4 |
use Carp; |
use Carp; |
5 |
use File::Glob qw(:globally :nocase); |
use File::Glob qw(:globally :nocase); |
6 |
|
|
|
use Data::Dumper; |
|
|
|
|
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 = 0.12; |
$VERSION = 0.24; |
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 (); |
84 |
read_fdt => 1, |
read_fdt => 1, |
85 |
include_deleted => 1, |
include_deleted => 1, |
86 |
hash_filter => sub { |
hash_filter => sub { |
87 |
my $v = shift; |
my ($v,$field_number) = @_; |
88 |
$v =~ s#foo#bar#g; |
$v =~ s#foo#bar#g; |
89 |
}, |
}, |
90 |
debug => 1, |
debug => 1, |
91 |
|
join_subfields_with => ' ; ', |
92 |
); |
); |
93 |
|
|
94 |
Options are described below: |
Options are described below: |
114 |
|
|
115 |
=item hash_filter |
=item hash_filter |
116 |
|
|
117 |
Filter code ref which will be used before data is converted to hash. |
Filter code ref which will be used before data is converted to hash. It will |
118 |
|
receive two arguments, whole line from current field (in C<< $_[0] >>) and |
119 |
|
field number (in C<< $_[1] >>). |
120 |
|
|
121 |
=item debug |
=item debug |
122 |
|
|
123 |
Dump a B<lot> of debugging output. |
Dump a B<lot> of debugging output even at level 1. For even more increase level. |
124 |
|
|
125 |
|
=item join_subfields_with |
126 |
|
|
127 |
|
Define delimiter which will be used to join repeatable subfields. This |
128 |
|
option is included to support lagacy application written against version |
129 |
|
older than 0.21 of this module. By default, it disabled. See L</to_hash>. |
130 |
|
|
131 |
|
=item ignore_empty_subfields |
132 |
|
|
133 |
|
Remove all empty subfields while reading from ISIS file. |
134 |
|
|
135 |
=back |
=back |
136 |
|
|
143 |
|
|
144 |
croak "new needs database name (isisdb) as argument!" unless ({@_}->{isisdb}); |
croak "new needs database name (isisdb) as argument!" unless ({@_}->{isisdb}); |
145 |
|
|
146 |
foreach my $v (qw{isisdb debug include_deleted hash_filter}) { |
foreach my $v (qw{isisdb debug include_deleted hash_filter join_subfields_with ignore_empty_subfields}) { |
147 |
$self->{$v} = {@_}->{$v}; |
$self->{$v} = {@_}->{$v} if defined({@_}->{$v}); |
148 |
} |
} |
149 |
|
|
150 |
my @isis_files = grep(/\.(FDT|MST|XRF|CNT)$/i,glob($self->{isisdb}."*")); |
my @isis_files = grep(/\.(FDT|MST|XRF|CNT)$/i,glob($self->{isisdb}."*")); |
164 |
} |
} |
165 |
} |
} |
166 |
|
|
167 |
print STDERR "## using files: ",join(" ",@isis_files),"\n" if ($self->{debug}); |
if ($self->{debug}) { |
168 |
|
print STDERR "## using files: ",join(" ",@isis_files),"\n"; |
169 |
|
eval "use Data::Dump"; |
170 |
|
|
171 |
|
if (! $@) { |
172 |
|
*Dumper = *Data::Dump::dump; |
173 |
|
} else { |
174 |
|
use Data::Dumper; |
175 |
|
} |
176 |
|
} |
177 |
|
|
178 |
# if you want to read .FDT file use read_fdt argument when creating class! |
# if you want to read .FDT file use read_fdt argument when creating class! |
179 |
if ($self->{read_fdt} && -e $self->{fdt_file}) { |
if ($self->{read_fdt} && -e $self->{fdt_file}) { |
222 |
read($self->{'fileMST'}, $buff, 4) || croak "can't read NXTMFN from MST: $!"; |
read($self->{'fileMST'}, $buff, 4) || croak "can't read NXTMFN from MST: $!"; |
223 |
$self->{'NXTMFN'}=unpack("V",$buff) || croak "NXTNFN is zero"; |
$self->{'NXTMFN'}=unpack("V",$buff) || croak "NXTNFN is zero"; |
224 |
|
|
225 |
print STDERR Dumper($self),"\n" if ($self->{debug}); |
print STDERR "## self ",Dumper($self),"\n" if ($self->{debug}); |
226 |
|
|
227 |
# open files for later |
# open files for later |
228 |
open($self->{'fileXRF'}, $self->{xrf_file}) || croak "can't open '$self->{xrf_file}': $!"; |
open($self->{'fileXRF'}, $self->{xrf_file}) || croak "can't open '$self->{xrf_file}': $!"; |
284 |
|
|
285 |
# read XRFMFB abd XRFMFP |
# read XRFMFB abd XRFMFP |
286 |
read($self->{'fileXRF'}, $buff, 4); |
read($self->{'fileXRF'}, $buff, 4); |
287 |
my $pointer=unpack("V",$buff) || croak "pointer is null"; |
my $pointer=unpack("V",$buff); |
288 |
|
if (! $pointer) { |
289 |
|
if ($self->{include_deleted}) { |
290 |
|
return; |
291 |
|
} else { |
292 |
|
warn "pointer for MFN $mfn is null\n"; |
293 |
|
return; |
294 |
|
} |
295 |
|
} |
296 |
|
|
297 |
# check for logically deleted record |
# check for logically deleted record |
298 |
if ($pointer & 0x80000000) { |
if ($pointer & 0x80000000) { |
388 |
# skip zero-sized fields |
# skip zero-sized fields |
389 |
next if ($FieldLEN[$i] == 0); |
next if ($FieldLEN[$i] == 0); |
390 |
|
|
391 |
push @{$self->{record}->{$FieldTAG[$i]}}, substr($buff,$FieldPOS[$i],$FieldLEN[$i]); |
my $v = substr($buff,$FieldPOS[$i],$FieldLEN[$i]); |
392 |
|
|
393 |
|
if ( $self->{ignore_empty_subfields} ) { |
394 |
|
$v =~ s/(\^\w)+(\^\w)/$2/g; |
395 |
|
$v =~ s/\^\w$//; # last on line? |
396 |
|
next if ($v eq ''); |
397 |
|
} |
398 |
|
|
399 |
|
push @{$self->{record}->{$FieldTAG[$i]}}, $v; |
400 |
} |
} |
401 |
|
|
402 |
$self->{'current_mfn'} = $mfn; |
$self->{'current_mfn'} = $mfn; |
406 |
return $self->{'record'}; |
return $self->{'record'}; |
407 |
} |
} |
408 |
|
|
409 |
|
=head2 mfn |
410 |
|
|
411 |
|
Returns current MFN position |
412 |
|
|
413 |
|
my $mfn = $isis->mfn; |
414 |
|
|
415 |
|
=cut |
416 |
|
|
417 |
|
# This function should be simple return $self->{current_mfn}, |
418 |
|
# but if new is called with _hack_mfn it becomes setter. |
419 |
|
# It's useful in tests when setting $isis->{record} directly |
420 |
|
|
421 |
|
sub mfn { |
422 |
|
my $self = shift; |
423 |
|
return $self->{current_mfn}; |
424 |
|
}; |
425 |
|
|
426 |
|
|
427 |
=head2 to_ascii |
=head2 to_ascii |
428 |
|
|
429 |
Returns ASCII output of record with specified MFN |
Returns ASCII output of record with specified MFN |
447 |
|
|
448 |
my $mfn = shift || croak "need MFN"; |
my $mfn = shift || croak "need MFN"; |
449 |
|
|
450 |
my $rec = $self->fetch($mfn); |
my $rec = $self->fetch($mfn) || return; |
451 |
|
|
452 |
my $out = "0\t$mfn"; |
my $out = "0\t$mfn"; |
453 |
|
|
503 |
} |
} |
504 |
], |
], |
505 |
|
|
506 |
|
In case there are repeatable subfields in record, this will create |
507 |
|
following structure: |
508 |
|
|
509 |
|
'900' => [ { |
510 |
|
'a' => [ 'foo', 'bar', 'baz' ], |
511 |
|
}] |
512 |
|
|
513 |
|
Or in more complex example of |
514 |
|
|
515 |
|
902 ^aa1^aa2^aa3^bb1^aa4^bb2^cc1^aa5 |
516 |
|
|
517 |
|
it will create |
518 |
|
|
519 |
|
902 => [ |
520 |
|
{ a => ["a1", "a2", "a3", "a4", "a5"], b => ["b1", "b2"], c => "c1" }, |
521 |
|
], |
522 |
|
|
523 |
|
This behaviour can be changed using C<join_subfields_with> option to L</new>, |
524 |
|
in which case C<to_hash> will always create single value for each subfield. |
525 |
|
This will change result to: |
526 |
|
|
527 |
|
|
528 |
|
|
529 |
This method will also create additional field C<000> with MFN. |
This method will also create additional field C<000> with MFN. |
530 |
|
|
531 |
|
There is also more elaborative way to call C<to_hash> like this: |
532 |
|
|
533 |
|
my $hash = $isis->to_hash({ |
534 |
|
mfn => 42, |
535 |
|
include_subfields => 1, |
536 |
|
}); |
537 |
|
|
538 |
|
Each option controll creation of hash: |
539 |
|
|
540 |
|
=over 4 |
541 |
|
|
542 |
|
=item mfn |
543 |
|
|
544 |
|
Specify MFN number of record |
545 |
|
|
546 |
|
=item include_subfields |
547 |
|
|
548 |
|
This option will create additional key in hash called C<subfields> which will |
549 |
|
have original record subfield order and index to that subfield like this: |
550 |
|
|
551 |
|
902 => [ { |
552 |
|
a => ["a1", "a2", "a3", "a4", "a5"], |
553 |
|
b => ["b1", "b2"], |
554 |
|
c => "c1", |
555 |
|
subfields => ["a", 0, "a", 1, "a", 2, "b", 0, "a", 3, "b", 1, "c", 0, "a", 4], |
556 |
|
} ], |
557 |
|
|
558 |
|
=item join_subfields_with |
559 |
|
|
560 |
|
Define delimiter which will be used to join repeatable subfields. You can |
561 |
|
specify option here instead in L</new> if you want to have per-record control. |
562 |
|
|
563 |
|
=item hash_filter |
564 |
|
|
565 |
|
You can override C<hash_filter> defined in L</new> using this option. |
566 |
|
|
567 |
|
=back |
568 |
|
|
569 |
=cut |
=cut |
570 |
|
|
571 |
sub to_hash { |
sub to_hash { |
572 |
my $self = shift; |
my $self = shift; |
573 |
|
|
574 |
|
|
575 |
my $mfn = shift || confess "need mfn!"; |
my $mfn = shift || confess "need mfn!"; |
576 |
|
my $arg; |
577 |
|
|
578 |
|
my $hash_filter = $self->{hash_filter}; |
579 |
|
|
580 |
|
if (ref($mfn) eq 'HASH') { |
581 |
|
$arg = $mfn; |
582 |
|
$mfn = $arg->{mfn} || confess "need mfn in arguments"; |
583 |
|
$hash_filter = $arg->{hash_filter} if ($arg->{hash_filter}); |
584 |
|
} |
585 |
|
|
586 |
# init record to include MFN as field 000 |
# init record to include MFN as field 000 |
587 |
my $rec = { '000' => [ $mfn ] }; |
my $rec = { '000' => [ $mfn ] }; |
588 |
|
|
589 |
my $row = $self->fetch($mfn); |
my $row = $self->fetch($mfn) || return; |
590 |
|
|
591 |
foreach my $k (keys %{$row}) { |
my $j_rs = $arg->{join_subfields_with} || $self->{join_subfields_with}; |
592 |
foreach my $l (@{$row->{$k}}) { |
$j_rs = $self->{join_subfields_with} unless(defined($j_rs)); |
593 |
|
my $i_sf = $arg->{include_subfields}; |
594 |
|
|
595 |
|
foreach my $f_nr (keys %{$row}) { |
596 |
|
foreach my $l (@{$row->{$f_nr}}) { |
597 |
|
|
598 |
# filter output |
# filter output |
599 |
$l = $self->{'hash_filter'}->($l) if ($self->{'hash_filter'}); |
$l = $hash_filter->($l, $f_nr) if ($hash_filter); |
600 |
|
next unless defined($l); |
601 |
|
|
602 |
my $val; |
my $val; |
603 |
|
my $r_sf; # repeatable subfields in this record |
604 |
|
|
605 |
# has identifiers? |
# has identifiers? |
606 |
($val->{'i1'},$val->{'i2'}) = ($1,$2) if ($l =~ s/^([01 #])([01 #])\^/\^/); |
($val->{'i1'},$val->{'i2'}) = ($1,$2) if ($l =~ s/^([01 #])([01 #])\^/\^/); |
609 |
if ($l =~ m/\^/) { |
if ($l =~ m/\^/) { |
610 |
foreach my $t (split(/\^/,$l)) { |
foreach my $t (split(/\^/,$l)) { |
611 |
next if (! $t); |
next if (! $t); |
612 |
$val->{substr($t,0,1)} = substr($t,1); |
my ($sf,$v) = (substr($t,0,1), substr($t,1)); |
613 |
|
# XXX this might be option, but why? |
614 |
|
next unless (defined($v) && $v ne ''); |
615 |
|
# warn "### $f_nr^$sf:$v",$/ if ($self->{debug} > 1); |
616 |
|
|
617 |
|
if (ref( $val->{$sf} ) eq 'ARRAY') { |
618 |
|
|
619 |
|
push @{ $val->{$sf} }, $v; |
620 |
|
|
621 |
|
# record repeatable subfield it it's offset |
622 |
|
push @{ $val->{subfields} }, ( $sf, $#{ $val->{$sf} } ) if (! $j_rs && $i_sf); |
623 |
|
$r_sf->{$sf}++; |
624 |
|
|
625 |
|
} elsif (defined( $val->{$sf} )) { |
626 |
|
|
627 |
|
# convert scalar field to array |
628 |
|
$val->{$sf} = [ $val->{$sf}, $v ]; |
629 |
|
|
630 |
|
push @{ $val->{subfields} }, ( $sf, 1 ) if (! $j_rs && $i_sf); |
631 |
|
$r_sf->{$sf}++; |
632 |
|
|
633 |
|
} else { |
634 |
|
$val->{$sf} = $v; |
635 |
|
push @{ $val->{subfields} }, ( $sf, 0 ) if ($i_sf); |
636 |
|
} |
637 |
} |
} |
638 |
} else { |
} else { |
639 |
$val = $l; |
$val = $l; |
640 |
} |
} |
641 |
|
|
642 |
push @{$rec->{$k}}, $val; |
if ($j_rs) { |
643 |
|
map { |
644 |
|
$val->{$_} = join($j_rs, @{ $val->{$_} }); |
645 |
|
} keys %$r_sf |
646 |
|
} |
647 |
|
|
648 |
|
push @{$rec->{$f_nr}}, $val; |
649 |
} |
} |
650 |
} |
} |
651 |
|
|
758 |
tested this against ouput of one C<isis.dll>-based application, but I don't |
tested this against ouput of one C<isis.dll>-based application, but I don't |
759 |
know any details about it's version. |
know any details about it's version. |
760 |
|
|
761 |
|
=head1 VERSIONS |
762 |
|
|
763 |
|
As this is young module, new features are added in subsequent version. It's |
764 |
|
a good idea to specify version when using this module like this: |
765 |
|
|
766 |
|
use Biblio::Isis 0.23 |
767 |
|
|
768 |
|
Below is list of changes in specific version of module (so you can target |
769 |
|
older versions if you really have to): |
770 |
|
|
771 |
|
=over 8 |
772 |
|
|
773 |
|
=item 0.24 |
774 |
|
|
775 |
|
Added C<ignore_empty_subfields> |
776 |
|
|
777 |
|
=item 0.23 |
778 |
|
|
779 |
|
Added C<hash_filter> to L</to_hash> |
780 |
|
|
781 |
|
Fixed bug with documented C<join_subfields_with> in L</new> which wasn't |
782 |
|
implemented |
783 |
|
|
784 |
|
=item 0.22 |
785 |
|
|
786 |
|
Added field number when calling C<hash_filter> |
787 |
|
|
788 |
|
=item 0.21 |
789 |
|
|
790 |
|
Added C<join_subfields_with> to L</new> and L</to_hash>. |
791 |
|
|
792 |
|
Added C<include_subfields> to L</to_hash>. |
793 |
|
|
794 |
|
=item 0.20 |
795 |
|
|
796 |
|
Added C<< $isis->mfn >>, support for repeatable subfields and |
797 |
|
C<< $isis->to_hash({ mfn => 42, ... }) >> calling convention |
798 |
|
|
799 |
|
=back |
800 |
|
|
801 |
=head1 AUTHOR |
=head1 AUTHOR |
802 |
|
|
803 |
Dobrica Pavlinusic |
Dobrica Pavlinusic |
819 |
|
|
820 |
=head1 SEE ALSO |
=head1 SEE ALSO |
821 |
|
|
822 |
|
L<Biblio::Isis::Manual> for CDS/ISIS manual appendix F, G and H which describe file format |
823 |
|
|
824 |
OpenIsis web site L<http://www.openisis.org> |
OpenIsis web site L<http://www.openisis.org> |
825 |
|
|
826 |
perl4lib site L<http://perl4lib.perl.org> |
perl4lib site L<http://perl4lib.perl.org> |