/[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 8 by dpavlin, Wed Dec 29 15:17:59 2004 UTC trunk/lib/Biblio/Isis.pm revision 56 by dpavlin, Sat Jul 8 16:03:52 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 Data::Dumper;  use File::Glob qw(:globally :nocase);
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     = 0.02;          $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 ();
# Line 18  BEGIN { Line 18  BEGIN {
18    
19  =head1 NAME  =head1 NAME
20    
21  IsisDB - Read CDS/ISIS database  Biblio::Isis - Read CDS/ISIS, WinISIS and IsisMarc database
22    
23  =head1 SYNOPSIS  =head1 SYNOPSIS
24    
25    use IsisDB    use Biblio::Isis;
26    my $isis = new IsisDB(  
27      my $isis = new Biblio::Isis(
28          isisdb => './cds/cds',          isisdb => './cds/cds',
29    );    );
30    
31      for(my $mfn = 1; $mfn <= $isis->count; $mfn++) {
32            print $isis->to_ascii($mfn),"\n";
33      }
34    
35  =head1 DESCRIPTION  =head1 DESCRIPTION
36    
37  This module will read CDS/ISIS databases and create hash values out of it.  This module will read ISIS databases created by DOS CDS/ISIS, WinIsis or
38  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>),
42    ASCII dump (using C<to_ascii>) or just hash with field names and packed
43    values (like C<^asomething^belse>).
44    
45    Unique feature of this module is ability to C<include_deleted> records.
46    It will also skip zero sized fields (OpenIsis has a bug in XS bindings, so
47    fields which are zero sized will be filled with random junk from memory).
48    
49    It also has support for identifiers (only if ISIS database is created by
50    IsisMarc), see C<to_hash>.
51    
52    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
54    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
56    OpenIsis, so reduced time in other parts of the code should compensate for
57    slower performance of this module (speed of reading ISIS database is
58    rarely an issue).
59    
60  =head1 METHODS  =head1 METHODS
61    
# Line 50  It can be used as perl-only alternative Line 75  It can be used as perl-only alternative
75  # some binary reads  # some binary reads
76  #  #
77    
 sub Read32 {  
         my $self = shift;  
   
         my $f = shift || die "Read32 needs file handle";  
         read($$f,$b,4) || die "can't read 4 bytes from $$f from position ".tell($f);  
         return unpack("l",$b);  
 }  
   
78  =head2 new  =head2 new
79    
80  Open CDS/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,
86            hash_filter => sub {
87                    my $v = shift;
88                    $v =~ s#foo#bar#g;
89            },
90          debug => 1,          debug => 1,
91   );   );
92    
# Line 74  Options are described below: Line 96  Options are described below:
96    
97  =item isisdb  =item isisdb
98    
99  Prefix path to CDS/ISIS. It should contain full or relative path to database  This is full or relative path to ISIS database files which include
100  and common prefix of C<.FDT>, C<.MST>, C<.CNT>, C<.XRF> and C<.MST> files.  common prefix of C<.MST>, and C<.XRF> and optionally C<.FDT> (if using
101    C<read_fdt> option) files.
102    
103    In this example it uses C<./cds/cds.MST> and related files.
104    
105  =item read_fdt  =item read_fdt
106    
107  Boolean flag to specify if field definition table should be read. It's off  Boolean flag to specify if field definition table should be read. It's off
108  by default.  by default.
109    
110    =item include_deleted
111    
112    Don't skip logically deleted records in ISIS.
113    
114    =item hash_filter
115    
116    Filter code ref which will be used before data is converted to hash.
117    
118  =item debug  =item debug
119    
120  Dump a C<lot> of debugging output.  Dump a B<lot> of debugging output even at level 1. For even more increase level.
121    
122  =back  =back
123    
 It will also set C<$isis-E<gt>{'maxmfn'}> which is maximum MFN stored in database.  
   
124  =cut  =cut
125    
126  sub new {  sub new {
# Line 97  sub new { Line 128  sub new {
128          my $self = {};          my $self = {};
129          bless($self, $class);          bless($self, $class);
130    
131          $self->{isisdb} = {@_}->{isisdb} || croak "new needs database name as argument!";          croak "new needs database name (isisdb) as argument!" unless ({@_}->{isisdb});
132    
133            foreach my $v (qw{isisdb debug include_deleted hash_filter}) {
134                    $self->{$v} = {@_}->{$v};
135            }
136    
137            my @isis_files = grep(/\.(FDT|MST|XRF|CNT)$/i,glob($self->{isisdb}."*"));
138    
139          $self->{debug} = {@_}->{debug};          foreach my $f (@isis_files) {
140                    my $ext = $1 if ($f =~ m/\.(\w\w\w)$/);
141                    $self->{lc($ext)."_file"} = $f;
142            }
143    
144            my @must_exist = qw(mst xrf);
145            push @must_exist, "fdt" if ($self->{read_fdt});
146    
147            foreach my $ext (@must_exist) {
148                    unless ($self->{$ext."_file"}) {
149                            carp "missing ",uc($ext)," file in ",$self->{isisdb};
150                            return;
151                    }
152            }
153    
154            if ($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 ({@_}->{read_fdt} && -e $self->{isisdb}.".FDT") {          if ($self->{read_fdt} && -e $self->{fdt_file}) {
167    
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, $self->{isisdb}.".FDT") || croak "can't read '$self->{isisdb}.FDT': $!";                  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);
# Line 126  sub new { Line 188  sub new {
188                          }                          }
189                  }                  }
190                                    
191                  close(fileFDT);                  close($fileFDT);
192          }          }
193    
194          # Get the Maximum MFN from $db.MST          # Get the Maximum MFN from $db.MST
195    
196          open(fileMST,$self->{isisdb}.".MST") || croak "can't read '$self->{isisdb}.MST': $!";          open($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
# Line 139  sub new { Line 202  sub new {
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(fileMST,4,0);          seek($self->{'fileMST'},4,0) || croak "can't seek to offset 0 in MST: $!";
         $self->{'NXTMFN'}=$self->Read32(\*fileMST) || carp "NXTNFN is zero";  
   
         # save maximum MFN  
         $self->{'maxmfn'} = $self->{'NXTMFN'} - 1;  
   
         close(fileMST);  
206    
207          # Get the index information from $db.CNT          my $buff;
     
         open(fileCNT, $self->{isisdb}.".CNT") || croak "can't read '$self->{isisdb}.CNT': $!";  
208    
209          # There is two 26 Bytes fixed lenght records          read($self->{'fileMST'}, $buff, 4) || croak "can't read NXTMFN from MST: $!";
210            $self->{'NXTMFN'}=unpack("V",$buff) || croak "NXTNFN is zero";
211    
212          #  0: IDTYPE    BTree type                              16          print STDERR "## self ",Dumper($self),"\n" if ($self->{debug});
         #  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);  
   
                 my $IDTYPE = shift @arr;  
                 foreach (@flds) {  
                         $self->{$IDTYPE}->{$_} = abs(shift @arr);  
                 }  
         }  
213    
214          my $buff;          # open files for later
215          read(fileCNT, $buff, 26);          open($self->{'fileXRF'}, $self->{xrf_file}) || croak "can't open '$self->{xrf_file}': $!";
216          $self->unpack_cnt($buff);          binmode($self->{'fileXRF'});
   
         read(fileCNT, $buff, 26);  
         $self->unpack_cnt($buff);  
217    
218            $self ? return $self : return undef;
219    }
220    
221          close(fileCNT);  =head2 count
222    
223          print Dumper($self) if ($self->{debug});  Return number of records in database
224    
225          # open files for later    print $isis->count;
         open($self->{'fileXRF'}, $self->{isisdb}.".XRF") || croak "can't open '$self->{isisdb}.XRF': $!";  
226    
227          open($self->{'fileMST'}, $self->{isisdb}.".MST") || croak "can't open '$self->{isisdb}.MST': $!";  =cut
228    
229          $self ? return $self : return undef;  sub count {
230            my $self = shift;
231            return $self->{'NXTMFN'} - 1;
232  }  }
233    
234  =head2 fetch  =head2 fetch
# Line 206  Read record with selected MFN Line 238  Read record with selected MFN
238    my $rec = $isis->fetch(55);    my $rec = $isis->fetch(55);
239    
240  Returns hash with keys which are field names and values are unpacked values  Returns hash with keys which are field names and values are unpacked values
241  for that field.  for that field like this:
242    
243      $rec = {
244        '210' => [ '^aNew York^cNew York University press^dcop. 1988' ],
245        '990' => [ '2140', '88', 'HAY' ],
246      };
247    
248  =cut  =cut
249    
# Line 215  sub fetch { Line 252  sub fetch {
252    
253          my $mfn = shift || croak "fetch needs MFN as argument!";          my $mfn = shift || croak "fetch needs MFN as argument!";
254    
255          print "fetch: $mfn\n" if ($self->{debug});          # is mfn allready in memory?
256            my $old_mfn = $self->{'current_mfn'} || -1;
257            return $self->{record} if ($mfn == $old_mfn);
258    
259            print STDERR "## fetch: $mfn\n" if ($self->{debug});
260    
261          # XXX check this?          # XXX check this?
262          my $mfnpos=($mfn+int(($mfn-1)/127))*4;          my $mfnpos=($mfn+int(($mfn-1)/127))*4;
263    
264          print "seeking to $mfnpos in file '$self->{isisdb}.XRF'\n" if ($self->{debug});          print STDERR "## seeking to $mfnpos in file '$self->{xrf_file}'\n" if ($self->{debug});
265          seek($self->{'fileXRF'},$mfnpos,0);          seek($self->{'fileXRF'},$mfnpos,0);
266    
267            my $buff;
268    
269            # delete old record
270            delete $self->{record};
271    
272          # read XRFMFB abd XRFMFP          # read XRFMFB abd XRFMFP
273          my $pointer=$self->Read32(\*{$self->{'fileXRF'}});          read($self->{'fileXRF'}, $buff, 4);
274            my $pointer=unpack("V",$buff);
275            if (! $pointer) {
276                    if ($self->{include_deleted}) {
277                            return;
278                    } else {
279                            warn "pointer for MFN $mfn is null\n";
280                            return;
281                    }
282            }
283    
284            # check for logically deleted record
285            if ($pointer & 0x80000000) {
286                    print STDERR "## record $mfn is logically deleted\n" if ($self->{debug});
287                    $self->{deleted} = $mfn;
288    
289                    return unless $self->{include_deleted};
290    
291                    # abs
292                    $pointer = ($pointer ^ 0xffffffff) + 1;
293            }
294    
295          my $XRFMFB = int($pointer/2048);          my $XRFMFB = int($pointer/2048);
296          my $XRFMFP = $pointer - ($XRFMFB*2048);          my $XRFMFP = $pointer - ($XRFMFB*2048);
297    
298          print "XRFMFB: $XRFMFB XRFMFP: $XRFMFP\n" if ($self->{debug});          # (XRFMFB - 1) * 512 + XRFMFP
299            # why do i have to do XRFMFP % 1024 ?
         # XXX fix this to be more readable!!  
         # e.g. (XRFMFB - 1) * 512 + XRFMFP  
300    
301          my $offset = $pointer;          my $blk_off = (($XRFMFB - 1) * 512) + ($XRFMFP % 512);
         my $offset2=int($offset/2048)-1;  
         my $offset22=int($offset/4096);  
         my $offset3=$offset-($offset22*4096);  
         if ($offset3>512) {  
                 $offset3=$offset3-2048;  
         }  
         my $offset4=($offset2*512)+$offset3;  
302    
303          print "$offset - $offset2 - $offset3 - $offset4\n" if ($self->{debug});          print STDERR "## pointer: $pointer XRFMFB: $XRFMFB XRFMFP: $XRFMFP offset: $blk_off\n" if ($self->{'debug'});
304    
305          # Get Record Information          # Get Record Information
306    
307          seek($self->{'fileMST'},$offset4,0);          seek($self->{'fileMST'},$blk_off,0) || croak "can't seek to $blk_off: $!";
308    
309            read($self->{'fileMST'}, $buff, 4) || croak "can't read 4 bytes at offset $blk_off from MST file: $!";
310            my $value=unpack("V",$buff);
311    
312          my $value=$self->Read32(\*{$self->{'fileMST'}});          print STDERR "## offset for rowid $value is $blk_off (blk $XRFMFB off $XRFMFP)\n" if ($self->{debug});
313    
314          if ($value!=$mfn) {          if ($value!=$mfn) {
315  print ("Error: The MFN:".$mfn." is not found in MST(".$value.")");                      if ($value == 0) {
316                  return -1;      # XXX deleted record?                          print STDERR "## record $mfn is physically deleted\n" if ($self->{debug});
317          }                          $self->{deleted} = $mfn;
318                            return;
319                    }
320    
321  #       $MFRL=$self->Read16($fileMST);                  carp "Error: MFN ".$mfn." not found in MST file, found $value";    
322  #       $MFBWB=$self->Read32($fileMST);                  return;
323  #       $MFBWP=$self->Read16($fileMST);          }
 #       $BASE=$self->Read16($fileMST);  
 #       $NVF=$self->Read16($fileMST);  
 #       $STATUS=$self->Read16($fileMST);  
324    
         my $buff;  
325          read($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});
330    
331            warn "MFRL $MFRL is not even number" unless ($MFRL % 2 == 0);
332    
333          print "MFRL: $MFRL MFBWB: $MFBWB MFBWP: $MFBWP BASE: $BASE NVF: $NVF STATUS: $STATUS\n" if ($self->{debug});          warn "BASE is not 18+6*NVF" unless ($BASE == 18 + 6 * $NVF);
334    
335          # Get Directory Format          # Get Directory Format
336    
# Line 278  print ("Error: The MFN:".$mfn." is not f Line 340  print ("Error: The MFN:".$mfn." is not f
340    
341          read($self->{'fileMST'}, $buff, 6 * $NVF);          read($self->{'fileMST'}, $buff, 6 * $NVF);
342    
343          my $fld_len = 0;          my $rec_len = 0;
344    
345          for (my $i = 0 ; $i < $NVF ; $i++) {          for (my $i = 0 ; $i < $NVF ; $i++) {
346    
347  #               $TAG=$self->Read16($fileMST);                  my ($TAG,$POS,$LEN) = unpack("vvv", substr($buff,$i * 6, 6));
 #               $POS=$self->Read16($fileMST);  
 #               $LEN=$self->Read16($fileMST);  
348    
349                  my ($TAG,$POS,$LEN) = unpack("sss", substr($buff,$i * 6, 6));                  print STDERR "## TAG: $TAG POS: $POS LEN: $LEN\n" if ($self->{debug});
   
                 print "TAG: $TAG POS: $POS LEN: $LEN\n" if ($self->{debug});  
350    
351                  # The TAG does not exists in .FDT so we set it to 0.                  # The TAG does not exists in .FDT so we set it to 0.
352                  #                  #
# Line 304  print ("Error: The MFN:".$mfn." is not f Line 362  print ("Error: The MFN:".$mfn." is not f
362                  push @FieldPOS,$POS;                  push @FieldPOS,$POS;
363                  push @FieldLEN,$LEN;                  push @FieldLEN,$LEN;
364    
365                  $fld_len += $LEN;                  $rec_len += $LEN;
366          }          }
367    
368          # Get Variable Fields          # Get Variable Fields
369    
370          delete $self->{record};          read($self->{'fileMST'},$buff,$rec_len);
371    
372          read($self->{'fileMST'},$buff,$fld_len);          print STDERR "## rec_len: $rec_len poc: ",tell($self->{'fileMST'})."\n" if ($self->{debug});
373    
374          for (my $i = 0 ; $i < $NVF ; $i++) {          for (my $i = 0 ; $i < $NVF ; $i++) {
375                    # skip zero-sized fields
376                    next if ($FieldLEN[$i] == 0);
377    
378                  push @{$self->{record}->{$FieldTAG[$i]}}, substr($buff,$FieldPOS[$i],$FieldLEN[$i]);                  push @{$self->{record}->{$FieldTAG[$i]}}, substr($buff,$FieldPOS[$i],$FieldLEN[$i]);
379          }          }
         close(fileMST);  
380    
381          # The record is marked for deletion          $self->{'current_mfn'} = $mfn;
         if ($STATUS==1) {  
                 return -1;  
         }  
382    
383          print Dumper($self) if ($self->{debug});          print STDERR Dumper($self),"\n" if ($self->{debug});
384    
385          return $self->{'record'};          return $self->{'record'};
386  }  }
387    
388    =head2 mfn
389    
390    Returns current MFN position
391    
392      my $mfn = $isis->mfn;
393    
394    =cut
395    
396    # This function should be simple return $self->{current_mfn},
397    # but if new is called with _hack_mfn it becomes setter.
398    # It's useful in tests when setting $isis->{record} directly
399    
400    sub mfn {
401            my $self = shift;
402            return $self->{current_mfn};
403    };
404    
405    
406  =head2 to_ascii  =head2 to_ascii
407    
408  Dump ascii output of selected MFN  Returns ASCII output of record with specified MFN
409    
410      print $isis->to_ascii(42);
411    
412    This outputs something like this:
413    
414      210   ^aNew York^cNew York University press^dcop. 1988
415      990   2140
416      990   88
417      990   HAY
418    
419    print $isis->to_ascii(55);  If C<read_fdt> is specified when calling C<new> it will display field names
420    from C<.FDT> file instead of numeric tags.
421    
422  =cut  =cut
423    
# Line 341  sub to_ascii { Line 426  sub to_ascii {
426    
427          my $mfn = shift || croak "need MFN";          my $mfn = shift || croak "need MFN";
428    
429          my $rec = $self->fetch($mfn);          my $rec = $self->fetch($mfn) || return;
430    
431          my $out = "0\t$mfn";          my $out = "0\t$mfn";
432    
433          foreach my $f (sort keys %{$rec}) {          foreach my $f (sort keys %{$rec}) {
434                  $out .= "\n$f\t".join("\n$f\t",@{$self->{record}->{$f}});                  my $fn = $self->tag_name($f);
435                    $out .= "\n$fn\t".join("\n$fn\t",@{$self->{record}->{$f}});
436          }          }
437    
438          $out .= "\n";          $out .= "\n";
# Line 354  sub to_ascii { Line 440  sub to_ascii {
440          return $out;          return $out;
441  }  }
442    
443  #  =head2 to_hash
444  # XXX porting from php left-over:  
445  #  Read record with specified MFN and convert it to hash
446  # do I *REALLY* need those methods, or should I use  
447  # $self->{something} directly?    my $hash = $isis->to_hash($mfn);
448  #  
449  # Probably direct usage is better!  It has ability to convert characters (using C<hash_filter>) from ISIS
450  #  database before creating structures enabling character re-mapping or quick
451    fix-up of data.
452    
453    This function returns hash which is like this:
454    
455      $hash = {
456        '210' => [
457                   {
458                     'c' => 'New York University press',
459                     'a' => 'New York',
460                     'd' => 'cop. 1988'
461                   }
462                 ],
463        '990' => [
464                   '2140',
465                   '88',
466                   'HAY'
467                 ],
468      };
469    
470    You can later use that hash to produce any output from ISIS data.
471    
472    If database is created using IsisMarc, it will also have to special fields
473    which will be used for identifiers, C<i1> and C<i2> like this:
474    
475      '200' => [
476                 {
477                   'i1' => '1',
478                   'i2' => ' '
479                   'a' => 'Goa',
480                   'f' => 'Valdo D\'Arienzo',
481                   'e' => 'tipografie e tipografi nel XVI secolo',
482                 }
483               ],
484    
485    In case there are repeatable subfields in record, this will create
486    following structure:
487    
488      '900' => [ {
489            'a' => [ 'foo', 'bar', 'baz' ],
490      }]
491    
492    This method will also create additional field C<000> with MFN.
493    
494    There is also more elaborative way to call C<to_hash> like this:
495    
496      my $hash = $isis->to_hash({
497            mfn => 42,
498            include_empty_subfields => 1,
499      });
500    
501    =cut
502    
503    sub to_hash {
504            my $self = shift;
505    
506    
507            my $mfn = shift || confess "need mfn!";
508            my $arg;
509    
510            if (ref($mfn) eq 'HASH') {
511                    $arg = $mfn;
512                    $mfn = $arg->{mfn} || confess "need mfn in arguments";
513            }
514    
515            # init record to include MFN as field 000
516            my $rec = { '000' => [ $mfn ] };
517    
518            my $row = $self->fetch($mfn) || return;
519    
520            foreach my $k (keys %{$row}) {
521                    foreach my $l (@{$row->{$k}}) {
522    
523                            # filter output
524                            if ($self->{'hash_filter'}) {
525                                    $l = $self->{'hash_filter'}->($l);
526                                    next unless defined($l);
527                            }
528    
529                            my $val;
530    
531                            # has identifiers?
532                            ($val->{'i1'},$val->{'i2'}) = ($1,$2) if ($l =~ s/^([01 #])([01 #])\^/\^/);
533    
534                            # has subfields?
535                            if ($l =~ m/\^/) {
536                                    foreach my $t (split(/\^/,$l)) {
537                                            next if (! $t);
538                                            my ($sf,$v) = (substr($t,0,1), substr($t,1));
539                                            # FIXME make this option !
540                                            next unless ($v);
541    #                                       warn "### $k^$sf:$v",$/ if ($self->{debug} > 1);
542    
543                                            # FIXME array return optional, by default unroll to ' ; '
544                                            if (ref( $val->{$sf} ) eq 'ARRAY') {
545    
546                                                    push @{ $val->{$sf} }, $v;
547                                            } elsif (defined( $val->{$sf} )) {
548                                                    # convert scalar field to array
549                                                    $val->{$sf} = [ $val->{$sf}, $v ];
550                                            } else {
551                                                    $val->{$sf} = $v;
552                                            }
553                                    }
554                            } else {
555                                    $val = $l;
556                            }
557    
558                            push @{$rec->{$k}}, $val;
559                    }
560            }
561    
562            return $rec;
563    }
564    
565    =head2 tag_name
566    
567    Return name of selected tag
568    
569     print $isis->tag_name('200');
570    
571    =cut
572    
573    sub tag_name {
574            my $self = shift;
575            my $tag = shift || return;
576            return $self->{'TagName'}->{$tag} || $tag;
577    }
578    
579    
580    =head2 read_cnt
581    
582  sub TagName {  Read content of C<.CNT> file and return hash containing it.
583    
584      print Dumper($isis->read_cnt);
585    
586    This function is not used by module (C<.CNT> files are not required for this
587    module to work), but it can be useful to examine your index (while debugging
588    for example).
589    
590    =cut
591    
592    sub read_cnt  {
593          my $self = shift;          my $self = shift;
594          return $self->{TagName};  
595            croak "missing CNT file in ",$self->{isisdb} unless ($self->{cnt_file});
596    
597            # Get the index information from $db.CNT
598      
599            open(my $fileCNT, $self->{cnt_file}) || croak "can't read '$self->{cnt_file}': $!";
600            binmode($fileCNT);
601    
602            my $buff;
603    
604            read($fileCNT, $buff, 26) || croak "can't read first table from CNT: $!";
605            $self->unpack_cnt($buff);
606    
607            read($fileCNT, $buff, 26) || croak "can't read second table from CNT: $!";
608            $self->unpack_cnt($buff);
609    
610            close($fileCNT);
611    
612            return $self->{cnt};
613  }  }
614    
615  sub NextMFN {  =head2 unpack_cnt
616    
617    Unpack one of two 26 bytes fixed length record in C<.CNT> file.
618    
619    Here is definition of record:
620    
621     off key        description                             size
622      0: IDTYPE     BTree type                              s
623      2: ORDN       Nodes Order                             s
624      4: ORDF       Leafs Order                             s
625      6: N          Number of Memory buffers for nodes      s
626      8: K          Number of buffers for first level index s
627     10: LIV        Current number of Index Levels          s
628     12: POSRX      Pointer to Root Record in N0x           l
629     16: NMAXPOS    Next Available position in N0x          l
630     20: FMAXPOS    Next available position in L0x          l
631     24: ABNORMAL   Formal BTree normality indicator        s
632     length: 26 bytes
633    
634    This will fill C<$self> object under C<cnt> with hash. It's used by C<read_cnt>.
635    
636    =cut
637    
638    sub unpack_cnt {
639          my $self = shift;          my $self = shift;
640          return $self->{NXTMFN};  
641            my @flds = qw(ORDN ORDF N K LIV POSRX NMAXPOS FMAXPOS ABNORMAL);
642    
643            my $buff = shift || return;
644            my @arr = unpack("vvvvvvVVVv", $buff);
645    
646            print STDERR "unpack_cnt: ",join(" ",@arr),"\n" if ($self->{'debug'});
647    
648            my $IDTYPE = shift @arr;
649            foreach (@flds) {
650                    $self->{cnt}->{$IDTYPE}->{$_} = abs(shift @arr);
651            }
652  }  }
653    
654  1;  1;
655    
656  =head1 BUGS  =head1 BUGS
657    
658  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
659    some variations in input databases which has been tested with this module.
660    When I was in doubt, I assumed that OpenIsis's implementation was right
661    (except for obvious bugs).
662    
663    However, every effort has been made to test this module with as much
664    databases (and programs that create them) as possible.
665    
666    I would be very greatful for success or failure reports about usage of this
667    module with databases from programs other than WinIsis and IsisMarc. I had
668    tested this against ouput of one C<isis.dll>-based application, but I don't
669    know any details about it's version.
670    
671    =head1 VERSIONS
672    
673    You can find version dependencies documented here
674    
675    =over 8
676    
677    =item 0.20
678    
679    Added C<< $isis->mfn >>, support for repeatable subfields and
680    C<< $isis->to_hash({ mfn => 42, ... }) >> calling convention
681    
682    =back
683    
684  =head1 AUTHOR  =head1 AUTHOR
685    
# Line 386  This module has been very lightly tested Line 688  This module has been very lightly tested
688          dpavlin@rot13.org          dpavlin@rot13.org
689          http://www.rot13.org/~dpavlin/          http://www.rot13.org/~dpavlin/
690    
691  This module is based heavily on code from LIBISIS.PHP - Library to read ISIS files V0.1.1  This module is based heavily on code from C<LIBISIS.PHP> library to read ISIS files V0.1.1
692  written in php and (c) 2000 Franck Martin - <franck@sopac.org> released under LGPL.  written in php and (c) 2000 Franck Martin <franck@sopac.org> and released under LGPL.
693    
694  =head1 COPYRIGHT  =head1 COPYRIGHT
695    
# Line 400  LICENSE file included with this module. Line 702  LICENSE file included with this module.
702    
703  =head1 SEE ALSO  =head1 SEE ALSO
704    
705  L<http://www.openisis.org|OpenIsis>, perl(1).  OpenIsis web site L<http://www.openisis.org>
706    
707    perl4lib site L<http://perl4lib.perl.org>
708    
Legend:
Removed from v.8  
changed lines
  Added in v.56
  ViewVC Help
Powered by ViewVC 1.1.26