1dbm(3UCB) SunOS/BSD Compatibility Library Functions dbm(3UCB)
2
3
4
6 dbm, dbminit, dbmclose, fetch, store, delete, firstkey, nextkey - data
7 base subroutines
8
10 /usr/ucb/cc [ flag ... ] file ... -ldbm
11 #include <dbm.h>
12
13 typedef struct {
14 char *dptr;
15 int dsize;
16 }datum;
17
18 int dbminit(file)
19 char *file;
20
21
22 int dbmclose();
23
24
25 datum fetch(key)
26 datum key;
27
28
29 int store( key, dat)
30 datum key, dat;
31
32
33 int delete(key)
34 datum key;
35
36
37 datum firstkey();
38
39
40 datum nextkey(key)
41 datum key;
42
43
45 The dbm() library has been superseded by ndbm (see ndbm(3C)).
46
47
48 These functions maintain key/content pairs in a data base. The func‐
49 tions will handle very large (a billion blocks) databases and will
50 access a keyed item in one or two file system accesses.
51
52
53 key/dat and their content are described by the datum typedef. A datum
54 specifies a string of dsize bytes pointed to by dptr. Arbitrary binary
55 data, as well as normal ASCII strings, are allowed. The data base is
56 stored in two files. One file is a directory containing a bit map and
57 has .dir as its suffix. The second file contains all data and has .pag
58 as its suffix.
59
60
61 Before a database can be accessed, it must be opened by dbminit(). At
62 the time of this call, the files file.dir and file.pag must exist. An
63 empty database is created by creating zero-length .dir and .pag files.
64
65
66 A database may be closed by calling dbmclose(). You must close a data‐
67 base before opening a new one.
68
69
70 Once open, the data stored under a key is accessed by fetch() and data
71 is placed under a key by store. A key (and its associated contents) is
72 deleted by delete(). A linear pass through all keys in a database may
73 be made, in an (apparently) random order, by use of firstkey() and nex‐
74 tkey(). firstkey() will return the first key in the database. With any
75 key nextkey() will return the next key in the database. This code will
76 traverse the data base:
77
78 for (key = firstkey; key.dptr != NULL; key = nextkey(key))
79
80
82 All functions that return an int indicate errors with negative values.
83 A zero return indicates no error. Routines that return a datum indicate
84 errors with a NULL (0) dptr.
85
87 ar(1), cat(1), cp(1), tar(1), ndbm(3C)
88
90 Use of these interfaces should be restricted to only applications writ‐
91 ten on BSD platforms. Use of these interfaces with any of the system
92 libraries or in multi-thread applications is unsupported.
93
94
95 The .pag file will contain holes so that its apparent size may be
96 larger than its actual content. Older versions of the UNIX operating
97 system may create real file blocks for these holes when touched. These
98 files cannot be copied by normal means ( cp(1), cat(1), tar(1), ar(1))
99 without filling in the holes.
100
101
102 dptr pointers returned by these subroutines point into static storage
103 that is changed by subsequent calls.
104
105
106 The sum of the sizes of a key/content pair must not exceed the internal
107 block size (currently 1024 bytes). Moreover all key/content pairs that
108 hash together must fit on a single block. store will return an error in
109 the event that a disk block fills with inseparable data.
110
111
112 delete() does not physically reclaim file space, although it does make
113 it available for reuse.
114
115
116 The order of keys presented by firstkey() and nextkey() depends on a
117 hashing function, not on anything interesting.
118
119
120 There are no interlocks and no reliable cache flushing; thus concurrent
121 updating and reading is risky.
122
123
124 The database files (file.dir and file.pag) are binary and are architec‐
125 ture-specific (for example, they depend on the architecture's byte
126 order.) These files are not guaranteed to be portable across architec‐
127 tures.
128
129
130
131SunOS 5.11 30 Oct 2007 dbm(3UCB)