Commit | Line | Data |
---|---|---|
a0d0e21e LW |
1 | package AnyDBM_File; |
2 | ||
3 | @ISA = qw(NDBM_File DB_File GDBM_File SDBM_File ODBM_File) unless @ISA; | |
4 | ||
5 | eval { require NDBM_File } || | |
6 | eval { require DB_File } || | |
7 | eval { require GDBM_File } || | |
8 | eval { require SDBM_File } || | |
9 | eval { require ODBM_File }; | |
f06db76b AD |
10 | |
11 | =head1 NAME | |
12 | ||
13 | AnyDBM_File - provide framework for multiple DBMs | |
14 | ||
15 | NDBM_File, ODBM_File, SDBM_File, GDBM_File - various DBM implementations | |
16 | ||
17 | =head1 SYNOPSIS | |
18 | ||
19 | use AnyDBM_File; | |
20 | ||
21 | =head1 DESCRIPTION | |
22 | ||
23 | This module is a "pure virtual base class"--it has nothing of its own. | |
24 | It's just there to inherit from one of the various DBM packages. It | |
25 | prefers ndbm for compatibility reasons with Perl 4, then Berkeley DB (See | |
26 | L<DB_File>), GDBM, SDBM (which is always there--it comes with Perl), and | |
27 | finally ODBM. This way old programs that used to use NDBM via dbmopen() | |
28 | can still do so, but new ones can reorder @ISA: | |
29 | ||
30 | @AnyDBM_File::ISA = qw(DB_File GDBM_File NDBM_File); | |
31 | ||
32 | Note, however, that an explicit use overrides the specified order: | |
33 | ||
34 | use GDBM_File; | |
35 | @AnyDBM_File::ISA = qw(DB_File GDBM_File NDBM_File); | |
36 | ||
37 | will only find GDBM_File. | |
38 | ||
39 | Having multiple DBM implementations makes it trivial to copy database formats: | |
40 | ||
41 | use POSIX; use NDBM_File; use DB_File; | |
c954a603 | 42 | tie %newhash, 'DB_File', $new_filename, O_CREAT|O_RDWR; |
43 | tie %oldhash, 'NDBM_File', $old_filename, 1, 0; | |
f06db76b AD |
44 | %newhash = %oldhash; |
45 | ||
46 | =head2 DBM Comparisons | |
47 | ||
48 | Here's a partial table of features the different packages offer: | |
49 | ||
50 | odbm ndbm sdbm gdbm bsd-db | |
51 | ---- ---- ---- ---- ------ | |
52 | Linkage comes w/ perl yes yes yes yes yes | |
53 | Src comes w/ perl no no yes no no | |
54 | Comes w/ many unix os yes yes[0] no no no | |
55 | Builds ok on !unix ? ? yes yes ? | |
56 | Code Size ? ? small big big | |
57 | Database Size ? ? small big? ok[1] | |
58 | Speed ? ? slow ok fast | |
59 | FTPable no no yes yes yes | |
60 | Easy to build N/A N/A yes yes ok[2] | |
61 | Size limits 1k 4k 1k[3] none none | |
62 | Byte-order independent no no no no yes | |
63 | Licensing restrictions ? ? no yes no | |
64 | ||
65 | ||
66 | =over 4 | |
67 | ||
68 | =item [0] | |
69 | ||
70 | on mixed universe machines, may be in the bsd compat library, | |
71 | which is often shunned. | |
72 | ||
73 | =item [1] | |
74 | ||
75 | Can be trimmed if you compile for one access method. | |
76 | ||
77 | =item [2] | |
78 | ||
79 | See L<DB_File>. | |
80 | Requires symbolic links. | |
81 | ||
82 | =item [3] | |
83 | ||
84 | By default, but can be redefined. | |
85 | ||
86 | =back | |
87 | ||
88 | =head1 SEE ALSO | |
89 | ||
90 | dbm(3), ndbm(3), DB_File(3) | |
91 | ||
92 | =cut |