-
Notifications
You must be signed in to change notification settings - Fork 34
/
sbwt_file_compression.pl
224 lines (169 loc) · 5.48 KB
/
sbwt_file_compression.pl
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
#!/usr/bin/perl
# Author: Daniel "Trizen" Șuteu
# Date: 09 November 2024
# https://github.com/trizen
# Compress/decompress files using SWAP transform + LZB + Burrows-Wheeler Transform (BWT) + Move-to-Front Transform + Run-length encoding + Huffman coding.
# Reference:
# Data Compression (Summer 2023) - Lecture 13 - BZip2
# https://youtube.com/watch?v=cvoZbBZ3M2A
use 5.036;
use Getopt::Std qw(getopts);
use File::Basename qw(basename);
use Compression::Util qw(:all);
use POSIX qw(ceil);
use constant {
PKGNAME => 'SBWT',
VERSION => '0.01',
FORMAT => 'sbwt',
CHUNK_SIZE => 1 << 17,
};
# Container signature
use constant SIGNATURE => uc(FORMAT) . chr(1);
sub usage {
my ($code) = @_;
print <<"EOH";
usage: $0 [options] [input file] [output file]
options:
-e : extract
-i <filename> : input filename
-o <filename> : output filename
-r : rewrite output
-v : version number
-h : this message
examples:
$0 document.txt
$0 document.txt archive.${\FORMAT}
$0 archive.${\FORMAT} document.txt
$0 -e -i archive.${\FORMAT} -o document.txt
EOH
exit($code // 0);
}
sub version {
printf("%s %s\n", PKGNAME, VERSION);
exit;
}
sub valid_archive {
my ($fh) = @_;
if (read($fh, (my $sig), length(SIGNATURE), 0) == length(SIGNATURE)) {
$sig eq SIGNATURE || return;
}
return 1;
}
sub main {
my %opt;
getopts('ei:o:vhr', \%opt);
$opt{h} && usage(0);
$opt{v} && version();
my ($input, $output) = @ARGV;
$input //= $opt{i} // usage(2);
$output //= $opt{o};
my $ext = qr{\.${\FORMAT}\z}io;
if ($opt{e} || $input =~ $ext) {
if (not defined $output) {
($output = basename($input)) =~ s{$ext}{}
|| die "$0: no output file specified!\n";
}
if (not $opt{r} and -e $output) {
print "'$output' already exists! -- Replace? [y/N] ";
<STDIN> =~ /^y/i || exit 17;
}
decompress_file($input, $output)
|| die "$0: error: decompression failed!\n";
}
elsif ($input !~ $ext || (defined($output) && $output =~ $ext)) {
$output //= basename($input) . '.' . FORMAT;
compress_file($input, $output)
|| die "$0: error: compression failed!\n";
}
else {
warn "$0: don't know what to do...\n";
usage(1);
}
}
sub swap_transform ($text, $extra = 1) {
my @bits;
my @arr = unpack('C*', $text);
my $k = 0;
foreach my $i (1 .. $#arr) {
if ($arr[$i] < $arr[$i - 1 - $k]) {
push @bits, 1;
unshift @arr, splice(@arr, $i, 1);
++$k if $extra;
}
else {
push @bits, 0;
}
}
return (pack('C*', @arr), \@bits);
}
sub reverse_swap_transform ($text, $bits) {
my @arr = unpack('C*', $text);
for (my $i = $#arr ; $i >= 0 ; --$i) {
if ($bits->[$i - 1] == 1) {
splice(@arr, $i, 0, shift(@arr));
}
}
pack('C*', @arr);
}
# Compress file
sub compress_file ($input, $output) {
open my $fh, '<:raw', $input
or die "Can't open file <<$input>> for reading: $!";
my $header = SIGNATURE;
# Open the output file for writing
open my $out_fh, '>:raw', $output
or die "Can't open file <<$output>> for write: $!";
# Print the header
print $out_fh $header;
# Compress data
while (read($fh, (my $chunk), CHUNK_SIZE)) {
local $Compression::Util::LZ_MIN_LEN = 512;
my ($t, $bits) = swap_transform(lzb_compress($chunk, \&lzss_encode_fast), 0);
my $vrle_bits = binary_vrl_encode(join('', @$bits));
if (length($vrle_bits) < scalar @$bits) {
say STDERR "With VLRE: ", length($vrle_bits), " < ", scalar(@$bits);
print $out_fh chr(1);
}
else {
say STDERR "Without VRLE: ", length($vrle_bits), " > ", scalar(@$bits);
$vrle_bits = join('', @$bits);
print $out_fh chr(0);
}
print $out_fh pack('N', length $vrle_bits);
my ($bwt, $idx) = bwt_encode($t);
print $out_fh pack('B*', $vrle_bits);
my ($mtf, $alphabet) = mtf_encode(string2symbols($bwt));
my $rle = zrle_encode($mtf);
print $out_fh (pack('N', $idx) . encode_alphabet($alphabet) . create_huffman_entry($rle));
}
# Close the file
close $out_fh;
}
# Decompress file
sub decompress_file ($input, $output) {
# Open and validate the input file
open my $fh, '<:raw', $input
or die "Can't open file <<$input>> for reading: $!";
valid_archive($fh) || die "$0: file `$input' is not a \U${\FORMAT}\E v${\VERSION} archive!\n";
# Open the output file
open my $out_fh, '>:raw', $output
or die "Can't open file <<$output>> for writing: $!";
while (!eof($fh)) {
my $with_vrle = ord(getc($fh));
my $bits_len = bytes2int($fh, 4);
my $bits = read_bits($fh, $bits_len);
$bits = binary_vrl_decode($bits) if $with_vrle;
my $idx = bytes2int($fh, 4);
my $alphabet = decode_alphabet($fh);
my $rle = decode_huffman_entry($fh);
my $mtf = zrle_decode($rle);
my $bwt = mtf_decode($mtf, $alphabet);
my $data = bwt_decode(pack('C*', @$bwt), $idx);
print $out_fh lzb_decompress(reverse_swap_transform($data, [split(//, $bits)]));
}
# Close the file
close $fh;
close $out_fh;
}
main();
exit(0);