1SNOBOL4DBM(1) General Commands Manual SNOBOL4DBM(1)
2
6 snobol4dbm - SNOBOL4 NDBM interface
7
10 -INCLUDE 'ndbm.sno'
11 dbhandle = DBM_OPEN(file,flags,mode)
13 DBM_CLOSE(dbhandle)
15 DBM_STORE(dbhandle,key,datum,flags)
17 datum = DBM_FETCH(dbhandle,key)
19 DBM_DELETE(dbhandle,key)
21 key = DBM_FIRSTKEY(dbhandle)
23 key = DBM_NEXTKEY(dbhandle)
25 DBM_ERROR(dbhandle)
27 DBM_CLEARERR(dbhandle)
29
32 "NDBM" (for New Data Base Managemnt) is an industry standard fast
33 hashed storage API first created in 4.3BSD, and included in the Unix 98
34 (SUSv2) standard. The original DBM API appeared in AT&T Research Unix
35 Version 7, and only allowed access to a single file at a time.
36 There are many different implementations of this API, including:
39 · The original BSD 4.3 ndbm, based on AT&T dbm. Found in commer‐
42 cial Un*x offerings.
43 · Berkeley DB v1 compatibility interface. Supplied with 4.4BSD
46 based systems: (Free|Open|Net)BSD, MacOS X.
47 · GNU DBM (GDBM) found in Linux distributions.
50 · SDBM, Ozan Yigit's Public Domain implementation of NDBM. Sup‐
53 plied with this distribution, and used as a last resort on Un*x
54 systems (and by default on non Un*x systems).
55 DBM_OPEN takes a filename (STRING), flags (either "R" for read-only,
58 "W" for write access, or "CW" to create and write a new file), and a
59 "mode" string, which defaults to "0660" (octal) and returns a database
60 handle which can be passed to the remaining functions.
61 DBM_CLOSE closes the database file. DBM_CLOSE MUST be called to ensure
64 that all your data is written.
65 DBM_STORE takes a database handle, key and datum strings, and a flag
68 (either DBM_INSERT to insert a new pair, or fail if the key already
69 exists, or DBM_REPLACE to insert or replace existing data). The key
70 and datum strings may contain an arbitrary series of characters.
71 DBM_FETCH returns the stored datum (if any) for the supplied key, or
74 fails.
75 DBM_DELETE deletes the stored datum (if any) for the supplied key, or
78 fails.
79 DBM_FIRSTKEY and subsequent calls to DBM_NEXTKEY allow you to traverse
82 all stored keys. The keys will be returned in arbitrary order, and the
83 routines will fail at the end of the traversal.
84 DBM_ERROR is a predicate which succeeds if the database handle is valid
87 and an I/O error has occurred on the file. DBM_CLEARERR is a predicate
88 which succeeds if if the database handle is valid, and has the side
89 effect of clearing the I/O error flag.
90
93 NDBM, GDBM, and SDBM create two files: filename.dir, filename.pag.
94 Berkeley DB creates a single filename.db file.
95
98 ndbm(3), dbopen(3), gdbm(3).
99
102 Philip L. Budne
103
106 None of the implementations allow concurrent read and write.
107 Some implementations (classic NDBM and SDBM) place limits on the total
110 size of key plus datum (typically slightly less than 1KB).
111 NOTE: Some implementations (classic NDBM and SDBM) create sparse file
114 which appear (to "ls -l") to be larger than they are (see "ls -s").
115 Copying such files may cause the non-allocated blocks to be "filled"
116 with zeroed disk blocks, and then the files really will be large!
117 Only GDBM provides locking to eliminate the possibility of file corrup‐
120 tion, or reading of incomplete data.
121 GDBM locking sometimes fails on NFS mounted partitions (particularly on
124 Linux systems). GDBM's NDBM interface does not provide a way to dis‐
125 able locking.
126 DBM databases accessed concurrently by multiple processes are tradi‐
129 tionally (re)created from text files and used for fast disk-based read-
130 only table lookups. Programs which need to update the file generate a
131 new temporary copy using a different name, and then rename the new
132 file(s), so that the next reader gets the new copies (existing readers
133 continue to see old data).
134Local 3 Dec 2005 SNOBOL4DBM(1)