2 |
|
|
3 |
use strict; |
use strict; |
4 |
use Carp; |
use Carp; |
5 |
use Data::Dumper; |
use Data::Dump qw/dump/; |
6 |
|
|
7 |
BEGIN { |
BEGIN { |
8 |
use Exporter (); |
use Exporter (); |
9 |
use vars qw ($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS); |
use vars qw ($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS); |
10 |
$VERSION = 0.08; |
$VERSION = 0.10; |
11 |
@ISA = qw (Exporter); |
@ISA = qw (Exporter); |
12 |
#Give a hoot don't pollute, do not export more than needed by default |
#Give a hoot don't pollute, do not export more than needed by default |
13 |
@EXPORT = qw (); |
@EXPORT = qw (); |
87 |
my $len = read($self->{fh}, $leader, 24); |
my $len = read($self->{fh}, $leader, 24); |
88 |
|
|
89 |
if ($len < 24) { |
if ($len < 24) { |
90 |
carp "short read of leader, aborting\n"; |
warn "short read of leader, aborting\n"; |
91 |
|
$self->{count}--; |
92 |
last; |
last; |
93 |
} |
} |
94 |
|
|
125 |
|
|
126 |
# skip to next record |
# skip to next record |
127 |
my $o = substr($leader,0,5); |
my $o = substr($leader,0,5); |
128 |
|
warn "# in record ", $self->{count}," record length isn't number but: ",dump($o),"\n" unless $o =~ m/^\d+$/; |
129 |
if ($o > 24) { |
if ($o > 24) { |
130 |
seek($self->{fh},$o-24,1) if ($o); |
seek($self->{fh},$o-24,1) if ($o); |
131 |
} else { |
} else { |
277 |
|
|
278 |
Read record with specified MFN and convert it to hash |
Read record with specified MFN and convert it to hash |
279 |
|
|
280 |
my $hash = $marc->to_hash($mfn); |
my $hash = $marc->to_hash( $mfn, include_subfields => 1, ); |
281 |
|
|
282 |
It has ability to convert characters (using C<hash_filter>) from MARC |
It has ability to convert characters (using C<hash_filter>) from MARC |
283 |
database before creating structures enabling character re-mapping or quick |
database before creating structures enabling character re-mapping or quick |
304 |
|
|
305 |
my $mfn = shift || confess "need mfn!"; |
my $mfn = shift || confess "need mfn!"; |
306 |
|
|
307 |
|
my $args = {@_}; |
308 |
|
|
309 |
# init record to include MFN as field 000 |
# init record to include MFN as field 000 |
310 |
my $rec = { '000' => [ $mfn ] }; |
my $rec = { '000' => [ $mfn ] }; |
311 |
|
|
312 |
my $row = $self->fetch($mfn) || return; |
my $row = $self->fetch($mfn) || return; |
313 |
|
|
314 |
foreach my $rec_nr (keys %{$row}) { |
foreach my $tag (keys %{$row}) { |
315 |
foreach my $l (@{$row->{$rec_nr}}) { |
foreach my $l (@{$row->{$tag}}) { |
316 |
|
|
317 |
# remove end marker |
# remove end marker |
318 |
$l =~ s/\x1E$//; |
$l =~ s/\x1E$//; |
319 |
|
|
320 |
# filter output |
# filter output |
321 |
$l = $self->{'hash_filter'}->($l, $rec_nr) if ($self->{'hash_filter'}); |
$l = $self->{'hash_filter'}->($l, $tag) if ($self->{'hash_filter'}); |
322 |
|
|
323 |
my $val; |
my $val; |
324 |
|
|
325 |
# has identifiers? |
# has identifiers? |
326 |
($val->{'i1'},$val->{'i2'}) = ($1,$2) if ($l =~ s/^([01 #])([01 #])\x1F/\x1F/); |
($val->{'i1'},$val->{'i2'}) = ($1,$2) if ($l =~ s/^([01 #])([01 #])\x1F/\x1F/); |
327 |
|
|
328 |
|
my $sf_usage; |
329 |
|
my @subfields; |
330 |
|
|
331 |
# has subfields? |
# has subfields? |
332 |
if ($l =~ m/\x1F/) { |
if ($l =~ m/\x1F/) { |
333 |
foreach my $t (split(/\x1F/,$l)) { |
foreach my $t (split(/\x1F/,$l)) { |
334 |
next if (! $t); |
next if (! $t); |
335 |
my $f = substr($t,0,1); |
my $f = substr($t,0,1); |
336 |
# repeatable subfileds. When we hit first one, |
|
337 |
# store CURRENT (up to that) in first repetition |
push @subfields, ( $f, $sf_usage->{$f}++ || 0 ); |
338 |
# of this record. Then, new record with same |
|
339 |
# identifiers will be created. |
# repeatable subfiled -- convert it to array |
340 |
if ($val->{$f}) { |
if ($val->{$f}) { |
341 |
push @{$rec->{$rec_nr}}, $val; |
if ( ref($val->{$f}) ne 'ARRAY' ) { |
342 |
$val = { |
$val->{$f} = [ $val->{$f}, $val ]; |
343 |
i1 => $val->{i1}, |
} else { |
344 |
i2 => $val->{i2}, |
push @{$val->{$f}}, $val; |
345 |
}; |
} |
346 |
} |
} |
347 |
$val->{substr($t,0,1)} = substr($t,1); |
$val->{substr($t,0,1)} = substr($t,1); |
348 |
} |
} |
349 |
|
$val->{subfields} = [ @subfields ] if $args->{include_subfields}; |
350 |
} else { |
} else { |
351 |
$val = $l; |
$val = $l; |
352 |
} |
} |
353 |
|
|
354 |
push @{$rec->{$rec_nr}}, $val; |
push @{$rec->{$tag}}, $val; |
355 |
} |
} |
356 |
} |
} |
357 |
|
|
385 |
1; |
1; |
386 |
__END__ |
__END__ |
387 |
|
|
388 |
|
=head1 UTF-8 ENCODING |
389 |
|
|
390 |
|
This module does nothing with encoding. But, since MARC format is byte |
391 |
|
oriented even when using UTF-8 which has variable number of bytes for each |
392 |
|
character, file is opened in binary mode. |
393 |
|
|
394 |
|
As a result, all scalars recturned to perl don't have utf-8 flag. Solution is |
395 |
|
to use C<hash_filter> and L<Encode> to decode utf-8 encoding like this: |
396 |
|
|
397 |
|
use Encode; |
398 |
|
|
399 |
|
my $marc = new MARC::Fast( |
400 |
|
marcdb => 'utf8.marc', |
401 |
|
hash_filter => sub { |
402 |
|
Encode::decode( 'utf-8', $_[0] ); |
403 |
|
}, |
404 |
|
); |
405 |
|
|
406 |
|
This will affect C<to_hash>, but C<fetch> will still return binary representation |
407 |
|
since it doesn't support C<hash_filter>. |
408 |
|
|
409 |
=head1 AUTHOR |
=head1 AUTHOR |
410 |
|
|
411 |
Dobrica Pavlinusic |
Dobrica Pavlinusic |