1 # DB_File.pm -- Perl 5 interface to Berkeley DB
3 # written by Paul Marquess (pmarquess@bfsec.bt.co.uk)
4 # last modified 20th Dec 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 $DB_RECNO
196 ($constname = $AUTOLOAD) =~ s/.*:://;
197 my $val = constant($constname, @_ ? $_[0] : 0);
199 if ($! =~ /Invalid/) {
200 $AutoLoader::AUTOLOAD = $AUTOLOAD;
201 goto &AutoLoader::AUTOLOAD;
204 my($pack,$file,$line) = caller;
205 croak "Your vendor has not defined DB macro $constname, used at $file line $line.
209 eval "sub $AUTOLOAD { $val }";
215 # Make all Fcntl O_XXX constants available for importing
217 my @O = grep /^O_/, @Fcntl::EXPORT;
218 Fcntl->import(@O); # first we import what we want to export
222 ## import borrowed from IO::File
223 ## exports Fcntl constants if available.
226 # my $callpkg = caller;
227 # Exporter::export $pkg, $callpkg, @_;
230 # Exporter::export 'Fcntl', $callpkg, '/^O_/';
234 bootstrap DB_File $VERSION;
236 # Preloaded methods go here. Autoload methods go after __END__, and are
237 # processed by the autosplit program.
239 sub tie_hash_or_array
242 my $tieHASH = ( (caller(1))[3] =~ /TIEHASH/ ) ;
244 $arg[4] = tied %{ $arg[4] }
245 if @arg >= 5 && ref $arg[4] && $arg[4] =~ /=HASH/ && tied %{ $arg[4] } ;
247 # make recno in Berkeley DB version 2 work like recno in version 1.
248 if ($db_version > 1 and defined $arg[4] and $arg[4] =~ /RECNO/ and
249 $arg[1] and ! -e $arg[1]) {
250 open(FH, ">$arg[1]") or return undef ;
252 chmod $arg[3] ? $arg[3] : 0666 , $arg[1] ;
255 DoTie_($tieHASH, @arg) ;
260 tie_hash_or_array(@_) ;
265 tie_hash_or_array(@_) ;
273 my $status = $self->seq($key, $value, R_FIRST());
276 while ($status == 0) {
278 $status = $self->seq($key, $value, R_NEXT());
280 foreach $key (reverse @keys) {
281 my $s = $self->del($key);
291 my $current_length = $self->length() ;
293 if ($length < $current_length) {
295 for ($key = $current_length - 1 ; $key >= $length ; -- $key)
298 elsif ($length > $current_length)
299 { $self->put($length-1, "") }
304 croak "Usage: \$db->get_dup(key [,flag])\n"
305 unless @_ == 2 or @_ == 3 ;
312 my $wantarray = wantarray ;
318 # iterate through the database until either EOF ($status == 0)
319 # or a different key is encountered ($key ne $origkey).
320 for ($status = $db->seq($key, $value, R_CURSOR()) ;
321 $status == 0 and $key eq $origkey ;
322 $status = $db->seq($key, $value, R_NEXT()) ) {
324 # save the value or count number of matches
327 { ++ $values{$value} }
329 { push (@values, $value) }
336 return ($wantarray ? ($flag ? %values : @values) : $counter) ;
345 DB_File - Perl5 access to Berkeley DB version 1.x
351 [$X =] tie %hash, 'DB_File', [$filename, $flags, $mode, $DB_HASH] ;
352 [$X =] tie %hash, 'DB_File', $filename, $flags, $mode, $DB_BTREE ;
353 [$X =] tie @array, 'DB_File', $filename, $flags, $mode, $DB_RECNO ;
355 $status = $X->del($key [, $flags]) ;
356 $status = $X->put($key, $value [, $flags]) ;
357 $status = $X->get($key, $value [, $flags]) ;
358 $status = $X->seq($key, $value, $flags) ;
359 $status = $X->sync([$flags]) ;
363 $count = $X->get_dup($key) ;
364 @list = $X->get_dup($key) ;
365 %list = $X->get_dup($key, 1) ;
379 B<DB_File> is a module which allows Perl programs to make use of the
380 facilities provided by Berkeley DB version 1.x (if you have a newer
381 version of DB, see L<Using DB_File with Berkeley DB version 2>). It is
382 assumed that you have a copy of the Berkeley DB manual pages at hand
383 when reading this documentation. The interface defined here mirrors the
384 Berkeley DB interface closely.
386 Berkeley DB is a C library which provides a consistent interface to a
387 number of database formats. B<DB_File> provides an interface to all
388 three of the database types currently supported by Berkeley DB.
396 This database type allows arbitrary key/value pairs to be stored in data
397 files. This is equivalent to the functionality provided by other
398 hashing packages like DBM, NDBM, ODBM, GDBM, and SDBM. Remember though,
399 the files created using DB_HASH are not compatible with any of the
400 other packages mentioned.
402 A default hashing algorithm, which will be adequate for most
403 applications, is built into Berkeley DB. If you do need to use your own
404 hashing algorithm it is possible to write your own in Perl and have
405 B<DB_File> use it instead.
409 The btree format allows arbitrary key/value pairs to be stored in a
410 sorted, balanced binary tree.
412 As with the DB_HASH format, it is possible to provide a user defined
413 Perl routine to perform the comparison of keys. By default, though, the
414 keys are stored in lexical order.
418 DB_RECNO allows both fixed-length and variable-length flat text files
419 to be manipulated using the same key/value pair interface as in DB_HASH
420 and DB_BTREE. In this case the key will consist of a record (line)
425 =head2 Using DB_File with Berkeley DB version 2
427 Although B<DB_File> is intended to be used with Berkeley DB version 1,
428 it can also be used with version 2. In this case the interface is
429 limited to the functionality provided by Berkeley DB 1.x. Anywhere the
430 version 2 interface differs, B<DB_File> arranges for it to work like
431 version 1. This feature allows B<DB_File> scripts that were built with
432 version 1 to be migrated to version 2 without any changes.
434 If you want to make use of the new features available in Berkeley DB
435 2.x, use the Perl module B<BerkeleyDB> instead.
437 At the time of writing this document the B<BerkeleyDB> module is still
438 alpha quality (the version number is < 1.0), and so unsuitable for use
439 in any serious development work. Once its version number is >= 1.0, it
440 is considered stable enough for real work.
442 B<Note:> The database file format has changed in Berkeley DB version 2.
443 If you cannot recreate your databases, you must dump any existing
444 databases with the C<db_dump185> utility that comes with Berkeley DB.
445 Once you have upgraded DB_File to use Berkeley DB version 2, your
446 databases can be recreated using C<db_load>. Refer to the Berkeley DB
447 documentation for further details.
449 Please read L<COPYRIGHT> before using version 2.x of Berkeley DB with
452 =head2 Interface to Berkeley DB
454 B<DB_File> allows access to Berkeley DB files using the tie() mechanism
455 in Perl 5 (for full details, see L<perlfunc/tie()>). This facility
456 allows B<DB_File> to access Berkeley DB files using either an
457 associative array (for DB_HASH & DB_BTREE file types) or an ordinary
458 array (for the DB_RECNO file type).
460 In addition to the tie() interface, it is also possible to access most
461 of the functions provided in the Berkeley DB API directly.
462 See L<THE API INTERFACE>.
464 =head2 Opening a Berkeley DB Database File
466 Berkeley DB uses the function dbopen() to open or create a database.
467 Here is the C prototype for dbopen():
470 dbopen (const char * file, int flags, int mode,
471 DBTYPE type, const void * openinfo)
473 The parameter C<type> is an enumeration which specifies which of the 3
474 interface methods (DB_HASH, DB_BTREE or DB_RECNO) is to be used.
475 Depending on which of these is actually chosen, the final parameter,
476 I<openinfo> points to a data structure which allows tailoring of the
477 specific interface method.
479 This interface is handled slightly differently in B<DB_File>. Here is
480 an equivalent call using B<DB_File>:
482 tie %array, 'DB_File', $filename, $flags, $mode, $DB_HASH ;
484 The C<filename>, C<flags> and C<mode> parameters are the direct
485 equivalent of their dbopen() counterparts. The final parameter $DB_HASH
486 performs the function of both the C<type> and C<openinfo> parameters in
489 In the example above $DB_HASH is actually a pre-defined reference to a
490 hash object. B<DB_File> has three of these pre-defined references.
491 Apart from $DB_HASH, there is also $DB_BTREE and $DB_RECNO.
493 The keys allowed in each of these pre-defined references is limited to
494 the names used in the equivalent C structure. So, for example, the
495 $DB_HASH reference will only allow keys called C<bsize>, C<cachesize>,
496 C<ffactor>, C<hash>, C<lorder> and C<nelem>.
498 To change one of these elements, just assign to it like this:
500 $DB_HASH->{'cachesize'} = 10000 ;
502 The three predefined variables $DB_HASH, $DB_BTREE and $DB_RECNO are
503 usually adequate for most applications. If you do need to create extra
504 instances of these objects, constructors are available for each file
507 Here are examples of the constructors and the valid options available
508 for DB_HASH, DB_BTREE and DB_RECNO respectively.
510 $a = new DB_File::HASHINFO ;
518 $b = new DB_File::BTREEINFO ;
528 $c = new DB_File::RECNOINFO ;
537 The values stored in the hashes above are mostly the direct equivalent
538 of their C counterpart. Like their C counterparts, all are set to a
539 default values - that means you don't have to set I<all> of the
540 values when you only want to change one. Here is an example:
542 $a = new DB_File::HASHINFO ;
543 $a->{'cachesize'} = 12345 ;
544 tie %y, 'DB_File', "filename", $flags, 0777, $a ;
546 A few of the options need extra discussion here. When used, the C
547 equivalent of the keys C<hash>, C<compare> and C<prefix> store pointers
548 to C functions. In B<DB_File> these keys are used to store references
549 to Perl subs. Below are templates for each of the subs:
555 # return the hash value for $data
561 my ($key, $key2) = @_ ;
563 # return 0 if $key1 eq $key2
564 # -1 if $key1 lt $key2
565 # 1 if $key1 gt $key2
566 return (-1 , 0 or 1) ;
571 my ($key, $key2) = @_ ;
573 # return number of bytes of $key2 which are
574 # necessary to determine that it is greater than $key1
578 See L<Changing the BTREE sort order> for an example of using the
581 If you are using the DB_RECNO interface and you intend making use of
582 C<bval>, you should check out L<The 'bval' Option>.
584 =head2 Default Parameters
586 It is possible to omit some or all of the final 4 parameters in the
587 call to C<tie> and let them take default values. As DB_HASH is the most
588 common file format used, the call:
590 tie %A, "DB_File", "filename" ;
594 tie %A, "DB_File", "filename", O_CREAT|O_RDWR, 0666, $DB_HASH ;
596 It is also possible to omit the filename parameter as well, so the
603 tie %A, "DB_File", undef, O_CREAT|O_RDWR, 0666, $DB_HASH ;
605 See L<In Memory Databases> for a discussion on the use of C<undef>
606 in place of a filename.
608 =head2 In Memory Databases
610 Berkeley DB allows the creation of in-memory databases by using NULL
611 (that is, a C<(char *)0> in C) in place of the filename. B<DB_File>
612 uses C<undef> instead of NULL to provide this functionality.
616 The DB_HASH file format is probably the most commonly used of the three
617 file formats that B<DB_File> supports. It is also very straightforward
620 =head2 A Simple Example
622 This example shows how to create a database, add key/value pairs to the
623 database, delete keys/value pairs and finally how to enumerate the
624 contents of the database.
628 use vars qw( %h $k $v ) ;
630 tie %h, "DB_File", "fruit", O_RDWR|O_CREAT, 0640, $DB_HASH
631 or die "Cannot open file 'fruit': $!\n";
633 # Add a few key/value pairs to the file
634 $h{"apple"} = "red" ;
635 $h{"orange"} = "orange" ;
636 $h{"banana"} = "yellow" ;
637 $h{"tomato"} = "red" ;
639 # Check for existence of a key
640 print "Banana Exists\n\n" if $h{"banana"} ;
642 # Delete a key/value pair.
645 # print the contents of the file
646 while (($k, $v) = each %h)
647 { print "$k -> $v\n" }
659 Note that the like ordinary associative arrays, the order of the keys
660 retrieved is in an apparently random order.
664 The DB_BTREE format is useful when you want to store data in a given
665 order. By default the keys will be stored in lexical order, but as you
666 will see from the example shown in the next section, it is very easy to
667 define your own sorting function.
669 =head2 Changing the BTREE sort order
671 This script shows how to override the default sorting algorithm that
672 BTREE uses. Instead of using the normal lexical ordering, a case
673 insensitive compare function will be used.
682 my ($key1, $key2) = @_ ;
683 "\L$key1" cmp "\L$key2" ;
686 # specify the Perl sub that will do the comparison
687 $DB_BTREE->{'compare'} = \&Compare ;
689 tie %h, "DB_File", "tree", O_RDWR|O_CREAT, 0640, $DB_BTREE
690 or die "Cannot open file 'tree': $!\n" ;
692 # Add a key/value pair to the file
693 $h{'Wall'} = 'Larry' ;
694 $h{'Smith'} = 'John' ;
695 $h{'mouse'} = 'mickey' ;
696 $h{'duck'} = 'donald' ;
701 # Cycle through the keys printing them in order.
702 # Note it is not necessary to sort the keys as
703 # the btree will have kept them in order automatically.
709 Here is the output from the code above.
715 There are a few point to bear in mind if you want to change the
716 ordering in a BTREE database:
722 The new compare function must be specified when you create the database.
726 You cannot change the ordering once the database has been created. Thus
727 you must use the same compare function every time you access the
732 =head2 Handling Duplicate Keys
734 The BTREE file type optionally allows a single key to be associated
735 with an arbitrary number of values. This option is enabled by setting
736 the flags element of C<$DB_BTREE> to R_DUP when creating the database.
738 There are some difficulties in using the tied hash interface if you
739 want to manipulate a BTREE database with duplicate keys. Consider this
745 use vars qw($filename %h ) ;
750 # Enable duplicate records
751 $DB_BTREE->{'flags'} = R_DUP ;
753 tie %h, "DB_File", $filename, O_RDWR|O_CREAT, 0640, $DB_BTREE
754 or die "Cannot open $filename: $!\n";
756 # Add some key/value pairs to the file
757 $h{'Wall'} = 'Larry' ;
758 $h{'Wall'} = 'Brick' ; # Note the duplicate key
759 $h{'Wall'} = 'Brick' ; # Note the duplicate key and value
760 $h{'Smith'} = 'John' ;
761 $h{'mouse'} = 'mickey' ;
763 # iterate through the associative array
764 # and print each key/value pair.
766 { print "$_ -> $h{$_}\n" }
778 As you can see 3 records have been successfully created with key C<Wall>
779 - the only thing is, when they are retrieved from the database they
780 I<seem> to have the same value, namely C<Larry>. The problem is caused
781 by the way that the associative array interface works. Basically, when
782 the associative array interface is used to fetch the value associated
783 with a given key, it will only ever retrieve the first value.
785 Although it may not be immediately obvious from the code above, the
786 associative array interface can be used to write values with duplicate
787 keys, but it cannot be used to read them back from the database.
789 The way to get around this problem is to use the Berkeley DB API method
790 called C<seq>. This method allows sequential access to key/value
791 pairs. See L<THE API INTERFACE> for details of both the C<seq> method
792 and the API in general.
794 Here is the script above rewritten using the C<seq> API method.
799 use vars qw($filename $x %h $status $key $value) ;
804 # Enable duplicate records
805 $DB_BTREE->{'flags'} = R_DUP ;
807 $x = tie %h, "DB_File", $filename, O_RDWR|O_CREAT, 0640, $DB_BTREE
808 or die "Cannot open $filename: $!\n";
810 # Add some key/value pairs to the file
811 $h{'Wall'} = 'Larry' ;
812 $h{'Wall'} = 'Brick' ; # Note the duplicate key
813 $h{'Wall'} = 'Brick' ; # Note the duplicate key and value
814 $h{'Smith'} = 'John' ;
815 $h{'mouse'} = 'mickey' ;
817 # iterate through the btree using seq
818 # and print each key/value pair.
820 for ($status = $x->seq($key, $value, R_FIRST) ;
822 $status = $x->seq($key, $value, R_NEXT) )
823 { print "$key -> $value\n" }
836 This time we have got all the key/value pairs, including the multiple
837 values associated with the key C<Wall>.
839 =head2 The get_dup() Method
841 B<DB_File> comes with a utility method, called C<get_dup>, to assist in
842 reading duplicate values from BTREE databases. The method can take the
845 $count = $x->get_dup($key) ;
846 @list = $x->get_dup($key) ;
847 %list = $x->get_dup($key, 1) ;
849 In a scalar context the method returns the number of values associated
850 with the key, C<$key>.
852 In list context, it returns all the values which match C<$key>. Note
853 that the values will be returned in an apparently random order.
855 In list context, if the second parameter is present and evaluates
856 TRUE, the method returns an associative array. The keys of the
857 associative array correspond to the values that matched in the BTREE
858 and the values of the array are a count of the number of times that
859 particular value occurred in the BTREE.
861 So assuming the database created above, we can use C<get_dup> like
864 my $cnt = $x->get_dup("Wall") ;
865 print "Wall occurred $cnt times\n" ;
867 my %hash = $x->get_dup("Wall", 1) ;
868 print "Larry is there\n" if $hash{'Larry'} ;
869 print "There are $hash{'Brick'} Brick Walls\n" ;
871 my @list = $x->get_dup("Wall") ;
872 print "Wall => [@list]\n" ;
874 @list = $x->get_dup("Smith") ;
875 print "Smith => [@list]\n" ;
877 @list = $x->get_dup("Dog") ;
878 print "Dog => [@list]\n" ;
883 Wall occurred 3 times
885 There are 2 Brick Walls
886 Wall => [Brick Brick Larry]
890 =head2 Matching Partial Keys
892 The BTREE interface has a feature which allows partial keys to be
893 matched. This functionality is I<only> available when the C<seq> method
894 is used along with the R_CURSOR flag.
896 $x->seq($key, $value, R_CURSOR) ;
898 Here is the relevant quote from the dbopen man page where it defines
899 the use of the R_CURSOR flag with seq:
901 Note, for the DB_BTREE access method, the returned key is not
902 necessarily an exact match for the specified key. The returned key
903 is the smallest key greater than or equal to the specified key,
904 permitting partial key matches and range searches.
906 In the example script below, the C<match> sub uses this feature to find
907 and print the first matching key/value pair given a partial key.
913 use vars qw($filename $x %h $st $key $value) ;
919 my $orig_key = $key ;
920 $x->seq($key, $value, R_CURSOR) ;
921 print "$orig_key\t-> $key\t-> $value\n" ;
927 $x = tie %h, "DB_File", $filename, O_RDWR|O_CREAT, 0640, $DB_BTREE
928 or die "Cannot open $filename: $!\n";
930 # Add some key/value pairs to the file
931 $h{'mouse'} = 'mickey' ;
932 $h{'Wall'} = 'Larry' ;
933 $h{'Walls'} = 'Brick' ;
934 $h{'Smith'} = 'John' ;
939 for ($st = $x->seq($key, $value, R_FIRST) ;
941 $st = $x->seq($key, $value, R_NEXT) )
943 { print "$key -> $value\n" }
945 print "\nPARTIAL MATCH\n" ;
969 DB_RECNO provides an interface to flat text files. Both variable and
970 fixed length records are supported.
972 In order to make RECNO more compatible with Perl the array offset for
973 all RECNO arrays begins at 0 rather than 1 as in Berkeley DB.
975 As with normal Perl arrays, a RECNO array can be accessed using
976 negative indexes. The index -1 refers to the last element of the array,
977 -2 the second last, and so on. Attempting to access an element before
978 the start of the array will raise a fatal run-time error.
980 =head2 The 'bval' Option
982 The operation of the bval option warrants some discussion. Here is the
983 definition of bval from the Berkeley DB 1.85 recno manual page:
985 The delimiting byte to be used to mark the end of a
986 record for variable-length records, and the pad charac-
987 ter for fixed-length records. If no value is speci-
988 fied, newlines (``\n'') are used to mark the end of
989 variable-length records and fixed-length records are
992 The second sentence is wrong. In actual fact bval will only default to
993 C<"\n"> when the openinfo parameter in dbopen is NULL. If a non-NULL
994 openinfo parameter is used at all, the value that happens to be in bval
995 will be used. That means you always have to specify bval when making
996 use of any of the options in the openinfo parameter. This documentation
997 error will be fixed in the next release of Berkeley DB.
999 That clarifies the situation with regards Berkeley DB itself. What
1000 about B<DB_File>? Well, the behavior defined in the quote above is
1001 quite useful, so B<DB_File> conforms it.
1003 That means that you can specify other options (e.g. cachesize) and
1004 still have bval default to C<"\n"> for variable length records, and
1005 space for fixed length records.
1007 =head2 A Simple Example
1009 Here is a simple example that uses RECNO.
1015 tie @h, "DB_File", "text", O_RDWR|O_CREAT, 0640, $DB_RECNO
1016 or die "Cannot open file 'text': $!\n" ;
1018 # Add a few key/value pairs to the file
1023 # Check for existence of a key
1024 print "Element 1 Exists with value $h[1]\n" if $h[1] ;
1026 # use a negative index
1027 print "The last element is $h[-1]\n" ;
1028 print "The 2nd last element is $h[-2]\n" ;
1032 Here is the output from the script:
1035 Element 1 Exists with value blue
1036 The last element is yellow
1037 The 2nd last element is blue
1039 =head2 Extra Methods
1041 If you are using a version of Perl earlier than 5.004_57, the tied
1042 array interface is quite limited. The example script above will work,
1043 but you won't be able to use C<push>, C<pop>, C<shift>, C<unshift>
1044 etc. with the tied array.
1046 To make the interface more useful for older versions of Perl, a number
1047 of methods are supplied with B<DB_File> to simulate the missing array
1048 operations. All these methods are accessed via the object returned from
1051 Here are the methods:
1055 =item B<$X-E<gt>push(list) ;>
1057 Pushes the elements of C<list> to the end of the array.
1059 =item B<$value = $X-E<gt>pop ;>
1061 Removes and returns the last element of the array.
1063 =item B<$X-E<gt>shift>
1065 Removes and returns the first element of the array.
1067 =item B<$X-E<gt>unshift(list) ;>
1069 Pushes the elements of C<list> to the start of the array.
1071 =item B<$X-E<gt>length>
1073 Returns the number of elements in the array.
1077 =head2 Another Example
1079 Here is a more complete example that makes use of some of the methods
1080 described above. It also makes use of the API interface directly (see
1081 L<THE API INTERFACE>).
1084 use vars qw(@h $H $file $i) ;
1092 $H = tie @h, "DB_File", $file, O_RDWR|O_CREAT, 0640, $DB_RECNO
1093 or die "Cannot open file $file: $!\n" ;
1095 # first create a text file to play with
1103 # Print the records in order.
1105 # The length method is needed here because evaluating a tied
1106 # array in a scalar context does not return the number of
1107 # elements in the array.
1109 print "\nORIGINAL\n" ;
1110 foreach $i (0 .. $H->length - 1) {
1111 print "$i: $h[$i]\n" ;
1114 # use the push & pop methods
1117 print "\nThe last record was [$a]\n" ;
1119 # and the shift & unshift methods
1121 $H->unshift("first") ;
1122 print "The first record was [$a]\n" ;
1124 # Use the API to add a new record after record 2.
1126 $H->put($i, "Newbie", R_IAFTER) ;
1128 # and a new record before record 1.
1130 $H->put($i, "New One", R_IBEFORE) ;
1135 # now print the records in reverse order
1136 print "\nREVERSE\n" ;
1137 for ($i = $H->length - 1 ; $i >= 0 ; -- $i)
1138 { print "$i: $h[$i]\n" }
1140 # same again, but use the API functions instead
1141 print "\nREVERSE again\n" ;
1142 my ($s, $k, $v) = (0, 0, 0) ;
1143 for ($s = $H->seq($k, $v, R_LAST) ;
1145 $s = $H->seq($k, $v, R_PREV))
1146 { print "$k: $v\n" }
1151 and this is what it outputs:
1160 The last record was [four]
1161 The first record was [zero]
1185 Rather than iterating through the array, C<@h> like this:
1189 it is necessary to use either this:
1191 foreach $i (0 .. $H->length - 1)
1195 for ($a = $H->get($k, $v, R_FIRST) ;
1197 $a = $H->get($k, $v, R_NEXT) )
1201 Notice that both times the C<put> method was used the record index was
1202 specified using a variable, C<$i>, rather than the literal value
1203 itself. This is because C<put> will return the record number of the
1204 inserted line via that parameter.
1208 =head1 THE API INTERFACE
1210 As well as accessing Berkeley DB using a tied hash or array, it is also
1211 possible to make direct use of most of the API functions defined in the
1212 Berkeley DB documentation.
1214 To do this you need to store a copy of the object returned from the tie.
1216 $db = tie %hash, "DB_File", "filename" ;
1218 Once you have done that, you can access the Berkeley DB API functions
1219 as B<DB_File> methods directly like this:
1221 $db->put($key, $value, R_NOOVERWRITE) ;
1223 B<Important:> If you have saved a copy of the object returned from
1224 C<tie>, the underlying database file will I<not> be closed until both
1225 the tied variable is untied and all copies of the saved object are
1229 $db = tie %hash, "DB_File", "filename"
1230 or die "Cannot tie filename: $!" ;
1235 See L<The untie() Gotcha> for more details.
1237 All the functions defined in L<dbopen> are available except for
1238 close() and dbopen() itself. The B<DB_File> method interface to the
1239 supported functions have been implemented to mirror the way Berkeley DB
1240 works whenever possible. In particular note that:
1246 The methods return a status value. All return 0 on success.
1247 All return -1 to signify an error and set C<$!> to the exact
1248 error code. The return code 1 generally (but not always) means that the
1249 key specified did not exist in the database.
1251 Other return codes are defined. See below and in the Berkeley DB
1252 documentation for details. The Berkeley DB documentation should be used
1253 as the definitive source.
1257 Whenever a Berkeley DB function returns data via one of its parameters,
1258 the equivalent B<DB_File> method does exactly the same.
1262 If you are careful, it is possible to mix API calls with the tied
1263 hash/array interface in the same piece of code. Although only a few of
1264 the methods used to implement the tied interface currently make use of
1265 the cursor, you should always assume that the cursor has been changed
1266 any time the tied hash/array interface is used. As an example, this
1267 code will probably not do what you expect:
1269 $X = tie %x, 'DB_File', $filename, O_RDWR|O_CREAT, 0777, $DB_BTREE
1270 or die "Cannot tie $filename: $!" ;
1272 # Get the first key/value pair and set the cursor
1273 $X->seq($key, $value, R_FIRST) ;
1275 # this line will modify the cursor
1276 $count = scalar keys %x ;
1278 # Get the second key/value pair.
1279 # oops, it didn't, it got the last key/value pair!
1280 $X->seq($key, $value, R_NEXT) ;
1282 The code above can be rearranged to get around the problem, like this:
1284 $X = tie %x, 'DB_File', $filename, O_RDWR|O_CREAT, 0777, $DB_BTREE
1285 or die "Cannot tie $filename: $!" ;
1287 # this line will modify the cursor
1288 $count = scalar keys %x ;
1290 # Get the first key/value pair and set the cursor
1291 $X->seq($key, $value, R_FIRST) ;
1293 # Get the second key/value pair.
1295 $X->seq($key, $value, R_NEXT) ;
1299 All the constants defined in L<dbopen> for use in the flags parameters
1300 in the methods defined below are also available. Refer to the Berkeley
1301 DB documentation for the precise meaning of the flags values.
1303 Below is a list of the methods available.
1307 =item B<$status = $X-E<gt>get($key, $value [, $flags]) ;>
1309 Given a key (C<$key>) this method reads the value associated with it
1310 from the database. The value read from the database is returned in the
1311 C<$value> parameter.
1313 If the key does not exist the method returns 1.
1315 No flags are currently defined for this method.
1317 =item B<$status = $X-E<gt>put($key, $value [, $flags]) ;>
1319 Stores the key/value pair in the database.
1321 If you use either the R_IAFTER or R_IBEFORE flags, the C<$key> parameter
1322 will have the record number of the inserted key/value pair set.
1324 Valid flags are R_CURSOR, R_IAFTER, R_IBEFORE, R_NOOVERWRITE and
1327 =item B<$status = $X-E<gt>del($key [, $flags]) ;>
1329 Removes all key/value pairs with key C<$key> from the database.
1331 A return code of 1 means that the requested key was not in the
1334 R_CURSOR is the only valid flag at present.
1336 =item B<$status = $X-E<gt>fd ;>
1338 Returns the file descriptor for the underlying database.
1340 See L<Locking Databases> for an example of how to make use of the
1341 C<fd> method to lock your database.
1343 =item B<$status = $X-E<gt>seq($key, $value, $flags) ;>
1345 This interface allows sequential retrieval from the database. See
1346 L<dbopen> for full details.
1348 Both the C<$key> and C<$value> parameters will be set to the key/value
1349 pair read from the database.
1351 The flags parameter is mandatory. The valid flag values are R_CURSOR,
1352 R_FIRST, R_LAST, R_NEXT and R_PREV.
1354 =item B<$status = $X-E<gt>sync([$flags]) ;>
1356 Flushes any cached buffers to disk.
1358 R_RECNOSYNC is the only valid flag at present.
1362 =head1 HINTS AND TIPS
1365 =head2 Locking Databases
1367 Concurrent access of a read-write database by several parties requires
1368 them all to use some kind of locking. Here's an example of Tom's that
1369 uses the I<fd> method to get the file descriptor, and then a careful
1370 open() to give something Perl will flock() for you. Run this repeatedly
1371 in the background to watch the locks granted in proper order.
1382 my($oldval, $fd, $db, %db, $value, $key);
1384 $key = shift || 'default';
1385 $value = shift || 'magic';
1389 $db = tie(%db, 'DB_File', '/tmp/foo.db', O_CREAT|O_RDWR, 0644)
1390 || die "dbcreat /tmp/foo.db $!";
1392 print "$$: db fd is $fd\n";
1393 open(DB_FH, "+<&=$fd") || die "dup $!";
1396 unless (flock (DB_FH, LOCK_SH | LOCK_NB)) {
1397 print "$$: CONTENTION; can't read during write update!
1398 Waiting for read lock ($!) ....";
1399 unless (flock (DB_FH, LOCK_SH)) { die "flock: $!" }
1401 print "$$: Read lock granted\n";
1403 $oldval = $db{$key};
1404 print "$$: Old value was $oldval\n";
1405 flock(DB_FH, LOCK_UN);
1407 unless (flock (DB_FH, LOCK_EX | LOCK_NB)) {
1408 print "$$: CONTENTION; must have exclusive lock!
1409 Waiting for write lock ($!) ....";
1410 unless (flock (DB_FH, LOCK_EX)) { die "flock: $!" }
1413 print "$$: Write lock granted\n";
1415 $db->sync; # to flush
1418 flock(DB_FH, LOCK_UN);
1422 print "$$: Updated db to $key=$value\n";
1424 =head2 Sharing Databases With C Applications
1426 There is no technical reason why a Berkeley DB database cannot be
1427 shared by both a Perl and a C application.
1429 The vast majority of problems that are reported in this area boil down
1430 to the fact that C strings are NULL terminated, whilst Perl strings are
1433 Here is a real example. Netscape 2.0 keeps a record of the locations you
1434 visit along with the time you last visited them in a DB_HASH database.
1435 This is usually stored in the file F<~/.netscape/history.db>. The key
1436 field in the database is the location string and the value field is the
1437 time the location was last visited stored as a 4 byte binary value.
1439 If you haven't already guessed, the location string is stored with a
1440 terminating NULL. This means you need to be careful when accessing the
1443 Here is a snippet of code that is loosely based on Tom Christiansen's
1444 I<ggh> script (available from your nearest CPAN archive in
1445 F<authors/id/TOMC/scripts/nshist.gz>).
1451 use vars qw( $dotdir $HISTORY %hist_db $href $binary_time $date ) ;
1452 $dotdir = $ENV{HOME} || $ENV{LOGNAME};
1454 $HISTORY = "$dotdir/.netscape/history.db";
1456 tie %hist_db, 'DB_File', $HISTORY
1457 or die "Cannot open $HISTORY: $!\n" ;;
1459 # Dump the complete database
1460 while ( ($href, $binary_time) = each %hist_db ) {
1462 # remove the terminating NULL
1463 $href =~ s/\x00$// ;
1465 # convert the binary time into a user friendly string
1466 $date = localtime unpack("V", $binary_time);
1467 print "$date $href\n" ;
1470 # check for the existence of a specific key
1471 # remember to add the NULL
1472 if ( $binary_time = $hist_db{"http://mox.perl.com/\x00"} ) {
1473 $date = localtime unpack("V", $binary_time) ;
1474 print "Last visited mox.perl.com on $date\n" ;
1477 print "Never visited mox.perl.com\n"
1482 =head2 The untie() Gotcha
1484 If you make use of the Berkeley DB API, it is I<very> strongly
1485 recommended that you read L<perltie/The untie Gotcha>.
1487 Even if you don't currently make use of the API interface, it is still
1490 Here is an example which illustrates the problem from a B<DB_File>
1499 $X = tie %x, 'DB_File', 'tst.fil' , O_RDWR|O_TRUNC
1500 or die "Cannot tie first time: $!" ;
1506 tie %x, 'DB_File', 'tst.fil' , O_RDWR|O_CREAT
1507 or die "Cannot tie second time: $!" ;
1511 When run, the script will produce this error message:
1513 Cannot tie second time: Invalid argument at bad.file line 14.
1515 Although the error message above refers to the second tie() statement
1516 in the script, the source of the problem is really with the untie()
1517 statement that precedes it.
1519 Having read L<perltie> you will probably have already guessed that the
1520 error is caused by the extra copy of the tied object stored in C<$X>.
1521 If you haven't, then the problem boils down to the fact that the
1522 B<DB_File> destructor, DESTROY, will not be called until I<all>
1523 references to the tied object are destroyed. Both the tied variable,
1524 C<%x>, and C<$X> above hold a reference to the object. The call to
1525 untie() will destroy the first, but C<$X> still holds a valid
1526 reference, so the destructor will not get called and the database file
1527 F<tst.fil> will remain open. The fact that Berkeley DB then reports the
1528 attempt to open a database that is alreday open via the catch-all
1529 "Invalid argument" doesn't help.
1531 If you run the script with the C<-w> flag the error message becomes:
1533 untie attempted while 1 inner references still exist at bad.file line 12.
1534 Cannot tie second time: Invalid argument at bad.file line 14.
1536 which pinpoints the real problem. Finally the script can now be
1537 modified to fix the original problem by destroying the API object
1546 $X = tie %x, 'DB_File', 'tst.fil' , O_RDWR|O_CREAT
1550 =head1 COMMON QUESTIONS
1552 =head2 Why is there Perl source in my database?
1554 If you look at the contents of a database file created by DB_File,
1555 there can sometimes be part of a Perl script included in it.
1557 This happens because Berkeley DB uses dynamic memory to allocate
1558 buffers which will subsequently be written to the database file. Being
1559 dynamic, the memory could have been used for anything before DB
1560 malloced it. As Berkeley DB doesn't clear the memory once it has been
1561 allocated, the unused portions will contain random junk. In the case
1562 where a Perl script gets written to the database, the random junk will
1563 correspond to an area of dynamic memory that happened to be used during
1564 the compilation of the script.
1566 Unless you don't like the possibility of there being part of your Perl
1567 scripts embedded in a database file, this is nothing to worry about.
1569 =head2 How do I store complex data structures with DB_File?
1571 Although B<DB_File> cannot do this directly, there is a module which
1572 can layer transparently over B<DB_File> to accomplish this feat.
1574 Check out the MLDBM module, available on CPAN in the directory
1575 F<modules/by-module/MLDBM>.
1577 =head2 What does "Invalid Argument" mean?
1579 You will get this error message when one of the parameters in the
1580 C<tie> call is wrong. Unfortunately there are quite a few parameters to
1581 get wrong, so it can be difficult to figure out which one it is.
1583 Here are a couple of possibilities:
1589 Attempting to reopen a database without closing it.
1593 Using the O_WRONLY flag.
1597 =head2 What does "Bareword 'DB_File' not allowed" mean?
1599 You will encounter this particular error message when you have the
1600 C<strict 'subs'> pragma (or the full strict pragma) in your script.
1601 Consider this script:
1606 tie %x, DB_File, "filename" ;
1608 Running it produces the error in question:
1610 Bareword "DB_File" not allowed while "strict subs" in use
1612 To get around the error, place the word C<DB_File> in either single or
1613 double quotes, like this:
1615 tie %x, "DB_File", "filename" ;
1617 Although it might seem like a real pain, it is really worth the effort
1618 of having a C<use strict> in all your scripts.
1622 Moved to the Changes file.
1626 Some older versions of Berkeley DB had problems with fixed length
1627 records using the RECNO file format. This problem has been fixed since
1628 version 1.85 of Berkeley DB.
1630 I am sure there are bugs in the code. If you do find any, or can
1631 suggest any enhancements, I would welcome your comments.
1635 B<DB_File> comes with the standard Perl source distribution. Look in
1636 the directory F<ext/DB_File>. Given the amount of time between releases
1637 of Perl the version that ships with Perl is quite likely to be out of
1638 date, so the most recent version can always be found on CPAN (see
1639 L<perlmod/CPAN> for details), in the directory
1640 F<modules/by-module/DB_File>.
1642 This version of B<DB_File> will work with either version 1.x or 2.x of
1643 Berkeley DB, but is limited to the functionality provided by version 1.
1645 The official web site for Berkeley DB is
1646 F<http://www.sleepycat.com/db>. The ftp equivalent is
1647 F<ftp.sleepycat.com:/pub>. Both versions 1 and 2 of Berkeley DB are
1650 Alternatively, Berkeley DB version 1 is available at your nearest CPAN
1651 archive in F<src/misc/db.1.85.tar.gz>.
1653 If you are running IRIX, then get Berkeley DB version 1 from
1654 F<http://reality.sgi.com/ariel>. It has the patches necessary to
1655 compile properly on IRIX 5.3.
1659 Copyright (c) 1997 Paul Marquess. All rights reserved. This program is
1660 free software; you can redistribute it and/or modify it under the same
1661 terms as Perl itself.
1663 Although B<DB_File> is covered by the Perl license, the library it
1664 makes use of, namely Berkeley DB, is not. Berkeley DB has its own
1665 copyright and its own license. Please take the time to read it.
1667 The license for Berkeley DB version 2, and how it relates to DB_File
1668 does need some extra clarification. Here are are few words taken from
1669 the Berkeley DB FAQ regarding the version 2 license:
1671 The major difference is that the license for DB 2.0, when
1672 downloaded from the net, requires that the software that
1673 uses DB 2.0 be freely redistributable.
1675 That means that if you want to use DB_File, and you have changed either
1676 the source for Berkeley DB or Perl, then the changes must be freely
1679 In the case of Perl, the term source refers to the complete source
1680 code for Perl (e.g. sv.c, toke.c, perl.h) and any external modules that
1681 you are using (e.g. DB_File, Tk).
1683 Note that any Perl scripts that you write are your property - this
1684 includes scripts that make use of DB_File. Neither the Perl license or
1685 the Berkeley DB license place any restriction on what you have to do
1688 If you are in any doubt about the license situation, contact either the
1689 Berkeley DB authors or the author of DB_File. See L<"AUTHOR"> for details.
1694 L<perl(1)>, L<dbopen(3)>, L<hash(3)>, L<recno(3)>, L<btree(3)>
1698 The DB_File interface was written by Paul Marquess
1699 E<lt>pmarquess@bfsec.bt.co.ukE<gt>.
1700 Questions about the DB system itself may be addressed to
1701 E<lt>db@sleepycat.com<gt>.