Updated for 2.1b2 distribution.
[python/dscho.git] / Doc / lib / libgdbm.tex
blob7e26b78f336ed9af995880681b0f0287a9704a02
1 \section{\module{gdbm} ---
2 GNU's reinterpretation of dbm}
4 \declaremodule{builtin}{gdbm}
5 \platform{Unix}
6 \modulesynopsis{GNU's reinterpretation of dbm.}
9 This module is quite similar to the \refmodule{dbm}\refbimodindex{dbm}
10 module, but uses \code{gdbm} instead to provide some additional
11 functionality. Please note that the file formats created by
12 \code{gdbm} and \code{dbm} are incompatible.
14 The \module{gdbm} module provides an interface to the GNU DBM
15 library. \code{gdbm} objects behave like mappings
16 (dictionaries), except that keys and values are always strings.
17 Printing a \code{gdbm} object doesn't print the keys and values, and
18 the \method{items()} and \method{values()} methods are not supported.
20 The module defines the following constant and functions:
22 \begin{excdesc}{error}
23 Raised on \code{gdbm}-specific errors, such as I/O errors.
24 \exception{KeyError} is raised for general mapping errors like
25 specifying an incorrect key.
26 \end{excdesc}
28 \begin{funcdesc}{open}{filename, \optional{flag, \optional{mode}}}
29 Open a \code{gdbm} database and return a \code{gdbm} object. The
30 \var{filename} argument is the name of the database file.
32 The optional \var{flag} argument can be
33 \code{'r'} (to open an existing database for reading only --- default),
34 \code{'w'} (to open an existing database for reading and writing),
35 \code{'c'} (which creates the database if it doesn't exist), or
36 \code{'n'} (which always creates a new empty database).
38 The following additional characters may be appended to the flag to
39 control how the database is opened:
41 \begin{itemize}
42 \item \code{'f'} --- Open the database in fast mode. Writes to the database
43 will not be syncronized.
44 \item \code{'s'} --- Synchronized mode. This will cause changes to the database
45 will be immediately written to the file.
46 \item \code{'u'} --- Do not lock database.
47 \end{itemize}
49 Not all flags are valid for all versions of \code{gdbm}. The
50 module constant \code{open_flags} is a string of supported flag
51 characters. The exception \exception{error} is raised if an invalid
52 flag is specified.
54 The optional \var{mode} argument is the \UNIX{} mode of the file, used
55 only when the database has to be created. It defaults to octal
56 \code{0666}.
57 \end{funcdesc}
59 In addition to the dictionary-like methods, \code{gdbm} objects have the
60 following methods:
62 \begin{funcdesc}{firstkey}{}
63 It's possible to loop over every key in the database using this method
64 and the \method{nextkey()} method. The traversal is ordered by
65 \code{gdbm}'s internal hash values, and won't be sorted by the key
66 values. This method returns the starting key.
67 \end{funcdesc}
69 \begin{funcdesc}{nextkey}{key}
70 Returns the key that follows \var{key} in the traversal. The
71 following code prints every key in the database \code{db}, without
72 having to create a list in memory that contains them all:
74 \begin{verbatim}
75 k = db.firstkey()
76 while k != None:
77 print k
78 k = db.nextkey(k)
79 \end{verbatim}
80 \end{funcdesc}
82 \begin{funcdesc}{reorganize}{}
83 If you have carried out a lot of deletions and would like to shrink
84 the space used by the \code{gdbm} file, this routine will reorganize
85 the database. \code{gdbm} will not shorten the length of a database
86 file except by using this reorganization; otherwise, deleted file
87 space will be kept and reused as new (key, value) pairs are added.
88 \end{funcdesc}
90 \begin{funcdesc}{sync}{}
91 When the database has been opened in fast mode, this method forces any
92 unwritten data to be written to the disk.
93 \end{funcdesc}
96 \begin{seealso}
97 \seemodule{anydbm}{Generic interface to \code{dbm}-style databases.}
98 \seemodule{whichdb}{Utility module used to determine the type of an
99 existing database.}
100 \end{seealso}