1 # DB_File.pm -- Perl 5 interface to Berkeley DB
3 # written by Paul Marquess (pmarquess@bfsec.bt.co.uk)
4 # last modified 20th Nov 1997
7 # Copyright (c) 1995, 1996, 1997 Paul Marquess. All rights reserved.
8 # This program is free software; you can redistribute it and/or
9 # modify it under the same terms as Perl itself.
12 package DB_File::HASHINFO ;
19 @DB_File::HASHINFO::ISA = qw(Tie::Hash);
34 bless { VALID => { map {$_, 1}
35 qw( bsize ffactor nelem cachesize hash lorder)
47 return $self->{GOT}{$key} if exists $self->{VALID}{$key} ;
50 croak "${pkg}::FETCH - Unknown element '$key'" ;
60 if ( exists $self->{VALID}{$key} )
62 $self->{GOT}{$key} = $value ;
67 croak "${pkg}::STORE - Unknown element '$key'" ;
75 if ( exists $self->{VALID}{$key} )
77 delete $self->{GOT}{$key} ;
82 croak "DB_File::HASHINFO::DELETE - Unknown element '$key'" ;
90 exists $self->{VALID}{$key} ;
98 croak ref($self) . " does not define the method ${method}" ;
101 sub FIRSTKEY { my $self = shift ; $self->NotHere("FIRSTKEY") }
102 sub NEXTKEY { my $self = shift ; $self->NotHere("NEXTKEY") }
103 sub CLEAR { my $self = shift ; $self->NotHere("CLEAR") }
105 package DB_File::RECNOINFO ;
109 @DB_File::RECNOINFO::ISA = qw(DB_File::HASHINFO) ;
115 bless { VALID => { map {$_, 1}
116 qw( bval cachesize psize flags lorder reclen bfname )
122 package DB_File::BTREEINFO ;
126 @DB_File::BTREEINFO::ISA = qw(DB_File::HASHINFO) ;
132 bless { VALID => { map {$_, 1}
133 qw( flags cachesize maxkeypage minkeypage psize
134 compare prefix lorder )
144 use vars qw($VERSION @ISA @EXPORT $AUTOLOAD $DB_BTREE $DB_HASH $DB_RECNO $db_version) ;
150 #typedef enum { DB_BTREE, DB_HASH, DB_RECNO } DBTYPE;
151 $DB_BTREE = new DB_File::BTREEINFO ;
152 $DB_HASH = new DB_File::HASHINFO ;
153 $DB_RECNO = new DB_File::RECNOINFO ;
159 @ISA = qw(Tie::Hash Exporter DynaLoader);
161 $DB_BTREE $DB_HASH $DB_RECNO
194 *FETCHSIZE = \&length;
198 ($constname = $AUTOLOAD) =~ s/.*:://;
199 my $val = constant($constname, @_ ? $_[0] : 0);
201 if ($! =~ /Invalid/) {
202 $AutoLoader::AUTOLOAD = $AUTOLOAD;
203 goto &AutoLoader::AUTOLOAD;
206 my($pack,$file,$line) = caller;
207 croak "Your vendor has not defined DB macro $constname, used at $file line $line.
211 eval "sub $AUTOLOAD { $val }";
217 # Make all Fcntl O_XXX constants available for importing
219 my @O = grep /^O_/, @Fcntl::EXPORT;
220 Fcntl->import(@O); # first we import what we want to export
224 ## import borrowed from IO::File
225 ## exports Fcntl constants if available.
228 # my $callpkg = caller;
229 # Exporter::export $pkg, $callpkg, @_;
232 # Exporter::export 'Fcntl', $callpkg, '/^O_/';
236 bootstrap DB_File $VERSION;
238 # Preloaded methods go here. Autoload methods go after __END__, and are
239 # processed by the autosplit program.
241 sub tie_hash_or_array
244 my $tieHASH = ( (caller(1))[3] =~ /TIEHASH/ ) ;
246 $arg[4] = tied %{ $arg[4] }
247 if @arg >= 5 && ref $arg[4] && $arg[4] =~ /=HASH/ && tied %{ $arg[4] } ;
249 # make recno in Berkeley DB version 2 work like recno in version 1.
250 if ($db_version > 1 and defined $arg[4] and $arg[4] =~ /RECNO/ and
251 $arg[1] and ! -e $arg[1]) {
252 open(FH, ">$arg[1]") or return undef ;
254 chmod $arg[3] ? $arg[3] : 0666 , $arg[1] ;
257 DoTie_($tieHASH, @arg) ;
262 tie_hash_or_array(@_) ;
267 tie_hash_or_array(@_) ;
274 my $status = $self->seq($key, $value, R_FIRST());
277 while ($status == 0) {
279 $status = $self->seq($key, $value, R_NEXT());
281 foreach $key (reverse @keys) {
282 my $s = $self->del($key);
288 croak "Usage: \$db->get_dup(key [,flag])\n"
289 unless @_ == 2 or @_ == 3 ;
296 my $wantarray = wantarray ;
302 # iterate through the database until either EOF ($status == 0)
303 # or a different key is encountered ($key ne $origkey).
304 for ($status = $db->seq($key, $value, R_CURSOR()) ;
305 $status == 0 and $key eq $origkey ;
306 $status = $db->seq($key, $value, R_NEXT()) ) {
308 # save the value or count number of matches
311 { ++ $values{$value} }
313 { push (@values, $value) }
320 return ($wantarray ? ($flag ? %values : @values) : $counter) ;
329 DB_File - Perl5 access to Berkeley DB version 1.x
335 [$X =] tie %hash, 'DB_File', [$filename, $flags, $mode, $DB_HASH] ;
336 [$X =] tie %hash, 'DB_File', $filename, $flags, $mode, $DB_BTREE ;
337 [$X =] tie @array, 'DB_File', $filename, $flags, $mode, $DB_RECNO ;
339 $status = $X->del($key [, $flags]) ;
340 $status = $X->put($key, $value [, $flags]) ;
341 $status = $X->get($key, $value [, $flags]) ;
342 $status = $X->seq($key, $value, $flags) ;
343 $status = $X->sync([$flags]) ;
347 $count = $X->get_dup($key) ;
348 @list = $X->get_dup($key) ;
349 %list = $X->get_dup($key, 1) ;
363 B<DB_File> is a module which allows Perl programs to make use of the
364 facilities provided by Berkeley DB version 1.x (if you have a newer
365 version of DB, see L<Using DB_File with Berkeley DB version 2>). It is
366 assumed that you have a copy of the Berkeley DB manual pages at hand
367 when reading this documentation. The interface defined here mirrors the
368 Berkeley DB interface closely.
370 Berkeley DB is a C library which provides a consistent interface to a
371 number of database formats. B<DB_File> provides an interface to all
372 three of the database types currently supported by Berkeley DB.
380 This database type allows arbitrary key/value pairs to be stored in data
381 files. This is equivalent to the functionality provided by other
382 hashing packages like DBM, NDBM, ODBM, GDBM, and SDBM. Remember though,
383 the files created using DB_HASH are not compatible with any of the
384 other packages mentioned.
386 A default hashing algorithm, which will be adequate for most
387 applications, is built into Berkeley DB. If you do need to use your own
388 hashing algorithm it is possible to write your own in Perl and have
389 B<DB_File> use it instead.
393 The btree format allows arbitrary key/value pairs to be stored in a
394 sorted, balanced binary tree.
396 As with the DB_HASH format, it is possible to provide a user defined
397 Perl routine to perform the comparison of keys. By default, though, the
398 keys are stored in lexical order.
402 DB_RECNO allows both fixed-length and variable-length flat text files
403 to be manipulated using the same key/value pair interface as in DB_HASH
404 and DB_BTREE. In this case the key will consist of a record (line)
409 =head2 Using DB_File with Berkeley DB version 2
411 Although B<DB_File> is intended to be used with Berkeley DB version 1,
412 it can also be used with version 2. In this case the interface is
413 limited to the functionality provided by Berkeley DB 1.x. Anywhere the
414 version 2 interface differs, B<DB_File> arranges for it to work like
415 version 1. This feature allows B<DB_File> scripts that were built with
416 version 1 to be migrated to version 2 without any changes.
418 If you want to make use of the new features available in Berkeley DB
419 2.x, use the Perl module B<BerkeleyDB> instead.
421 At the time of writing this document the B<BerkeleyDB> module is still
422 alpha quality (the version number is < 1.0), and so unsuitable for use
423 in any serious development work. Once its version number is >= 1.0, it
424 is considered stable enough for real work.
426 B<Note:> The database file format has changed in Berkeley DB version 2.
427 If you cannot recreate your databases, you must dump any existing
428 databases with the C<db_dump185> utility that comes with Berkeley DB.
429 Once you have upgraded DB_File to use Berkeley DB version 2, your
430 databases can be recreated using C<db_load>. Refer to the Berkeley DB
431 documentation for further details.
433 Please read L<COPYRIGHT> before using version 2.x of Berkeley DB with
436 =head2 Interface to Berkeley DB
438 B<DB_File> allows access to Berkeley DB files using the tie() mechanism
439 in Perl 5 (for full details, see L<perlfunc/tie()>). This facility
440 allows B<DB_File> to access Berkeley DB files using either an
441 associative array (for DB_HASH & DB_BTREE file types) or an ordinary
442 array (for the DB_RECNO file type).
444 In addition to the tie() interface, it is also possible to access most
445 of the functions provided in the Berkeley DB API directly.
446 See L<THE API INTERFACE>.
448 =head2 Opening a Berkeley DB Database File
450 Berkeley DB uses the function dbopen() to open or create a database.
451 Here is the C prototype for dbopen():
454 dbopen (const char * file, int flags, int mode,
455 DBTYPE type, const void * openinfo)
457 The parameter C<type> is an enumeration which specifies which of the 3
458 interface methods (DB_HASH, DB_BTREE or DB_RECNO) is to be used.
459 Depending on which of these is actually chosen, the final parameter,
460 I<openinfo> points to a data structure which allows tailoring of the
461 specific interface method.
463 This interface is handled slightly differently in B<DB_File>. Here is
464 an equivalent call using B<DB_File>:
466 tie %array, 'DB_File', $filename, $flags, $mode, $DB_HASH ;
468 The C<filename>, C<flags> and C<mode> parameters are the direct
469 equivalent of their dbopen() counterparts. The final parameter $DB_HASH
470 performs the function of both the C<type> and C<openinfo> parameters in
473 In the example above $DB_HASH is actually a pre-defined reference to a
474 hash object. B<DB_File> has three of these pre-defined references.
475 Apart from $DB_HASH, there is also $DB_BTREE and $DB_RECNO.
477 The keys allowed in each of these pre-defined references is limited to
478 the names used in the equivalent C structure. So, for example, the
479 $DB_HASH reference will only allow keys called C<bsize>, C<cachesize>,
480 C<ffactor>, C<hash>, C<lorder> and C<nelem>.
482 To change one of these elements, just assign to it like this:
484 $DB_HASH->{'cachesize'} = 10000 ;
486 The three predefined variables $DB_HASH, $DB_BTREE and $DB_RECNO are
487 usually adequate for most applications. If you do need to create extra
488 instances of these objects, constructors are available for each file
491 Here are examples of the constructors and the valid options available
492 for DB_HASH, DB_BTREE and DB_RECNO respectively.
494 $a = new DB_File::HASHINFO ;
502 $b = new DB_File::BTREEINFO ;
512 $c = new DB_File::RECNOINFO ;
521 The values stored in the hashes above are mostly the direct equivalent
522 of their C counterpart. Like their C counterparts, all are set to a
523 default values - that means you don't have to set I<all> of the
524 values when you only want to change one. Here is an example:
526 $a = new DB_File::HASHINFO ;
527 $a->{'cachesize'} = 12345 ;
528 tie %y, 'DB_File', "filename", $flags, 0777, $a ;
530 A few of the options need extra discussion here. When used, the C
531 equivalent of the keys C<hash>, C<compare> and C<prefix> store pointers
532 to C functions. In B<DB_File> these keys are used to store references
533 to Perl subs. Below are templates for each of the subs:
539 # return the hash value for $data
545 my ($key, $key2) = @_ ;
547 # return 0 if $key1 eq $key2
548 # -1 if $key1 lt $key2
549 # 1 if $key1 gt $key2
550 return (-1 , 0 or 1) ;
555 my ($key, $key2) = @_ ;
557 # return number of bytes of $key2 which are
558 # necessary to determine that it is greater than $key1
562 See L<Changing the BTREE sort order> for an example of using the
565 If you are using the DB_RECNO interface and you intend making use of
566 C<bval>, you should check out L<The 'bval' Option>.
568 =head2 Default Parameters
570 It is possible to omit some or all of the final 4 parameters in the
571 call to C<tie> and let them take default values. As DB_HASH is the most
572 common file format used, the call:
574 tie %A, "DB_File", "filename" ;
578 tie %A, "DB_File", "filename", O_CREAT|O_RDWR, 0666, $DB_HASH ;
580 It is also possible to omit the filename parameter as well, so the
587 tie %A, "DB_File", undef, O_CREAT|O_RDWR, 0666, $DB_HASH ;
589 See L<In Memory Databases> for a discussion on the use of C<undef>
590 in place of a filename.
592 =head2 In Memory Databases
594 Berkeley DB allows the creation of in-memory databases by using NULL
595 (that is, a C<(char *)0> in C) in place of the filename. B<DB_File>
596 uses C<undef> instead of NULL to provide this functionality.
600 The DB_HASH file format is probably the most commonly used of the three
601 file formats that B<DB_File> supports. It is also very straightforward
604 =head2 A Simple Example
606 This example shows how to create a database, add key/value pairs to the
607 database, delete keys/value pairs and finally how to enumerate the
608 contents of the database.
612 use vars qw( %h $k $v ) ;
614 tie %h, "DB_File", "fruit", O_RDWR|O_CREAT, 0640, $DB_HASH
615 or die "Cannot open file 'fruit': $!\n";
617 # Add a few key/value pairs to the file
618 $h{"apple"} = "red" ;
619 $h{"orange"} = "orange" ;
620 $h{"banana"} = "yellow" ;
621 $h{"tomato"} = "red" ;
623 # Check for existence of a key
624 print "Banana Exists\n\n" if $h{"banana"} ;
626 # Delete a key/value pair.
629 # print the contents of the file
630 while (($k, $v) = each %h)
631 { print "$k -> $v\n" }
643 Note that the like ordinary associative arrays, the order of the keys
644 retrieved is in an apparently random order.
648 The DB_BTREE format is useful when you want to store data in a given
649 order. By default the keys will be stored in lexical order, but as you
650 will see from the example shown in the next section, it is very easy to
651 define your own sorting function.
653 =head2 Changing the BTREE sort order
655 This script shows how to override the default sorting algorithm that
656 BTREE uses. Instead of using the normal lexical ordering, a case
657 insensitive compare function will be used.
666 my ($key1, $key2) = @_ ;
667 "\L$key1" cmp "\L$key2" ;
670 # specify the Perl sub that will do the comparison
671 $DB_BTREE->{'compare'} = \&Compare ;
673 tie %h, "DB_File", "tree", O_RDWR|O_CREAT, 0640, $DB_BTREE
674 or die "Cannot open file 'tree': $!\n" ;
676 # Add a key/value pair to the file
677 $h{'Wall'} = 'Larry' ;
678 $h{'Smith'} = 'John' ;
679 $h{'mouse'} = 'mickey' ;
680 $h{'duck'} = 'donald' ;
685 # Cycle through the keys printing them in order.
686 # Note it is not necessary to sort the keys as
687 # the btree will have kept them in order automatically.
693 Here is the output from the code above.
699 There are a few point to bear in mind if you want to change the
700 ordering in a BTREE database:
706 The new compare function must be specified when you create the database.
710 You cannot change the ordering once the database has been created. Thus
711 you must use the same compare function every time you access the
716 =head2 Handling Duplicate Keys
718 The BTREE file type optionally allows a single key to be associated
719 with an arbitrary number of values. This option is enabled by setting
720 the flags element of C<$DB_BTREE> to R_DUP when creating the database.
722 There are some difficulties in using the tied hash interface if you
723 want to manipulate a BTREE database with duplicate keys. Consider this
729 use vars qw($filename %h ) ;
734 # Enable duplicate records
735 $DB_BTREE->{'flags'} = R_DUP ;
737 tie %h, "DB_File", $filename, O_RDWR|O_CREAT, 0640, $DB_BTREE
738 or die "Cannot open $filename: $!\n";
740 # Add some key/value pairs to the file
741 $h{'Wall'} = 'Larry' ;
742 $h{'Wall'} = 'Brick' ; # Note the duplicate key
743 $h{'Wall'} = 'Brick' ; # Note the duplicate key and value
744 $h{'Smith'} = 'John' ;
745 $h{'mouse'} = 'mickey' ;
747 # iterate through the associative array
748 # and print each key/value pair.
750 { print "$_ -> $h{$_}\n" }
762 As you can see 3 records have been successfully created with key C<Wall>
763 - the only thing is, when they are retrieved from the database they
764 I<seem> to have the same value, namely C<Larry>. The problem is caused
765 by the way that the associative array interface works. Basically, when
766 the associative array interface is used to fetch the value associated
767 with a given key, it will only ever retrieve the first value.
769 Although it may not be immediately obvious from the code above, the
770 associative array interface can be used to write values with duplicate
771 keys, but it cannot be used to read them back from the database.
773 The way to get around this problem is to use the Berkeley DB API method
774 called C<seq>. This method allows sequential access to key/value
775 pairs. See L<THE API INTERFACE> for details of both the C<seq> method
776 and the API in general.
778 Here is the script above rewritten using the C<seq> API method.
783 use vars qw($filename $x %h $status $key $value) ;
788 # Enable duplicate records
789 $DB_BTREE->{'flags'} = R_DUP ;
791 $x = tie %h, "DB_File", $filename, O_RDWR|O_CREAT, 0640, $DB_BTREE
792 or die "Cannot open $filename: $!\n";
794 # Add some key/value pairs to the file
795 $h{'Wall'} = 'Larry' ;
796 $h{'Wall'} = 'Brick' ; # Note the duplicate key
797 $h{'Wall'} = 'Brick' ; # Note the duplicate key and value
798 $h{'Smith'} = 'John' ;
799 $h{'mouse'} = 'mickey' ;
801 # iterate through the btree using seq
802 # and print each key/value pair.
804 for ($status = $x->seq($key, $value, R_FIRST) ;
806 $status = $x->seq($key, $value, R_NEXT) )
807 { print "$key -> $value\n" }
820 This time we have got all the key/value pairs, including the multiple
821 values associated with the key C<Wall>.
823 =head2 The get_dup() Method
825 B<DB_File> comes with a utility method, called C<get_dup>, to assist in
826 reading duplicate values from BTREE databases. The method can take the
829 $count = $x->get_dup($key) ;
830 @list = $x->get_dup($key) ;
831 %list = $x->get_dup($key, 1) ;
833 In a scalar context the method returns the number of values associated
834 with the key, C<$key>.
836 In list context, it returns all the values which match C<$key>. Note
837 that the values will be returned in an apparently random order.
839 In list context, if the second parameter is present and evaluates
840 TRUE, the method returns an associative array. The keys of the
841 associative array correspond to the values that matched in the BTREE
842 and the values of the array are a count of the number of times that
843 particular value occurred in the BTREE.
845 So assuming the database created above, we can use C<get_dup> like
848 my $cnt = $x->get_dup("Wall") ;
849 print "Wall occurred $cnt times\n" ;
851 my %hash = $x->get_dup("Wall", 1) ;
852 print "Larry is there\n" if $hash{'Larry'} ;
853 print "There are $hash{'Brick'} Brick Walls\n" ;
855 my @list = $x->get_dup("Wall") ;
856 print "Wall => [@list]\n" ;
858 @list = $x->get_dup("Smith") ;
859 print "Smith => [@list]\n" ;
861 @list = $x->get_dup("Dog") ;
862 print "Dog => [@list]\n" ;
867 Wall occurred 3 times
869 There are 2 Brick Walls
870 Wall => [Brick Brick Larry]
874 =head2 Matching Partial Keys
876 The BTREE interface has a feature which allows partial keys to be
877 matched. This functionality is I<only> available when the C<seq> method
878 is used along with the R_CURSOR flag.
880 $x->seq($key, $value, R_CURSOR) ;
882 Here is the relevant quote from the dbopen man page where it defines
883 the use of the R_CURSOR flag with seq:
885 Note, for the DB_BTREE access method, the returned key is not
886 necessarily an exact match for the specified key. The returned key
887 is the smallest key greater than or equal to the specified key,
888 permitting partial key matches and range searches.
890 In the example script below, the C<match> sub uses this feature to find
891 and print the first matching key/value pair given a partial key.
897 use vars qw($filename $x %h $st $key $value) ;
903 my $orig_key = $key ;
904 $x->seq($key, $value, R_CURSOR) ;
905 print "$orig_key\t-> $key\t-> $value\n" ;
911 $x = tie %h, "DB_File", $filename, O_RDWR|O_CREAT, 0640, $DB_BTREE
912 or die "Cannot open $filename: $!\n";
914 # Add some key/value pairs to the file
915 $h{'mouse'} = 'mickey' ;
916 $h{'Wall'} = 'Larry' ;
917 $h{'Walls'} = 'Brick' ;
918 $h{'Smith'} = 'John' ;
923 for ($st = $x->seq($key, $value, R_FIRST) ;
925 $st = $x->seq($key, $value, R_NEXT) )
927 { print "$key -> $value\n" }
929 print "\nPARTIAL MATCH\n" ;
953 DB_RECNO provides an interface to flat text files. Both variable and
954 fixed length records are supported.
956 In order to make RECNO more compatible with Perl the array offset for
957 all RECNO arrays begins at 0 rather than 1 as in Berkeley DB.
959 As with normal Perl arrays, a RECNO array can be accessed using
960 negative indexes. The index -1 refers to the last element of the array,
961 -2 the second last, and so on. Attempting to access an element before
962 the start of the array will raise a fatal run-time error.
964 =head2 The 'bval' Option
966 The operation of the bval option warrants some discussion. Here is the
967 definition of bval from the Berkeley DB 1.85 recno manual page:
969 The delimiting byte to be used to mark the end of a
970 record for variable-length records, and the pad charac-
971 ter for fixed-length records. If no value is speci-
972 fied, newlines (``\n'') are used to mark the end of
973 variable-length records and fixed-length records are
976 The second sentence is wrong. In actual fact bval will only default to
977 C<"\n"> when the openinfo parameter in dbopen is NULL. If a non-NULL
978 openinfo parameter is used at all, the value that happens to be in bval
979 will be used. That means you always have to specify bval when making
980 use of any of the options in the openinfo parameter. This documentation
981 error will be fixed in the next release of Berkeley DB.
983 That clarifies the situation with regards Berkeley DB itself. What
984 about B<DB_File>? Well, the behavior defined in the quote above is
985 quite useful, so B<DB_File> conforms it.
987 That means that you can specify other options (e.g. cachesize) and
988 still have bval default to C<"\n"> for variable length records, and
989 space for fixed length records.
991 =head2 A Simple Example
993 Here is a simple example that uses RECNO.
999 tie @h, "DB_File", "text", O_RDWR|O_CREAT, 0640, $DB_RECNO
1000 or die "Cannot open file 'text': $!\n" ;
1002 # Add a few key/value pairs to the file
1007 # Check for existence of a key
1008 print "Element 1 Exists with value $h[1]\n" if $h[1] ;
1010 # use a negative index
1011 print "The last element is $h[-1]\n" ;
1012 print "The 2nd last element is $h[-2]\n" ;
1016 Here is the output from the script:
1019 Element 1 Exists with value blue
1020 The last element is yellow
1021 The 2nd last element is blue
1023 =head2 Extra Methods
1025 As you can see from the example above, the tied array interface is
1026 quite limited. To make the interface more useful, a number of methods
1027 are supplied with B<DB_File> to simulate the standard array operations
1028 that are not currently implemented in Perl's tied array interface. All
1029 these methods are accessed via the object returned from the tie call.
1031 Here are the methods:
1035 =item B<$X-E<gt>push(list) ;>
1037 Pushes the elements of C<list> to the end of the array.
1039 =item B<$value = $X-E<gt>pop ;>
1041 Removes and returns the last element of the array.
1043 =item B<$X-E<gt>shift>
1045 Removes and returns the first element of the array.
1047 =item B<$X-E<gt>unshift(list) ;>
1049 Pushes the elements of C<list> to the start of the array.
1051 =item B<$X-E<gt>length>
1053 Returns the number of elements in the array.
1057 =head2 Another Example
1059 Here is a more complete example that makes use of some of the methods
1060 described above. It also makes use of the API interface directly (see
1061 L<THE API INTERFACE>).
1064 use vars qw(@h $H $file $i) ;
1072 $H = tie @h, "DB_File", $file, O_RDWR|O_CREAT, 0640, $DB_RECNO
1073 or die "Cannot open file $file: $!\n" ;
1075 # first create a text file to play with
1083 # Print the records in order.
1085 # The length method is needed here because evaluating a tied
1086 # array in a scalar context does not return the number of
1087 # elements in the array.
1089 print "\nORIGINAL\n" ;
1090 foreach $i (0 .. $H->length - 1) {
1091 print "$i: $h[$i]\n" ;
1094 # use the push & pop methods
1097 print "\nThe last record was [$a]\n" ;
1099 # and the shift & unshift methods
1101 $H->unshift("first") ;
1102 print "The first record was [$a]\n" ;
1104 # Use the API to add a new record after record 2.
1106 $H->put($i, "Newbie", R_IAFTER) ;
1108 # and a new record before record 1.
1110 $H->put($i, "New One", R_IBEFORE) ;
1115 # now print the records in reverse order
1116 print "\nREVERSE\n" ;
1117 for ($i = $H->length - 1 ; $i >= 0 ; -- $i)
1118 { print "$i: $h[$i]\n" }
1120 # same again, but use the API functions instead
1121 print "\nREVERSE again\n" ;
1122 my ($s, $k, $v) = (0, 0, 0) ;
1123 for ($s = $H->seq($k, $v, R_LAST) ;
1125 $s = $H->seq($k, $v, R_PREV))
1126 { print "$k: $v\n" }
1131 and this is what it outputs:
1140 The last record was [four]
1141 The first record was [zero]
1165 Rather than iterating through the array, C<@h> like this:
1169 it is necessary to use either this:
1171 foreach $i (0 .. $H->length - 1)
1175 for ($a = $H->get($k, $v, R_FIRST) ;
1177 $a = $H->get($k, $v, R_NEXT) )
1181 Notice that both times the C<put> method was used the record index was
1182 specified using a variable, C<$i>, rather than the literal value
1183 itself. This is because C<put> will return the record number of the
1184 inserted line via that parameter.
1188 =head1 THE API INTERFACE
1190 As well as accessing Berkeley DB using a tied hash or array, it is also
1191 possible to make direct use of most of the API functions defined in the
1192 Berkeley DB documentation.
1194 To do this you need to store a copy of the object returned from the tie.
1196 $db = tie %hash, "DB_File", "filename" ;
1198 Once you have done that, you can access the Berkeley DB API functions
1199 as B<DB_File> methods directly like this:
1201 $db->put($key, $value, R_NOOVERWRITE) ;
1203 B<Important:> If you have saved a copy of the object returned from
1204 C<tie>, the underlying database file will I<not> be closed until both
1205 the tied variable is untied and all copies of the saved object are
1209 $db = tie %hash, "DB_File", "filename"
1210 or die "Cannot tie filename: $!" ;
1215 See L<The untie() Gotcha> for more details.
1217 All the functions defined in L<dbopen> are available except for
1218 close() and dbopen() itself. The B<DB_File> method interface to the
1219 supported functions have been implemented to mirror the way Berkeley DB
1220 works whenever possible. In particular note that:
1226 The methods return a status value. All return 0 on success.
1227 All return -1 to signify an error and set C<$!> to the exact
1228 error code. The return code 1 generally (but not always) means that the
1229 key specified did not exist in the database.
1231 Other return codes are defined. See below and in the Berkeley DB
1232 documentation for details. The Berkeley DB documentation should be used
1233 as the definitive source.
1237 Whenever a Berkeley DB function returns data via one of its parameters,
1238 the equivalent B<DB_File> method does exactly the same.
1242 If you are careful, it is possible to mix API calls with the tied
1243 hash/array interface in the same piece of code. Although only a few of
1244 the methods used to implement the tied interface currently make use of
1245 the cursor, you should always assume that the cursor has been changed
1246 any time the tied hash/array interface is used. As an example, this
1247 code will probably not do what you expect:
1249 $X = tie %x, 'DB_File', $filename, O_RDWR|O_CREAT, 0777, $DB_BTREE
1250 or die "Cannot tie $filename: $!" ;
1252 # Get the first key/value pair and set the cursor
1253 $X->seq($key, $value, R_FIRST) ;
1255 # this line will modify the cursor
1256 $count = scalar keys %x ;
1258 # Get the second key/value pair.
1259 # oops, it didn't, it got the last key/value pair!
1260 $X->seq($key, $value, R_NEXT) ;
1262 The code above can be rearranged to get around the problem, like this:
1264 $X = tie %x, 'DB_File', $filename, O_RDWR|O_CREAT, 0777, $DB_BTREE
1265 or die "Cannot tie $filename: $!" ;
1267 # this line will modify the cursor
1268 $count = scalar keys %x ;
1270 # Get the first key/value pair and set the cursor
1271 $X->seq($key, $value, R_FIRST) ;
1273 # Get the second key/value pair.
1275 $X->seq($key, $value, R_NEXT) ;
1279 All the constants defined in L<dbopen> for use in the flags parameters
1280 in the methods defined below are also available. Refer to the Berkeley
1281 DB documentation for the precise meaning of the flags values.
1283 Below is a list of the methods available.
1287 =item B<$status = $X-E<gt>get($key, $value [, $flags]) ;>
1289 Given a key (C<$key>) this method reads the value associated with it
1290 from the database. The value read from the database is returned in the
1291 C<$value> parameter.
1293 If the key does not exist the method returns 1.
1295 No flags are currently defined for this method.
1297 =item B<$status = $X-E<gt>put($key, $value [, $flags]) ;>
1299 Stores the key/value pair in the database.
1301 If you use either the R_IAFTER or R_IBEFORE flags, the C<$key> parameter
1302 will have the record number of the inserted key/value pair set.
1304 Valid flags are R_CURSOR, R_IAFTER, R_IBEFORE, R_NOOVERWRITE and
1307 =item B<$status = $X-E<gt>del($key [, $flags]) ;>
1309 Removes all key/value pairs with key C<$key> from the database.
1311 A return code of 1 means that the requested key was not in the
1314 R_CURSOR is the only valid flag at present.
1316 =item B<$status = $X-E<gt>fd ;>
1318 Returns the file descriptor for the underlying database.
1320 See L<Locking Databases> for an example of how to make use of the
1321 C<fd> method to lock your database.
1323 =item B<$status = $X-E<gt>seq($key, $value, $flags) ;>
1325 This interface allows sequential retrieval from the database. See
1326 L<dbopen> for full details.
1328 Both the C<$key> and C<$value> parameters will be set to the key/value
1329 pair read from the database.
1331 The flags parameter is mandatory. The valid flag values are R_CURSOR,
1332 R_FIRST, R_LAST, R_NEXT and R_PREV.
1334 =item B<$status = $X-E<gt>sync([$flags]) ;>
1336 Flushes any cached buffers to disk.
1338 R_RECNOSYNC is the only valid flag at present.
1342 =head1 HINTS AND TIPS
1345 =head2 Locking Databases
1347 Concurrent access of a read-write database by several parties requires
1348 them all to use some kind of locking. Here's an example of Tom's that
1349 uses the I<fd> method to get the file descriptor, and then a careful
1350 open() to give something Perl will flock() for you. Run this repeatedly
1351 in the background to watch the locks granted in proper order.
1362 my($oldval, $fd, $db, %db, $value, $key);
1364 $key = shift || 'default';
1365 $value = shift || 'magic';
1369 $db = tie(%db, 'DB_File', '/tmp/foo.db', O_CREAT|O_RDWR, 0644)
1370 || die "dbcreat /tmp/foo.db $!";
1372 print "$$: db fd is $fd\n";
1373 open(DB_FH, "+<&=$fd") || die "dup $!";
1376 unless (flock (DB_FH, LOCK_SH | LOCK_NB)) {
1377 print "$$: CONTENTION; can't read during write update!
1378 Waiting for read lock ($!) ....";
1379 unless (flock (DB_FH, LOCK_SH)) { die "flock: $!" }
1381 print "$$: Read lock granted\n";
1383 $oldval = $db{$key};
1384 print "$$: Old value was $oldval\n";
1385 flock(DB_FH, LOCK_UN);
1387 unless (flock (DB_FH, LOCK_EX | LOCK_NB)) {
1388 print "$$: CONTENTION; must have exclusive lock!
1389 Waiting for write lock ($!) ....";
1390 unless (flock (DB_FH, LOCK_EX)) { die "flock: $!" }
1393 print "$$: Write lock granted\n";
1395 $db->sync; # to flush
1398 flock(DB_FH, LOCK_UN);
1402 print "$$: Updated db to $key=$value\n";
1404 =head2 Sharing Databases With C Applications
1406 There is no technical reason why a Berkeley DB database cannot be
1407 shared by both a Perl and a C application.
1409 The vast majority of problems that are reported in this area boil down
1410 to the fact that C strings are NULL terminated, whilst Perl strings are
1413 Here is a real example. Netscape 2.0 keeps a record of the locations you
1414 visit along with the time you last visited them in a DB_HASH database.
1415 This is usually stored in the file F<~/.netscape/history.db>. The key
1416 field in the database is the location string and the value field is the
1417 time the location was last visited stored as a 4 byte binary value.
1419 If you haven't already guessed, the location string is stored with a
1420 terminating NULL. This means you need to be careful when accessing the
1423 Here is a snippet of code that is loosely based on Tom Christiansen's
1424 I<ggh> script (available from your nearest CPAN archive in
1425 F<authors/id/TOMC/scripts/nshist.gz>).
1431 use vars qw( $dotdir $HISTORY %hist_db $href $binary_time $date ) ;
1432 $dotdir = $ENV{HOME} || $ENV{LOGNAME};
1434 $HISTORY = "$dotdir/.netscape/history.db";
1436 tie %hist_db, 'DB_File', $HISTORY
1437 or die "Cannot open $HISTORY: $!\n" ;;
1439 # Dump the complete database
1440 while ( ($href, $binary_time) = each %hist_db ) {
1442 # remove the terminating NULL
1443 $href =~ s/\x00$// ;
1445 # convert the binary time into a user friendly string
1446 $date = localtime unpack("V", $binary_time);
1447 print "$date $href\n" ;
1450 # check for the existence of a specific key
1451 # remember to add the NULL
1452 if ( $binary_time = $hist_db{"http://mox.perl.com/\x00"} ) {
1453 $date = localtime unpack("V", $binary_time) ;
1454 print "Last visited mox.perl.com on $date\n" ;
1457 print "Never visited mox.perl.com\n"
1462 =head2 The untie() Gotcha
1464 If you make use of the Berkeley DB API, it is I<very> strongly
1465 recommended that you read L<perltie/The untie Gotcha>.
1467 Even if you don't currently make use of the API interface, it is still
1470 Here is an example which illustrates the problem from a B<DB_File>
1479 $X = tie %x, 'DB_File', 'tst.fil' , O_RDWR|O_TRUNC
1480 or die "Cannot tie first time: $!" ;
1486 tie %x, 'DB_File', 'tst.fil' , O_RDWR|O_CREAT
1487 or die "Cannot tie second time: $!" ;
1491 When run, the script will produce this error message:
1493 Cannot tie second time: Invalid argument at bad.file line 14.
1495 Although the error message above refers to the second tie() statement
1496 in the script, the source of the problem is really with the untie()
1497 statement that precedes it.
1499 Having read L<perltie> you will probably have already guessed that the
1500 error is caused by the extra copy of the tied object stored in C<$X>.
1501 If you haven't, then the problem boils down to the fact that the
1502 B<DB_File> destructor, DESTROY, will not be called until I<all>
1503 references to the tied object are destroyed. Both the tied variable,
1504 C<%x>, and C<$X> above hold a reference to the object. The call to
1505 untie() will destroy the first, but C<$X> still holds a valid
1506 reference, so the destructor will not get called and the database file
1507 F<tst.fil> will remain open. The fact that Berkeley DB then reports the
1508 attempt to open a database that is alreday open via the catch-all
1509 "Invalid argument" doesn't help.
1511 If you run the script with the C<-w> flag the error message becomes:
1513 untie attempted while 1 inner references still exist at bad.file line 12.
1514 Cannot tie second time: Invalid argument at bad.file line 14.
1516 which pinpoints the real problem. Finally the script can now be
1517 modified to fix the original problem by destroying the API object
1526 $X = tie %x, 'DB_File', 'tst.fil' , O_RDWR|O_CREAT
1530 =head1 COMMON QUESTIONS
1532 =head2 Why is there Perl source in my database?
1534 If you look at the contents of a database file created by DB_File,
1535 there can sometimes be part of a Perl script included in it.
1537 This happens because Berkeley DB uses dynamic memory to allocate
1538 buffers which will subsequently be written to the database file. Being
1539 dynamic, the memory could have been used for anything before DB
1540 malloced it. As Berkeley DB doesn't clear the memory once it has been
1541 allocated, the unused portions will contain random junk. In the case
1542 where a Perl script gets written to the database, the random junk will
1543 correspond to an area of dynamic memory that happened to be used during
1544 the compilation of the script.
1546 Unless you don't like the possibility of there being part of your Perl
1547 scripts embedded in a database file, this is nothing to worry about.
1549 =head2 How do I store complex data structures with DB_File?
1551 Although B<DB_File> cannot do this directly, there is a module which
1552 can layer transparently over B<DB_File> to accomplish this feat.
1554 Check out the MLDBM module, available on CPAN in the directory
1555 F<modules/by-module/MLDBM>.
1557 =head2 What does "Invalid Argument" mean?
1559 You will get this error message when one of the parameters in the
1560 C<tie> call is wrong. Unfortunately there are quite a few parameters to
1561 get wrong, so it can be difficult to figure out which one it is.
1563 Here are a couple of possibilities:
1569 Attempting to reopen a database without closing it.
1573 Using the O_WRONLY flag.
1577 =head2 What does "Bareword 'DB_File' not allowed" mean?
1579 You will encounter this particular error message when you have the
1580 C<strict 'subs'> pragma (or the full strict pragma) in your script.
1581 Consider this script:
1586 tie %x, DB_File, "filename" ;
1588 Running it produces the error in question:
1590 Bareword "DB_File" not allowed while "strict subs" in use
1592 To get around the error, place the word C<DB_File> in either single or
1593 double quotes, like this:
1595 tie %x, "DB_File", "filename" ;
1597 Although it might seem like a real pain, it is really worth the effort
1598 of having a C<use strict> in all your scripts.
1602 Moved to the Changes file.
1606 Some older versions of Berkeley DB had problems with fixed length
1607 records using the RECNO file format. This problem has been fixed since
1608 version 1.85 of Berkeley DB.
1610 I am sure there are bugs in the code. If you do find any, or can
1611 suggest any enhancements, I would welcome your comments.
1615 B<DB_File> comes with the standard Perl source distribution. Look in
1616 the directory F<ext/DB_File>. Given the amount of time between releases
1617 of Perl the version that ships with Perl is quite likely to be out of
1618 date, so the most recent version can always be found on CPAN (see
1619 L<perlmod/CPAN> for details), in the directory
1620 F<modules/by-module/DB_File>.
1622 This version of B<DB_File> will work with either version 1.x or 2.x of
1623 Berkeley DB, but is limited to the functionality provided by version 1.
1625 The official web site for Berkeley DB is
1626 F<http://www.sleepycat.com/db>. The ftp equivalent is
1627 F<ftp.sleepycat.com:/pub>. Both versions 1 and 2 of Berkeley DB are
1630 Alternatively, Berkeley DB version 1 is available at your nearest CPAN
1631 archive in F<src/misc/db.1.85.tar.gz>.
1633 If you are running IRIX, then get Berkeley DB version 1 from
1634 F<http://reality.sgi.com/ariel>. It has the patches necessary to
1635 compile properly on IRIX 5.3.
1639 Copyright (c) 1997 Paul Marquess. All rights reserved. This program is
1640 free software; you can redistribute it and/or modify it under the same
1641 terms as Perl itself.
1643 Although B<DB_File> is covered by the Perl license, the library it
1644 makes use of, namely Berkeley DB, is not. Berkeley DB has its own
1645 copyright and its own license. Please take the time to read it.
1647 The license for Berkeley DB version 2, and how it relates to DB_File
1648 does need some extra clarification. Here are are few words taken from
1649 the Berkeley DB FAQ regarding the version 2 license:
1651 The major difference is that the license for DB 2.0, when
1652 downloaded from the net, requires that the software that
1653 uses DB 2.0 be freely redistributable.
1655 That means that if you want to use DB_File, and you have changed either
1656 the source for Berkeley DB or Perl, then the changes must be freely
1659 In the case of Perl, the term source refers to the complete source
1660 code for Perl (e.g. sv.c, toke.c, perl.h) and any external modules that
1661 you are using (e.g. DB_File, Tk).
1663 Note that any Perl scripts that you write are your property - this
1664 includes scripts that make use of DB_File. Neither the Perl license or
1665 the Berkeley DB license place any restriction on what you have to do
1668 If you are in any doubt about the license situation, contact either the
1669 Berkeley DB authors or the author of DB_File. See L<"AUTHOR"> for details.
1674 L<perl(1)>, L<dbopen(3)>, L<hash(3)>, L<recno(3)>, L<btree(3)>
1678 The DB_File interface was written by Paul Marquess
1679 E<lt>pmarquess@bfsec.bt.co.ukE<gt>.
1680 Questions about the DB system itself may be addressed to
1681 E<lt>db@sleepycat.com<gt>.