/[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 9 by dpavlin, Wed Dec 29 16:01:41 2004 UTC trunk/lib/Biblio/Isis.pm revision 70 by dpavlin, Fri May 18 20:26:01 2007 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     = 0.24_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,$field_number) = @_;
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. It will
118    receive two arguments, whole line from current field (in C<< $_[0] >>) and
119    field number (in C<< $_[1] >>).
120    
121  =item debug  =item debug
122    
123  Dump a C<lot> of debugging output.  Dump a B<lot> of debugging output even at level 1. For even more increase level.
124    
125  =item include_deleted  =item join_subfields_with
126    
127  Don't skip logically deleted records.  Define delimiter which will be used to join repeatable subfields. This
128    option is included to support lagacy application written against version
129    older than 0.21 of this module. By default, it disabled. See L</to_hash>.
130    
131  =back  =item ignore_empty_subfields
132    
133  It will also set C<$isis-E<gt>{'maxmfn'}> which is maximum MFN stored in database.  Remove all empty subfields while reading from ISIS file.
134    
135    =back
136    
137  =cut  =cut
138    
# Line 104  sub new { Line 143  sub new {
143    
144          croak "new needs database name (isisdb) as argument!" unless ({@_}->{isisdb});          croak "new needs database name (isisdb) as argument!" unless ({@_}->{isisdb});
145    
146          foreach my $v (qw{isisdb debug include_deleted}) {          foreach my $v (qw{isisdb debug include_deleted hash_filter join_subfields_with ignore_empty_subfields}) {
147                  $self->{$v} = {@_}->{$v};                  $self->{$v} = {@_}->{$v} if defined({@_}->{$v});
148            }
149    
150            my @isis_files = grep(/\.(FDT|MST|XRF|CNT)$/i,glob($self->{isisdb}."*"));
151    
152            foreach my $f (@isis_files) {
153                    my $ext = $1 if ($f =~ m/\.(\w\w\w)$/);
154                    $self->{lc($ext)."_file"} = $f;
155            }
156    
157            my @must_exist = qw(mst xrf);
158            push @must_exist, "fdt" if ($self->{read_fdt});
159    
160            foreach my $ext (@must_exist) {
161                    unless ($self->{$ext."_file"}) {
162                            carp "missing ",uc($ext)," file in ",$self->{isisdb};
163                            return;
164                    }
165            }
166    
167            if ($self->{debug}) {
168                    print STDERR "## using files: ",join(" ",@isis_files),"\n";
169                    eval "use Data::Dump";
170    
171                    if (! $@) {
172                            *Dumper = *Data::Dump::dump;
173                    } else {
174                            use Data::Dumper;
175                    }
176          }          }
177    
178          # 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!
179          if ({@_}->{read_fdt} && -e $self->{isisdb}.".FDT") {          if ($self->{read_fdt} && -e $self->{fdt_file}) {
180    
181                  # read the $db.FDT file for tags                  # read the $db.FDT file for tags
182                  my $fieldzone=0;                  my $fieldzone=0;
183    
184                  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}': $!";
185                    binmode($fileFDT);
186    
187                  while (<fileFDT>) {                  while (<$fileFDT>) {
188                          chomp;                          chomp;
189                          if ($fieldzone) {                          if ($fieldzone) {
190                                  my $name=substr($_,0,30);                                  my $name=substr($_,0,30);
# Line 133  sub new { Line 201  sub new {
201                          }                          }
202                  }                  }
203                                    
204                  close(fileFDT);                  close($fileFDT);
205          }          }
206    
207          # Get the Maximum MFN from $db.MST          # Get the Maximum MFN from $db.MST
208    
209          open(fileMST,$self->{isisdb}.".MST") || croak "can't read '$self->{isisdb}.MST': $!";          open($self->{'fileMST'}, $self->{mst_file}) || croak "can't open '$self->{mst_file}': $!";
210            binmode($self->{'fileMST'});
211    
212          # MST format:   (* = 32 bit signed)          # MST format:   (* = 32 bit signed)
213          # CTLMFN*       always 0          # CTLMFN*       always 0
# Line 146  sub new { Line 215  sub new {
215          # NXTMFB*       last block allocated to master file          # NXTMFB*       last block allocated to master file
216          # NXTMFP        offset to next available position in last block          # NXTMFP        offset to next available position in last block
217          # MFTYPE        always 0 for user db file (1 for system)          # MFTYPE        always 0 for user db file (1 for system)
218          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";  
219    
220          # 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': $!";  
   
         # There is two 26 Bytes fixed lenght records  
221    
222          #  0: IDTYPE    BTree type                              16          read($self->{'fileMST'}, $buff, 4) || croak "can't read NXTMFN from MST: $!";
223          #  2: ORDN      Nodes Order                             16          $self->{'NXTMFN'}=unpack("V",$buff) || croak "NXTNFN is zero";
         #  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);  
                 }  
         }  
224    
225          my $buff;          print STDERR "## self ",Dumper($self),"\n" if ($self->{debug});
         read(fileCNT, $buff, 26);  
         $self->unpack_cnt($buff);  
226    
227          read(fileCNT, $buff, 26);          # open files for later
228          $self->unpack_cnt($buff);          open($self->{'fileXRF'}, $self->{xrf_file}) || croak "can't open '$self->{xrf_file}': $!";
229            binmode($self->{'fileXRF'});
230    
231            $self ? return $self : return undef;
232    }
233    
234          close(fileCNT);  =head2 count
235    
236          print Dumper($self) if ($self->{debug});  Return number of records in database
237    
238          # open files for later    print $isis->count;
         open($self->{'fileXRF'}, $self->{isisdb}.".XRF") || croak "can't open '$self->{isisdb}.XRF': $!";  
239    
240          open($self->{'fileMST'}, $self->{isisdb}.".MST") || croak "can't open '$self->{isisdb}.MST': $!";  =cut
241    
242          $self ? return $self : return undef;  sub count {
243            my $self = shift;
244            return $self->{'NXTMFN'} - 1;
245  }  }
246    
247  =head2 fetch  =head2 fetch
# Line 215  Read record with selected MFN Line 251  Read record with selected MFN
251    my $rec = $isis->fetch(55);    my $rec = $isis->fetch(55);
252    
253  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
254  for that field.  for that field like this:
255    
256      $rec = {
257        '210' => [ '^aNew York^cNew York University press^dcop. 1988' ],
258        '990' => [ '2140', '88', 'HAY' ],
259      };
260    
261  =cut  =cut
262    
# Line 224  sub fetch { Line 265  sub fetch {
265    
266          my $mfn = shift || croak "fetch needs MFN as argument!";          my $mfn = shift || croak "fetch needs MFN as argument!";
267    
268          print "fetch: $mfn\n" if ($self->{debug});          # is mfn allready in memory?
269            my $old_mfn = $self->{'current_mfn'} || -1;
270            return $self->{record} if ($mfn == $old_mfn);
271    
272            print STDERR "## fetch: $mfn\n" if ($self->{debug});
273    
274          # XXX check this?          # XXX check this?
275          my $mfnpos=($mfn+int(($mfn-1)/127))*4;          my $mfnpos=($mfn+int(($mfn-1)/127))*4;
276    
277          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 ($self->{debug});
278          seek($self->{'fileXRF'},$mfnpos,0);          seek($self->{'fileXRF'},$mfnpos,0);
279    
280            my $buff;
281    
282            # delete old record
283            delete $self->{record};
284    
285          # read XRFMFB abd XRFMFP          # read XRFMFB abd XRFMFP
286          my $pointer=$self->Read32(\*{$self->{'fileXRF'}});          read($self->{'fileXRF'}, $buff, 4);
287            my $pointer=unpack("V",$buff);
288            if (! $pointer) {
289                    if ($self->{include_deleted}) {
290                            return;
291                    } else {
292                            warn "pointer for MFN $mfn is null\n";
293                            return;
294                    }
295            }
296    
297            # check for logically deleted record
298            if ($pointer & 0x80000000) {
299                    print STDERR "## record $mfn is logically deleted\n" if ($self->{debug});
300                    $self->{deleted} = $mfn;
301    
302                    return unless $self->{include_deleted};
303    
304                    # abs
305                    $pointer = ($pointer ^ 0xffffffff) + 1;
306            }
307    
308          my $XRFMFB = int($pointer/2048);          my $XRFMFB = int($pointer/2048);
309          my $XRFMFP = $pointer - ($XRFMFB*2048);          my $XRFMFP = $pointer - ($XRFMFB*2048);
310    
311          print "XRFMFB: $XRFMFB XRFMFP: $XRFMFP\n" if ($self->{debug});          # (XRFMFB - 1) * 512 + XRFMFP
312            # why do i have to do XRFMFP % 1024 ?
313    
314          # XXX fix this to be more readable!!          my $blk_off = (($XRFMFB - 1) * 512) + ($XRFMFP % 512);
         # e.g. (XRFMFB - 1) * 512 + XRFMFP  
315    
316          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});  
317    
318          # Get Record Information          # Get Record Information
319    
320          seek($self->{'fileMST'},$offset4,0);          seek($self->{'fileMST'},$blk_off,0) || croak "can't seek to $blk_off: $!";
321    
322            read($self->{'fileMST'}, $buff, 4) || croak "can't read 4 bytes at offset $blk_off from MST file: $!";
323            my $value=unpack("V",$buff);
324    
325          my $value=$self->Read32(\*{$self->{'fileMST'}});          print STDERR "## offset for rowid $value is $blk_off (blk $XRFMFB off $XRFMFP)\n" if ($self->{debug});
326    
327          if ($value!=$mfn) {          if ($value!=$mfn) {
328  print ("Error: The MFN:".$mfn." is not found in MST(".$value.")");                      if ($value == 0) {
329                  return -1;      # XXX deleted record?                          print STDERR "## record $mfn is physically deleted\n" if ($self->{debug});
330          }                          $self->{deleted} = $mfn;
331                            return;
332                    }
333    
334  #       $MFRL=$self->Read16($fileMST);                  carp "Error: MFN ".$mfn." not found in MST file, found $value";    
335  #       $MFBWB=$self->Read32($fileMST);                  return;
336  #       $MFBWP=$self->Read16($fileMST);          }
 #       $BASE=$self->Read16($fileMST);  
 #       $NVF=$self->Read16($fileMST);  
 #       $STATUS=$self->Read16($fileMST);  
337    
         my $buff;  
338          read($self->{'fileMST'}, $buff, 14);          read($self->{'fileMST'}, $buff, 14);
339    
340          my ($MFRL,$MFBWB,$MFBWP,$BASE,$NVF,$STATUS) = unpack("slssss", $buff);          my ($MFRL,$MFBWB,$MFBWP,$BASE,$NVF,$STATUS) = unpack("vVvvvv", $buff);
341    
342          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});
343    
344          # delete old record          warn "MFRL $MFRL is not even number" unless ($MFRL % 2 == 0);
         delete $self->{record};  
345    
346          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;  
         }  
347    
348          # Get Directory Format          # Get Directory Format
349    
# Line 295  print ("Error: The MFN:".$mfn." is not f Line 353  print ("Error: The MFN:".$mfn." is not f
353    
354          read($self->{'fileMST'}, $buff, 6 * $NVF);          read($self->{'fileMST'}, $buff, 6 * $NVF);
355    
356          my $fld_len = 0;          my $rec_len = 0;
357    
358          for (my $i = 0 ; $i < $NVF ; $i++) {          for (my $i = 0 ; $i < $NVF ; $i++) {
359    
360  #               $TAG=$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));  
361    
362                  print "TAG: $TAG POS: $POS LEN: $LEN\n" if ($self->{debug});                  print STDERR "## TAG: $TAG POS: $POS LEN: $LEN\n" if ($self->{debug});
363    
364                  # 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.
365                  #                  #
# Line 321  print ("Error: The MFN:".$mfn." is not f Line 375  print ("Error: The MFN:".$mfn." is not f
375                  push @FieldPOS,$POS;                  push @FieldPOS,$POS;
376                  push @FieldLEN,$LEN;                  push @FieldLEN,$LEN;
377    
378                  $fld_len += $LEN;                  $rec_len += $LEN;
379          }          }
380    
381          # Get Variable Fields          # Get Variable Fields
382    
383          read($self->{'fileMST'},$buff,$fld_len);          read($self->{'fileMST'},$buff,$rec_len);
384    
385            print STDERR "## rec_len: $rec_len poc: ",tell($self->{'fileMST'})."\n" if ($self->{debug});
386    
387          for (my $i = 0 ; $i < $NVF ; $i++) {          for (my $i = 0 ; $i < $NVF ; $i++) {
388                  push @{$self->{record}->{$FieldTAG[$i]}}, substr($buff,$FieldPOS[$i],$FieldLEN[$i]);                  # skip zero-sized fields
389                    next if ($FieldLEN[$i] == 0);
390    
391                    my $v = substr($buff,$FieldPOS[$i],$FieldLEN[$i]);
392    
393                    if ( $self->{ignore_empty_subfields} ) {
394                            $v =~ s/(\^\w)+(\^\w)/$2/g;
395                            $v =~ s/\^\w$//;                        # last on line?
396                            next if ($v eq '');
397                    }
398    
399                    push @{$self->{record}->{$FieldTAG[$i]}}, $v;
400          }          }
         close(fileMST);  
401    
402          print Dumper($self) if ($self->{debug});          $self->{'current_mfn'} = $mfn;
403    
404            print STDERR Dumper($self),"\n" if ($self->{debug});
405    
406          return $self->{'record'};          return $self->{'record'};
407  }  }
408    
409    =head2 mfn
410    
411    Returns current MFN position
412    
413      my $mfn = $isis->mfn;
414    
415    =cut
416    
417    # This function should be simple return $self->{current_mfn},
418    # but if new is called with _hack_mfn it becomes setter.
419    # It's useful in tests when setting $isis->{record} directly
420    
421    sub mfn {
422            my $self = shift;
423            return $self->{current_mfn};
424    };
425    
426    
427  =head2 to_ascii  =head2 to_ascii
428    
429  Dump ascii output of selected MFN  Returns ASCII output of record with specified MFN
430    
431      print $isis->to_ascii(42);
432    
433    print $isis->to_ascii(55);  This outputs something like this:
434    
435      210   ^aNew York^cNew York University press^dcop. 1988
436      990   2140
437      990   88
438      990   HAY
439    
440    If C<read_fdt> is specified when calling C<new> it will display field names
441    from C<.FDT> file instead of numeric tags.
442    
443  =cut  =cut
444    
# Line 351  sub to_ascii { Line 447  sub to_ascii {
447    
448          my $mfn = shift || croak "need MFN";          my $mfn = shift || croak "need MFN";
449    
450          my $rec = $self->fetch($mfn);          my $rec = $self->fetch($mfn) || return;
451    
452          my $out = "0\t$mfn";          my $out = "0\t$mfn";
453    
454          foreach my $f (sort keys %{$rec}) {          foreach my $f (sort keys %{$rec}) {
455                  $out .= "\n$f\t".join("\n$f\t",@{$self->{record}->{$f}});                  my $fn = $self->tag_name($f);
456                    $out .= "\n$fn\t".join("\n$fn\t",@{$self->{record}->{$f}});
457          }          }
458    
459          $out .= "\n";          $out .= "\n";
# Line 364  sub to_ascii { Line 461  sub to_ascii {
461          return $out;          return $out;
462  }  }
463    
464  #  =head2 to_hash
465  # XXX porting from php left-over:  
466  #  Read record with specified MFN and convert it to hash
467  # do I *REALLY* need those methods, or should I use  
468  # $self->{something} directly?    my $hash = $isis->to_hash($mfn);
469  #  
470  # Probably direct usage is better!  It has ability to convert characters (using C<hash_filter>) from ISIS
471  #  database before creating structures enabling character re-mapping or quick
472    fix-up of data.
473    
474    This function returns hash which is like this:
475    
476      $hash = {
477        '210' => [
478                   {
479                     'c' => 'New York University press',
480                     'a' => 'New York',
481                     'd' => 'cop. 1988'
482                   }
483                 ],
484        '990' => [
485                   '2140',
486                   '88',
487                   'HAY'
488                 ],
489      };
490    
491    You can later use that hash to produce any output from ISIS data.
492    
493    If database is created using IsisMarc, it will also have to special fields
494    which will be used for identifiers, C<i1> and C<i2> like this:
495    
496      '200' => [
497                 {
498                   'i1' => '1',
499                   'i2' => ' '
500                   'a' => 'Goa',
501                   'f' => 'Valdo D\'Arienzo',
502                   'e' => 'tipografie e tipografi nel XVI secolo',
503                 }
504               ],
505    
506    In case there are repeatable subfields in record, this will create
507    following structure:
508    
509      '900' => [ {
510            'a' => [ 'foo', 'bar', 'baz' ],
511      }]
512    
513    Or in more complex example of
514    
515      902   ^aa1^aa2^aa3^bb1^aa4^bb2^cc1^aa5
516    
517    it will create
518    
519      902   => [
520            { a => ["a1", "a2", "a3", "a4", "a5"], b => ["b1", "b2"], c => "c1" },
521      ],
522    
523    This behaviour can be changed using C<join_subfields_with> option to L</new>,
524    in which case C<to_hash> will always create single value for each subfield.
525    This will change result to:
526    
527    
528    
529  sub TagName {  This method will also create additional field C<000> with MFN.
530    
531    There is also more elaborative way to call C<to_hash> like this:
532    
533      my $hash = $isis->to_hash({
534            mfn => 42,
535            include_subfields => 1,
536      });
537    
538    Each option controll creation of hash:
539    
540    =over 4
541    
542    =item mfn
543    
544    Specify MFN number of record
545    
546    =item include_subfields
547    
548    This option will create additional key in hash called C<subfields> which will
549    have original record subfield order and index to that subfield like this:
550    
551      902   => [ {
552            a => ["a1", "a2", "a3", "a4", "a5"],
553            b => ["b1", "b2"],
554            c => "c1",
555            subfields => ["a", 0, "a", 1, "a", 2, "b", 0, "a", 3, "b", 1, "c", 0, "a", 4],
556      } ],
557    
558    =item join_subfields_with
559    
560    Define delimiter which will be used to join repeatable subfields. You can
561    specify option here instead in L</new> if you want to have per-record control.
562    
563    =item hash_filter
564    
565    You can override C<hash_filter> defined in L</new> using this option.
566    
567    =back
568    
569    =cut
570    
571    sub to_hash {
572            my $self = shift;
573    
574    
575            my $mfn = shift || confess "need mfn!";
576            my $arg;
577    
578            my $hash_filter = $self->{hash_filter};
579    
580            if (ref($mfn) eq 'HASH') {
581                    $arg = $mfn;
582                    $mfn = $arg->{mfn} || confess "need mfn in arguments";
583                    $hash_filter = $arg->{hash_filter} if ($arg->{hash_filter});
584            }
585    
586            # init record to include MFN as field 000
587            my $rec = { '000' => [ $mfn ] };
588    
589            my $row = $self->fetch($mfn) || return;
590    
591            my $j_rs = $arg->{join_subfields_with} || $self->{join_subfields_with};
592            $j_rs = $self->{join_subfields_with} unless(defined($j_rs));
593            my $i_sf = $arg->{include_subfields};
594    
595            foreach my $f_nr (keys %{$row}) {
596                    foreach my $l (@{$row->{$f_nr}}) {
597    
598                            # filter output
599                            $l = $hash_filter->($l, $f_nr) if ($hash_filter);
600                            next unless defined($l);
601    
602                            my $val;
603                            my $r_sf;       # repeatable subfields in this record
604    
605                            # has identifiers?
606                            ($val->{'i1'},$val->{'i2'}) = ($1,$2) if ($l =~ s/^([01 #])([01 #])\^/\^/);
607    
608                            # has subfields?
609                            if ($l =~ m/\^/) {
610                                    foreach my $t (split(/\^/,$l)) {
611                                            next if (! $t);
612                                            my ($sf,$v) = (substr($t,0,1), substr($t,1));
613                                            # XXX this might be option, but why?
614                                            next unless (defined($v) && $v ne '');
615    #                                       warn "### $f_nr^$sf:$v",$/ if ($self->{debug} > 1);
616    
617                                            if (ref( $val->{$sf} ) eq 'ARRAY') {
618    
619                                                    push @{ $val->{$sf} }, $v;
620    
621                                                    # record repeatable subfield it it's offset
622                                                    push @{ $val->{subfields} }, ( $sf, $#{ $val->{$sf} } ) if (! $j_rs && $i_sf);
623                                                    $r_sf->{$sf}++;
624    
625                                            } elsif (defined( $val->{$sf} )) {
626    
627                                                    # convert scalar field to array
628                                                    $val->{$sf} = [ $val->{$sf}, $v ];
629    
630                                                    push @{ $val->{subfields} }, ( $sf, 1 ) if (! $j_rs && $i_sf);
631                                                    $r_sf->{$sf}++;
632    
633                                            } else {
634                                                    $val->{$sf} = $v;
635                                                    push @{ $val->{subfields} }, ( $sf, 0 ) if ($i_sf);
636                                            }
637                                    }
638                            } else {
639                                    $val = $l;
640                            }
641    
642                            if ($j_rs) {
643                                    map {
644                                            $val->{$_} = join($j_rs, @{ $val->{$_} });
645                                    } keys %$r_sf
646                            }
647    
648                            push @{$rec->{$f_nr}}, $val;
649                    }
650            }
651    
652            return $rec;
653    }
654    
655    =head2 tag_name
656    
657    Return name of selected tag
658    
659     print $isis->tag_name('200');
660    
661    =cut
662    
663    sub tag_name {
664            my $self = shift;
665            my $tag = shift || return;
666            return $self->{'TagName'}->{$tag} || $tag;
667    }
668    
669    
670    =head2 read_cnt
671    
672    Read content of C<.CNT> file and return hash containing it.
673    
674      print Dumper($isis->read_cnt);
675    
676    This function is not used by module (C<.CNT> files are not required for this
677    module to work), but it can be useful to examine your index (while debugging
678    for example).
679    
680    =cut
681    
682    sub read_cnt  {
683          my $self = shift;          my $self = shift;
684          return $self->{TagName};  
685            croak "missing CNT file in ",$self->{isisdb} unless ($self->{cnt_file});
686    
687            # Get the index information from $db.CNT
688      
689            open(my $fileCNT, $self->{cnt_file}) || croak "can't read '$self->{cnt_file}': $!";
690            binmode($fileCNT);
691    
692            my $buff;
693    
694            read($fileCNT, $buff, 26) || croak "can't read first table from CNT: $!";
695            $self->unpack_cnt($buff);
696    
697            read($fileCNT, $buff, 26) || croak "can't read second table from CNT: $!";
698            $self->unpack_cnt($buff);
699    
700            close($fileCNT);
701    
702            return $self->{cnt};
703  }  }
704    
705  sub NextMFN {  =head2 unpack_cnt
706    
707    Unpack one of two 26 bytes fixed length record in C<.CNT> file.
708    
709    Here is definition of record:
710    
711     off key        description                             size
712      0: IDTYPE     BTree type                              s
713      2: ORDN       Nodes Order                             s
714      4: ORDF       Leafs Order                             s
715      6: N          Number of Memory buffers for nodes      s
716      8: K          Number of buffers for first level index s
717     10: LIV        Current number of Index Levels          s
718     12: POSRX      Pointer to Root Record in N0x           l
719     16: NMAXPOS    Next Available position in N0x          l
720     20: FMAXPOS    Next available position in L0x          l
721     24: ABNORMAL   Formal BTree normality indicator        s
722     length: 26 bytes
723    
724    This will fill C<$self> object under C<cnt> with hash. It's used by C<read_cnt>.
725    
726    =cut
727    
728    sub unpack_cnt {
729          my $self = shift;          my $self = shift;
730          return $self->{NXTMFN};  
731            my @flds = qw(ORDN ORDF N K LIV POSRX NMAXPOS FMAXPOS ABNORMAL);
732    
733            my $buff = shift || return;
734            my @arr = unpack("vvvvvvVVVv", $buff);
735    
736            print STDERR "unpack_cnt: ",join(" ",@arr),"\n" if ($self->{'debug'});
737    
738            my $IDTYPE = shift @arr;
739            foreach (@flds) {
740                    $self->{cnt}->{$IDTYPE}->{$_} = abs(shift @arr);
741            }
742  }  }
743    
744  1;  1;
745    
746  =head1 BUGS  =head1 BUGS
747    
748  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
749    some variations in input databases which has been tested with this module.
750    When I was in doubt, I assumed that OpenIsis's implementation was right
751    (except for obvious bugs).
752    
753    However, every effort has been made to test this module with as much
754    databases (and programs that create them) as possible.
755    
756    I would be very greatful for success or failure reports about usage of this
757    module with databases from programs other than WinIsis and IsisMarc. I had
758    tested this against ouput of one C<isis.dll>-based application, but I don't
759    know any details about it's version.
760    
761    =head1 VERSIONS
762    
763    As this is young module, new features are added in subsequent version. It's
764    a good idea to specify version when using this module like this:
765    
766      use Biblio::Isis 0.23
767    
768    Below is list of changes in specific version of module (so you can target
769    older versions if you really have to):
770    
771    =over 8
772    
773    =item 0.24
774    
775    Added C<ignore_empty_subfields>
776    
777    =item 0.23
778    
779    Added C<hash_filter> to L</to_hash>
780    
781    Fixed bug with documented C<join_subfields_with> in L</new> which wasn't
782    implemented
783    
784    =item 0.22
785    
786    Added field number when calling C<hash_filter>
787    
788    =item 0.21
789    
790    Added C<join_subfields_with> to L</new> and L</to_hash>.
791    
792    Added C<include_subfields> to L</to_hash>.
793    
794    =item 0.20
795    
796    Added C<< $isis->mfn >>, support for repeatable subfields and
797    C<< $isis->to_hash({ mfn => 42, ... }) >> calling convention
798    
799    =back
800    
801  =head1 AUTHOR  =head1 AUTHOR
802    
# Line 396  This module has been very lightly tested Line 805  This module has been very lightly tested
805          dpavlin@rot13.org          dpavlin@rot13.org
806          http://www.rot13.org/~dpavlin/          http://www.rot13.org/~dpavlin/
807    
808  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
809  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.
810    
811  =head1 COPYRIGHT  =head1 COPYRIGHT
812    
# Line 410  LICENSE file included with this module. Line 819  LICENSE file included with this module.
819    
820  =head1 SEE ALSO  =head1 SEE ALSO
821    
822  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
823    
824    OpenIsis web site L<http://www.openisis.org>
825    
826    perl4lib site L<http://perl4lib.perl.org>
827    
Legend:
Removed from v.9  
changed lines
  Added in v.70
  ViewVC Help
Powered by ViewVC 1.1.26