1 |
package IsisDB; |
package Biblio::Isis; |
2 |
use strict; |
use strict; |
3 |
|
|
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 = 0.09; |
$VERSION = 0.20; |
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 (); |
18 |
|
|
19 |
=head1 NAME |
=head1 NAME |
20 |
|
|
21 |
IsisDB - Read CDS/ISIS, WinISIS and IsisMarc database |
Biblio::Isis - Read CDS/ISIS, WinISIS and IsisMarc database |
22 |
|
|
23 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
24 |
|
|
25 |
use IsisDB; |
use Biblio::Isis; |
26 |
|
|
27 |
my $isis = new IsisDB( |
my $isis = new Biblio::Isis( |
28 |
isisdb => './cds/cds', |
isisdb => './cds/cds', |
29 |
); |
); |
30 |
|
|
79 |
|
|
80 |
Open ISIS database |
Open ISIS database |
81 |
|
|
82 |
my $isis = new IsisDB( |
my $isis = new Biblio::Isis( |
83 |
isisdb => './cds/cds', |
isisdb => './cds/cds', |
84 |
read_fdt => 1, |
read_fdt => 1, |
85 |
include_deleted => 1, |
include_deleted => 1, |
145 |
push @must_exist, "fdt" if ($self->{read_fdt}); |
push @must_exist, "fdt" if ($self->{read_fdt}); |
146 |
|
|
147 |
foreach my $ext (@must_exist) { |
foreach my $ext (@must_exist) { |
148 |
croak "missing ",uc($ext)," file in ",$self->{isisdb} unless ($self->{$ext."_file"}); |
unless ($self->{$ext."_file"}) { |
149 |
|
carp "missing ",uc($ext)," file in ",$self->{isisdb}; |
150 |
|
return; |
151 |
|
} |
152 |
} |
} |
153 |
|
|
154 |
print STDERR "## using files: ",join(" ",@isis_files),"\n" if ($self->{debug}); |
if ($self->{debug}) { |
155 |
|
print STDERR "## using files: ",join(" ",@isis_files),"\n"; |
156 |
|
eval "use Data::Dump"; |
157 |
|
|
158 |
|
if (! $@) { |
159 |
|
*Dumper = *Data::Dump::dump; |
160 |
|
} else { |
161 |
|
use Data::Dumper; |
162 |
|
} |
163 |
|
} |
164 |
|
|
165 |
# if you want to read .FDT file use read_fdt argument when creating class! |
# if you want to read .FDT file use read_fdt argument when creating class! |
166 |
if ($self->{read_fdt} && -e $self->{fdt_file}) { |
if ($self->{read_fdt} && -e $self->{fdt_file}) { |
168 |
# read the $db.FDT file for tags |
# read the $db.FDT file for tags |
169 |
my $fieldzone=0; |
my $fieldzone=0; |
170 |
|
|
171 |
open(fileFDT, $self->{fdt_file}) || croak "can't read '$self->{fdt_file}': $!"; |
open(my $fileFDT, $self->{fdt_file}) || croak "can't read '$self->{fdt_file}': $!"; |
172 |
|
binmode($fileFDT); |
173 |
|
|
174 |
while (<fileFDT>) { |
while (<$fileFDT>) { |
175 |
chomp; |
chomp; |
176 |
if ($fieldzone) { |
if ($fieldzone) { |
177 |
my $name=substr($_,0,30); |
my $name=substr($_,0,30); |
188 |
} |
} |
189 |
} |
} |
190 |
|
|
191 |
close(fileFDT); |
close($fileFDT); |
192 |
} |
} |
193 |
|
|
194 |
# Get the Maximum MFN from $db.MST |
# Get the Maximum MFN from $db.MST |
195 |
|
|
196 |
open($self->{'fileMST'}, $self->{mst_file}) || croak "can't open '$self->{mst_file}': $!"; |
open($self->{'fileMST'}, $self->{mst_file}) || croak "can't open '$self->{mst_file}': $!"; |
197 |
|
binmode($self->{'fileMST'}); |
198 |
|
|
199 |
# MST format: (* = 32 bit signed) |
# MST format: (* = 32 bit signed) |
200 |
# CTLMFN* always 0 |
# CTLMFN* always 0 |
202 |
# NXTMFB* last block allocated to master file |
# NXTMFB* last block allocated to master file |
203 |
# NXTMFP offset to next available position in last block |
# NXTMFP offset to next available position in last block |
204 |
# MFTYPE always 0 for user db file (1 for system) |
# MFTYPE always 0 for user db file (1 for system) |
205 |
seek($self->{'fileMST'},4,0); |
seek($self->{'fileMST'},4,0) || croak "can't seek to offset 0 in MST: $!"; |
206 |
|
|
207 |
my $buff; |
my $buff; |
208 |
|
|
209 |
read($self->{'fileMST'}, $buff, 4); |
read($self->{'fileMST'}, $buff, 4) || croak "can't read NXTMFN from MST: $!"; |
210 |
$self->{'NXTMFN'}=unpack("l",$buff) || carp "NXTNFN is zero"; |
$self->{'NXTMFN'}=unpack("V",$buff) || croak "NXTNFN is zero"; |
211 |
|
|
212 |
|
print STDERR "## self ",Dumper($self),"\n" if ($self->{debug}); |
|
|
|
|
|
|
|
print STDERR Dumper($self),"\n" if ($self->{debug}); |
|
213 |
|
|
214 |
# open files for later |
# open files for later |
215 |
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}': $!"; |
216 |
|
binmode($self->{'fileXRF'}); |
217 |
|
|
218 |
$self ? return $self : return undef; |
$self ? return $self : return undef; |
219 |
} |
} |
231 |
return $self->{'NXTMFN'} - 1; |
return $self->{'NXTMFN'} - 1; |
232 |
} |
} |
233 |
|
|
|
=head2 read_cnt |
|
|
|
|
|
Read content of C<.CNT> file and return hash containing it. |
|
|
|
|
|
print Dumper($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(fileCNT, $self->{cnt_file}) || croak "can't read '$self->{cnt_file}': $!"; |
|
|
|
|
|
my $buff; |
|
|
|
|
|
read(fileCNT, $buff, 26); |
|
|
$self->unpack_cnt($buff); |
|
|
|
|
|
read(fileCNT, $buff, 26); |
|
|
$self->unpack_cnt($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("ssssssllls", $buff); |
|
|
|
|
|
print STDERR "unpack_cnt: ",join(" ",@arr),"\n" if ($self->{'debug'}); |
|
|
|
|
|
my $IDTYPE = shift @arr; |
|
|
foreach (@flds) { |
|
|
$self->{cnt}->{$IDTYPE}->{$_} = abs(shift @arr); |
|
|
} |
|
|
} |
|
|
|
|
234 |
=head2 fetch |
=head2 fetch |
235 |
|
|
236 |
Read record with selected MFN |
Read record with selected MFN |
271 |
|
|
272 |
# read XRFMFB abd XRFMFP |
# read XRFMFB abd XRFMFP |
273 |
read($self->{'fileXRF'}, $buff, 4); |
read($self->{'fileXRF'}, $buff, 4); |
274 |
my $pointer=unpack("l",$buff) || carp "pointer is null"; |
my $pointer=unpack("V",$buff); |
275 |
|
if (! $pointer) { |
276 |
|
if ($self->{include_deleted}) { |
277 |
|
return; |
278 |
|
} else { |
279 |
|
warn "pointer for MFN $mfn is null\n"; |
280 |
|
return; |
281 |
|
} |
282 |
|
} |
283 |
|
|
284 |
# check for logically deleted record |
# check for logically deleted record |
285 |
if ($pointer < 0) { |
if ($pointer & 0x80000000) { |
286 |
print STDERR "## record $mfn is logically deleted\n" if ($self->{debug}); |
print STDERR "## record $mfn is logically deleted\n" if ($self->{debug}); |
287 |
$self->{deleted} = $mfn; |
$self->{deleted} = $mfn; |
288 |
|
|
289 |
return unless $self->{include_deleted}; |
return unless $self->{include_deleted}; |
290 |
|
|
291 |
$pointer = abs($pointer); |
# abs |
292 |
|
$pointer = ($pointer ^ 0xffffffff) + 1; |
293 |
} |
} |
294 |
|
|
295 |
my $XRFMFB = int($pointer/2048); |
my $XRFMFB = int($pointer/2048); |
304 |
|
|
305 |
# Get Record Information |
# Get Record Information |
306 |
|
|
307 |
seek($self->{'fileMST'},$blk_off,0); |
seek($self->{'fileMST'},$blk_off,0) || croak "can't seek to $blk_off: $!"; |
308 |
|
|
309 |
read($self->{'fileMST'}, $buff, 4); |
read($self->{'fileMST'}, $buff, 4) || croak "can't read 4 bytes at offset $blk_off from MST file: $!"; |
310 |
my $value=unpack("l",$buff); |
my $value=unpack("V",$buff); |
311 |
|
|
312 |
print STDERR "## offset for rowid $value is $blk_off (blk $XRFMFB off $XRFMFP)\n" if ($self->{debug}); |
print STDERR "## offset for rowid $value is $blk_off (blk $XRFMFB off $XRFMFP)\n" if ($self->{debug}); |
313 |
|
|
324 |
|
|
325 |
read($self->{'fileMST'}, $buff, 14); |
read($self->{'fileMST'}, $buff, 14); |
326 |
|
|
327 |
my ($MFRL,$MFBWB,$MFBWP,$BASE,$NVF,$STATUS) = unpack("slssss", $buff); |
my ($MFRL,$MFBWB,$MFBWP,$BASE,$NVF,$STATUS) = unpack("vVvvvv", $buff); |
328 |
|
|
329 |
print STDERR "## MFRL: $MFRL MFBWB: $MFBWB MFBWP: $MFBWP BASE: $BASE NVF: $NVF STATUS: $STATUS\n" if ($self->{debug}); |
print STDERR "## MFRL: $MFRL MFBWB: $MFBWB MFBWP: $MFBWP BASE: $BASE NVF: $NVF STATUS: $STATUS\n" if ($self->{debug}); |
330 |
|
|
344 |
|
|
345 |
for (my $i = 0 ; $i < $NVF ; $i++) { |
for (my $i = 0 ; $i < $NVF ; $i++) { |
346 |
|
|
347 |
my ($TAG,$POS,$LEN) = unpack("sss", substr($buff,$i * 6, 6)); |
my ($TAG,$POS,$LEN) = unpack("vvv", substr($buff,$i * 6, 6)); |
348 |
|
|
349 |
print STDERR "## TAG: $TAG POS: $POS LEN: $LEN\n" if ($self->{debug}); |
print STDERR "## TAG: $TAG POS: $POS LEN: $LEN\n" if ($self->{debug}); |
350 |
|
|
408 |
|
|
409 |
my $mfn = shift || croak "need MFN"; |
my $mfn = shift || croak "need MFN"; |
410 |
|
|
411 |
my $rec = $self->fetch($mfn); |
my $rec = $self->fetch($mfn) || return; |
412 |
|
|
413 |
my $out = "0\t$mfn"; |
my $out = "0\t$mfn"; |
414 |
|
|
464 |
} |
} |
465 |
], |
], |
466 |
|
|
467 |
|
In case there are repeatable subfields in record, this will create |
468 |
|
following structure: |
469 |
|
|
470 |
|
'900' => [ { |
471 |
|
'a' => [ 'foo', 'bar', 'baz' ], |
472 |
|
}] |
473 |
|
|
474 |
This method will also create additional field C<000> with MFN. |
This method will also create additional field C<000> with MFN. |
475 |
|
|
476 |
=cut |
=cut |
483 |
# init record to include MFN as field 000 |
# init record to include MFN as field 000 |
484 |
my $rec = { '000' => [ $mfn ] }; |
my $rec = { '000' => [ $mfn ] }; |
485 |
|
|
486 |
my $row = $self->fetch($mfn); |
my $row = $self->fetch($mfn) || return; |
487 |
|
|
488 |
foreach my $k (keys %{$row}) { |
foreach my $k (keys %{$row}) { |
489 |
foreach my $l (@{$row->{$k}}) { |
foreach my $l (@{$row->{$k}}) { |
490 |
|
|
491 |
# filter output |
# filter output |
492 |
$l = $self->{'hash_filter'}->($l) if ($self->{'hash_filter'}); |
if ($self->{'hash_filter'}) { |
493 |
|
$l = $self->{'hash_filter'}->($l); |
494 |
|
next unless defined($l); |
495 |
|
} |
496 |
|
|
497 |
my $val; |
my $val; |
498 |
|
|
503 |
if ($l =~ m/\^/) { |
if ($l =~ m/\^/) { |
504 |
foreach my $t (split(/\^/,$l)) { |
foreach my $t (split(/\^/,$l)) { |
505 |
next if (! $t); |
next if (! $t); |
506 |
$val->{substr($t,0,1)} = substr($t,1); |
my ($sf,$v) = (substr($t,0,1), substr($t,1)); |
507 |
|
warn "### $k^$sf:$v",$/ if ($self->{debug} > 1); |
508 |
|
if (ref( $val->{$sf} ) eq 'ARRAY') { |
509 |
|
push @{ $val->{$sf} }, $v; |
510 |
|
} elsif (defined( $val->{$sf} )) { |
511 |
|
# convert scalar field to array |
512 |
|
$val->{$sf} = [ $val->{$sf}, $v ]; |
513 |
|
} else { |
514 |
|
$val->{$sf} = $v; |
515 |
|
} |
516 |
} |
} |
517 |
} else { |
} else { |
518 |
$val = $l; |
$val = $l; |
539 |
return $self->{'TagName'}->{$tag} || $tag; |
return $self->{'TagName'}->{$tag} || $tag; |
540 |
} |
} |
541 |
|
|
542 |
|
|
543 |
|
=head2 read_cnt |
544 |
|
|
545 |
|
Read content of C<.CNT> file and return hash containing it. |
546 |
|
|
547 |
|
print Dumper($isis->read_cnt); |
548 |
|
|
549 |
|
This function is not used by module (C<.CNT> files are not required for this |
550 |
|
module to work), but it can be useful to examine your index (while debugging |
551 |
|
for example). |
552 |
|
|
553 |
|
=cut |
554 |
|
|
555 |
|
sub read_cnt { |
556 |
|
my $self = shift; |
557 |
|
|
558 |
|
croak "missing CNT file in ",$self->{isisdb} unless ($self->{cnt_file}); |
559 |
|
|
560 |
|
# Get the index information from $db.CNT |
561 |
|
|
562 |
|
open(my $fileCNT, $self->{cnt_file}) || croak "can't read '$self->{cnt_file}': $!"; |
563 |
|
binmode($fileCNT); |
564 |
|
|
565 |
|
my $buff; |
566 |
|
|
567 |
|
read($fileCNT, $buff, 26) || croak "can't read first table from CNT: $!"; |
568 |
|
$self->unpack_cnt($buff); |
569 |
|
|
570 |
|
read($fileCNT, $buff, 26) || croak "can't read second table from CNT: $!"; |
571 |
|
$self->unpack_cnt($buff); |
572 |
|
|
573 |
|
close($fileCNT); |
574 |
|
|
575 |
|
return $self->{cnt}; |
576 |
|
} |
577 |
|
|
578 |
|
=head2 unpack_cnt |
579 |
|
|
580 |
|
Unpack one of two 26 bytes fixed length record in C<.CNT> file. |
581 |
|
|
582 |
|
Here is definition of record: |
583 |
|
|
584 |
|
off key description size |
585 |
|
0: IDTYPE BTree type s |
586 |
|
2: ORDN Nodes Order s |
587 |
|
4: ORDF Leafs Order s |
588 |
|
6: N Number of Memory buffers for nodes s |
589 |
|
8: K Number of buffers for first level index s |
590 |
|
10: LIV Current number of Index Levels s |
591 |
|
12: POSRX Pointer to Root Record in N0x l |
592 |
|
16: NMAXPOS Next Available position in N0x l |
593 |
|
20: FMAXPOS Next available position in L0x l |
594 |
|
24: ABNORMAL Formal BTree normality indicator s |
595 |
|
length: 26 bytes |
596 |
|
|
597 |
|
This will fill C<$self> object under C<cnt> with hash. It's used by C<read_cnt>. |
598 |
|
|
599 |
|
=cut |
600 |
|
|
601 |
|
sub unpack_cnt { |
602 |
|
my $self = shift; |
603 |
|
|
604 |
|
my @flds = qw(ORDN ORDF N K LIV POSRX NMAXPOS FMAXPOS ABNORMAL); |
605 |
|
|
606 |
|
my $buff = shift || return; |
607 |
|
my @arr = unpack("vvvvvvVVVv", $buff); |
608 |
|
|
609 |
|
print STDERR "unpack_cnt: ",join(" ",@arr),"\n" if ($self->{'debug'}); |
610 |
|
|
611 |
|
my $IDTYPE = shift @arr; |
612 |
|
foreach (@flds) { |
613 |
|
$self->{cnt}->{$IDTYPE}->{$_} = abs(shift @arr); |
614 |
|
} |
615 |
|
} |
616 |
|
|
617 |
1; |
1; |
618 |
|
|
619 |
=head1 BUGS |
=head1 BUGS |