656 |
$log->debug("transform: |$from| -> |$to|"); |
my $d = "|$from| -> |$to|"; |
657 |
|
$log->debug("transform: $d"); |
658 |
|
|
659 |
my $regex = _get_regex($sf,$from,$to); |
my $regex = _get_regex($sf,$from,$to); |
660 |
push @{ $regexpes->{$f} }, $regex; |
push @{ $regexpes->{$f} }, { regex => $regex, debug => $d }; |
661 |
$log->debug("regex: $regex"); |
$log->debug("regex: $regex"); |
662 |
} |
} |
663 |
} |
} |
668 |
|
|
669 |
=head2 modify_file_regexps |
=head2 modify_file_regexps |
670 |
|
|
671 |
Generate hash with regexpes to be applied using l<filter> from |
Generate hash with regexpes to be applied using L<filter> from |
672 |
pseudo hash/yaml format for regex mappings. |
pseudo hash/yaml format for regex mappings. |
673 |
|
|
674 |
It should be obvious: |
It should be obvious: |
718 |
$log->debug("transform: |$from| -> |$to|"); |
$log->debug("transform: |$from| -> |$to|"); |
719 |
|
|
720 |
my $regex = _get_regex($sf,$from,$to); |
my $regex = _get_regex($sf,$from,$to); |
721 |
push @{ $regexpes->{$f} }, $regex; |
push @{ $regexpes->{$f} }, { |
722 |
|
regex => $regex, |
723 |
|
file => $modify_path, |
724 |
|
line => $., |
725 |
|
}; |
726 |
$log->debug("regex: $regex"); |
$log->debug("regex: $regex"); |
727 |
|
} else { |
728 |
|
die "can't parse: $_"; |
729 |
} |
} |
730 |
} |
} |
731 |
|
|