/[Biblio-Isis]/trunk/lib/Biblio/Isis.pm
This is repository of my old source code which isn't updated any more. Go to git.rot13.org for current projects!
ViewVC logotype

Diff of /trunk/lib/Biblio/Isis.pm

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

trunk/IsisDB.pm revision 18 by dpavlin, Thu Dec 30 22:40:53 2004 UTC trunk/lib/Biblio/Isis.pm revision 61 by dpavlin, Sun Jul 9 21:36:33 2006 UTC
# Line 1  Line 1 
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     = 0.07;          $VERSION     = 0.22_1;
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 ();
# Line 20  BEGIN { Line 18  BEGIN {
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    
31    for(my $mfn = 1; $mfn <= $isis->{'maxmfn'}; $mfn++) {    for(my $mfn = 1; $mfn <= $isis->count; $mfn++) {
32          print $isis->to_ascii($mfn),"\n";          print $isis->to_ascii($mfn),"\n";
33    }    }
34    
35  =head1 DESCRIPTION  =head1 DESCRIPTION
36    
37  This module will read ISIS databases created by DOS CDS/ISIS, WinIsis or  This module will read ISIS databases created by DOS CDS/ISIS, WinIsis or
38  IsisMarc. It can be used as perl-only alternative to OpenIsis module.  IsisMarc. It can be used as perl-only alternative to OpenIsis module which
39    seems to depriciate it's old C<XS> bindings for perl.
40    
41  It can create hash values from data in ISIS database (using C<to_hash>),  It can create hash values from data in ISIS database (using C<to_hash>),
42  ASCII dump (using C<to_ascii>) or just hash with field names and packed  ASCII dump (using C<to_ascii>) or just hash with field names and packed
# Line 50  fields which are zero sized will be fill Line 49  fields which are zero sized will be fill
49  It also has support for identifiers (only if ISIS database is created by  It also has support for identifiers (only if ISIS database is created by
50  IsisMarc), see C<to_hash>.  IsisMarc), see C<to_hash>.
51    
52  This will module will always be slower than OpenIsis module which use C  This module will always be slower than OpenIsis module which use C
53  library. However, since it's written in perl, it's platform independent (so  library. However, since it's written in perl, it's platform independent (so
54  you don't need C compiler), and can be easily modified. I hope that it  you don't need C compiler), and can be easily modified. I hope that it
55  creates data structures which are easier to use than ones created by  creates data structures which are easier to use than ones created by
# Line 80  rarely an issue). Line 79  rarely an issue).
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,
# Line 89  Open ISIS database Line 88  Open ISIS database
88                  $v =~ s#foo#bar#g;                  $v =~ s#foo#bar#g;
89          },          },
90          debug => 1,          debug => 1,
91            join_subfields_with => ' ; ',
92   );   );
93    
94  Options are described below:  Options are described below:
# Line 118  Filter code ref which will be used befor Line 118  Filter code ref which will be used befor
118    
119  =item debug  =item debug
120    
121  Dump a B<lot> of debugging output.  Dump a B<lot> of debugging output even at level 1. For even more increase level.
122    
123  =back  =item join_subfields_with
124    
125  It will also set C<$isis-E<gt>{'maxmfn'}> which is maximum MFN stored in database.  Define delimiter which will be used to join repeatable subfields. This
126    option is included to support lagacy application written against version
127    older than 0.21 of this module. By default, it disabled. See L</to_hash>.
128    
129    =back
130    
131  =cut  =cut
132    
# Line 148  sub new { Line 152  sub new {
152          push @must_exist, "fdt" if ($self->{read_fdt});          push @must_exist, "fdt" if ($self->{read_fdt});
153    
154          foreach my $ext (@must_exist) {          foreach my $ext (@must_exist) {
155                  confess "missing ",uc($ext)," file in ",$self->{isisdb} unless ($self->{$ext."_file"});                  unless ($self->{$ext."_file"}) {
156                            carp "missing ",uc($ext)," file in ",$self->{isisdb};
157                            return;
158                    }
159          }          }
160    
161          print STDERR "## using files: ",join(" ",@isis_files),"\n" if ($self->{debug});          if ($self->{debug}) {
162                    print STDERR "## using files: ",join(" ",@isis_files),"\n";
163                    eval "use Data::Dump";
164    
165                    if (! $@) {
166                            *Dumper = *Data::Dump::dump;
167                    } else {
168                            use Data::Dumper;
169                    }
170            }
171    
172          # if you want to read .FDT file use read_fdt argument when creating class!          # if you want to read .FDT file use read_fdt argument when creating class!
173          if ($self->{read_fdt} && -e $self->{fdt_file}) {          if ($self->{read_fdt} && -e $self->{fdt_file}) {
# Line 159  sub new { Line 175  sub new {
175                  # read the $db.FDT file for tags                  # read the $db.FDT file for tags
176                  my $fieldzone=0;                  my $fieldzone=0;
177    
178                  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}': $!";
179                    binmode($fileFDT);
180    
181                  while (<fileFDT>) {                  while (<$fileFDT>) {
182                          chomp;                          chomp;
183                          if ($fieldzone) {                          if ($fieldzone) {
184                                  my $name=substr($_,0,30);                                  my $name=substr($_,0,30);
# Line 178  sub new { Line 195  sub new {
195                          }                          }
196                  }                  }
197                                    
198                  close(fileFDT);                  close($fileFDT);
199          }          }
200    
201          # Get the Maximum MFN from $db.MST          # Get the Maximum MFN from $db.MST
202    
203          open($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}': $!";
204            binmode($self->{'fileMST'});
205    
206          # MST format:   (* = 32 bit signed)          # MST format:   (* = 32 bit signed)
207          # CTLMFN*       always 0          # CTLMFN*       always 0
# Line 191  sub new { Line 209  sub new {
209          # NXTMFB*       last block allocated to master file          # NXTMFB*       last block allocated to master file
210          # NXTMFP        offset to next available position in last block          # NXTMFP        offset to next available position in last block
211          # MFTYPE        always 0 for user db file (1 for system)          # MFTYPE        always 0 for user db file (1 for system)
212          seek($self->{'fileMST'},4,0);          seek($self->{'fileMST'},4,0) || croak "can't seek to offset 0 in MST: $!";
213    
214          my $buff;          my $buff;
215    
216          read($self->{'fileMST'}, $buff, 4);          read($self->{'fileMST'}, $buff, 4) || croak "can't read NXTMFN from MST: $!";
217          $self->{'NXTMFN'}=unpack("l",$buff) || carp "NXTNFN is zero";          $self->{'NXTMFN'}=unpack("V",$buff) || croak "NXTNFN is zero";
   
         # save maximum MFN  
         $self->{'maxmfn'} = $self->{'NXTMFN'} - 1;  
   
   
218    
219            print STDERR "## self ",Dumper($self),"\n" if ($self->{debug});
         print STDERR Dumper($self),"\n" if ($self->{debug});  
220    
221          # open files for later          # open files for later
222          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}': $!";
223            binmode($self->{'fileXRF'});
224    
225          $self ? return $self : return undef;          $self ? return $self : return undef;
226  }  }
227    
228  =head2 read_cnt  =head2 count
229    
230  This function is not really used by module, but can be useful to find info  Return number of records in database
 about your index (if debugging it for example).  
231    
232    print Dumper($isis->read_cnt);    print $isis->count;
233    
234  =cut  =cut
235    
236  sub read_cnt  {  sub count {
237          my $self = shift;          my $self = shift;
238            return $self->{'NXTMFN'} - 1;
         confess "missing CNT file in ",$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}': $!";  
   
         # There is two 26 Bytes fixed lenght records  
   
         #  0: IDTYPE    BTree type                              16  
         #  2: ORDN      Nodes Order                             16  
         #  4: ORDF      Leafs Order                             16  
         #  6: N         Number of Memory buffers for nodes      16  
         #  8: K         Number of buffers for first level index 16  
         # 10: LIV       Current number of Index Levels          16  
         # 12: POSRX*    Pointer to Root Record in N0x           32  
         # 16: NMAXPOS*  Next Available position in N0x          32  
         # 20: FMAXPOS*  Next available position in L0x          32  
         # 24: ABNORMAL  Formal BTree normality indicator        16  
         # length: 26 bytes  
   
         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);  
                 }  
         }  
   
         my $buff;  
   
         read(fileCNT, $buff, 26);  
         $self->unpack_cnt($buff);  
   
         read(fileCNT, $buff, 26);  
         $self->unpack_cnt($buff);  
   
         close(fileCNT);  
   
         return $self->{cnt};  
239  }  }
240    
241  =head2 fetch  =head2 fetch
# Line 296  sub fetch { Line 261  sub fetch {
261    
262          # is mfn allready in memory?          # is mfn allready in memory?
263          my $old_mfn = $self->{'current_mfn'} || -1;          my $old_mfn = $self->{'current_mfn'} || -1;
264          return if ($mfn == $old_mfn);          return $self->{record} if ($mfn == $old_mfn);
265    
266          print STDERR "## fetch: $mfn\n" if ($self->{debug});          print STDERR "## fetch: $mfn\n" if ($self->{debug});
267    
# Line 308  sub fetch { Line 273  sub fetch {
273    
274          my $buff;          my $buff;
275    
276            # delete old record
277            delete $self->{record};
278    
279          # read XRFMFB abd XRFMFP          # read XRFMFB abd XRFMFP
280          read($self->{'fileXRF'}, $buff, 4);          read($self->{'fileXRF'}, $buff, 4);
281          my $pointer=unpack("l",$buff) || carp "pointer is null";          my $pointer=unpack("V",$buff);
282            if (! $pointer) {
283                    if ($self->{include_deleted}) {
284                            return;
285                    } else {
286                            warn "pointer for MFN $mfn is null\n";
287                            return;
288                    }
289            }
290    
291            # check for logically deleted record
292            if ($pointer & 0x80000000) {
293                    print STDERR "## record $mfn is logically deleted\n" if ($self->{debug});
294                    $self->{deleted} = $mfn;
295    
296                    return unless $self->{include_deleted};
297    
298                    # abs
299                    $pointer = ($pointer ^ 0xffffffff) + 1;
300            }
301    
302          my $XRFMFB = int($pointer/2048);          my $XRFMFB = int($pointer/2048);
303          my $XRFMFP = $pointer - ($XRFMFB*2048);          my $XRFMFP = $pointer - ($XRFMFB*2048);
304    
   
305          # (XRFMFB - 1) * 512 + XRFMFP          # (XRFMFB - 1) * 512 + XRFMFP
306          # why do i have to do XRFMFP % 1024 ?          # why do i have to do XRFMFP % 1024 ?
307    
308          my $blk_off = (($XRFMFB - 1) * 512) + ($XRFMFP % 1024);          my $blk_off = (($XRFMFB - 1) * 512) + ($XRFMFP % 512);
309    
310          print STDERR "## pointer: $pointer XRFMFB: $XRFMFB XRFMFP: $XRFMFP offset: $blk_off\n" if ($self->{'debug'});          print STDERR "## pointer: $pointer XRFMFB: $XRFMFB XRFMFP: $XRFMFP offset: $blk_off\n" if ($self->{'debug'});
311    
312          # Get Record Information          # Get Record Information
313    
314          seek($self->{'fileMST'},$blk_off,0);          seek($self->{'fileMST'},$blk_off,0) || croak "can't seek to $blk_off: $!";
315    
316          read($self->{'fileMST'}, $buff, 4);          read($self->{'fileMST'}, $buff, 4) || croak "can't read 4 bytes at offset $blk_off from MST file: $!";
317          my $value=unpack("l",$buff);          my $value=unpack("V",$buff);
318    
319          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});
320    
321          if ($value!=$mfn) {          if ($value!=$mfn) {
322                  carp "Error: MFN ".$mfn." not found in MST(".$value.")";                      if ($value == 0) {
323                  #return;                # XXX deleted record?                          print STDERR "## record $mfn is physically deleted\n" if ($self->{debug});
324          }                          $self->{deleted} = $mfn;
325                            return;
326                    }
327    
328  #       $MFRL=$self->Read16($fileMST);                  carp "Error: MFN ".$mfn." not found in MST file, found $value";    
329  #       $MFBWB=$self->Read32($fileMST);                  return;
330  #       $MFBWP=$self->Read16($fileMST);          }
 #       $BASE=$self->Read16($fileMST);  
 #       $NVF=$self->Read16($fileMST);  
 #       $STATUS=$self->Read16($fileMST);  
331    
332          read($self->{'fileMST'}, $buff, 14);          read($self->{'fileMST'}, $buff, 14);
333    
334          my ($MFRL,$MFBWB,$MFBWP,$BASE,$NVF,$STATUS) = unpack("slssss", $buff);          my ($MFRL,$MFBWB,$MFBWP,$BASE,$NVF,$STATUS) = unpack("vVvvvv", $buff);
335    
336          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});
337    
338          # delete old record          warn "MFRL $MFRL is not even number" unless ($MFRL % 2 == 0);
         delete $self->{record};  
   
         ## FIXME this is a bug  
         if (! $self->{'include_deleted'} && $MFRL < 0) {  
                 print "## logically deleted record $mfn, skipping...\n" if ($self->{debug});  
                 return;  
         }  
339    
340          warn "BASE is not 18+6*NVF" unless ($BASE == 18 + 6 * $NVF);          warn "BASE is not 18+6*NVF" unless ($BASE == 18 + 6 * $NVF);
341    
# Line 373  sub fetch { Line 351  sub fetch {
351    
352          for (my $i = 0 ; $i < $NVF ; $i++) {          for (my $i = 0 ; $i < $NVF ; $i++) {
353    
354  #               $TAG=$self->Read16($fileMST);                  my ($TAG,$POS,$LEN) = unpack("vvv", substr($buff,$i * 6, 6));
 #               $POS=$self->Read16($fileMST);  
 #               $LEN=$self->Read16($fileMST);  
   
                 my ($TAG,$POS,$LEN) = unpack("sss", substr($buff,$i * 6, 6));  
355    
356                  print STDERR "## TAG: $TAG POS: $POS LEN: $LEN\n" if ($self->{debug});                  print STDERR "## TAG: $TAG POS: $POS LEN: $LEN\n" if ($self->{debug});
357    
# Line 413  sub fetch { Line 387  sub fetch {
387    
388          $self->{'current_mfn'} = $mfn;          $self->{'current_mfn'} = $mfn;
389    
390          print Dumper($self),"\n" if ($self->{debug});          print STDERR Dumper($self),"\n" if ($self->{debug});
391    
392          return $self->{'record'};          return $self->{'record'};
393  }  }
394    
395    =head2 mfn
396    
397    Returns current MFN position
398    
399      my $mfn = $isis->mfn;
400    
401    =cut
402    
403    # This function should be simple return $self->{current_mfn},
404    # but if new is called with _hack_mfn it becomes setter.
405    # It's useful in tests when setting $isis->{record} directly
406    
407    sub mfn {
408            my $self = shift;
409            return $self->{current_mfn};
410    };
411    
412    
413  =head2 to_ascii  =head2 to_ascii
414    
415  Dump ASCII output of record with specified MFN  Returns ASCII output of record with specified MFN
416    
417    print $isis->to_ascii(42);    print $isis->to_ascii(42);
418    
419  It outputs something like this:  This outputs something like this:
420    
421    210   ^aNew York^cNew York University press^dcop. 1988    210   ^aNew York^cNew York University press^dcop. 1988
422    990   2140    990   2140
# Line 441  sub to_ascii { Line 433  sub to_ascii {
433    
434          my $mfn = shift || croak "need MFN";          my $mfn = shift || croak "need MFN";
435    
436          my $rec = $self->fetch($mfn);          my $rec = $self->fetch($mfn) || return;
437    
438          my $out = "0\t$mfn";          my $out = "0\t$mfn";
439    
# Line 461  Read record with specified MFN and conve Line 453  Read record with specified MFN and conve
453    
454    my $hash = $isis->to_hash($mfn);    my $hash = $isis->to_hash($mfn);
455    
456  It has ability to convert characters (using C<hash_filter> from ISIS  It has ability to convert characters (using C<hash_filter>) from ISIS
457  database before creating structures enabling character re-mapping or quick  database before creating structures enabling character re-mapping or quick
458  fix-up of data.  fix-up of data.
459    
# Line 497  which will be used for identifiers, C<i1 Line 489  which will be used for identifiers, C<i1
489               }               }
490             ],             ],
491    
492    In case there are repeatable subfields in record, this will create
493    following structure:
494    
495      '900' => [ {
496            'a' => [ 'foo', 'bar', 'baz' ],
497      }]
498    
499    Or in more complex example of
500    
501      902   ^aa1^aa2^aa3^bb1^aa4^bb2^cc1^aa5
502    
503    it will create
504    
505      902   => [
506            { a => ["a1", "a2", "a3", "a4", "a5"], b => ["b1", "b2"], c => "c1" },
507      ],
508    
509    This behaviour can be changed using C<join_subfields_with> option to L</new>,
510    in which case C<to_hash> will always create single value for each subfield.
511    This will change result to:
512    
513    
514    
515  This method will also create additional field C<000> with MFN.  This method will also create additional field C<000> with MFN.
516    
517    There is also more elaborative way to call C<to_hash> like this:
518    
519      my $hash = $isis->to_hash({
520            mfn => 42,
521            include_subfields => 1,
522            regexps => [
523                    's/something/else/g',
524            ],
525      });
526    
527    Each option controll creation of hash:
528    
529    =over 4
530    
531    =item mfn
532    
533    Specify MFN number of record
534    
535    =item include_subfields
536    
537    This option will create additional key in hash called C<subfields> which will
538    have original record subfield order and index to that subfield like this:
539    
540      902   => [ {
541            a => ["a1", "a2", "a3", "a4", "a5"],
542            b => ["b1", "b2"],
543            c => "c1",
544            subfields => ["a", 0, "a", 1, "a", 2, "b", 0, "a", 3, "b", 1, "c", 0, "a", 4],
545      } ],
546    
547    =item join_subfields_with
548    
549    Define delimiter which will be used to join repeatable subfields. You can
550    specify option here instead in L</new> if you want to have per-record control.
551    
552    =back
553    
554  =cut  =cut
555    
556  sub to_hash {  sub to_hash {
557          my $self = shift;          my $self = shift;
558    
559    
560          my $mfn = shift || confess "need mfn!";          my $mfn = shift || confess "need mfn!";
561            my $arg;
562    
563            if (ref($mfn) eq 'HASH') {
564                    $arg = $mfn;
565                    $mfn = $arg->{mfn} || confess "need mfn in arguments";
566            }
567    
568            confess "regexps must be HASH" if ($arg->{regexps} && ref($arg->{regexps}) ne 'HASH');
569    
570          # init record to include MFN as field 000          # init record to include MFN as field 000
571          my $rec = { '000' => [ $mfn ] };          my $rec = { '000' => [ $mfn ] };
572    
573          my $row = $self->fetch($mfn);          my $row = $self->fetch($mfn) || return;
574    
575            my $j_rs = $arg->{join_subfields_with};
576            $j_rs = $self->{join_subfields_with} unless(defined($j_rs));
577            my $i_sf = $arg->{include_subfields};
578    
579          foreach my $k (keys %{$row}) {          foreach my $f_nr (keys %{$row}) {
580                  foreach my $l (@{$row->{$k}}) {                  foreach my $l (@{$row->{$f_nr}}) {
581    
582                          # filter output                          # filter output
583                          $l = $self->{'hash_filter'}->($l) if ($self->{'hash_filter'});                          if ($self->{'hash_filter'}) {
584                                    $l = $self->{'hash_filter'}->($l);
585                                    next unless defined($l);
586                            }
587    
588                            # apply regexps
589                            if ($arg->{regexps} && defined($arg->{regexps}->{$f_nr})) {
590                                    confess "regexps->{$f_nr} must be ARRAY" if (ref($arg->{regexps}->{$f_nr}) ne 'ARRAY');
591                                    my $c = 0;
592                                    foreach my $r (@{ $arg->{regexps}->{$f_nr} }) {
593                                            while ( eval '$l =~ ' . $r ) { $c++ };
594                                    }
595                                    warn "## field $f_nr triggered $c regexpes\n" if ($c && $self->{debug});
596                            }
597    
598                          my $val;                          my $val;
599                            my $r_sf;       # repeatable subfields in this record
600    
601                          # has identifiers?                          # has identifiers?
602                          ($val->{'i1'},$val->{'i2'}) = ($1,$2) if ($l =~ s/^([01 #])([01 #])//);                          ($val->{'i1'},$val->{'i2'}) = ($1,$2) if ($l =~ s/^([01 #])([01 #])\^/\^/);
603    
604                          # has subfields?                          # has subfields?
605                          if ($l =~ m/\^/) {                          if ($l =~ m/\^/) {
606                                  foreach my $t (split(/\^/,$l)) {                                  foreach my $t (split(/\^/,$l)) {
607                                          next if (! $t);                                          next if (! $t);
608                                          $val->{substr($t,0,1)} = substr($t,1);                                          my ($sf,$v) = (substr($t,0,1), substr($t,1));
609                                            # XXX this might be option, but why?
610                                            next unless ($v);
611    #                                       warn "### $f_nr^$sf:$v",$/ if ($self->{debug} > 1);
612    
613                                            if (ref( $val->{$sf} ) eq 'ARRAY') {
614    
615                                                    push @{ $val->{$sf} }, $v;
616    
617                                                    # record repeatable subfield it it's offset
618                                                    push @{ $val->{subfields} }, ( $sf, $#{ $val->{$sf} } ) if (! $j_rs && $i_sf);
619                                                    $r_sf->{$sf}++;
620    
621                                            } elsif (defined( $val->{$sf} )) {
622    
623                                                    # convert scalar field to array
624                                                    $val->{$sf} = [ $val->{$sf}, $v ];
625    
626                                                    push @{ $val->{subfields} }, ( $sf, 1 ) if (! $j_rs && $i_sf);
627                                                    $r_sf->{$sf}++;
628    
629                                            } else {
630                                                    $val->{$sf} = $v;
631                                                    push @{ $val->{subfields} }, ( $sf, 0 ) if ($i_sf);
632                                            }
633                                  }                                  }
634                          } else {                          } else {
635                                  $val = $l;                                  $val = $l;
636                          }                          }
637    
638                          push @{$rec->{$k}}, $val;                          if ($j_rs) {
639                                    map {
640                                            $val->{$_} = join($j_rs, @{ $val->{$_} });
641                                    } keys %$r_sf
642                            }
643    
644                            push @{$rec->{$f_nr}}, $val;
645                  }                  }
646          }          }
647    
# Line 553  sub tag_name { Line 662  sub tag_name {
662          return $self->{'TagName'}->{$tag} || $tag;          return $self->{'TagName'}->{$tag} || $tag;
663  }  }
664    
665    
666    =head2 read_cnt
667    
668    Read content of C<.CNT> file and return hash containing it.
669    
670      print Dumper($isis->read_cnt);
671    
672    This function is not used by module (C<.CNT> files are not required for this
673    module to work), but it can be useful to examine your index (while debugging
674    for example).
675    
676    =cut
677    
678    sub read_cnt  {
679            my $self = shift;
680    
681            croak "missing CNT file in ",$self->{isisdb} unless ($self->{cnt_file});
682    
683            # Get the index information from $db.CNT
684      
685            open(my $fileCNT, $self->{cnt_file}) || croak "can't read '$self->{cnt_file}': $!";
686            binmode($fileCNT);
687    
688            my $buff;
689    
690            read($fileCNT, $buff, 26) || croak "can't read first table from CNT: $!";
691            $self->unpack_cnt($buff);
692    
693            read($fileCNT, $buff, 26) || croak "can't read second table from CNT: $!";
694            $self->unpack_cnt($buff);
695    
696            close($fileCNT);
697    
698            return $self->{cnt};
699    }
700    
701    =head2 unpack_cnt
702    
703    Unpack one of two 26 bytes fixed length record in C<.CNT> file.
704    
705    Here is definition of record:
706    
707     off key        description                             size
708      0: IDTYPE     BTree type                              s
709      2: ORDN       Nodes Order                             s
710      4: ORDF       Leafs Order                             s
711      6: N          Number of Memory buffers for nodes      s
712      8: K          Number of buffers for first level index s
713     10: LIV        Current number of Index Levels          s
714     12: POSRX      Pointer to Root Record in N0x           l
715     16: NMAXPOS    Next Available position in N0x          l
716     20: FMAXPOS    Next available position in L0x          l
717     24: ABNORMAL   Formal BTree normality indicator        s
718     length: 26 bytes
719    
720    This will fill C<$self> object under C<cnt> with hash. It's used by C<read_cnt>.
721    
722    =cut
723    
724    sub unpack_cnt {
725            my $self = shift;
726    
727            my @flds = qw(ORDN ORDF N K LIV POSRX NMAXPOS FMAXPOS ABNORMAL);
728    
729            my $buff = shift || return;
730            my @arr = unpack("vvvvvvVVVv", $buff);
731    
732            print STDERR "unpack_cnt: ",join(" ",@arr),"\n" if ($self->{'debug'});
733    
734            my $IDTYPE = shift @arr;
735            foreach (@flds) {
736                    $self->{cnt}->{$IDTYPE}->{$_} = abs(shift @arr);
737            }
738    }
739    
740  1;  1;
741    
742  =head1 BUGS  =head1 BUGS
743    
744  This module has been very lightly tested. Use with caution and report bugs.  Some parts of CDS/ISIS documentation are not detailed enough to exmplain
745    some variations in input databases which has been tested with this module.
746    When I was in doubt, I assumed that OpenIsis's implementation was right
747    (except for obvious bugs).
748    
749    However, every effort has been made to test this module with as much
750    databases (and programs that create them) as possible.
751    
752    I would be very greatful for success or failure reports about usage of this
753    module with databases from programs other than WinIsis and IsisMarc. I had
754    tested this against ouput of one C<isis.dll>-based application, but I don't
755    know any details about it's version.
756    
757    =head1 VERSIONS
758    
759    As this is young module, new features are added in subsequent version. It's
760    a good idea to specify version when using this module like this:
761    
762      use Biblio::Isis 0.21
763    
764    Below is list of changes in specific version of module (so you can target
765    older versions if you really have to):
766    
767    =over 8
768    
769    =item 0.21
770    
771    Added C<join_subfields_with> to L</new> and L</to_hash>.
772    
773    Added C<include_subfields> to L</to_hash>.
774    
775    =item 0.20
776    
777    Added C<< $isis->mfn >>, support for repeatable subfields and
778    C<< $isis->to_hash({ mfn => 42, ... }) >> calling convention
779    
780    =back
781    
782  =head1 AUTHOR  =head1 AUTHOR
783    
# Line 580  LICENSE file included with this module. Line 800  LICENSE file included with this module.
800    
801  =head1 SEE ALSO  =head1 SEE ALSO
802    
803    L<Biblio::Isis::Manual> for CDS/ISIS manual appendix F, G and H which describe file format
804    
805  OpenIsis web site L<http://www.openisis.org>  OpenIsis web site L<http://www.openisis.org>
806    
807  perl4lib site L<http://perl4lib.perl.org>  perl4lib site L<http://perl4lib.perl.org>
Legend:
Removed from v.18  
changed lines
  Added in v.61
  ViewVC Help
Powered by ViewVC 1.1.26