1 |
package IsisDB; |
package Biblio::Isis; |
2 |
use strict; |
use strict; |
3 |
|
|
4 |
use Carp; |
use Carp; |
9 |
BEGIN { |
BEGIN { |
10 |
use Exporter (); |
use Exporter (); |
11 |
use vars qw ($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS); |
use vars qw ($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS); |
12 |
$VERSION = 0.09; |
$VERSION = 0.13; |
13 |
@ISA = qw (Exporter); |
@ISA = qw (Exporter); |
14 |
#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 |
15 |
@EXPORT = qw (); |
@EXPORT = qw (); |
20 |
|
|
21 |
=head1 NAME |
=head1 NAME |
22 |
|
|
23 |
IsisDB - Read CDS/ISIS, WinISIS and IsisMarc database |
Biblio::Isis - Read CDS/ISIS, WinISIS and IsisMarc database |
24 |
|
|
25 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
26 |
|
|
27 |
use IsisDB; |
use Biblio::Isis; |
28 |
|
|
29 |
my $isis = new IsisDB( |
my $isis = new Biblio::Isis( |
30 |
isisdb => './cds/cds', |
isisdb => './cds/cds', |
31 |
); |
); |
32 |
|
|
81 |
|
|
82 |
Open ISIS database |
Open ISIS database |
83 |
|
|
84 |
my $isis = new IsisDB( |
my $isis = new Biblio::Isis( |
85 |
isisdb => './cds/cds', |
isisdb => './cds/cds', |
86 |
read_fdt => 1, |
read_fdt => 1, |
87 |
include_deleted => 1, |
include_deleted => 1, |
147 |
push @must_exist, "fdt" if ($self->{read_fdt}); |
push @must_exist, "fdt" if ($self->{read_fdt}); |
148 |
|
|
149 |
foreach my $ext (@must_exist) { |
foreach my $ext (@must_exist) { |
150 |
croak "missing ",uc($ext)," file in ",$self->{isisdb} unless ($self->{$ext."_file"}); |
unless ($self->{$ext."_file"}) { |
151 |
|
carp "missing ",uc($ext)," file in ",$self->{isisdb}; |
152 |
|
return; |
153 |
|
} |
154 |
} |
} |
155 |
|
|
156 |
print STDERR "## using files: ",join(" ",@isis_files),"\n" if ($self->{debug}); |
print STDERR "## using files: ",join(" ",@isis_files),"\n" if ($self->{debug}); |
195 |
# NXTMFB* last block allocated to master file |
# NXTMFB* last block allocated to master file |
196 |
# NXTMFP offset to next available position in last block |
# NXTMFP offset to next available position in last block |
197 |
# MFTYPE always 0 for user db file (1 for system) |
# MFTYPE always 0 for user db file (1 for system) |
198 |
seek($self->{'fileMST'},4,0) || carp "can't seek to offset 0 in MST: $!"; |
seek($self->{'fileMST'},4,0) || croak "can't seek to offset 0 in MST: $!"; |
199 |
|
|
200 |
my $buff; |
my $buff; |
201 |
|
|
202 |
read($self->{'fileMST'}, $buff, 4) || carp "can't read NXTMFN from MST: $!"; |
read($self->{'fileMST'}, $buff, 4) || croak "can't read NXTMFN from MST: $!"; |
203 |
$self->{'NXTMFN'}=unpack("V",$buff) || carp "NXTNFN is zero"; |
$self->{'NXTMFN'}=unpack("V",$buff) || croak "NXTNFN is zero"; |
204 |
|
|
205 |
print STDERR Dumper($self),"\n" if ($self->{debug}); |
print STDERR Dumper($self),"\n" if ($self->{debug}); |
206 |
|
|
224 |
return $self->{'NXTMFN'} - 1; |
return $self->{'NXTMFN'} - 1; |
225 |
} |
} |
226 |
|
|
|
=head2 read_cnt |
|
|
|
|
|
Read content of C<.CNT> file and return hash containing it. |
|
|
|
|
|
print Dumper($isis->read_cnt); |
|
|
|
|
|
This function is not used by module (C<.CNT> files are not required for this |
|
|
module to work), but it can be useful to examine your index (while debugging |
|
|
for example). |
|
|
|
|
|
=cut |
|
|
|
|
|
sub read_cnt { |
|
|
my $self = shift; |
|
|
|
|
|
croak "missing CNT file in ",$self->{isisdb} unless ($self->{cnt_file}); |
|
|
|
|
|
# Get the index information from $db.CNT |
|
|
|
|
|
open(my $fileCNT, $self->{cnt_file}) || carp "can't read '$self->{cnt_file}': $!"; |
|
|
binmode($fileCNT); |
|
|
|
|
|
my $buff; |
|
|
|
|
|
read($fileCNT, $buff, 26) || carp "can't read first table from CNT: $!"; |
|
|
$self->unpack_cnt($buff); |
|
|
|
|
|
read($fileCNT, $buff, 26) || carp "can't read second table from CNT: $!"; |
|
|
$self->unpack_cnt($buff); |
|
|
|
|
|
close($fileCNT); |
|
|
|
|
|
return $self->{cnt}; |
|
|
} |
|
|
|
|
|
=head2 unpack_cnt |
|
|
|
|
|
Unpack one of two 26 bytes fixed length record in C<.CNT> file. |
|
|
|
|
|
Here is definition of record: |
|
|
|
|
|
off key description size |
|
|
0: IDTYPE BTree type s |
|
|
2: ORDN Nodes Order s |
|
|
4: ORDF Leafs Order s |
|
|
6: N Number of Memory buffers for nodes s |
|
|
8: K Number of buffers for first level index s |
|
|
10: LIV Current number of Index Levels s |
|
|
12: POSRX Pointer to Root Record in N0x l |
|
|
16: NMAXPOS Next Available position in N0x l |
|
|
20: FMAXPOS Next available position in L0x l |
|
|
24: ABNORMAL Formal BTree normality indicator s |
|
|
length: 26 bytes |
|
|
|
|
|
This will fill C<$self> object under C<cnt> with hash. It's used by C<read_cnt>. |
|
|
|
|
|
=cut |
|
|
|
|
|
sub unpack_cnt { |
|
|
my $self = shift; |
|
|
|
|
|
my @flds = qw(ORDN ORDF N K LIV POSRX NMAXPOS FMAXPOS ABNORMAL); |
|
|
|
|
|
my $buff = shift || return; |
|
|
my @arr = unpack("vvvvvvVVVv", $buff); |
|
|
|
|
|
print STDERR "unpack_cnt: ",join(" ",@arr),"\n" if ($self->{'debug'}); |
|
|
|
|
|
my $IDTYPE = shift @arr; |
|
|
foreach (@flds) { |
|
|
$self->{cnt}->{$IDTYPE}->{$_} = abs(shift @arr); |
|
|
} |
|
|
} |
|
|
|
|
227 |
=head2 fetch |
=head2 fetch |
228 |
|
|
229 |
Read record with selected MFN |
Read record with selected MFN |
264 |
|
|
265 |
# read XRFMFB abd XRFMFP |
# read XRFMFB abd XRFMFP |
266 |
read($self->{'fileXRF'}, $buff, 4); |
read($self->{'fileXRF'}, $buff, 4); |
267 |
my $pointer=unpack("V",$buff) || carp "pointer is null"; |
my $pointer=unpack("V",$buff); |
268 |
|
if (! $pointer) { |
269 |
|
if ($self->{include_deleted}) { |
270 |
|
return; |
271 |
|
} else { |
272 |
|
warn "pointer for MFN $mfn is null\n"; |
273 |
|
return; |
274 |
|
} |
275 |
|
} |
276 |
|
|
277 |
# check for logically deleted record |
# check for logically deleted record |
278 |
if ($pointer & 0x80000000) { |
if ($pointer & 0x80000000) { |
401 |
|
|
402 |
my $mfn = shift || croak "need MFN"; |
my $mfn = shift || croak "need MFN"; |
403 |
|
|
404 |
my $rec = $self->fetch($mfn); |
my $rec = $self->fetch($mfn) || return; |
405 |
|
|
406 |
my $out = "0\t$mfn"; |
my $out = "0\t$mfn"; |
407 |
|
|
469 |
# init record to include MFN as field 000 |
# init record to include MFN as field 000 |
470 |
my $rec = { '000' => [ $mfn ] }; |
my $rec = { '000' => [ $mfn ] }; |
471 |
|
|
472 |
my $row = $self->fetch($mfn); |
my $row = $self->fetch($mfn) || return; |
473 |
|
|
474 |
foreach my $k (keys %{$row}) { |
foreach my $k (keys %{$row}) { |
475 |
foreach my $l (@{$row->{$k}}) { |
foreach my $l (@{$row->{$k}}) { |
513 |
return $self->{'TagName'}->{$tag} || $tag; |
return $self->{'TagName'}->{$tag} || $tag; |
514 |
} |
} |
515 |
|
|
516 |
|
|
517 |
|
=head2 read_cnt |
518 |
|
|
519 |
|
Read content of C<.CNT> file and return hash containing it. |
520 |
|
|
521 |
|
print Dumper($isis->read_cnt); |
522 |
|
|
523 |
|
This function is not used by module (C<.CNT> files are not required for this |
524 |
|
module to work), but it can be useful to examine your index (while debugging |
525 |
|
for example). |
526 |
|
|
527 |
|
=cut |
528 |
|
|
529 |
|
sub read_cnt { |
530 |
|
my $self = shift; |
531 |
|
|
532 |
|
croak "missing CNT file in ",$self->{isisdb} unless ($self->{cnt_file}); |
533 |
|
|
534 |
|
# Get the index information from $db.CNT |
535 |
|
|
536 |
|
open(my $fileCNT, $self->{cnt_file}) || croak "can't read '$self->{cnt_file}': $!"; |
537 |
|
binmode($fileCNT); |
538 |
|
|
539 |
|
my $buff; |
540 |
|
|
541 |
|
read($fileCNT, $buff, 26) || croak "can't read first table from CNT: $!"; |
542 |
|
$self->unpack_cnt($buff); |
543 |
|
|
544 |
|
read($fileCNT, $buff, 26) || croak "can't read second table from CNT: $!"; |
545 |
|
$self->unpack_cnt($buff); |
546 |
|
|
547 |
|
close($fileCNT); |
548 |
|
|
549 |
|
return $self->{cnt}; |
550 |
|
} |
551 |
|
|
552 |
|
=head2 unpack_cnt |
553 |
|
|
554 |
|
Unpack one of two 26 bytes fixed length record in C<.CNT> file. |
555 |
|
|
556 |
|
Here is definition of record: |
557 |
|
|
558 |
|
off key description size |
559 |
|
0: IDTYPE BTree type s |
560 |
|
2: ORDN Nodes Order s |
561 |
|
4: ORDF Leafs Order s |
562 |
|
6: N Number of Memory buffers for nodes s |
563 |
|
8: K Number of buffers for first level index s |
564 |
|
10: LIV Current number of Index Levels s |
565 |
|
12: POSRX Pointer to Root Record in N0x l |
566 |
|
16: NMAXPOS Next Available position in N0x l |
567 |
|
20: FMAXPOS Next available position in L0x l |
568 |
|
24: ABNORMAL Formal BTree normality indicator s |
569 |
|
length: 26 bytes |
570 |
|
|
571 |
|
This will fill C<$self> object under C<cnt> with hash. It's used by C<read_cnt>. |
572 |
|
|
573 |
|
=cut |
574 |
|
|
575 |
|
sub unpack_cnt { |
576 |
|
my $self = shift; |
577 |
|
|
578 |
|
my @flds = qw(ORDN ORDF N K LIV POSRX NMAXPOS FMAXPOS ABNORMAL); |
579 |
|
|
580 |
|
my $buff = shift || return; |
581 |
|
my @arr = unpack("vvvvvvVVVv", $buff); |
582 |
|
|
583 |
|
print STDERR "unpack_cnt: ",join(" ",@arr),"\n" if ($self->{'debug'}); |
584 |
|
|
585 |
|
my $IDTYPE = shift @arr; |
586 |
|
foreach (@flds) { |
587 |
|
$self->{cnt}->{$IDTYPE}->{$_} = abs(shift @arr); |
588 |
|
} |
589 |
|
} |
590 |
|
|
591 |
1; |
1; |
592 |
|
|
593 |
=head1 BUGS |
=head1 BUGS |