1 # DB_File.pm -- Perl 5 interface to Berkeley DB
3 # written by Paul Marquess (pmarquess@bfsec.bt.co.uk)
4 # last modified 12th Mar 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 ${method}" ;
101 sub DESTROY { undef %{$_[0]} }
102 sub FIRSTKEY { my $self = shift ; $self->NotHere("FIRSTKEY") }
103 sub NEXTKEY { my $self = shift ; $self->NotHere("NEXTKEY") }
104 sub CLEAR { my $self = shift ; $self->NotHere("CLEAR") }
106 package DB_File::RECNOINFO ;
110 @DB_File::RECNOINFO::ISA = qw(DB_File::HASHINFO) ;
116 bless { VALID => { map {$_, 1}
117 qw( bval cachesize psize flags lorder reclen bfname )
123 package DB_File::BTREEINFO ;
127 @DB_File::BTREEINFO::ISA = qw(DB_File::HASHINFO) ;
133 bless { VALID => { map {$_, 1}
134 qw( flags cachesize maxkeypage minkeypage psize
135 compare prefix lorder )
145 use vars qw($VERSION @ISA @EXPORT $AUTOLOAD $DB_BTREE $DB_HASH $DB_RECNO) ;
151 #typedef enum { DB_BTREE, DB_HASH, DB_RECNO } DBTYPE;
152 $DB_BTREE = new DB_File::BTREEINFO ;
153 $DB_HASH = new DB_File::HASHINFO ;
154 $DB_RECNO = new DB_File::RECNOINFO ;
160 @ISA = qw(Tie::Hash Exporter DynaLoader);
162 $DB_BTREE $DB_HASH $DB_RECNO
197 ($constname = $AUTOLOAD) =~ s/.*:://;
198 my $val = constant($constname, @_ ? $_[0] : 0);
200 if ($! =~ /Invalid/) {
201 $AutoLoader::AUTOLOAD = $AUTOLOAD;
202 goto &AutoLoader::AUTOLOAD;
205 my($pack,$file,$line) = caller;
206 croak "Your vendor has not defined DB macro $constname, used at $file line $line.
210 eval "sub $AUTOLOAD { $val }";
215 # import borrowed from IO::File
216 # exports Fcntl constants if available.
219 my $callpkg = caller;
220 Exporter::export $pkg, $callpkg, @_;
223 Exporter::export 'Fcntl', $callpkg, '/^O_/';
227 bootstrap DB_File $VERSION;
229 # Preloaded methods go here. Autoload methods go after __END__, and are
230 # processed by the autosplit program.
236 $arg[4] = tied %{ $arg[4] }
237 if @arg >= 5 && ref $arg[4] && $arg[4] =~ /=HASH/ && tied %{ $arg[4] } ;
242 *TIEARRAY = \&TIEHASH ;
246 croak "Usage: \$db->get_dup(key [,flag])\n"
247 unless @_ == 2 or @_ == 3 ;
254 my $wantarray = wantarray ;
260 # iterate through the database until either EOF ($status == 0)
261 # or a different key is encountered ($key ne $origkey).
262 for ($status = $db->seq($key, $value, R_CURSOR()) ;
263 $status == 0 and $key eq $origkey ;
264 $status = $db->seq($key, $value, R_NEXT()) ) {
266 # save the value or count number of matches
269 { ++ $values{$value} }
271 { push (@values, $value) }
278 return ($wantarray ? ($flag ? %values : @values) : $counter) ;
287 DB_File - Perl5 access to Berkeley DB
293 [$X =] tie %hash, 'DB_File', [$filename, $flags, $mode, $DB_HASH] ;
294 [$X =] tie %hash, 'DB_File', $filename, $flags, $mode, $DB_BTREE ;
295 [$X =] tie @array, 'DB_File', $filename, $flags, $mode, $DB_RECNO ;
297 $status = $X->del($key [, $flags]) ;
298 $status = $X->put($key, $value [, $flags]) ;
299 $status = $X->get($key, $value [, $flags]) ;
300 $status = $X->seq($key, $value, $flags) ;
301 $status = $X->sync([$flags]) ;
305 $count = $X->get_dup($key) ;
306 @list = $X->get_dup($key) ;
307 %list = $X->get_dup($key, 1) ;
321 B<DB_File> is a module which allows Perl programs to make use of the
322 facilities provided by Berkeley DB. If you intend to use this
323 module you should really have a copy of the Berkeley DB manual pages at
324 hand. The interface defined here mirrors the Berkeley DB interface
327 Please note that this module will only work with version 1.x of
328 Berkeley DB. Once Berkeley DB version 2 is released, B<DB_File> will be
329 upgraded to work with it.
331 Berkeley DB is a C library which provides a consistent interface to a
332 number of database formats. B<DB_File> provides an interface to all
333 three of the database types currently supported by Berkeley DB.
341 This database type allows arbitrary key/value pairs to be stored in data
342 files. This is equivalent to the functionality provided by other
343 hashing packages like DBM, NDBM, ODBM, GDBM, and SDBM. Remember though,
344 the files created using DB_HASH are not compatible with any of the
345 other packages mentioned.
347 A default hashing algorithm, which will be adequate for most
348 applications, is built into Berkeley DB. If you do need to use your own
349 hashing algorithm it is possible to write your own in Perl and have
350 B<DB_File> use it instead.
354 The btree format allows arbitrary key/value pairs to be stored in a
355 sorted, balanced binary tree.
357 As with the DB_HASH format, it is possible to provide a user defined
358 Perl routine to perform the comparison of keys. By default, though, the
359 keys are stored in lexical order.
363 DB_RECNO allows both fixed-length and variable-length flat text files
364 to be manipulated using the same key/value pair interface as in DB_HASH
365 and DB_BTREE. In this case the key will consist of a record (line)
370 =head2 Interface to Berkeley DB
372 B<DB_File> allows access to Berkeley DB files using the tie() mechanism
373 in Perl 5 (for full details, see L<perlfunc/tie()>). This facility
374 allows B<DB_File> to access Berkeley DB files using either an
375 associative array (for DB_HASH & DB_BTREE file types) or an ordinary
376 array (for the DB_RECNO file type).
378 In addition to the tie() interface, it is also possible to access most
379 of the functions provided in the Berkeley DB API directly.
380 See L<THE API INTERFACE>.
382 =head2 Opening a Berkeley DB Database File
384 Berkeley DB uses the function dbopen() to open or create a database.
385 Here is the C prototype for dbopen():
388 dbopen (const char * file, int flags, int mode,
389 DBTYPE type, const void * openinfo)
391 The parameter C<type> is an enumeration which specifies which of the 3
392 interface methods (DB_HASH, DB_BTREE or DB_RECNO) is to be used.
393 Depending on which of these is actually chosen, the final parameter,
394 I<openinfo> points to a data structure which allows tailoring of the
395 specific interface method.
397 This interface is handled slightly differently in B<DB_File>. Here is
398 an equivalent call using B<DB_File>:
400 tie %array, 'DB_File', $filename, $flags, $mode, $DB_HASH ;
402 The C<filename>, C<flags> and C<mode> parameters are the direct
403 equivalent of their dbopen() counterparts. The final parameter $DB_HASH
404 performs the function of both the C<type> and C<openinfo> parameters in
407 In the example above $DB_HASH is actually a pre-defined reference to a
408 hash object. B<DB_File> has three of these pre-defined references.
409 Apart from $DB_HASH, there is also $DB_BTREE and $DB_RECNO.
411 The keys allowed in each of these pre-defined references is limited to
412 the names used in the equivalent C structure. So, for example, the
413 $DB_HASH reference will only allow keys called C<bsize>, C<cachesize>,
414 C<ffactor>, C<hash>, C<lorder> and C<nelem>.
416 To change one of these elements, just assign to it like this:
418 $DB_HASH->{'cachesize'} = 10000 ;
420 The three predefined variables $DB_HASH, $DB_BTREE and $DB_RECNO are
421 usually adequate for most applications. If you do need to create extra
422 instances of these objects, constructors are available for each file
425 Here are examples of the constructors and the valid options available
426 for DB_HASH, DB_BTREE and DB_RECNO respectively.
428 $a = new DB_File::HASHINFO ;
436 $b = new DB_File::BTREEINFO ;
446 $c = new DB_File::RECNOINFO ;
455 The values stored in the hashes above are mostly the direct equivalent
456 of their C counterpart. Like their C counterparts, all are set to a
457 default values - that means you don't have to set I<all> of the
458 values when you only want to change one. Here is an example:
460 $a = new DB_File::HASHINFO ;
461 $a->{'cachesize'} = 12345 ;
462 tie %y, 'DB_File', "filename", $flags, 0777, $a ;
464 A few of the options need extra discussion here. When used, the C
465 equivalent of the keys C<hash>, C<compare> and C<prefix> store pointers
466 to C functions. In B<DB_File> these keys are used to store references
467 to Perl subs. Below are templates for each of the subs:
473 # return the hash value for $data
479 my ($key, $key2) = @_ ;
481 # return 0 if $key1 eq $key2
482 # -1 if $key1 lt $key2
483 # 1 if $key1 gt $key2
484 return (-1 , 0 or 1) ;
489 my ($key, $key2) = @_ ;
491 # return number of bytes of $key2 which are
492 # necessary to determine that it is greater than $key1
496 See L<Changing the BTREE sort order> for an example of using the
499 If you are using the DB_RECNO interface and you intend making use of
500 C<bval>, you should check out L<The 'bval' Option>.
502 =head2 Default Parameters
504 It is possible to omit some or all of the final 4 parameters in the
505 call to C<tie> and let them take default values. As DB_HASH is the most
506 common file format used, the call:
508 tie %A, "DB_File", "filename" ;
512 tie %A, "DB_File", "filename", O_CREAT|O_RDWR, 0666, $DB_HASH ;
514 It is also possible to omit the filename parameter as well, so the
521 tie %A, "DB_File", undef, O_CREAT|O_RDWR, 0666, $DB_HASH ;
523 See L<In Memory Databases> for a discussion on the use of C<undef>
524 in place of a filename.
526 =head2 In Memory Databases
528 Berkeley DB allows the creation of in-memory databases by using NULL
529 (that is, a C<(char *)0> in C) in place of the filename. B<DB_File>
530 uses C<undef> instead of NULL to provide this functionality.
534 The DB_HASH file format is probably the most commonly used of the three
535 file formats that B<DB_File> supports. It is also very straightforward
538 =head2 A Simple Example
540 This example shows how to create a database, add key/value pairs to the
541 database, delete keys/value pairs and finally how to enumerate the
542 contents of the database.
546 use vars qw( %h $k $v ) ;
548 tie %h, "DB_File", "fruit", O_RDWR|O_CREAT, 0640, $DB_HASH
549 or die "Cannot open file 'fruit': $!\n";
551 # Add a few key/value pairs to the file
552 $h{"apple"} = "red" ;
553 $h{"orange"} = "orange" ;
554 $h{"banana"} = "yellow" ;
555 $h{"tomato"} = "red" ;
557 # Check for existence of a key
558 print "Banana Exists\n\n" if $h{"banana"} ;
560 # Delete a key/value pair.
563 # print the contents of the file
564 while (($k, $v) = each %h)
565 { print "$k -> $v\n" }
577 Note that the like ordinary associative arrays, the order of the keys
578 retrieved is in an apparently random order.
582 The DB_BTREE format is useful when you want to store data in a given
583 order. By default the keys will be stored in lexical order, but as you
584 will see from the example shown in the next section, it is very easy to
585 define your own sorting function.
587 =head2 Changing the BTREE sort order
589 This script shows how to override the default sorting algorithm that
590 BTREE uses. Instead of using the normal lexical ordering, a case
591 insensitive compare function will be used.
600 my ($key1, $key2) = @_ ;
601 "\L$key1" cmp "\L$key2" ;
604 # specify the Perl sub that will do the comparison
605 $DB_BTREE->{'compare'} = \&Compare ;
607 tie %h, "DB_File", "tree", O_RDWR|O_CREAT, 0640, $DB_BTREE
608 or die "Cannot open file 'tree': $!\n" ;
610 # Add a key/value pair to the file
611 $h{'Wall'} = 'Larry' ;
612 $h{'Smith'} = 'John' ;
613 $h{'mouse'} = 'mickey' ;
614 $h{'duck'} = 'donald' ;
619 # Cycle through the keys printing them in order.
620 # Note it is not necessary to sort the keys as
621 # the btree will have kept them in order automatically.
627 Here is the output from the code above.
633 There are a few point to bear in mind if you want to change the
634 ordering in a BTREE database:
640 The new compare function must be specified when you create the database.
644 You cannot change the ordering once the database has been created. Thus
645 you must use the same compare function every time you access the
650 =head2 Handling Duplicate Keys
652 The BTREE file type optionally allows a single key to be associated
653 with an arbitrary number of values. This option is enabled by setting
654 the flags element of C<$DB_BTREE> to R_DUP when creating the database.
656 There are some difficulties in using the tied hash interface if you
657 want to manipulate a BTREE database with duplicate keys. Consider this
663 use vars qw($filename %h ) ;
668 # Enable duplicate records
669 $DB_BTREE->{'flags'} = R_DUP ;
671 tie %h, "DB_File", $filename, O_RDWR|O_CREAT, 0640, $DB_BTREE
672 or die "Cannot open $filename: $!\n";
674 # Add some key/value pairs to the file
675 $h{'Wall'} = 'Larry' ;
676 $h{'Wall'} = 'Brick' ; # Note the duplicate key
677 $h{'Wall'} = 'Brick' ; # Note the duplicate key and value
678 $h{'Smith'} = 'John' ;
679 $h{'mouse'} = 'mickey' ;
681 # iterate through the associative array
682 # and print each key/value pair.
684 { print "$_ -> $h{$_}\n" }
696 As you can see 3 records have been successfully created with key C<Wall>
697 - the only thing is, when they are retrieved from the database they
698 I<seem> to have the same value, namely C<Larry>. The problem is caused
699 by the way that the associative array interface works. Basically, when
700 the associative array interface is used to fetch the value associated
701 with a given key, it will only ever retrieve the first value.
703 Although it may not be immediately obvious from the code above, the
704 associative array interface can be used to write values with duplicate
705 keys, but it cannot be used to read them back from the database.
707 The way to get around this problem is to use the Berkeley DB API method
708 called C<seq>. This method allows sequential access to key/value
709 pairs. See L<THE API INTERFACE> for details of both the C<seq> method
710 and the API in general.
712 Here is the script above rewritten using the C<seq> API method.
717 use vars qw($filename $x %h $status $key $value) ;
722 # Enable duplicate records
723 $DB_BTREE->{'flags'} = R_DUP ;
725 $x = tie %h, "DB_File", $filename, O_RDWR|O_CREAT, 0640, $DB_BTREE
726 or die "Cannot open $filename: $!\n";
728 # Add some key/value pairs to the file
729 $h{'Wall'} = 'Larry' ;
730 $h{'Wall'} = 'Brick' ; # Note the duplicate key
731 $h{'Wall'} = 'Brick' ; # Note the duplicate key and value
732 $h{'Smith'} = 'John' ;
733 $h{'mouse'} = 'mickey' ;
735 # iterate through the btree using seq
736 # and print each key/value pair.
738 for ($status = $x->seq($key, $value, R_FIRST) ;
740 $status = $x->seq($key, $value, R_NEXT) )
741 { print "$key -> $value\n" }
754 This time we have got all the key/value pairs, including the multiple
755 values associated with the key C<Wall>.
757 =head2 The get_dup() Method
759 B<DB_File> comes with a utility method, called C<get_dup>, to assist in
760 reading duplicate values from BTREE databases. The method can take the
763 $count = $x->get_dup($key) ;
764 @list = $x->get_dup($key) ;
765 %list = $x->get_dup($key, 1) ;
767 In a scalar context the method returns the number of values associated
768 with the key, C<$key>.
770 In list context, it returns all the values which match C<$key>. Note
771 that the values will be returned in an apparently random order.
773 In list context, if the second parameter is present and evaluates
774 TRUE, the method returns an associative array. The keys of the
775 associative array correspond to the values that matched in the BTREE
776 and the values of the array are a count of the number of times that
777 particular value occurred in the BTREE.
779 So assuming the database created above, we can use C<get_dup> like
782 my $cnt = $x->get_dup("Wall") ;
783 print "Wall occurred $cnt times\n" ;
785 my %hash = $x->get_dup("Wall", 1) ;
786 print "Larry is there\n" if $hash{'Larry'} ;
787 print "There are $hash{'Brick'} Brick Walls\n" ;
789 my @list = $x->get_dup("Wall") ;
790 print "Wall => [@list]\n" ;
792 @list = $x->get_dup("Smith") ;
793 print "Smith => [@list]\n" ;
795 @list = $x->get_dup("Dog") ;
796 print "Dog => [@list]\n" ;
801 Wall occurred 3 times
803 There are 2 Brick Walls
804 Wall => [Brick Brick Larry]
808 =head2 Matching Partial Keys
810 The BTREE interface has a feature which allows partial keys to be
811 matched. This functionality is I<only> available when the C<seq> method
812 is used along with the R_CURSOR flag.
814 $x->seq($key, $value, R_CURSOR) ;
816 Here is the relevant quote from the dbopen man page where it defines
817 the use of the R_CURSOR flag with seq:
819 Note, for the DB_BTREE access method, the returned key is not
820 necessarily an exact match for the specified key. The returned key
821 is the smallest key greater than or equal to the specified key,
822 permitting partial key matches and range searches.
824 In the example script below, the C<match> sub uses this feature to find
825 and print the first matching key/value pair given a partial key.
831 use vars qw($filename $x %h $st $key $value) ;
837 my $orig_key = $key ;
838 $x->seq($key, $value, R_CURSOR) ;
839 print "$orig_key\t-> $key\t-> $value\n" ;
845 $x = tie %h, "DB_File", $filename, O_RDWR|O_CREAT, 0640, $DB_BTREE
846 or die "Cannot open $filename: $!\n";
848 # Add some key/value pairs to the file
849 $h{'mouse'} = 'mickey' ;
850 $h{'Wall'} = 'Larry' ;
851 $h{'Walls'} = 'Brick' ;
852 $h{'Smith'} = 'John' ;
857 for ($st = $x->seq($key, $value, R_FIRST) ;
859 $st = $x->seq($key, $value, R_NEXT) )
861 { print "$key -> $value\n" }
863 print "\nPARTIAL MATCH\n" ;
887 DB_RECNO provides an interface to flat text files. Both variable and
888 fixed length records are supported.
890 In order to make RECNO more compatible with Perl the array offset for
891 all RECNO arrays begins at 0 rather than 1 as in Berkeley DB.
893 As with normal Perl arrays, a RECNO array can be accessed using
894 negative indexes. The index -1 refers to the last element of the array,
895 -2 the second last, and so on. Attempting to access an element before
896 the start of the array will raise a fatal run-time error.
898 =head2 The 'bval' Option
900 The operation of the bval option warrants some discussion. Here is the
901 definition of bval from the Berkeley DB 1.85 recno manual page:
903 The delimiting byte to be used to mark the end of a
904 record for variable-length records, and the pad charac-
905 ter for fixed-length records. If no value is speci-
906 fied, newlines (``\n'') are used to mark the end of
907 variable-length records and fixed-length records are
910 The second sentence is wrong. In actual fact bval will only default to
911 C<"\n"> when the openinfo parameter in dbopen is NULL. If a non-NULL
912 openinfo parameter is used at all, the value that happens to be in bval
913 will be used. That means you always have to specify bval when making
914 use of any of the options in the openinfo parameter. This documentation
915 error will be fixed in the next release of Berkeley DB.
917 That clarifies the situation with regards Berkeley DB itself. What
918 about B<DB_File>? Well, the behavior defined in the quote above is
919 quite useful, so B<DB_File> conforms it.
921 That means that you can specify other options (e.g. cachesize) and
922 still have bval default to C<"\n"> for variable length records, and
923 space for fixed length records.
925 =head2 A Simple Example
927 Here is a simple example that uses RECNO.
933 tie @h, "DB_File", "text", O_RDWR|O_CREAT, 0640, $DB_RECNO
934 or die "Cannot open file 'text': $!\n" ;
936 # Add a few key/value pairs to the file
941 # Check for existence of a key
942 print "Element 1 Exists with value $h[1]\n" if $h[1] ;
944 # use a negative index
945 print "The last element is $h[-1]\n" ;
946 print "The 2nd last element is $h[-2]\n" ;
950 Here is the output from the script:
953 Element 1 Exists with value blue
954 The last element is yellow
955 The 2nd last element is blue
959 As you can see from the example above, the tied array interface is
960 quite limited. To make the interface more useful, a number of methods
961 are supplied with B<DB_File> to simulate the standard array operations
962 that are not currently implemented in Perl's tied array interface. All
963 these methods are accessed via the object returned from the tie call.
965 Here are the methods:
969 =item B<$X-E<gt>push(list) ;>
971 Pushes the elements of C<list> to the end of the array.
973 =item B<$value = $X-E<gt>pop ;>
975 Removes and returns the last element of the array.
977 =item B<$X-E<gt>shift>
979 Removes and returns the first element of the array.
981 =item B<$X-E<gt>unshift(list) ;>
983 Pushes the elements of C<list> to the start of the array.
985 =item B<$X-E<gt>length>
987 Returns the number of elements in the array.
991 =head2 Another Example
993 Here is a more complete example that makes use of some of the methods
994 described above. It also makes use of the API interface directly (see
995 L<THE API INTERFACE>).
998 use vars qw(@h $H $file $i) ;
1006 $H = tie @h, "DB_File", $file, O_RDWR|O_CREAT, 0640, $DB_RECNO
1007 or die "Cannot open file $file: $!\n" ;
1009 # first create a text file to play with
1017 # Print the records in order.
1019 # The length method is needed here because evaluating a tied
1020 # array in a scalar context does not return the number of
1021 # elements in the array.
1023 print "\nORIGINAL\n" ;
1024 foreach $i (0 .. $H->length - 1) {
1025 print "$i: $h[$i]\n" ;
1028 # use the push & pop methods
1031 print "\nThe last record was [$a]\n" ;
1033 # and the shift & unshift methods
1035 $H->unshift("first") ;
1036 print "The first record was [$a]\n" ;
1038 # Use the API to add a new record after record 2.
1040 $H->put($i, "Newbie", R_IAFTER) ;
1042 # and a new record before record 1.
1044 $H->put($i, "New One", R_IBEFORE) ;
1049 # now print the records in reverse order
1050 print "\nREVERSE\n" ;
1051 for ($i = $H->length - 1 ; $i >= 0 ; -- $i)
1052 { print "$i: $h[$i]\n" }
1054 # same again, but use the API functions instead
1055 print "\nREVERSE again\n" ;
1056 my ($s, $k, $v) = (0, 0, 0) ;
1057 for ($s = $H->seq($k, $v, R_LAST) ;
1059 $s = $H->seq($k, $v, R_PREV))
1060 { print "$k: $v\n" }
1065 and this is what it outputs:
1074 The last record was [four]
1075 The first record was [zero]
1099 Rather than iterating through the array, C<@h> like this:
1103 it is necessary to use either this:
1105 foreach $i (0 .. $H->length - 1)
1109 for ($a = $H->get($k, $v, R_FIRST) ;
1111 $a = $H->get($k, $v, R_NEXT) )
1115 Notice that both times the C<put> method was used the record index was
1116 specified using a variable, C<$i>, rather than the literal value
1117 itself. This is because C<put> will return the record number of the
1118 inserted line via that parameter.
1122 =head1 THE API INTERFACE
1124 As well as accessing Berkeley DB using a tied hash or array, it is also
1125 possible to make direct use of most of the API functions defined in the
1126 Berkeley DB documentation.
1128 To do this you need to store a copy of the object returned from the tie.
1130 $db = tie %hash, "DB_File", "filename" ;
1132 Once you have done that, you can access the Berkeley DB API functions
1133 as B<DB_File> methods directly like this:
1135 $db->put($key, $value, R_NOOVERWRITE) ;
1137 B<Important:> If you have saved a copy of the object returned from
1138 C<tie>, the underlying database file will I<not> be closed until both
1139 the tied variable is untied and all copies of the saved object are
1143 $db = tie %hash, "DB_File", "filename"
1144 or die "Cannot tie filename: $!" ;
1149 See L<The untie() Gotcha> for more details.
1151 All the functions defined in L<dbopen> are available except for
1152 close() and dbopen() itself. The B<DB_File> method interface to the
1153 supported functions have been implemented to mirror the way Berkeley DB
1154 works whenever possible. In particular note that:
1160 The methods return a status value. All return 0 on success.
1161 All return -1 to signify an error and set C<$!> to the exact
1162 error code. The return code 1 generally (but not always) means that the
1163 key specified did not exist in the database.
1165 Other return codes are defined. See below and in the Berkeley DB
1166 documentation for details. The Berkeley DB documentation should be used
1167 as the definitive source.
1171 Whenever a Berkeley DB function returns data via one of its parameters,
1172 the equivalent B<DB_File> method does exactly the same.
1176 If you are careful, it is possible to mix API calls with the tied
1177 hash/array interface in the same piece of code. Although only a few of
1178 the methods used to implement the tied interface currently make use of
1179 the cursor, you should always assume that the cursor has been changed
1180 any time the tied hash/array interface is used. As an example, this
1181 code will probably not do what you expect:
1183 $X = tie %x, 'DB_File', $filename, O_RDWR|O_CREAT, 0777, $DB_BTREE
1184 or die "Cannot tie $filename: $!" ;
1186 # Get the first key/value pair and set the cursor
1187 $X->seq($key, $value, R_FIRST) ;
1189 # this line will modify the cursor
1190 $count = scalar keys %x ;
1192 # Get the second key/value pair.
1193 # oops, it didn't, it got the last key/value pair!
1194 $X->seq($key, $value, R_NEXT) ;
1196 The code above can be rearranged to get around the problem, like this:
1198 $X = tie %x, 'DB_File', $filename, O_RDWR|O_CREAT, 0777, $DB_BTREE
1199 or die "Cannot tie $filename: $!" ;
1201 # this line will modify the cursor
1202 $count = scalar keys %x ;
1204 # Get the first key/value pair and set the cursor
1205 $X->seq($key, $value, R_FIRST) ;
1207 # Get the second key/value pair.
1209 $X->seq($key, $value, R_NEXT) ;
1213 All the constants defined in L<dbopen> for use in the flags parameters
1214 in the methods defined below are also available. Refer to the Berkeley
1215 DB documentation for the precise meaning of the flags values.
1217 Below is a list of the methods available.
1221 =item B<$status = $X-E<gt>get($key, $value [, $flags]) ;>
1223 Given a key (C<$key>) this method reads the value associated with it
1224 from the database. The value read from the database is returned in the
1225 C<$value> parameter.
1227 If the key does not exist the method returns 1.
1229 No flags are currently defined for this method.
1231 =item B<$status = $X-E<gt>put($key, $value [, $flags]) ;>
1233 Stores the key/value pair in the database.
1235 If you use either the R_IAFTER or R_IBEFORE flags, the C<$key> parameter
1236 will have the record number of the inserted key/value pair set.
1238 Valid flags are R_CURSOR, R_IAFTER, R_IBEFORE, R_NOOVERWRITE and
1241 =item B<$status = $X-E<gt>del($key [, $flags]) ;>
1243 Removes all key/value pairs with key C<$key> from the database.
1245 A return code of 1 means that the requested key was not in the
1248 R_CURSOR is the only valid flag at present.
1250 =item B<$status = $X-E<gt>fd ;>
1252 Returns the file descriptor for the underlying database.
1254 See L<Locking Databases> for an example of how to make use of the
1255 C<fd> method to lock your database.
1257 =item B<$status = $X-E<gt>seq($key, $value, $flags) ;>
1259 This interface allows sequential retrieval from the database. See
1260 L<dbopen> for full details.
1262 Both the C<$key> and C<$value> parameters will be set to the key/value
1263 pair read from the database.
1265 The flags parameter is mandatory. The valid flag values are R_CURSOR,
1266 R_FIRST, R_LAST, R_NEXT and R_PREV.
1268 =item B<$status = $X-E<gt>sync([$flags]) ;>
1270 Flushes any cached buffers to disk.
1272 R_RECNOSYNC is the only valid flag at present.
1276 =head1 HINTS AND TIPS
1279 =head2 Locking Databases
1281 Concurrent access of a read-write database by several parties requires
1282 them all to use some kind of locking. Here's an example of Tom's that
1283 uses the I<fd> method to get the file descriptor, and then a careful
1284 open() to give something Perl will flock() for you. Run this repeatedly
1285 in the background to watch the locks granted in proper order.
1296 my($oldval, $fd, $db, %db, $value, $key);
1298 $key = shift || 'default';
1299 $value = shift || 'magic';
1303 $db = tie(%db, 'DB_File', '/tmp/foo.db', O_CREAT|O_RDWR, 0644)
1304 || die "dbcreat /tmp/foo.db $!";
1306 print "$$: db fd is $fd\n";
1307 open(DB_FH, "+<&=$fd") || die "dup $!";
1310 unless (flock (DB_FH, LOCK_SH | LOCK_NB)) {
1311 print "$$: CONTENTION; can't read during write update!
1312 Waiting for read lock ($!) ....";
1313 unless (flock (DB_FH, LOCK_SH)) { die "flock: $!" }
1315 print "$$: Read lock granted\n";
1317 $oldval = $db{$key};
1318 print "$$: Old value was $oldval\n";
1319 flock(DB_FH, LOCK_UN);
1321 unless (flock (DB_FH, LOCK_EX | LOCK_NB)) {
1322 print "$$: CONTENTION; must have exclusive lock!
1323 Waiting for write lock ($!) ....";
1324 unless (flock (DB_FH, LOCK_EX)) { die "flock: $!" }
1327 print "$$: Write lock granted\n";
1329 $db->sync; # to flush
1332 flock(DB_FH, LOCK_UN);
1336 print "$$: Updated db to $key=$value\n";
1338 =head2 Sharing Databases With C Applications
1340 There is no technical reason why a Berkeley DB database cannot be
1341 shared by both a Perl and a C application.
1343 The vast majority of problems that are reported in this area boil down
1344 to the fact that C strings are NULL terminated, whilst Perl strings are
1347 Here is a real example. Netscape 2.0 keeps a record of the locations you
1348 visit along with the time you last visited them in a DB_HASH database.
1349 This is usually stored in the file F<~/.netscape/history.db>. The key
1350 field in the database is the location string and the value field is the
1351 time the location was last visited stored as a 4 byte binary value.
1353 If you haven't already guessed, the location string is stored with a
1354 terminating NULL. This means you need to be careful when accessing the
1357 Here is a snippet of code that is loosely based on Tom Christiansen's
1358 I<ggh> script (available from your nearest CPAN archive in
1359 F<authors/id/TOMC/scripts/nshist.gz>).
1365 use vars qw( $dotdir $HISTORY %hist_db $href $binary_time $date ) ;
1366 $dotdir = $ENV{HOME} || $ENV{LOGNAME};
1368 $HISTORY = "$dotdir/.netscape/history.db";
1370 tie %hist_db, 'DB_File', $HISTORY
1371 or die "Cannot open $HISTORY: $!\n" ;;
1373 # Dump the complete database
1374 while ( ($href, $binary_time) = each %hist_db ) {
1376 # remove the terminating NULL
1377 $href =~ s/\x00$// ;
1379 # convert the binary time into a user friendly string
1380 $date = localtime unpack("V", $binary_time);
1381 print "$date $href\n" ;
1384 # check for the existence of a specific key
1385 # remember to add the NULL
1386 if ( $binary_time = $hist_db{"http://mox.perl.com/\x00"} ) {
1387 $date = localtime unpack("V", $binary_time) ;
1388 print "Last visited mox.perl.com on $date\n" ;
1391 print "Never visited mox.perl.com\n"
1396 =head2 The untie() Gotcha
1398 If you make use of the Berkeley DB API, it is I<very> strongly
1399 recommended that you read L<perltie/The untie Gotcha>.
1401 Even if you don't currently make use of the API interface, it is still
1404 Here is an example which illustrates the problem from a B<DB_File>
1413 $X = tie %x, 'DB_File', 'tst.fil' , O_RDWR|O_TRUNC
1414 or die "Cannot tie first time: $!" ;
1420 tie %x, 'DB_File', 'tst.fil' , O_RDWR|O_CREAT
1421 or die "Cannot tie second time: $!" ;
1425 When run, the script will produce this error message:
1427 Cannot tie second time: Invalid argument at bad.file line 14.
1429 Although the error message above refers to the second tie() statement
1430 in the script, the source of the problem is really with the untie()
1431 statement that precedes it.
1433 Having read L<perltie> you will probably have already guessed that the
1434 error is caused by the extra copy of the tied object stored in C<$X>.
1435 If you haven't, then the problem boils down to the fact that the
1436 B<DB_File> destructor, DESTROY, will not be called until I<all>
1437 references to the tied object are destroyed. Both the tied variable,
1438 C<%x>, and C<$X> above hold a reference to the object. The call to
1439 untie() will destroy the first, but C<$X> still holds a valid
1440 reference, so the destructor will not get called and the database file
1441 F<tst.fil> will remain open. The fact that Berkeley DB then reports the
1442 attempt to open a database that is alreday open via the catch-all
1443 "Invalid argument" doesn't help.
1445 If you run the script with the C<-w> flag the error message becomes:
1447 untie attempted while 1 inner references still exist at bad.file line 12.
1448 Cannot tie second time: Invalid argument at bad.file line 14.
1450 which pinpoints the real problem. Finally the script can now be
1451 modified to fix the original problem by destroying the API object
1460 $X = tie %x, 'DB_File', 'tst.fil' , O_RDWR|O_CREAT
1464 =head1 COMMON QUESTIONS
1466 =head2 Why is there Perl source in my database?
1468 If you look at the contents of a database file created by DB_File,
1469 there can sometimes be part of a Perl script included in it.
1471 This happens because Berkeley DB uses dynamic memory to allocate
1472 buffers which will subsequently be written to the database file. Being
1473 dynamic, the memory could have been used for anything before DB
1474 malloced it. As Berkeley DB doesn't clear the memory once it has been
1475 allocated, the unused portions will contain random junk. In the case
1476 where a Perl script gets written to the database, the random junk will
1477 correspond to an area of dynamic memory that happened to be used during
1478 the compilation of the script.
1480 Unless you don't like the possibility of there being part of your Perl
1481 scripts embedded in a database file, this is nothing to worry about.
1483 =head2 How do I store complex data structures with DB_File?
1485 Although B<DB_File> cannot do this directly, there is a module which
1486 can layer transparently over B<DB_File> to accomplish this feat.
1488 Check out the MLDBM module, available on CPAN in the directory
1489 F<modules/by-module/MLDBM>.
1491 =head2 What does "Invalid Argument" mean?
1493 You will get this error message when one of the parameters in the
1494 C<tie> call is wrong. Unfortunately there are quite a few parameters to
1495 get wrong, so it can be difficult to figure out which one it is.
1497 Here are a couple of possibilities:
1503 Attempting to reopen a database without closing it.
1507 Using the O_WRONLY flag.
1511 =head2 What does "Bareword 'DB_File' not allowed" mean?
1513 You will encounter this particular error message when you have the
1514 C<strict 'subs'> pragma (or the full strict pragma) in your script.
1515 Consider this script:
1520 tie %x, DB_File, "filename" ;
1522 Running it produces the error in question:
1524 Bareword "DB_File" not allowed while "strict subs" in use
1526 To get around the error, place the word C<DB_File> in either single or
1527 double quotes, like this:
1529 tie %x, "DB_File", "filename" ;
1531 Although it might seem like a real pain, it is really worth the effort
1532 of having a C<use strict> in all your scripts.
1544 When B<DB_File> is opening a database file it no longer terminates the
1545 process if I<dbopen> returned an error. This allows file protection
1546 errors to be caught at run time. Thanks to Judith Grass
1547 E<lt>grass@cybercash.comE<gt> for spotting the bug.
1551 Added prototype support for multiple btree compare callbacks.
1555 B<DB_File> has been in use for over a year. To reflect that, the
1556 version number has been incremented to 1.0.
1558 Added complete support for multiple concurrent callbacks.
1560 Using the I<push> method on an empty list didn't work properly. This
1565 Fixed a core dump problem with SunOS.
1567 The return value from TIEHASH wasn't set to NULL when dbopen returned
1572 Merged OS/2 specific code into DB_File.xs
1574 Removed some redundant code in DB_File.xs.
1576 Documentation update.
1578 Allow negative subscripts with RECNO interface.
1580 Changed the default flags from O_RDWR to O_CREAT|O_RDWR.
1582 The example code which showed how to lock a database needed a call to
1583 C<sync> added. Without it the resultant database file was empty.
1585 Added get_dup method.
1589 Documentation update.
1591 B<DB_File> now imports the constants (O_RDWR, O_CREAT etc.) from Fcntl
1594 The standard hash function C<exists> is now supported.
1596 Modified the behavior of get_dup. When it returns an associative
1597 array, the value is the count of the number of matching BTREE values.
1601 Minor documentation changes.
1603 Fixed a bug in hash_cb. Patches supplied by Dave Hammen,
1604 E<lt>hammen@gothamcity.jsc.nasa.govE<gt>.
1606 Fixed a bug with the constructors for DB_File::HASHINFO,
1607 DB_File::BTREEINFO and DB_File::RECNOINFO. Also tidied up the
1608 constructors to make them C<-w> clean.
1610 Reworked part of the test harness to be more locale friendly.
1614 Made all scripts in the documentation C<strict> and C<-w> clean.
1616 Added logic to F<DB_File.xs> to allow the module to be built after Perl
1621 Minor namespace cleanup: Localized C<PrintBtree>.
1625 Fixed bug with RECNO, where bval wasn't defaulting to "\n".
1629 Documented operation of bval.
1633 Minor bug fix in DB_File::HASHINFO, DB_File::RECNOINFO and
1636 Changed default mode to 0666.
1640 Fixed fd method so that it still returns -1 for in-memory files when db
1645 Documented the untie gotcha.
1649 Documented the incompatibility with version 2 of Berkeley DB.
1655 Some older versions of Berkeley DB had problems with fixed length
1656 records using the RECNO file format. The newest version at the time of
1657 writing was 1.85 - this seems to have fixed the problems with RECNO.
1659 I am sure there are bugs in the code. If you do find any, or can
1660 suggest any enhancements, I would welcome your comments.
1664 B<DB_File> comes with the standard Perl source distribution. Look in
1665 the directory F<ext/DB_File>.
1667 This version of B<DB_File> will only work with version 1.x of Berkeley
1668 DB. It is I<not> yet compatible with version 2.
1670 Version 1 of Berkeley DB is available at your nearest CPAN archive (see
1671 L<perlmod/"CPAN"> for a list) in F<src/misc/db.1.85.tar.gz>, or via the
1672 host F<ftp.cs.berkeley.edu> in F</ucb/4bsd/db.tar.gz>. Alternatively,
1673 check out the Berkeley DB home page at F<http://www.bostic.com/db>. It
1674 is I<not> under the GPL.
1676 If you are running IRIX, then get Berkeley DB from
1677 F<http://reality.sgi.com/ariel>. It has the patches necessary to
1678 compile properly on IRIX 5.3.
1680 As of January 1997, version 1.86 of Berkeley DB is available from the
1681 Berkeley DB home page. Although this release does fix a number of bugs
1682 that were present in 1.85 you should be aware of the following
1683 information (taken from the Berkeley DB home page) before you consider
1686 DB version 1.86 includes a new implementation of the hash access
1687 method that fixes a variety of hashing problems found in DB version
1688 1.85. We are making it available as an interim solution until DB
1691 PLEASE NOTE: the underlying file format for the hash access method
1692 changed between version 1.85 and version 1.86, so you will have to
1693 dump and reload all of your databases to convert from version 1.85
1694 to version 1.86. If you do not absolutely require the fixes from
1695 version 1.86, we strongly urge you to wait until DB 2.0 is released
1696 before upgrading from 1.85.
1701 L<perl(1)>, L<dbopen(3)>, L<hash(3)>, L<recno(3)>, L<btree(3)>
1705 The DB_File interface was written by Paul Marquess
1706 E<lt>pmarquess@bfsec.bt.co.ukE<gt>.
1707 Questions about the DB system itself may be addressed to Keith Bostic
1708 E<lt>bostic@cs.berkeley.eduE<gt>.