/[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 10 by dpavlin, Wed Dec 29 16:04:07 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 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     = 0.03;          $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 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,
         debug => 1,  
85          include_deleted => 1,          include_deleted => 1,
86            hash_filter => sub {
87                    my $v = shift;
88                    $v =~ s#foo#bar#g;
89            },
90            debug => 1,
91            join_subfields_with => ' ; ',
92   );   );
93    
94  Options are described below:  Options are described below:
# Line 75  Options are described below: Line 97  Options are described below:
97    
98  =item isisdb  =item isisdb
99    
100  Prefix path to CDS/ISIS. It should contain full or relative path to database  This is full or relative path to ISIS database files which include
101  and common prefix of C<.FDT>, C<.MST>, C<.CNT>, C<.XRF> and C<.MST> files.  common prefix of C<.MST>, and C<.XRF> and optionally C<.FDT> (if using
102    C<read_fdt> option) files.
103    
104    In this example it uses C<./cds/cds.MST> and related files.
105    
106  =item read_fdt  =item read_fdt
107    
108  Boolean flag to specify if field definition table should be read. It's off  Boolean flag to specify if field definition table should be read. It's off
109  by default.  by default.
110    
111    =item include_deleted
112    
113    Don't skip logically deleted records in ISIS.
114    
115    =item hash_filter
116    
117    Filter code ref which will be used before data is converted to hash.
118    
119  =item debug  =item debug
120    
121  Dump a C<lot> of debugging output.  Dump a B<lot> of debugging output even at level 1. For even more increase level.
122    
123  =item include_deleted  =item join_subfields_with
124    
125  Don't skip logically deleted records.  Define delimiter which will be used to join repeatable subfields. This
126    option is included to support lagacy application written against version
127    older than 0.21 of this module. By default, it disabled. See L</to_hash>.
128    
129  =back  =back
130    
 It will also set C<$isis-E<gt>{'maxmfn'}> which is maximum MFN stored in database.  
   
131  =cut  =cut
132    
133  sub new {  sub new {
# Line 104  sub new { Line 137  sub new {
137    
138          croak "new needs database name (isisdb) as argument!" unless ({@_}->{isisdb});          croak "new needs database name (isisdb) as argument!" unless ({@_}->{isisdb});
139    
140          foreach my $v (qw{isisdb debug include_deleted}) {          foreach my $v (qw{isisdb debug include_deleted hash_filter}) {
141                  $self->{$v} = {@_}->{$v};                  $self->{$v} = {@_}->{$v};
142          }          }
143    
144            my @isis_files = grep(/\.(FDT|MST|XRF|CNT)$/i,glob($self->{isisdb}."*"));
145    
146            foreach my $f (@isis_files) {
147                    my $ext = $1 if ($f =~ m/\.(\w\w\w)$/);
148                    $self->{lc($ext)."_file"} = $f;
149            }
150    
151            my @must_exist = qw(mst xrf);
152            push @must_exist, "fdt" if ($self->{read_fdt});
153    
154            foreach my $ext (@must_exist) {
155                    unless ($self->{$ext."_file"}) {
156                            carp "missing ",uc($ext)," file in ",$self->{isisdb};
157                            return;
158                    }
159            }
160    
161            if ($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 ({@_}->{read_fdt} && -e $self->{isisdb}.".FDT") {          if ($self->{read_fdt} && -e $self->{fdt_file}) {
174    
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, $self->{isisdb}.".FDT") || croak "can't read '$self->{isisdb}.FDT': $!";                  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 133  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 $db.MST
202    
203          open(fileMST,$self->{isisdb}.".MST") || croak "can't read '$self->{isisdb}.MST': $!";          open($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 146  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(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";  
213    
214          # save maximum MFN          my $buff;
         $self->{'maxmfn'} = $self->{'NXTMFN'} - 1;  
   
         close(fileMST);  
   
         # Get the index information from $db.CNT  
     
         open(fileCNT, $self->{isisdb}.".CNT") || croak "can't read '$self->{isisdb}.CNT': $!";  
215    
216          # There is two 26 Bytes fixed lenght records          read($self->{'fileMST'}, $buff, 4) || croak "can't read NXTMFN from MST: $!";
217            $self->{'NXTMFN'}=unpack("V",$buff) || croak "NXTNFN is zero";
218    
219          #  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);  
   
                 print "unpack_cnt: ",join(" ",@arr),"\n" if ($self->{'debug'});  
   
                 my $IDTYPE = shift @arr;  
                 foreach (@flds) {  
                         $self->{$IDTYPE}->{$_} = abs(shift @arr);  
                 }  
         }  
220    
221          my $buff;          # open files for later
222          read(fileCNT, $buff, 26);          open($self->{'fileXRF'}, $self->{xrf_file}) || croak "can't open '$self->{xrf_file}': $!";
223          $self->unpack_cnt($buff);          binmode($self->{'fileXRF'});
   
         read(fileCNT, $buff, 26);  
         $self->unpack_cnt($buff);  
224    
225            $self ? return $self : return undef;
226    }
227    
228          close(fileCNT);  =head2 count
229    
230          print Dumper($self) if ($self->{debug});  Return number of records in database
231    
232          # open files for later    print $isis->count;
         open($self->{'fileXRF'}, $self->{isisdb}.".XRF") || croak "can't open '$self->{isisdb}.XRF': $!";  
233    
234          open($self->{'fileMST'}, $self->{isisdb}.".MST") || croak "can't open '$self->{isisdb}.MST': $!";  =cut
235    
236          $self ? return $self : return undef;  sub count {
237            my $self = shift;
238            return $self->{'NXTMFN'} - 1;
239  }  }
240    
241  =head2 fetch  =head2 fetch
# Line 215  Read record with selected MFN Line 245  Read record with selected MFN
245    my $rec = $isis->fetch(55);    my $rec = $isis->fetch(55);
246    
247  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
248  for that field.  for that field like this:
249    
250      $rec = {
251        '210' => [ '^aNew York^cNew York University press^dcop. 1988' ],
252        '990' => [ '2140', '88', 'HAY' ],
253      };
254    
255  =cut  =cut
256    
# Line 224  sub fetch { Line 259  sub fetch {
259    
260          my $mfn = shift || croak "fetch needs MFN as argument!";          my $mfn = shift || croak "fetch needs MFN as argument!";
261    
262          print "fetch: $mfn\n" if ($self->{debug});          # is mfn allready in memory?
263            my $old_mfn = $self->{'current_mfn'} || -1;
264            return $self->{record} if ($mfn == $old_mfn);
265    
266            print STDERR "## fetch: $mfn\n" if ($self->{debug});
267    
268          # XXX check this?          # XXX check this?
269          my $mfnpos=($mfn+int(($mfn-1)/127))*4;          my $mfnpos=($mfn+int(($mfn-1)/127))*4;
270    
271          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 ($self->{debug});
272          seek($self->{'fileXRF'},$mfnpos,0);          seek($self->{'fileXRF'},$mfnpos,0);
273    
274            my $buff;
275    
276            # delete old record
277            delete $self->{record};
278    
279          # read XRFMFB abd XRFMFP          # read XRFMFB abd XRFMFP
280          my $pointer=$self->Read32(\*{$self->{'fileXRF'}});          read($self->{'fileXRF'}, $buff, 4);
281            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          print "XRFMFB: $XRFMFB XRFMFP: $XRFMFP\n" if ($self->{debug});          # (XRFMFB - 1) * 512 + XRFMFP
306            # why do i have to do XRFMFP % 1024 ?
307    
308          # XXX fix this to be more readable!!          my $blk_off = (($XRFMFB - 1) * 512) + ($XRFMFP % 512);
         # e.g. (XRFMFB - 1) * 512 + XRFMFP  
309    
310          my $offset = $pointer;          print STDERR "## pointer: $pointer XRFMFB: $XRFMFB XRFMFP: $XRFMFP offset: $blk_off\n" if ($self->{'debug'});
         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;  
   
         print "$offset - $offset2 - $offset3 - $offset4\n" if ($self->{debug});  
311    
312          # Get Record Information          # Get Record Information
313    
314          seek($self->{'fileMST'},$offset4,0);          seek($self->{'fileMST'},$blk_off,0) || croak "can't seek to $blk_off: $!";
315    
316            read($self->{'fileMST'}, $buff, 4) || croak "can't read 4 bytes at offset $blk_off from MST file: $!";
317            my $value=unpack("V",$buff);
318    
319          my $value=$self->Read32(\*{$self->{'fileMST'}});          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  print ("Error: The MFN:".$mfn." is not found in MST(".$value.")");                      if ($value == 0) {
323                  return -1;      # 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    
         my $buff;  
332          read($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 "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};  
339    
340          if (! $self->{'include_deleted'} && $MFRL < 0) {          warn "BASE is not 18+6*NVF" unless ($BASE == 18 + 6 * $NVF);
                 print "## logically deleted record $mfn, skipping...\n" if ($self->{debug});  
                 return;  
         }  
341    
342          # Get Directory Format          # Get Directory Format
343    
# Line 295  print ("Error: The MFN:".$mfn." is not f Line 347  print ("Error: The MFN:".$mfn." is not f
347    
348          read($self->{'fileMST'}, $buff, 6 * $NVF);          read($self->{'fileMST'}, $buff, 6 * $NVF);
349    
350          my $fld_len = 0;          my $rec_len = 0;
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 "TAG: $TAG POS: $POS LEN: $LEN\n" if ($self->{debug});                  print STDERR "## TAG: $TAG POS: $POS LEN: $LEN\n" if ($self->{debug});
357    
358                  # 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.
359                  #                  #
# Line 321  print ("Error: The MFN:".$mfn." is not f Line 369  print ("Error: The MFN:".$mfn." is not f
369                  push @FieldPOS,$POS;                  push @FieldPOS,$POS;
370                  push @FieldLEN,$LEN;                  push @FieldLEN,$LEN;
371    
372                  $fld_len += $LEN;                  $rec_len += $LEN;
373          }          }
374    
375          # Get Variable Fields          # Get Variable Fields
376    
377          read($self->{'fileMST'},$buff,$fld_len);          read($self->{'fileMST'},$buff,$rec_len);
378    
379            print STDERR "## rec_len: $rec_len poc: ",tell($self->{'fileMST'})."\n" if ($self->{debug});
380    
381          for (my $i = 0 ; $i < $NVF ; $i++) {          for (my $i = 0 ; $i < $NVF ; $i++) {
382                  # skip zero-sized fields                  # skip zero-sized fields
# Line 334  print ("Error: The MFN:".$mfn." is not f Line 384  print ("Error: The MFN:".$mfn." is not f
384    
385                  push @{$self->{record}->{$FieldTAG[$i]}}, substr($buff,$FieldPOS[$i],$FieldLEN[$i]);                  push @{$self->{record}->{$FieldTAG[$i]}}, substr($buff,$FieldPOS[$i],$FieldLEN[$i]);
386          }          }
         close(fileMST);  
387    
388          print Dumper($self) if ($self->{debug});          $self->{'current_mfn'} = $mfn;
389    
390            print STDERR Dumper($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 selected MFN  Returns ASCII output of record with specified MFN
416    
417      print $isis->to_ascii(42);
418    
419    This outputs something like this:
420    
421    print $isis->to_ascii(55);    210   ^aNew York^cNew York University press^dcop. 1988
422      990   2140
423      990   88
424      990   HAY
425    
426    If C<read_fdt> is specified when calling C<new> it will display field names
427    from C<.FDT> file instead of numeric tags.
428    
429  =cut  =cut
430    
# Line 354  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    
440          foreach my $f (sort keys %{$rec}) {          foreach my $f (sort keys %{$rec}) {
441                  $out .= "\n$f\t".join("\n$f\t",@{$self->{record}->{$f}});                  my $fn = $self->tag_name($f);
442                    $out .= "\n$fn\t".join("\n$fn\t",@{$self->{record}->{$f}});
443          }          }
444    
445          $out .= "\n";          $out .= "\n";
# Line 367  sub to_ascii { Line 447  sub to_ascii {
447          return $out;          return $out;
448  }  }
449    
450  #  =head2 to_hash
451  # XXX porting from php left-over:  
452  #  Read record with specified MFN and convert it to hash
453  # do I *REALLY* need those methods, or should I use  
454  # $self->{something} directly?    my $hash = $isis->to_hash($mfn);
455  #  
456  # Probably direct usage is better!  It has ability to convert characters (using C<hash_filter>) from ISIS
457  #  database before creating structures enabling character re-mapping or quick
458    fix-up of data.
459    
460    This function returns hash which is like this:
461    
462      $hash = {
463        '210' => [
464                   {
465                     'c' => 'New York University press',
466                     'a' => 'New York',
467                     'd' => 'cop. 1988'
468                   }
469                 ],
470        '990' => [
471                   '2140',
472                   '88',
473                   'HAY'
474                 ],
475      };
476    
477    You can later use that hash to produce any output from ISIS data.
478    
479    If database is created using IsisMarc, it will also have to special fields
480    which will be used for identifiers, C<i1> and C<i2> like this:
481    
482      '200' => [
483                 {
484                   'i1' => '1',
485                   'i2' => ' '
486                   'a' => 'Goa',
487                   'f' => 'Valdo D\'Arienzo',
488                   'e' => 'tipografie e tipografi nel XVI secolo',
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  sub TagName {    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.
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
555    
556    sub to_hash {
557          my $self = shift;          my $self = shift;
558          return $self->{TagName};  
559    
560            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
571            my $rec = { '000' => [ $mfn ] };
572    
573            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 $f_nr (keys %{$row}) {
580                    foreach my $l (@{$row->{$f_nr}}) {
581    
582                            # filter output
583                            if ($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;
599                            my $r_sf;       # repeatable subfields in this record
600    
601                            # has identifiers?
602                            ($val->{'i1'},$val->{'i2'}) = ($1,$2) if ($l =~ s/^([01 #])([01 #])\^/\^/);
603    
604                            # has subfields?
605                            if ($l =~ m/\^/) {
606                                    foreach my $t (split(/\^/,$l)) {
607                                            next if (! $t);
608                                            my ($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 {
635                                    $val = $l;
636                            }
637    
638                            if ($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    
648            return $rec;
649  }  }
650    
651  sub NextMFN {  =head2 tag_name
652    
653    Return name of selected tag
654    
655     print $isis->tag_name('200');
656    
657    =cut
658    
659    sub tag_name {
660          my $self = shift;          my $self = shift;
661          return $self->{NXTMFN};          my $tag = shift || return;
662            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 399  This module has been very lightly tested Line 786  This module has been very lightly tested
786          dpavlin@rot13.org          dpavlin@rot13.org
787          http://www.rot13.org/~dpavlin/          http://www.rot13.org/~dpavlin/
788    
789  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
790  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.
791    
792  =head1 COPYRIGHT  =head1 COPYRIGHT
793    
# Line 413  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<http://www.openisis.org|OpenIsis>, perl(1).  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>
806    
807    perl4lib site L<http://perl4lib.perl.org>
808    
Legend:
Removed from v.10  
changed lines
  Added in v.61
  ViewVC Help
Powered by ViewVC 1.1.26