line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
#!/usr/bin/perl |
2
|
|
|
|
|
|
|
|
3
|
|
|
|
|
|
|
package Net::OnlineCode; |
4
|
|
|
|
|
|
|
|
5
|
|
|
|
|
|
|
# play nicely as a CPAN module |
6
|
|
|
|
|
|
|
|
7
|
4
|
|
|
4
|
|
32954
|
use strict; |
|
4
|
|
|
|
|
5
|
|
|
4
|
|
|
|
|
127
|
|
8
|
4
|
|
|
4
|
|
15
|
use warnings; |
|
4
|
|
|
|
|
5
|
|
|
4
|
|
|
|
|
125
|
|
9
|
4
|
|
|
4
|
|
16
|
use vars qw(@ISA @EXPORT_OK @EXPORT %EXPORT_TAGS $VERSION); |
|
4
|
|
|
|
|
6
|
|
|
4
|
|
|
|
|
483
|
|
10
|
|
|
|
|
|
|
|
11
|
4
|
|
|
4
|
|
22
|
use constant DEBUG => 0; |
|
4
|
|
|
|
|
4
|
|
|
4
|
|
|
|
|
309
|
|
12
|
4
|
|
|
4
|
|
49
|
use constant ASSERT => 1; |
|
4
|
|
|
|
|
6
|
|
|
4
|
|
|
|
|
639
|
|
13
|
|
|
|
|
|
|
|
14
|
|
|
|
|
|
|
require Exporter; |
15
|
|
|
|
|
|
|
|
16
|
|
|
|
|
|
|
@ISA = qw(Exporter); |
17
|
|
|
|
|
|
|
|
18
|
|
|
|
|
|
|
our @export_xor = qw (xor_strings safe_xor_strings fast_xor_strings); |
19
|
|
|
|
|
|
|
our @export_default = qw(); |
20
|
|
|
|
|
|
|
|
21
|
|
|
|
|
|
|
%EXPORT_TAGS = ( all => [ @export_default, @export_xor ], |
22
|
|
|
|
|
|
|
xor => [ @export_xor ], |
23
|
|
|
|
|
|
|
); |
24
|
|
|
|
|
|
|
@EXPORT_OK = ( @{ $EXPORT_TAGS{'all'} } ); |
25
|
|
|
|
|
|
|
@EXPORT = (); |
26
|
|
|
|
|
|
|
$VERSION = '0.04'; |
27
|
|
|
|
|
|
|
|
28
|
|
|
|
|
|
|
# Use XS for fast xors (TODO: make this optional) |
29
|
|
|
|
|
|
|
require XSLoader; |
30
|
|
|
|
|
|
|
XSLoader::load('Net::OnlineCode', $VERSION); |
31
|
|
|
|
|
|
|
|
32
|
|
|
|
|
|
|
# on to our stuff ... |
33
|
|
|
|
|
|
|
|
34
|
|
|
|
|
|
|
|
35
|
|
|
|
|
|
|
# Codec parameters |
36
|
|
|
|
|
|
|
# q is the number of message blocks that each auxiliary block will |
37
|
|
|
|
|
|
|
# link to |
38
|
|
|
|
|
|
|
# e epsilon, the degree of "suboptimality". Unlike Reed-Solomon or |
39
|
|
|
|
|
|
|
# Rabin's Information Dispersal Algorithm, Online Codes are not |
40
|
|
|
|
|
|
|
# optimal. This means that slightly more data needs to be generated |
41
|
|
|
|
|
|
|
# than either of these two codes. Also, whereas optimal codes |
42
|
|
|
|
|
|
|
# guarantee that a certain fraction of the "check" blocks/digits |
43
|
|
|
|
|
|
|
# suffice to reconstruct the original message, online codes only |
44
|
|
|
|
|
|
|
# guarantee that it can be reconstructed with a certain |
45
|
|
|
|
|
|
|
# probability |
46
|
|
|
|
|
|
|
# |
47
|
|
|
|
|
|
|
# Together with the number of blocks, n, these two variables define |
48
|
|
|
|
|
|
|
# the online code such that (1+qe)n check blocks are sufficient to |
49
|
|
|
|
|
|
|
# reconstruct the original message with a probability of 1 - (e/2) ** |
50
|
|
|
|
|
|
|
# (q+1). |
51
|
|
|
|
|
|
|
# |
52
|
|
|
|
|
|
|
|
53
|
4
|
|
|
4
|
|
18
|
use Carp; |
|
4
|
|
|
|
|
6
|
|
|
4
|
|
|
|
|
296
|
|
54
|
4
|
|
|
4
|
|
2062
|
use POSIX qw(ceil floor); |
|
4
|
|
|
|
|
22839
|
|
|
4
|
|
|
|
|
27
|
|
55
|
4
|
|
|
4
|
|
6845
|
use Digest::SHA qw(sha1 sha1_hex); |
|
4
|
|
|
|
|
13772
|
|
|
4
|
|
|
|
|
389
|
|
56
|
4
|
|
|
4
|
|
71
|
use Fcntl; |
|
4
|
|
|
|
|
6
|
|
|
4
|
|
|
|
|
10751
|
|
57
|
|
|
|
|
|
|
|
58
|
|
|
|
|
|
|
|
59
|
|
|
|
|
|
|
# Constructor for the base class |
60
|
|
|
|
|
|
|
# |
61
|
|
|
|
|
|
|
# This takes the three parameters that define the Online Code scheme, |
62
|
|
|
|
|
|
|
# corrects the value of epsilon if needed (see below) and then derives |
63
|
|
|
|
|
|
|
# the following: |
64
|
|
|
|
|
|
|
# |
65
|
|
|
|
|
|
|
# * max degree variable (F) |
66
|
|
|
|
|
|
|
# * number of auxiliary blocks (0.55 *qen) |
67
|
|
|
|
|
|
|
# * probability distribution p_1, p2, ... , p_F |
68
|
|
|
|
|
|
|
|
69
|
|
|
|
|
|
|
sub new { |
70
|
|
|
|
|
|
|
|
71
|
2
|
|
|
2
|
0
|
12
|
my $class = shift; |
72
|
|
|
|
|
|
|
# The default parameters used here for q and e (epsilon) are as |
73
|
|
|
|
|
|
|
# suggested in the paper "Rateless Codes and Big Downloads" by Petar |
74
|
|
|
|
|
|
|
# Maymounkov and David Maziere. Note that the value of e may be |
75
|
|
|
|
|
|
|
# overridden with a higher value if the lower value doesn't satisfy |
76
|
|
|
|
|
|
|
# max_degree(epsilon) > ceil(0.55 * q.e.mblocks) |
77
|
2
|
|
|
|
|
44
|
my %args = ( |
78
|
|
|
|
|
|
|
e => 0.01, |
79
|
|
|
|
|
|
|
q => 3, |
80
|
|
|
|
|
|
|
mblocks => undef, |
81
|
|
|
|
|
|
|
expand_aux => 0, |
82
|
|
|
|
|
|
|
e_warning => 0, |
83
|
|
|
|
|
|
|
|
84
|
|
|
|
|
|
|
# We don't use or store any RNG parameter that's been |
85
|
|
|
|
|
|
|
# passed into the constructor. |
86
|
|
|
|
|
|
|
@_ |
87
|
|
|
|
|
|
|
); |
88
|
|
|
|
|
|
|
|
89
|
|
|
|
|
|
|
|
90
|
2
|
|
|
|
|
6
|
my ($q,$e,$mblocks) = @args{qw(q e mblocks)}; |
91
|
|
|
|
|
|
|
|
92
|
2
|
50
|
|
|
|
7
|
unless (defined $args{mblocks}) { |
93
|
0
|
|
|
|
|
0
|
carp __PACKAGE__ . ": mblocks => (# message blocks) must be set\n"; |
94
|
0
|
|
|
|
|
0
|
return undef; |
95
|
|
|
|
|
|
|
} |
96
|
|
|
|
|
|
|
|
97
|
2
|
|
|
|
|
3
|
print "Net::OnlineCode mblocks = $mblocks\n" if DEBUG; |
98
|
|
|
|
|
|
|
|
99
|
2
|
|
|
|
|
3
|
my $P = undef; |
100
|
2
|
|
|
|
|
2
|
my $e_changed = 0; |
101
|
|
|
|
|
|
|
|
102
|
|
|
|
|
|
|
# how many auxiliary blocks would this scheme need? |
103
|
2
|
|
|
|
|
6
|
my $ablocks = _count_auxiliary($q,$e,$mblocks); |
104
|
|
|
|
|
|
|
|
105
|
|
|
|
|
|
|
# does epsilon value need updating? |
106
|
2
|
|
|
|
|
5
|
my $f = _max_degree($e); |
107
|
|
|
|
|
|
|
|
108
|
2
|
100
|
|
|
|
24
|
if ($f > $mblocks + $ablocks) { |
109
|
|
|
|
|
|
|
|
110
|
1
|
|
|
|
|
2
|
$e_changed = 1; |
111
|
|
|
|
|
|
|
|
112
|
1
|
50
|
|
|
|
5
|
if ($args{e_warning}) { |
113
|
0
|
|
|
|
|
0
|
print "E CHANGED!!\nWas: $e\n"; |
114
|
0
|
|
|
|
|
0
|
print "Gave F value of $f\n"; |
115
|
|
|
|
|
|
|
} |
116
|
|
|
|
|
|
|
|
117
|
|
|
|
|
|
|
# use a binary search to find a new epsilon such that |
118
|
|
|
|
|
|
|
# get_max_degree($epsilon) <= mblocks + ablocks (ie, n) |
119
|
1
|
|
|
|
|
2
|
my $epsilon = $e; |
120
|
|
|
|
|
|
|
|
121
|
|
|
|
|
|
|
local *eval_f = sub { |
122
|
11
|
|
|
11
|
|
11
|
my $t = shift; |
123
|
11
|
|
|
|
|
27
|
return _max_degree(1/(1 + exp(-$t))); |
124
|
1
|
|
|
|
|
7
|
}; |
125
|
|
|
|
|
|
|
|
126
|
1
|
|
|
|
|
4
|
my $l = -log(1/$e - 1); |
127
|
1
|
|
|
|
|
2
|
my $r = $l + 1; |
128
|
|
|
|
|
|
|
|
129
|
|
|
|
|
|
|
# expand right side of search until we get F <= n' |
130
|
1
|
|
|
|
|
2
|
while (eval_f($r) > $mblocks + $ablocks) { |
131
|
|
|
|
|
|
|
# $r = $l + ($r - $l) * 2; |
132
|
1
|
|
|
|
|
2
|
$r = 2 * $r - $l; |
133
|
|
|
|
|
|
|
} |
134
|
|
|
|
|
|
|
|
135
|
|
|
|
|
|
|
# binary search between left and right to find a suitable lower |
136
|
|
|
|
|
|
|
# value of epsilon still satisfying F <= n' |
137
|
1
|
|
|
|
|
4
|
while ($r - $l > 0.01) { |
138
|
8
|
|
|
|
|
5
|
my $m = ($l + $r) / 2; |
139
|
8
|
100
|
|
|
|
10
|
if (eval_f($m) > $mblocks + $ablocks) { |
140
|
3
|
|
|
|
|
5
|
$l = $m; |
141
|
|
|
|
|
|
|
} else { |
142
|
5
|
|
|
|
|
10
|
$r = $m; |
143
|
|
|
|
|
|
|
} |
144
|
|
|
|
|
|
|
} |
145
|
|
|
|
|
|
|
|
146
|
|
|
|
|
|
|
# update e and ablocks |
147
|
1
|
|
|
|
|
2
|
$epsilon = 1/(1 + exp(-$r)); |
148
|
1
|
|
|
|
|
2
|
$f = eval_f($r); |
149
|
|
|
|
|
|
|
#$f=_max_degree($epsilon); |
150
|
1
|
50
|
|
|
|
3
|
carp __PACKAGE__ . ": increased epsilon value from $e to $epsilon\n" |
151
|
|
|
|
|
|
|
if $args{e_warning}; |
152
|
1
|
|
|
|
|
1
|
$e = $epsilon; |
153
|
1
|
|
|
|
|
2
|
$ablocks = _count_auxiliary($q,$e,$mblocks); |
154
|
|
|
|
|
|
|
|
155
|
1
|
50
|
|
|
|
5
|
if ($args{e_warning}) { |
156
|
|
|
|
|
|
|
|
157
|
0
|
|
|
|
|
0
|
print "Is now: $e\n"; |
158
|
0
|
|
|
|
|
0
|
print "New F: $f\n"; |
159
|
|
|
|
|
|
|
} |
160
|
|
|
|
|
|
|
|
161
|
|
|
|
|
|
|
} |
162
|
|
|
|
|
|
|
|
163
|
|
|
|
|
|
|
# calculate the probability distribution |
164
|
2
|
|
|
|
|
2
|
print "new: mblocks=$mblocks, ablocks=$ablocks, q=$q\n" if DEBUG; |
165
|
2
|
|
|
|
|
7
|
$P = _probability_distribution($mblocks + $ablocks,$e); |
166
|
|
|
|
|
|
|
|
167
|
2
|
50
|
|
|
|
10
|
die "Wrong number of elements in probability distribution (got " |
168
|
|
|
|
|
|
|
. scalar(@$P) . ", expecting $f)\n" |
169
|
|
|
|
|
|
|
unless @$P == $f; |
170
|
|
|
|
|
|
|
|
171
|
2
|
|
|
|
|
28
|
my $self = { q => $q, e => $e, f => $f, P => $P, |
172
|
|
|
|
|
|
|
mblocks => $mblocks, ablocks => $ablocks, |
173
|
|
|
|
|
|
|
coblocks => $mblocks + $ablocks, |
174
|
|
|
|
|
|
|
chblocks => 0, expand_aux=> $args{expand_aux}, |
175
|
|
|
|
|
|
|
e_changed => $e_changed, unique => {}, |
176
|
|
|
|
|
|
|
}; |
177
|
|
|
|
|
|
|
|
178
|
2
|
|
|
|
|
3
|
print "expand_aux => $self->{expand_aux}\n" if DEBUG; |
179
|
|
|
|
|
|
|
|
180
|
2
|
|
|
|
|
22
|
bless $self, $class; |
181
|
|
|
|
|
|
|
|
182
|
|
|
|
|
|
|
} |
183
|
|
|
|
|
|
|
|
184
|
|
|
|
|
|
|
# while it probably doesn't matter too much to the encoder whether the |
185
|
|
|
|
|
|
|
# supplied e value needed to be changed, if the receiver plugs the |
186
|
|
|
|
|
|
|
# received value of e into the constructor and it ends up changing, |
187
|
|
|
|
|
|
|
# there will be a problem with receiving the file. |
188
|
|
|
|
|
|
|
sub e_changed { |
189
|
0
|
|
|
0
|
0
|
0
|
return shift ->{e_changed}; |
190
|
|
|
|
|
|
|
} |
191
|
|
|
|
|
|
|
|
192
|
|
|
|
|
|
|
# convenience accessor functions |
193
|
|
|
|
|
|
|
sub get_mblocks { # count message blocks; passed into new |
194
|
2
|
|
|
2
|
0
|
10
|
return shift -> {mblocks}; |
195
|
|
|
|
|
|
|
} |
196
|
|
|
|
|
|
|
|
197
|
|
|
|
|
|
|
sub get_ablocks { # count auxiliary blocks; set in new |
198
|
2
|
|
|
2
|
0
|
11
|
return shift -> {ablocks}; |
199
|
|
|
|
|
|
|
} |
200
|
|
|
|
|
|
|
|
201
|
|
|
|
|
|
|
sub get_coblocks { # count composite blocks |
202
|
2
|
|
|
2
|
0
|
3
|
my $self = shift; |
203
|
2
|
|
|
|
|
14
|
return $self->{mblocks} + $self->{ablocks}; |
204
|
|
|
|
|
|
|
} |
205
|
|
|
|
|
|
|
|
206
|
|
|
|
|
|
|
# count checkblocks |
207
|
|
|
|
|
|
|
sub get_chblocks { |
208
|
0
|
|
|
0
|
0
|
0
|
return shift->{chblocks} |
209
|
|
|
|
|
|
|
} |
210
|
|
|
|
|
|
|
|
211
|
|
|
|
|
|
|
sub get_q { # q == reliability factor |
212
|
0
|
|
|
0
|
0
|
0
|
return shift -> {q}; |
213
|
|
|
|
|
|
|
} |
214
|
|
|
|
|
|
|
|
215
|
|
|
|
|
|
|
sub get_e { # e == suboptimality factor |
216
|
2
|
|
|
2
|
0
|
806
|
return shift -> {e}; |
217
|
|
|
|
|
|
|
} |
218
|
|
|
|
|
|
|
|
219
|
|
|
|
|
|
|
sub get_epsilon { # epsilon == e, as above |
220
|
0
|
|
|
0
|
0
|
0
|
return shift -> {e}; |
221
|
|
|
|
|
|
|
} |
222
|
|
|
|
|
|
|
|
223
|
|
|
|
|
|
|
sub get_f { # f == max (check block) degree |
224
|
2
|
|
|
2
|
0
|
40
|
return shift -> {f}; |
225
|
|
|
|
|
|
|
} |
226
|
|
|
|
|
|
|
|
227
|
|
|
|
|
|
|
sub get_P { # P == probability distribution |
228
|
0
|
|
|
0
|
0
|
0
|
return shift -> {P}; # (array ref) |
229
|
|
|
|
|
|
|
} |
230
|
|
|
|
|
|
|
|
231
|
|
|
|
|
|
|
|
232
|
|
|
|
|
|
|
# "Private" routines |
233
|
|
|
|
|
|
|
|
234
|
|
|
|
|
|
|
# calculate how many auxiliary blocks need to be generated for a given |
235
|
|
|
|
|
|
|
# code setup |
236
|
|
|
|
|
|
|
sub _count_auxiliary { |
237
|
3
|
|
|
3
|
|
5
|
my ($q, $e, $n) = @_; |
238
|
|
|
|
|
|
|
|
239
|
3
|
|
|
|
|
24
|
my $count = int(ceil(0.55 * $q * $e * $n)); |
240
|
3
|
|
|
|
|
6
|
my $delta = 0.55 * $e; |
241
|
|
|
|
|
|
|
|
242
|
3
|
|
|
|
|
3
|
warn "failure probability " . ($delta ** $q) . "\n" if DEBUG; |
243
|
|
|
|
|
|
|
#$count = int(ceil($q * $delta * $n)); |
244
|
|
|
|
|
|
|
|
245
|
|
|
|
|
|
|
# Is it better to change q or the number of aux blocks if q is too |
246
|
|
|
|
|
|
|
# big? It's certainly easier to keep the q value and increase the |
247
|
|
|
|
|
|
|
# number of aux blocks, as I'm doing here, and may even be the right |
248
|
|
|
|
|
|
|
# thing to do rather than ignoring the user's q value. |
249
|
3
|
50
|
|
|
|
7
|
if ($count < $q) { |
250
|
0
|
|
|
|
|
0
|
$count = $q; |
251
|
|
|
|
|
|
|
# warn "updated _count_auxiliary output value to $q\n"; |
252
|
|
|
|
|
|
|
} |
253
|
3
|
|
|
|
|
4
|
return $count; |
254
|
|
|
|
|
|
|
} |
255
|
|
|
|
|
|
|
|
256
|
|
|
|
|
|
|
# The max degree specifies the maximum number of blocks to be XORed |
257
|
|
|
|
|
|
|
# together. This parameter is named F. |
258
|
|
|
|
|
|
|
sub _max_degree { |
259
|
|
|
|
|
|
|
|
260
|
15
|
|
|
15
|
|
14
|
my $epsilon = shift; |
261
|
|
|
|
|
|
|
|
262
|
15
|
|
|
|
|
30
|
my $quotient = (2 * log ($epsilon / 2)) / |
263
|
|
|
|
|
|
|
(log (1 - $epsilon / 2)); |
264
|
|
|
|
|
|
|
|
265
|
15
|
|
|
|
|
11
|
my $delta = 0.55 * $epsilon; |
266
|
|
|
|
|
|
|
#$quotient = (log ($epsilon) + log($delta)) / (log (1 - $epsilon)); |
267
|
|
|
|
|
|
|
|
268
|
15
|
|
|
|
|
30
|
return int(ceil($quotient)); |
269
|
|
|
|
|
|
|
} |
270
|
|
|
|
|
|
|
|
271
|
|
|
|
|
|
|
# Functions relating to probability distribution |
272
|
|
|
|
|
|
|
# |
273
|
|
|
|
|
|
|
# From the wikipedia page: |
274
|
|
|
|
|
|
|
# |
275
|
|
|
|
|
|
|
# http://en.wikipedia.org/wiki/Online_codes |
276
|
|
|
|
|
|
|
# |
277
|
|
|
|
|
|
|
# During the inner coding step the algorithm selects some number of |
278
|
|
|
|
|
|
|
# composite messages at random and XORs them together to form a check |
279
|
|
|
|
|
|
|
# block. In order for the algorithm to work correctly, both the number |
280
|
|
|
|
|
|
|
# of blocks to be XORed together and their distribution over composite |
281
|
|
|
|
|
|
|
# blocks must follow a particular probability distribution. |
282
|
|
|
|
|
|
|
# |
283
|
|
|
|
|
|
|
# Consult the references for the implementation details. |
284
|
|
|
|
|
|
|
# |
285
|
|
|
|
|
|
|
# The probability distribution is designed to map a random number in |
286
|
|
|
|
|
|
|
# the range [0,1) and return a degree i between 1 and F. The |
287
|
|
|
|
|
|
|
# probability distribution depends on a single input, n, which is the |
288
|
|
|
|
|
|
|
# number of blocks in the original message. The fixed values for q and |
289
|
|
|
|
|
|
|
# epsilon are also used. |
290
|
|
|
|
|
|
|
# |
291
|
|
|
|
|
|
|
# This code includes two changes from that described in the wikipedia |
292
|
|
|
|
|
|
|
# page. |
293
|
|
|
|
|
|
|
# |
294
|
|
|
|
|
|
|
# 1) Rather than returning an array of individual probabilities p_i, |
295
|
|
|
|
|
|
|
# the array includes the cumulative probabilities. For example, if |
296
|
|
|
|
|
|
|
# the p_i probabilities were: |
297
|
|
|
|
|
|
|
# (0.1, 0.2, 0.3, 0.2, 0.1, 0.1) |
298
|
|
|
|
|
|
|
# then the returned array would be: |
299
|
|
|
|
|
|
|
# (0.1, 0.3, 0.6, 0.8, 0.9, 1) (last element always has value 1) |
300
|
|
|
|
|
|
|
# This is done simply to make selecting a value based on the random |
301
|
|
|
|
|
|
|
# number more efficient, but the underlying probability distribution |
302
|
|
|
|
|
|
|
# is the same. |
303
|
|
|
|
|
|
|
# 2) Handling edge cases. These are: |
304
|
|
|
|
|
|
|
# a) the case where n = 1; and |
305
|
|
|
|
|
|
|
# b) the case where F > n |
306
|
|
|
|
|
|
|
# In both cases, the default value for epsilon cannot be used, so a |
307
|
|
|
|
|
|
|
# more suitable value is calculated. |
308
|
|
|
|
|
|
|
# |
309
|
|
|
|
|
|
|
# The return value is an array containing: |
310
|
|
|
|
|
|
|
# |
311
|
|
|
|
|
|
|
# * the max degree F |
312
|
|
|
|
|
|
|
# * a possibly updated value of epsilon |
313
|
|
|
|
|
|
|
# * the F values of the (cumulative) probability distribution |
314
|
|
|
|
|
|
|
|
315
|
|
|
|
|
|
|
sub _probability_distribution { |
316
|
|
|
|
|
|
|
|
317
|
2
|
|
|
2
|
|
2
|
my ($nblocks,$epsilon) = @_; # nblocks = number of *composite* blocks! |
318
|
|
|
|
|
|
|
|
319
|
2
|
|
|
|
|
2
|
print "generating probability distribution from nblocks $nblocks, e $epsilon\n" |
320
|
|
|
|
|
|
|
if DEBUG; |
321
|
|
|
|
|
|
|
|
322
|
2
|
|
|
|
|
3
|
my $f = _max_degree($epsilon); |
323
|
|
|
|
|
|
|
|
324
|
|
|
|
|
|
|
# after code reorganisation, this shouldn't happen: |
325
|
2
|
50
|
|
|
|
5
|
if ($f > $nblocks) { |
326
|
0
|
|
|
|
|
0
|
croak "BUG: " .__PACKAGE__ . " - epsilon still too small!\n"; |
327
|
|
|
|
|
|
|
} |
328
|
|
|
|
|
|
|
|
329
|
|
|
|
|
|
|
# probability distribution |
330
|
|
|
|
|
|
|
|
331
|
|
|
|
|
|
|
# Calculate the sum of the sequence: |
332
|
|
|
|
|
|
|
# |
333
|
|
|
|
|
|
|
# 1 + 1/F |
334
|
|
|
|
|
|
|
# p_1 = 1 - --------- |
335
|
|
|
|
|
|
|
# 1 + e |
336
|
|
|
|
|
|
|
# |
337
|
|
|
|
|
|
|
# |
338
|
|
|
|
|
|
|
# F . (1 - p_1) |
339
|
|
|
|
|
|
|
# p_i = --------------------- |
340
|
|
|
|
|
|
|
# (F - 1) . (i^2 - i) |
341
|
|
|
|
|
|
|
# |
342
|
|
|
|
|
|
|
# Since the i term is the only thing that changes for each p_i, I |
343
|
|
|
|
|
|
|
# optimise the calculation by keeping a fixed term involving only p |
344
|
|
|
|
|
|
|
# and f with a variable one involving i, then dividing as |
345
|
|
|
|
|
|
|
# appropriate. |
346
|
|
|
|
|
|
|
|
347
|
2
|
|
|
|
|
5
|
my $p1 = 1 - (1 + 1/$f)/(1 + $epsilon); |
348
|
2
|
|
|
|
|
4
|
my $pfterm = (1-$p1) * $f / ($f - 1); |
349
|
|
|
|
|
|
|
|
350
|
2
|
50
|
|
|
|
5
|
die "p1 is negative\n" if $p1 < 0; |
351
|
|
|
|
|
|
|
|
352
|
|
|
|
|
|
|
# hard-code simple cases where f = 1 or 2 |
353
|
2
|
50
|
|
|
|
6
|
if ($f == 1) { |
|
|
50
|
|
|
|
|
|
354
|
0
|
|
|
|
|
0
|
return [1]; |
355
|
|
|
|
|
|
|
} elsif ($f == 2) { |
356
|
0
|
|
|
|
|
0
|
return [$p1, 1]; |
357
|
|
|
|
|
|
|
} |
358
|
|
|
|
|
|
|
|
359
|
|
|
|
|
|
|
# calculate sum(p_i) for 2 <= i < F. |
360
|
|
|
|
|
|
|
# p_i=F is simply set to 1 to avoid rounding errors in the sum |
361
|
2
|
|
|
|
|
3
|
my $sum = $p1; |
362
|
2
|
|
|
|
|
167
|
my @P = ($sum); |
363
|
|
|
|
|
|
|
|
364
|
2
|
|
|
|
|
5
|
my $i = 2; |
365
|
2
|
|
|
|
|
6
|
while ($i < $f) { |
366
|
2515
|
|
|
|
|
2109
|
my $iterm = $i * ($i - 1); |
367
|
2515
|
|
|
|
|
1818
|
my $p_i = $pfterm / $iterm; |
368
|
|
|
|
|
|
|
|
369
|
2515
|
|
|
|
|
1783
|
$sum += $p_i; |
370
|
|
|
|
|
|
|
|
371
|
2515
|
50
|
|
|
|
3459
|
die "p_$i is negative\n" if $p_i < 0; |
372
|
|
|
|
|
|
|
|
373
|
2515
|
|
|
|
|
2117
|
push @P, $sum; |
374
|
2515
|
|
|
|
|
3453
|
$i++; |
375
|
|
|
|
|
|
|
} |
376
|
|
|
|
|
|
|
|
377
|
2
|
|
|
|
|
5
|
if (DEBUG) { |
378
|
|
|
|
|
|
|
# Make sure of the assumption that the sum of terms approaches 1. |
379
|
|
|
|
|
|
|
# If the "rounding error" below is not a very small number, we |
380
|
|
|
|
|
|
|
# know there is a problem with the assumption! |
381
|
|
|
|
|
|
|
my $p_last = $sum + $pfterm / ($f * $f - $f); |
382
|
|
|
|
|
|
|
my $absdiff = abs (1 - $p_last); |
383
|
|
|
|
|
|
|
warn "Absolute difference of 1,sum to p_F = $absdiff\n" if $absdiff >1e-8; |
384
|
|
|
|
|
|
|
} |
385
|
|
|
|
|
|
|
|
386
|
2
|
|
|
|
|
262
|
return [@P,1]; |
387
|
|
|
|
|
|
|
} |
388
|
|
|
|
|
|
|
|
389
|
|
|
|
|
|
|
|
390
|
|
|
|
|
|
|
# Using Floyd's algorithm instead of Fisher-Yates shuffle. Picks k |
391
|
|
|
|
|
|
|
# distinct elements from the range [start, start + n - 1]. Avoids the |
392
|
|
|
|
|
|
|
# need to copy/re-initialise an array of size n every time we make a |
393
|
|
|
|
|
|
|
# new check block. |
394
|
|
|
|
|
|
|
sub floyd { |
395
|
0
|
|
|
0
|
0
|
0
|
my ($rng, $start, $n, $k) = @_; |
396
|
0
|
|
|
|
|
0
|
my %set; |
397
|
0
|
|
|
|
|
0
|
my ($j, $t) = ($n - $k); |
398
|
0
|
|
|
|
|
0
|
while ($j < $n) { |
399
|
0
|
|
|
|
|
0
|
$t = floor($rng->rand($j + 1)); |
400
|
0
|
0
|
|
|
|
0
|
if (!exists($set{$t + $start})) { |
401
|
0
|
|
|
|
|
0
|
$set{$t + $start} = undef; |
402
|
|
|
|
|
|
|
} else { |
403
|
0
|
|
|
|
|
0
|
$set{$j + $start} = undef; |
404
|
|
|
|
|
|
|
} |
405
|
0
|
|
|
|
|
0
|
++$j; |
406
|
|
|
|
|
|
|
} |
407
|
|
|
|
|
|
|
# die "Floyd didn't pick $k elements" if ASSERT and $k != (keys %set); |
408
|
0
|
|
|
|
|
0
|
return keys %set; |
409
|
|
|
|
|
|
|
} |
410
|
|
|
|
|
|
|
|
411
|
|
|
|
|
|
|
# Routine to calculate the auxiliary block -> message block* mapping. |
412
|
|
|
|
|
|
|
# The passed rng object must already have been seeded, and both sender |
413
|
|
|
|
|
|
|
# and receiver should use the same seed. Returns [[..],[..],..] |
414
|
|
|
|
|
|
|
# representing which message blocks each of the auxiliary block is |
415
|
|
|
|
|
|
|
# composed of. |
416
|
|
|
|
|
|
|
|
417
|
|
|
|
|
|
|
sub auxiliary_mapping { |
418
|
|
|
|
|
|
|
|
419
|
0
|
|
|
0
|
0
|
0
|
my $self = shift; |
420
|
0
|
|
|
|
|
0
|
my $rng = shift; |
421
|
|
|
|
|
|
|
|
422
|
0
|
0
|
|
|
|
0
|
croak "auxiliary_mapping: rng is not a reference\n" unless ref($rng); |
423
|
|
|
|
|
|
|
|
424
|
|
|
|
|
|
|
# hash slices: powerful, but syntax is sometimes confusing |
425
|
0
|
|
|
|
|
0
|
my ($mblocks,$ablocks,$q) = @{$self}{"mblocks","ablocks","q"}; |
|
0
|
|
|
|
|
0
|
|
426
|
|
|
|
|
|
|
|
427
|
0
|
|
|
|
|
0
|
my $aux_mapping = []; |
428
|
|
|
|
|
|
|
|
429
|
0
|
|
|
|
|
0
|
my $ab_string = pack "L*", ($mblocks .. $mblocks + $ablocks -1); |
430
|
|
|
|
|
|
|
|
431
|
|
|
|
|
|
|
# list of empty hashes |
432
|
0
|
|
|
|
|
0
|
my @hashes; |
433
|
0
|
|
|
|
|
0
|
for (0 .. $mblocks + $ablocks -1) { $hashes[$_] = {}; } |
|
0
|
|
|
|
|
0
|
|
434
|
|
|
|
|
|
|
|
435
|
|
|
|
|
|
|
# Use an unrolled version of Floyd's algorithm for the default case |
436
|
|
|
|
|
|
|
# where q=3 |
437
|
0
|
0
|
|
|
|
0
|
if ($q == 3) { |
438
|
0
|
|
|
|
|
0
|
my ($a,$b,$c); |
439
|
0
|
|
|
|
|
0
|
for my $msg (0 .. $mblocks - 1) { |
440
|
0
|
|
|
|
|
0
|
$a = $mblocks + floor($rng->rand($ablocks - 2)); |
441
|
0
|
|
|
|
|
0
|
$hashes[$a] ->{$msg}=undef; |
442
|
0
|
|
|
|
|
0
|
$hashes[$msg]->{$a} =undef; |
443
|
0
|
|
|
|
|
0
|
$b = $mblocks + floor($rng->rand($ablocks - 1)); |
444
|
0
|
0
|
|
|
|
0
|
$b = $mblocks + $ablocks - 2 if $b == $a; |
445
|
0
|
|
|
|
|
0
|
$hashes[$b] ->{$msg}=undef; |
446
|
0
|
|
|
|
|
0
|
$hashes[$msg]->{$b} =undef; |
447
|
0
|
|
|
|
|
0
|
$c = $mblocks + floor($rng->rand($ablocks)); |
448
|
0
|
0
|
0
|
|
|
0
|
$c = $mblocks + $ablocks - 1 if $c == $a or $c == $b; |
449
|
0
|
|
|
|
|
0
|
$hashes[$c] ->{$msg}=undef; |
450
|
0
|
|
|
|
|
0
|
$hashes[$msg]->{$c} =undef; |
451
|
|
|
|
|
|
|
} |
452
|
|
|
|
|
|
|
} else { |
453
|
0
|
|
|
|
|
0
|
for my $msg (0 .. $mblocks - 1) { |
454
|
0
|
|
|
|
|
0
|
foreach my $aux (floyd($rng, $mblocks, $ablocks, $q)) { |
455
|
0
|
|
|
|
|
0
|
$hashes[$aux]->{$msg}=undef; |
456
|
0
|
|
|
|
|
0
|
$hashes[$msg]->{$aux}=undef; |
457
|
|
|
|
|
|
|
} |
458
|
|
|
|
|
|
|
} |
459
|
|
|
|
|
|
|
} |
460
|
|
|
|
|
|
|
|
461
|
|
|
|
|
|
|
# convert list of hashes into a list of lists |
462
|
0
|
|
|
|
|
0
|
for my $i (0 .. $mblocks + $ablocks -1) { |
463
|
0
|
|
|
|
|
0
|
print "map $i: " . (join " ", keys %{$hashes[$i]}) . "\n" if DEBUG; |
464
|
0
|
|
|
|
|
0
|
push @$aux_mapping, [ keys %{$hashes[$i]} ]; |
|
0
|
|
|
|
|
0
|
|
465
|
|
|
|
|
|
|
} |
466
|
|
|
|
|
|
|
|
467
|
|
|
|
|
|
|
# save and return aux_mapping |
468
|
0
|
|
|
|
|
0
|
$self->{aux_mapping} = $aux_mapping; |
469
|
|
|
|
|
|
|
} |
470
|
|
|
|
|
|
|
|
471
|
|
|
|
|
|
|
# Until I get the auto expand_aux working, this will have to do |
472
|
|
|
|
|
|
|
sub blklist_to_msglist { |
473
|
|
|
|
|
|
|
|
474
|
0
|
|
|
0
|
0
|
0
|
my ($self,@xor_list) = @_; |
475
|
|
|
|
|
|
|
|
476
|
0
|
|
|
|
|
0
|
my $mblocks = $self->{mblocks}; |
477
|
|
|
|
|
|
|
|
478
|
0
|
|
|
|
|
0
|
my %blocks; |
479
|
0
|
|
|
|
|
0
|
while (@xor_list) { |
480
|
0
|
|
|
|
|
0
|
my $entry = shift(@xor_list); |
481
|
0
|
0
|
|
|
|
0
|
if ($entry < $mblocks) { # is it a message block index? |
482
|
|
|
|
|
|
|
# toggle entry in the hash |
483
|
0
|
0
|
|
|
|
0
|
if (exists($blocks{$entry})) { |
484
|
0
|
|
|
|
|
0
|
delete $blocks{$entry}; |
485
|
|
|
|
|
|
|
} else { |
486
|
0
|
|
|
|
|
0
|
$blocks{$entry}= undef; |
487
|
|
|
|
|
|
|
} |
488
|
|
|
|
|
|
|
} else { |
489
|
|
|
|
|
|
|
# aux block : push all message blocks it's composed of |
490
|
0
|
|
|
|
|
0
|
my @expansion = @{$self->{aux_mapping}->[$entry]}; |
|
0
|
|
|
|
|
0
|
|
491
|
0
|
|
|
|
|
0
|
if (DEBUG) { |
492
|
|
|
|
|
|
|
print "expand_aux: expanding $entry to " . |
493
|
|
|
|
|
|
|
(join " ", @expansion) . "\n"; |
494
|
|
|
|
|
|
|
} |
495
|
0
|
|
|
|
|
0
|
push @xor_list, @expansion; |
496
|
|
|
|
|
|
|
} |
497
|
|
|
|
|
|
|
} |
498
|
0
|
|
|
|
|
0
|
return keys %blocks; |
499
|
|
|
|
|
|
|
} |
500
|
|
|
|
|
|
|
|
501
|
|
|
|
|
|
|
# Calculate the composition of a single check block based on the |
502
|
|
|
|
|
|
|
# supplied RNG. Returns a reference to a list of composite blocks |
503
|
|
|
|
|
|
|
# indices. |
504
|
|
|
|
|
|
|
|
505
|
|
|
|
|
|
|
sub checkblock_mapping { |
506
|
|
|
|
|
|
|
|
507
|
0
|
|
|
0
|
0
|
0
|
my $self = shift; |
508
|
0
|
|
|
|
|
0
|
my $rng = shift; |
509
|
|
|
|
|
|
|
|
510
|
0
|
0
|
|
|
|
0
|
croak "rng is not an object reference\n" unless ref($rng); |
511
|
|
|
|
|
|
|
|
512
|
0
|
|
|
|
|
0
|
my ($mblocks,$coblocks,$P) = @{$self}{"mblocks","coblocks","P"}; |
|
0
|
|
|
|
|
0
|
|
513
|
0
|
|
|
|
|
0
|
my @coblocks; |
514
|
|
|
|
|
|
|
|
515
|
|
|
|
|
|
|
# use weighted distribution to find how many blocks to link |
516
|
0
|
|
|
|
|
0
|
my $i = 0; |
517
|
0
|
|
|
|
|
0
|
my $r = $rng->rand; |
518
|
0
|
|
|
|
|
0
|
++$i while($r > $P->[$i]); # terminates since r < P[last] |
519
|
0
|
|
|
|
|
0
|
++$i; |
520
|
|
|
|
|
|
|
|
521
|
|
|
|
|
|
|
# select i composite blocks uniformly |
522
|
0
|
|
|
|
|
0
|
@coblocks = floyd($rng, 0, $coblocks , $i); |
523
|
|
|
|
|
|
|
|
524
|
0
|
|
|
|
|
0
|
if (ASSERT) { |
525
|
0
|
0
|
|
|
|
0
|
die "checkblock_mapping: created empty check block\n!" unless @coblocks; |
526
|
|
|
|
|
|
|
} |
527
|
|
|
|
|
|
|
|
528
|
0
|
|
|
|
|
0
|
print "CHECKblock mapping: " . (join " ", @coblocks) . "\n" if DEBUG; |
529
|
|
|
|
|
|
|
|
530
|
0
|
|
|
|
|
0
|
return \@coblocks; |
531
|
|
|
|
|
|
|
|
532
|
|
|
|
|
|
|
} |
533
|
|
|
|
|
|
|
|
534
|
|
|
|
|
|
|
# non-method sub for xoring a source string (passed by reference) with |
535
|
|
|
|
|
|
|
# one or more target strings. I may reimplement this using XS later to |
536
|
|
|
|
|
|
|
# make it more efficient, but will keep a pure-perl version with this |
537
|
|
|
|
|
|
|
# name. |
538
|
|
|
|
|
|
|
sub safe_xor_strings { |
539
|
|
|
|
|
|
|
|
540
|
4
|
|
|
4
|
0
|
17
|
my $source = shift; |
541
|
|
|
|
|
|
|
|
542
|
|
|
|
|
|
|
# die if user forgot to pass in a reference (to allow updating) or |
543
|
|
|
|
|
|
|
# called $self->safe_xor_strings by mistake |
544
|
4
|
50
|
|
|
|
14
|
croak "xor_strings: arg 1 should be a reference to a SCALAR!\n" |
545
|
|
|
|
|
|
|
unless ref($source) eq "SCALAR"; |
546
|
|
|
|
|
|
|
|
547
|
4
|
|
|
|
|
6
|
my $len = length ($$source); |
548
|
|
|
|
|
|
|
|
549
|
4
|
50
|
|
|
|
9
|
croak "xor_strings: source string can't have zero length!\n" |
550
|
|
|
|
|
|
|
unless $len; |
551
|
|
|
|
|
|
|
|
552
|
4
|
|
|
|
|
9
|
foreach my $target (@_) { |
553
|
4
|
50
|
|
|
|
11
|
croak "xor_strings: targets not all same size as source\n" |
554
|
|
|
|
|
|
|
unless length($target) == $len; |
555
|
4
|
|
|
|
|
11
|
map { substr ($$source, $_, 1) ^= substr ($target, $_, 1) } |
|
28
|
|
|
|
|
79
|
|
556
|
|
|
|
|
|
|
(0 .. $len-1); |
557
|
|
|
|
|
|
|
} |
558
|
|
|
|
|
|
|
|
559
|
4
|
|
|
|
|
18
|
return $$source; |
560
|
|
|
|
|
|
|
} |
561
|
|
|
|
|
|
|
|
562
|
|
|
|
|
|
|
# Later, xor_strings could be replaced with an C version with reduced |
563
|
|
|
|
|
|
|
# error checking, so make a backward-compatible version and an |
564
|
|
|
|
|
|
|
# explicit fast/unsafe version. |
565
|
0
|
|
|
0
|
0
|
|
sub xor_strings { safe_xor_strings(@_) } |
566
|
|
|
|
|
|
|
#sub fast_xor_strings { safe_xor_strings(@_) } # implemented in OnlineCode.xs. |
567
|
|
|
|
|
|
|
|
568
|
|
|
|
|
|
|
|
569
|
|
|
|
|
|
|
1; |
570
|
|
|
|
|
|
|
|
571
|
|
|
|
|
|
|
__END__ |