1Data::Entropy::Source(3U)ser Contributed Perl DocumentatiDoanta::Entropy::Source(3)
2
6 Data::Entropy::Source - encapsulated source of entropy
7
9 use Data::Entropy::Source;
10 $source = Data::Entropy::Source->new($handle, "sysread");
12 $c = $source->get_octet;
14 $str = $source->get_bits(17);
15 $i = $source->get_int(12345);
16 $i = $source->get_int(Math::BigInt->new("1000000000000"));
17 $j = $source->get_prob(1, 2);
18
20 An object of this class encapsulates a source of entropy (randomness).
21 Methods allow entropy to be dispensed in any quantity required, even
22 fractional bits. An entropy source object should not normally be used
23 directly. Rather, it should be used to support higher-level entropy-
24 consuming algorithms, such as those in Data::Entropy::Algorithms.
25 This type of object is constructed as a layer over a raw entropy source
27 which does not supply methods to extract arbitrary amounts of entropy.
28 The raw entropy source is expected to dispense only entire octets at a
29 time. The /dev/random devices on some versions of Unix constitute such
30 a source, for example. The raw entropy source is accessed via the
31 "IO::Handle" interface. This interface may be supplied by classes
32 other than "IO::Handle" itself, as is done for example by
33 "Data::Entropy::RawSource::CryptCounter".
34 If two entropy sources of this class are given exactly the same raw
36 entropy data, for example by reading from the same file, and exactly
37 the same sequence of "get_" method calls is made to them, then they
38 will return exactly the same values from those calls. (Calls with
39 numerical arguments that have the same numerical value but are of
40 different types count as the same for this purpose.) This means that a
41 run of an entropy-using algorithm can be made completely deterministic
42 if desired.
43
45 Data::Entropy::Source->new(RAW_SOURCE, READ_STYLE)
46 Constructs and returns an entropy source object based on the given
47 raw source. RAW_SOURCE must be an I/O handle referring to a source
48 of entropy that can be read one octet at a time. Specifically, it
49 must support either the "getc" or "sysread" method described in
50 IO::Handle. READ_STYLE must be a string, either "getc" or
51 "sysread", indicating which method should be used to read from the
52 raw source. No methods other than the one specified will ever be
53 called on the raw source handle, so a full implementation of
54 "IO::Handle" is not required.
55 The "sysread" method should be used with /dev/random and its ilk,
57 because buffering would be very wasteful of entropy and might
58 consequently block other processes that require entropy. "getc"
59 should be preferred when reading entropy from a regular file, and
60 it is the more convenient interface to implement when a non-I/O
61 object is being used for the handle.
62
64 $source->get_octet
65 Returns an octet of entropy, as a string of length one. This
66 provides direct access to the raw entropy source.
67 $source->get_bits(NBITS)
69 Returns NBITS bits of entropy, as a string of octets. If NBITS is
70 not a multiple of eight then the last octet in the string has its
71 most significant bits set to zero.
72 $source->get_int(LIMIT)
74 LIMIT must be a positive integer. Returns a uniformly-distributed
75 random number between zero inclusive and LIMIT exclusive. LIMIT
76 may be either a native integer, a "Math::BigInt" object, or an
77 integer-valued "Math::BigRat" object; the returned number is of the
78 same type.
79 This method dispenses a non-integer number of bits of entropy. For
81 example, if LIMIT is 10 then the result contains approximately 3.32
82 bits of entropy. The minimum non-zero amount of entropy that can
83 be obtained is 1 bit, with LIMIT = 2.
84 $source->get_prob(PROB0, PROB1)
86 PROB0 and PROB1 must be non-negative integers, not both zero. They
87 may each be either a native integer, a "Math::BigInt" object, or an
88 integer-valued "Math::BigRat" objects; types may be mixed. Returns
89 either 0 or 1, with relative probabilities PROB0 and PROB1. That
90 is, the probability of returning 0 is PROB0/(PROB0+PROB1), and the
91 probability of returning 1 is PROB1/(PROB0+PROB1).
92 This method dispenses a fraction of a bit of entropy. The maximum
94 amount of entropy that can be obtained is 1 bit, with PROB0 =
95 PROB1. The more different the probabilities are the less entropy
96 is obtained. For example, if PROB0 = 1 and PROB1 = 2 then the
97 result contains approximately 0.918 bits of entropy.
98
100 Data::Entropy, Data::Entropy::Algorithms,
101 Data::Entropy::RawSource::CryptCounter,
102 Data::Entropy::RawSource::Local, Data::Entropy::RawSource::RandomOrg,
103 IO::Handle
104
106 Andrew Main (Zefram) <zefram@fysh.org>
107
109 Copyright (C) 2006, 2007, 2009, 2011 Andrew Main (Zefram)
110 <zefram@fysh.org>
111
113 This module is free software; you can redistribute it and/or modify it
114 under the same terms as Perl itself.
115perl v5.32.0 2020-07-28 Data::Entropy::Source(3)