LZW compression: Difference between revisions
Applesoft BASIC
(Applesoft BASIC) |
|||
(120 intermediate revisions by 55 users not shown) | |||
Line 1:
{{task|Compression}}
The Lempel-Ziv-Welch (LZW) algorithm provides loss-less data compression.
You can read a complete description of it in the [[wp:Lempel-Ziv-Welch|Wikipedia article]] on the subject. It was patented, but it entered the public domain in 2004.
<br><br>
=={{header|11l}}==
{{trans|Python}}
<syntaxhighlight lang="11l">F compress(uncompressed)
V dict_size = 256
V dictionary = Dict((0 .< dict_size).map(i -> (String(Char(code' i)), i)))
V w = ‘’
[Int] result
L(c) uncompressed
V wc = w‘’c
I wc C dictionary
w = wc
E
result.append(dictionary[w])
dictionary[wc] = dict_size
dict_size++
w = c
I !w.empty
result.append(dictionary[w])
R result
F decompress([Int] &compressed)
V dict_size = 256
V dictionary = Dict((0 .< dict_size).map(i -> (i, String(Char(code' i)))))
V result = ‘’
V w = String(Char(code' compressed.pop(0)))
result ‘’= w
L(k) compressed
V entry = ‘’
I k C dictionary
entry = dictionary[k]
E I k == dict_size
entry = w‘’w[0]
E
exit(‘Bad compressed k: ’k)
result ‘’= entry
dictionary[dict_size] = w‘’entry[0]
dict_size++
w = entry
R result
V compressed = compress(‘TOBEORNOTTOBEORTOBEORNOT’)
print(compressed)
print(decompress(&compressed))</syntaxhighlight>
=={{header|Ada}}==
Line 8 ⟶ 57:
lzw.ads:
<
MAX_CODE : constant := 4095;
Line 18 ⟶ 67:
function Decompress (Data : in Compressed_Data) return String;
end LZW;</
lzw.adb:
<
with Ada.Strings.Unbounded;
Line 136 ⟶ 185:
end Decompress;
end LZW;</
test.adb:
<
with Ada.Text_IO;
Line 159 ⟶ 208:
Text_IO.Put_Line (Cleartext);
end;
end Test;</
=={{header|Applesoft BASIC}}==
{{trans|BBC BASIC}}
<syntaxhighlight lang="BASIC"> 0 PLAIN$ = "TOBEORNOTTOBEORTOBEORNOT"
10 GOSUB 200"ENCODE PLAIN$
20 FOR I = 1 TO LEN (ENCODE$) STEP 2
30 PRINT S$ ASC ( MID$ (ENCODE$,I)) + 256 * ASC ( MID$ (ENCODE$,I + 1));
40 LET S$ = " "
50 NEXT
60 PRINT
70 GOSUB 300"DECODE ENCODE$
80 PRINT PLAIN$;
90 END
100 FOR C = 0 TO 1E9 STEP 0
110 IF I > S THEN RETURN
120 FOR D = 1 TO L - 1
130 IF W$ < > DICT$(D) THEN NEXT D
140 IF D > = L THEN RETURN
150 LET I = I + 1
160 LET W$ = W$ + MID$ (PLAIN$,I,1)
170 LET C = D
180 NEXT C
190 RETURN
REM ENCODE PLAIN$ RETURN ENCODE$
200 IF NOT DI THEN DIM DICT$(4095)
210 FOR I = 0 TO 255:DICT$(I) = CHR$ (I): NEXT
220 LET DI = 1 : L = I : S = LEN (PLAIN$):ENCODE$ = ""
230 LET W$ = LEFT$ (PLAIN$,1)
240 FOR I = 1 TO 1E9 STEP 0
250 GOSUB 100
260 LET DICT$(L) = W$:L = L + 1:W$ = RIGHT$ (W$,1)
270 LET C% = C / 256:ENCODE$ = ENCODE$ + CHR$ (C - C% * 256) + CHR$ (C%)
280 IF I < = S THEN NEXT I
290 RETURN
REM DECODE ENCODE$ RETURN PLAIN$
300 IF NOT DI THEN DIM DICT$(4095)
310 FOR I = 0 TO 255:DICT$(I) = CHR$ (I): NEXT
320 LET DI = 1 : L = I
330 FOR I = L TO 4095:DICT$(I) = "": NEXT
340 LET C = ASC (ENCODE$) + 256 * ASC ( MID$ (ENCODE$,2))
350 LET W$ = DICT$(C)
360 LET PLAIN$ = W$
370 LET S = LEN (ENCODE$)
380 IF S < 4 THEN RETURN
400 FOR I = 3 TO S STEP 2
410 LET C = ASC ( MID$ (ENCODE$,I)) + 256 * ASC ( MID$ (ENCODE$,I + 1))
420 IF C < L THEN T$ = DICT$(C)
430 IF C > = L THEN T$ = W$ + LEFT$ (W$,1)
440 LET PLAIN$ = PLAIN$ + T$
450 LET DICT$(L) = W$ + LEFT$ (T$,1)
460 LET L = L + 1
470 LET W$ = T$
480 NEXT
490 RETURN</syntaxhighlight>
{{out}}
<pre>84 79 66 69 79 82 78 79 84 256 258 260 265 259 261 263
TOBEORNOTTOBEORTOBEORNOT</pre>
=={{header|AWK}}==
copy the following to standard input
<syntaxhighlight>
==comp
TOBEORNOTTOBEORTOBEORNOT
To be or not to be that is the question!
"There is nothing permanent except change." --- Heraclitus [540 -- 475 BCE]
==decomp
84,79,66,69,79,82,78,79,84,256,258,260,265,259,261,263
84,111,32,98,101,32,111,114,32,110,111,116,32,116,257,259,268,104,97,267,105,115,272,260,113,117,101,115,116,105,111,110,33
34,84,104,101,114,101,32,105,115,32,110,111,116,104,105,110,103,32,112,259,109,97,110,101,110,116,32,101,120,99,101,112,281,99,104,277,103,101,46,34,32,296,45,298,296,32,72,259,97,99,108,105,116,117,264,32,91,53,52,48,32,299,52,55,53,32,66,67,69,93
</syntaxhighlight>
<syntaxhighlight lang="AWK">
# ported from Python
BEGIN {
is_comp = 0
is_decomp = 0
}
{
if ($0 == "") next
if ($1 == "==comp") {
is_comp = 1
is_decomp = 0
print "\ncompressing..."
next
}
if ($1 == "==decomp") {
is_comp = 0
is_decomp = 1
print "\ndecompressing..."
next
}
if (is_comp) print compress($0)
if (is_decomp) print decompress($0)
}
function compress(str, dict_size, i, dictionary, w, result, len, uncompressed, c, wc ) {
dict_size = 256
for (i = 0; i <= dict_size; i++) dictionary[chr(i)] = i
w = ""
result = ""
len = split(str, uncompressed, "")
for (i = 1; i <= len; i++) {
c = uncompressed[i]
wc = w c
if (wc in dictionary) w = wc
else {
result = result "," dictionary[w]
dictionary[wc] = dict_size++
w = c
}
}
if (length(w)) result = result "," dictionary[w]
return substr(result,2)
return "[" substr(result,2) "]"
}
function decompress(str) {
dict_size = 256
for (i = 0; i <= dict_size; i++) dictionary[i] = chr(i)
result = ""
len = split(str, compressed, ",")
w = chr(compressed[1])
result = result w
for (i = 2; i <= len; i++) {
k = compressed[i]
if (k in dictionary) entry = dictionary[k]
else if (k == dict_size) entry = w substr(w,1,1)
else { entry = "*" }
result = result entry
dictionary[dict_size++] = w substr(entry,1,1)
w = entry
}
return result
}
function chr(c) {
return sprintf("%c", c + 0)
}</syntaxhighlight>
=={{header|Arturo}}==
<syntaxhighlight lang="rebol">compress: function [str][
dict: #[]
loop 0..255 'i -> dict\[to :char i]: i
w: ""
result: new []
loop str 'c [
wc: w ++ c
if? key? dict wc -> w: wc
else [
'result ++ dict\[w]
dict\[wc]: size dict
w: to :string c
]
]
if 0 < size w -> 'result ++ dict\[w]
return result
]
decompress: function [compressed][
dict: #[]
arr: new compressed
loop 0..255 'i -> dict\[i]: to :string to :char i
w: dict\[first arr]
remove 'arr .index 0
result: w
loop arr 'k [
entry: ""
if? key? dict k -> entry: dict\[k]
else [
if? k = size dict -> entry: w ++ first w
else -> panic ~"Error with compressed: |k|"
]
'result ++ entry
dict\[size dict]: w ++ first entry
w: entry
]
return result
]
compressed: compress "TOBEORNOTTOBEORTOBEORNOT"
print "Compressed:"
print compressed
print ""
decompressed: decompress compressed
print "Decompressed:"
print decompressed</syntaxhighlight>
{{out}}
<pre>Compressed:
84 79 66 69 79 82 78 79 84 256 258 260 265 259 261 263
Decompressed:
TOBEORNOTOBEORNOTOOBEORTOOBEORNOT</pre>
=={{header|BaCon}}==
<syntaxhighlight lang="bacon">CONST lzw_data$ = "TOBEORNOTTOBEORTOBEORNOT"
PRINT "LZWData: ", lzw_data$
encoded$ = Encode_LZW$(lzw_data$)
PRINT "Encoded: ", encoded$
PRINT "Decoded: ", Decode_LZW$(encoded$)
'----------------------------------------------------------
FUNCTION Encode_LZW$(sample$)
LOCAL dict ASSOC int
LOCAL ch$, buf$, result$
LOCAL nr, x
FOR nr = 0 TO 255
dict(CHR$(nr)) = nr
NEXT
FOR x = 1 TO LEN(sample$)
ch$ = MID$(sample$, x, 1)
IF dict(buf$ & ch$) THEN
buf$ = buf$ & ch$
ELSE
result$ = APPEND$(result$, 0, STR$(dict(buf$)))
dict(buf$ & ch$) = nr
INCR nr
buf$ = ch$
END IF
NEXT
result$ = APPEND$(result$, 0, STR$(dict(buf$)))
RETURN result$
END FUNCTION
'----------------------------------------------------------
FUNCTION Decode_LZW$(sample$)
LOCAL list$ ASSOC STRING
LOCAL old$, ch$, x$, out$, result$
LOCAL nr
FOR nr = 0 TO 255
list$(STR$(nr)) = CHR$(nr)
NEXT
old$ = TOKEN$(sample$, 1)
ch$ = list$(old$)
result$ = ch$
FOR x$ IN LAST$(sample$, 1)
IF NOT(LEN(list$(x$))) THEN
out$ = list$(old$)
out$ = out$ & ch$
ELSE
out$ = list$(x$)
END IF
result$ = result$ & out$
ch$ = LEFT$(out$, 1)
list$(STR$(nr)) = list$(old$) & ch$
INCR nr
old$ = x$
NEXT
RETURN result$
END FUNCTION</syntaxhighlight>
{{out}}
<pre>LZWData: TOBEORNOTTOBEORTOBEORNOT
Encoded: 84 79 66 69 79 82 78 79 84 256 258 260 265 259 261 263
Decoded: TOBEORNOTTOBEORTOBEORNOT
</pre>
=={{header|BBC BASIC}}==
{{works with|BBC BASIC for Windows}}
Uses fixed bit-width (16 bits) and initial dictionary size = 256.
<
encodeLZW$ = FNencodeLZW(plaintext$)
FOR i% = 1 TO LEN(encodeLZW$) STEP 2
Line 211 ⟶ 555:
w$ = t$
NEXT
= o$</
{{out}}
<pre>
Line 223 ⟶ 567:
Code 256 for clear table, 257 for end of data,
everything else are either byte values (<256) or code values.
'''WARNING: This code appears to have come from a GIF codec that has been modified to meet the requirements of this page, provided that the decoder works with the encoder to produce correct output. For writing GIF files the write_bits subroutine is wrong for Little Endian systems (it may be wrong for Big Endian as well.) The encoder also increases the number of bits in the variable length GIF-LZW after the N-2 code, whereas this must be done after N-1 to produce a working GIF file (just looking at the encoder, it's easy to see how this mistake could be made.)'''
<syntaxhighlight lang="c">#include <stdio.h>
#include <stdlib.h>
#include <string.h>
Line 235 ⟶ 582:
void* mem_alloc(size_t item_size, size_t n_item)
{
}
void* mem_extend(void *m, size_t new_n)
{
}
inline void _clear(void *m)
{
}
#define _new(type, n)
#define _del(m)
#define _len(m)
#define _setsize(m, n)
#define _extend(m)
Line 268 ⟶ 615:
typedef uint16_t ushort;
#define M_CLR
#define M_EOD
#define M_NEW
/* encode and decode dictionary structures.
Line 277 ⟶ 624:
then dict[97].next['b'] = 387, dict[387].next['c'] = 1022, etc. */
typedef struct {
} lzw_enc_t;
/* for decoding, dictionary contains index of whatever prefix index plus trailing
byte. i.e. like previous example,
the "back" element is used for temporarily chaining indices when resolving
a code to bytes
*/
typedef struct {
} lzw_dec_t;
byte* lzw_encode(byte *in, int max_bits)
{
}
}
}
}
byte* lzw_decode(byte *in)
{
}
}
}
}
}
}
}
}
bail:
}
int main()
{
}
}</
=={{header|C sharp}}==
{{trans|Java}}
<syntaxhighlight lang="c sharp">using System;
using System.Collections.Generic;
using System.Text;
namespace LZW
{
public class Program
{
public static void Main(string[] args)
{
List<int> compressed = Compress("TOBEORNOTTOBEORTOBEORNOT");
Console.WriteLine(string.Join(", ", compressed));
string decompressed = Decompress(compressed);
Console.WriteLine(decompressed);
}
public static List<int> Compress(string uncompressed)
{
// build the dictionary
Dictionary<string, int> dictionary = new Dictionary<string, int>();
for (int i = 0; i < 256; i++)
dictionary.Add(((char)i).ToString(), i);
string w = string.Empty;
List<int> compressed = new List<int>();
foreach (char c in uncompressed)
{
string wc = w + c;
if (dictionary.ContainsKey(wc))
{
w = wc;
}
else
{
// write w to output
compressed.Add(dictionary[w]);
// wc is a new sequence; add it to the dictionary
dictionary.Add(wc, dictionary.Count);
w = c.ToString();
}
}
// write remaining output if necessary
if (!string.IsNullOrEmpty(w))
compressed.Add(dictionary[w]);
return compressed;
}
public static string Decompress(List<int> compressed)
{
// build the dictionary
Dictionary<int, string> dictionary = new Dictionary<int, string>();
for (int i = 0; i < 256; i++)
dictionary.Add(i, ((char)i).ToString());
string w = dictionary[compressed[0]];
compressed.RemoveAt(0);
StringBuilder decompressed = new StringBuilder(w);
foreach (int k in compressed)
{
string entry = null;
if (dictionary.ContainsKey(k))
entry = dictionary[k];
else if (k == dictionary.Count)
entry = w + w[0];
decompressed.Append(entry);
// new sequence; add it to the dictionary
dictionary.Add(dictionary.Count, w + entry[0]);
w = entry;
}
return decompressed.ToString();
}
}
}</syntaxhighlight>
{{out}}
<pre>84, 79, 66, 69, 79, 82, 78, 79, 84, 256, 258, 260, 265, 259, 261, 263
TOBEORNOTTOBEORTOBEORNOT</pre>
=={{header|C++}}==
{{trans|D}}
<syntaxhighlight lang="cpp">#include <string>
#include <map>
// Compress a string to a list of output symbols.
// The result will be written to the output iterator
// starting at "result"; the final iterator is returned.
template <typename Iterator>
Iterator compress(const std::string &uncompressed, Iterator result) {
// Build the dictionary.
int dictSize = 256;
std::map<std::string,int> dictionary;
for (int i = 0; i < 256; i++)
dictionary[std::string(1, i)] = i;
std::string w;
for (std::string::const_iterator it = uncompressed.begin();
it != uncompressed.end(); ++it) {
char c = *it;
std::string wc = w + c;
if (dictionary.count(wc))
w = wc;
else {
*result++ = dictionary[w];
// Add wc to the dictionary.
dictionary[wc] = dictSize++;
w = std::string(1, c);
}
}
// Output the code for w.
if (!w.empty())
*result++ = dictionary[w];
return result;
}
// Decompress a list of output ks to a string.
// "begin" and "end" must form a valid range of ints
template <typename Iterator>
std::string decompress(Iterator begin, Iterator end) {
// Build the dictionary.
int dictSize = 256;
std::map<int,std::string> dictionary;
for (int i = 0; i < 256; i++)
dictionary[i] = std::string(1, i);
std::string w(1, *begin++);
std::string result = w;
std::string entry;
for ( ; begin != end; begin++) {
int k = *begin;
if (dictionary.count(k))
entry = dictionary[k];
else if (k == dictSize)
entry = w + w[0];
else
throw "Bad compressed k";
result += entry;
// Add w+entry[0] to the dictionary.
dictionary[dictSize++] = w + entry[0];
w = entry;
}
return result;
}
#include <iostream>
#include <iterator>
#include <vector>
int main() {
std::vector<int> compressed;
compress("TOBEORNOTTOBEORTOBEORNOT", std::back_inserter(compressed));
copy(compressed.begin(), compressed.end(), std::ostream_iterator<int>(std::cout, ", "));
std::cout << std::endl;
std::string decompressed = decompress(compressed.begin(), compressed.end());
std::cout << decompressed << std::endl;
return 0;
}</syntaxhighlight>
=={{header|Clojure}}==
<syntaxhighlight lang="lisp">(defn make-dict []
(let [vals (range 0 256)]
(zipmap (map (comp #'list #'char) vals) vals)))
(defn compress [#^String text]
(loop [t (seq text)
r '()
w '()
dict (make-dict)
s 256]
(let [c (first t)]
(if c
(let [wc (cons c w)]
(if (get dict wc)
(recur (rest t) r wc dict s)
(recur (rest t) (cons (get dict w) r) (list c) (assoc dict wc s) (inc s))))
(reverse (if w (cons (get dict w) r) r))))))
(compress "TOBEORNOTTOBEORTOBEORNOT")</syntaxhighlight>
{{out}}
<syntaxhighlight lang="lisp">(84 79 66 69 79 82 78 79 84 256 258 260 265 259 261 263)</syntaxhighlight>
=={{header|CoffeeScript}}==
This only does the encoding step for now.
<
lzw = (s) ->
dct = {} # map substrings to codes between 256 and 4096
Line 522 ⟶ 1,064:
console.log lzw "TOBEORNOTTOBEORTOBEORNOT"
</syntaxhighlight>
{{out}}
<pre>
Line 554 ⟶ 1,096:
The exact encoding used is dependent upon the user's locale (<code>LC_CTYPE</code> on Unix).
<
vector-append))
(defun vector-append (old new &optional (start2 0) end2)
Line 642 ⟶ 1,184:
for entry = (or (gethash k dictionary)
(if (equalp k dictionary-size)
(
(error "bad compresed entry at pos ~S" i)))
do (vector-append result entry)
Line 663 ⟶ 1,205:
(assert (equal #2=(lzw-decompress-to-string (lzw-compress string)) string) ()
"Can't compress ~S properly, got ~S instead" string #2#)
t)</
And the format used:
<
T
CL-USER> (lzw-compress "TOBEORNOTTOBEORTOBEORNOT")
#(84 79 66 69 79 82 78 79 84 256 258 260 265 259 261 263)
CL-USER> (lzw-decompress-to-string *)
"TOBEORNOTTOBEORTOBEORNOT"</
=={{header|D}}==
===Simpler Version===
<
auto compress(in string original) pure nothrow {
Line 909 ⟶ 1,257:
auto comp = "TOBEORNOTTOBEORTOBEORNOT".compress;
writeln(comp, "\n", comp.decompress);
}</
{{out}}
<pre>[84, 79, 66, 69, 79, 82, 78, 79, 84, 256, 258, 260, 265, 259, 261, 263]
Line 916 ⟶ 1,264:
===More Refined Version===
This longer version is a little more efficient and it uses stronger static typing.
<
import std.array: empty;
Line 1,010 ⟶ 1,358:
compressed.writeln;
LZW.decompress(compressed).assumeUTF.writeln;
}</
{{out}}
<pre>[84, 79, 66, 69, 79, 82, 78, 79, 84, 256, 258, 260, 265, 259, 261, 263]
Line 1,018 ⟶ 1,366:
{{trans|C}}
This code retains part of the style of the original C code.
<
CLR = 256, // Clear table marker.
EOD = 257, // End-of-data marker.
Line 1,239 ⟶ 1,587:
"Decoded OK.".writeln;
}</
{{out}}
<pre>Input size: 206403
Line 1,247 ⟶ 1,595:
=={{header|Dylan}}==
<
Synopsis: LZW implementation for Rosetta code
Line 1,289 ⟶ 1,637:
end;
format-out("%=\n", compress("TOBEORNOTTOBEORTOBEORNOT"))</
=={{header|Eiffel}}==
<syntaxhighlight lang="eiffel">
class
APPLICATION
create
make
feature {NONE}
make
local
test: LINKED_LIST [INTEGER]
do
create test.make
test := compress ("TOBEORNOTTOBEORTOBEORNOT")
across
test as t
loop
io.put_string (t.item.out + " ")
end
io.new_line
io.put_string (decompress (test))
end
decompress (compressed: LINKED_LIST [INTEGER]): STRING
--Decompressed version of 'compressed'.
local
dictsize, i, k: INTEGER
dictionary: HASH_TABLE [STRING, INTEGER]
w, entry: STRING
char: CHARACTER_8
do
dictsize := 256
create dictionary.make (300)
create entry.make_empty
create Result.make_empty
from
i := 0
until
i > 256
loop
char := i.to_character_8
dictionary.put (char.out, i)
i := i + 1
end
w := compressed.first.to_character_8.out
compressed.go_i_th (1)
compressed.remove
Result := w
from
k := 1
until
k > compressed.count
loop
if attached dictionary.at (compressed [k]) as ata then
entry := ata
elseif compressed [k] = dictsize then
entry := w + w.at (1).out
else
io.put_string ("EXEPTION")
end
Result := Result + entry
dictsize := dictsize + 1
dictionary.put (w + entry.at (1).out, dictsize)
w := entry
k := k + 1
end
end
compress (uncompressed: STRING): LINKED_LIST [INTEGER]
-- Compressed version of 'uncompressed'.
local
dictsize: INTEGER
dictionary: HASH_TABLE [INTEGER, STRING]
i: INTEGER
w, wc: STRING
char: CHARACTER_8
do
dictsize := 256
create dictionary.make (256)
create w.make_empty
from
i := 0
until
i > 256
loop
char := i.to_character_8
dictionary.put (i, char.out)
i := i + 1
end
create Result.make
from
i := 1
until
i > uncompressed.count
loop
wc := w + uncompressed [i].out
if dictionary.has (wc) then
w := wc
else
Result.extend (dictionary.at (w))
dictSize := dictSize + 1
dictionary.put (dictSize, wc)
w := "" + uncompressed [i].out
end
i := i + 1
end
if w.count > 0 then
Result.extend (dictionary.at (w))
end
end
end
</syntaxhighlight>
{{out}}
<pre>
84 79 66 69 79 82 78 79 84 257 259 261 266 260 262 264
TOBEORNOTTOBEORTOBEORNOT
</pre>
=={{header|Elixir}}==
{{trans|Erlang}}
<syntaxhighlight lang="elixir">defmodule LZW do
@encode_map Enum.into(0..255, Map.new, &{[&1],&1})
@decode_map Enum.into(0..255, Map.new, &{&1,[&1]})
def encode(str), do: encode(to_char_list(str), @encode_map, 256, [])
defp encode([h], d, _, out), do: Enum.reverse([d[[h]] | out])
defp encode([h|t], d, free, out) do
val = d[[h]]
find_match(t, [h], val, d, free, out)
end
defp find_match([h|t], l, lastval, d, free, out) do
case Map.fetch(d, [h|l]) do
{:ok, val} -> find_match(t, [h|l], val, d, free, out)
:error -> d1 = Map.put(d, [h|l], free)
encode([h|t], d1, free+1, [lastval | out])
end
end
defp find_match([], _, lastval, _, _, out), do: Enum.reverse([lastval | out])
def decode([h|t]) do
val = @decode_map[h]
decode(t, val, 256, @decode_map, val)
end
defp decode([], _, _, _, l), do: Enum.reverse(l) |> to_string
defp decode([h|t], old, free, d, l) do
val = if h == free, do: old ++ [List.first(old)], else: d[h]
add = [List.last(val) | old]
d1 = Map.put(d, free, add)
decode(t, val, free+1, d1, val++l)
end
end
str = "TOBEORNOTTOBEORTOBEORNOT"
IO.inspect enc = LZW.encode(str)
IO.inspect dec = LZW.decode(enc)
IO.inspect str == dec</syntaxhighlight>
{{out}}
<pre>
[84, 79, 66, 69, 79, 82, 78, 79, 84, 256, 258, 260, 265, 259, 261, 263]
"TOBEORNOTTOBEORTOBEORNOT"
true
</pre>
=={{header|EMal}}==
<syntaxhighlight lang="emal">
type LzwCompression
fun compress ← List by text uncompressed
List output ← int[]
text working ← Text.EMPTY
Map symbolTable ← text%int[].with(256, <int i|text%int(chr(i) => i))
for each text c in uncompressed
text augmented ← working + c
if symbolTable.has(augmented)
working ← augmented
else
symbolTable.insert(augmented, symbolTable.length)
int i ← symbolTable[working]
output.append(i)
working ← c
end
end
if not working.isEmpty()
int i ← symbolTable[working]
output.append(i)
end
return output
end
fun decompress ← text by List compressed
Map symbolTable ← int%text[].with(256, <int i|int%text(i => chr(i)))
text working ← symbolTable[compressed[0]]
text output ← *working
for each int i in compressed.extract(1)
text s
if symbolTable.has(i)
s ← symbolTable[i]
else if i æ symbolTable.length # cScSc problem
s ← working + working[0]
else
error(65, "Error decompressing")
end
output.append(s)
symbolTable.insert(symbolTable.length, working + s[0])
working ← s
end
return output
end
List compressed = compress("TOBEORNOTTOBEORTOBEORNOT")
writeLine(compressed)
text decompressed = decompress(compressed)
writeLine(decompressed)
</syntaxhighlight>
{{out}}
<pre>
[84,79,66,69,79,82,78,79,84,256,258,260,265,259,261,263]
TOBEORNOTTOBEORTOBEORNOT
</pre>
=={{header|Erlang}}==
<
-export([test/0, encode/1, decode/1]).
Line 1,301 ⟶ 1,873:
Str = "TOBEORNOTTOBEORTOBEORNOT",
[84,79,66,69,79,82,78,79,84,256,258,260,265,259,261,263] =
Str = decode(encode(Str)),
ok.
Line 1,318 ⟶ 1,890:
find_match([H|T], L, LastVal, D, Free, Out) ->
case dict:find([H|L], D) of
end;
find_match([], _, LastVal, _, _, Out) ->
Line 1,348 ⟶ 1,920:
init1(0, D) -> D;
init1(N, D) -> D1 = dict:store(N,[N],D), init1(N-1, D1).</
=={{header|Forth}}==
{{works with|GNU Forth|0.6.2}}
<
\ current string fragment
Line 1,370 ⟶ 1,942:
: free-dict forth-wordlist set-current ;
: in-dict? ( key len -- ? )
dict search-wordlist dup if nip then ;
Line 1,437 ⟶ 2,009:
abort" bad symbol!"
then then
entry count type
entry 1+ c@ w+c
w count add-symbol
Line 1,450 ⟶ 2,022:
out out-size @ decompress cr
\ TOBEORNOTTOBEORTOBEORNOT</
=={{header|Fortran}}==
<syntaxhighlight lang="fortran">
!
! lzw_shared_parameters.f90
!
! LZW Common Variables Used by Coder and Decoder
!
! Author: Pedro Garcia Freitas <sawp@sawp.com.br>
! May, 2011
!
! License: Creative Commons http://creativecommons.org/licenses/by-nc-nd/3.0/
!
MODULE LZW_SHARED_PARAMETERS
IMPLICIT NONE
!
! PARAMETER definitions
!
INTEGER , PARAMETER :: COMPILER_INTEGER_SIZE = 32 , BITS = 12 , FILEIN = 66 , &
& FILEOUT = 99 , MAX_VALUE = (2**BITS) - 1 , &
& MAX_CODE = MAX_VALUE - 1 , MAX_DICTIONARY_SIZE = 5021 , &
& SYMBOL_SIZE = 8 , MISSING_BITS = COMPILER_INTEGER_SIZE - &
& SYMBOL_SIZE
!
! Local variables
!
INTEGER , DIMENSION(0:MAX_DICTIONARY_SIZE) :: concatenatedsymbols
INTEGER , DIMENSION(0:MAX_DICTIONARY_SIZE) :: prefixcodes
INTEGER :: the_status = 0
! change this if compiler dont use 32 bits for integer
END MODULE LZW_SHARED_PARAMETERS
!
! codecIO.f90
!
! bit IO routines for coder and encoder.
!
! Author: Pedro Garcia Freitas <sawp@sawp.com.br>
! May, 2011
!
! License: Creative Commons http://creativecommons.org/licenses/by-nc-nd/3.0/
!
MODULE CODECIO
USE LZW_SHARED_PARAMETERS
IMPLICIT NONE
!
CONTAINS
SUBROUTINE SETOUTPUTCODE(Code)
IMPLICIT NONE
!
! Dummy arguments
!
INTEGER :: Code
INTENT (IN) Code
!
! Local variables
!
INTEGER :: buffer
INTEGER :: outputbitbuffer = 0
INTEGER :: outputbitcount = 0
INTEGER :: shift
INTEGER :: shiftedsymbol
!
shift = COMPILER_INTEGER_SIZE - BITS - outputbitcount
shiftedsymbol = ISHFT(Code , shift)
outputbitbuffer = IOR(outputbitbuffer , shiftedsymbol)
outputbitcount = outputbitcount + BITS
DO WHILE(outputbitcount >= SYMBOL_SIZE)
! IF( outputbitcount<SYMBOL_SIZE )EXIT
buffer = ISHFT(outputbitbuffer , -MISSING_BITS)
CALL SETRAWBYTE(buffer)
outputbitbuffer = ISHFT(outputbitbuffer , SYMBOL_SIZE)
outputbitcount = outputbitcount - SYMBOL_SIZE
END DO
RETURN
END SUBROUTINE SETOUTPUTCODE
SUBROUTINE SETRAWBYTE(Symbol)
IMPLICIT NONE
!
! Dummy arguments
!
INTEGER :: Symbol
INTENT (IN) Symbol
!
CALL FPUTC(FILEOUT , ACHAR(Symbol))
END SUBROUTINE SETRAWBYTE
FUNCTION GETRAWBYTE()
IMPLICIT NONE
!
! Dummy arguments
!
INTEGER :: GETRAWBYTE
!
! Local variables
!
CHARACTER :: bufferedbyte
!
CALL FGETC(FILEIN , bufferedbyte , THE_status)
GETRAWBYTE = IACHAR(bufferedbyte)
END FUNCTION GETRAWBYTE
FUNCTION GETINPUTCODE()
IMPLICIT NONE
!
! Dummy arguments
!
INTEGER :: GETINPUTCODE
!
! Local variables
!
INTEGER :: inputbitbuffer = 0
INTEGER :: inputbitcounter = 0
INTEGER :: integerinputbuff
INTEGER :: returnn
INTEGER :: shiftedbit
!
DO WHILE( inputbitcounter <= MISSING_BITS )
! IF( inputbitcounter>MISSING_BITS )EXIT
integerinputbuff = GETRAWBYTE()
shiftedbit = ISHFT(integerinputbuff , MISSING_BITS - inputbitcounter)
inputbitbuffer = IOR(inputbitbuffer , shiftedbit)
inputbitcounter = inputbitcounter + SYMBOL_SIZE
END DO
returnn = ISHFT(inputbitbuffer , BITS - COMPILER_INTEGER_SIZE)
inputbitbuffer = ISHFT(inputbitbuffer , BITS)
inputbitcounter = inputbitcounter - BITS
GETINPUTCODE = returnn
RETURN
END FUNCTION GETINPUTCODE
end module codecIO
! lzw_encoder.f90
!
! LZW Coder (Compressor)
!
! Author: Pedro Garcia Freitas <sawp@sawp.com.br>
! May, 2011
!
! License: Creative Commons http://creativecommons.org/licenses/by-nc-nd/3.0/
!
MODULE LZW_ENCODER
USE LZW_SHARED_PARAMETERS
USE CODECIO
IMPLICIT NONE
!
! PARAMETER definitions
!
INTEGER , PARAMETER :: HASH_SHIFT = BITS - SYMBOL_SIZE
!
! Local variables
!
INTEGER , DIMENSION(0:MAX_DICTIONARY_SIZE) :: symbolvalues
CONTAINS
SUBROUTINE COMPRESS()
IMPLICIT NONE
!
! Local variables
!
INTEGER :: codedsymbol
INTEGER :: my_index
INTEGER :: nextsymbol
INTEGER :: symbol
CHARACTER :: bufferedbyte
!
nextsymbol = COMPILER_INTEGER_SIZE*SYMBOL_SIZE
SYMbolvalues(:) = -1
! codedsymbol = GETRAWBYTE()
CALL FGETC(FILEIN , bufferedbyte , THE_status)
codedsymbol = IACHAR(bufferedbyte)
!Can be hand optimized - optimization
DO WHILE(THE_status == 0)
! symbol = GETRAWBYTE() ! Manual inline of function
CALL FGETC(FILEIN , bufferedbyte , THE_status)
symbol = IACHAR(bufferedbyte)
IF( THE_status/=0 )CYCLE
my_index = GETPOSITIONONDICTIONARY(codedsymbol , symbol)
IF( SYMbolvalues(my_index)/= - 1 )THEN
codedsymbol = SYMbolvalues(my_index)
ELSE
IF( nextsymbol<=MAX_CODE )THEN
SYMbolvalues(my_index) = nextsymbol
nextsymbol = nextsymbol + 1
PREfixcodes(my_index) = codedsymbol
CONcatenatedsymbols(my_index) = symbol
END IF
CALL SETOUTPUTCODE(codedsymbol)
codedsymbol = symbol
END IF
END DO
CALL SETOUTPUTCODE(codedsymbol)
CALL SETOUTPUTCODE(MAX_VALUE)
CALL SETOUTPUTCODE(0)
END SUBROUTINE COMPRESS
function getPositionOnDictionary(hashPrefix, hashSymbol)
integer, intent(in) :: hashPrefix
integer, intent(in) :: hashSymbol
integer :: getPositionOnDictionary
integer :: index
integer :: offset
index = ishft(hashSymbol, HASH_SHIFT)
index = ieor(index, hashPrefix)
if (index == 0) then
offset = 1
else
offset = MAX_DICTIONARY_SIZE - index
endif
do
if (symbolValues(index) == -1) then
getPositionOnDictionary = index
exit
endif
if (prefixCodes(index) == hashPrefix .and. &
& concatenatedSymbols(index) == hashSymbol) then
getPositionOnDictionary = index
exit
endif
index = index - offset
if (index < 0) then
index = index + MAX_DICTIONARY_SIZE
endif
end do
return
end function
end module LZW_Encoder
! lzw_decoder.f90
!
! LZW Decoder (Expanssor)
!
! Author: Pedro Garcia Freitas <sawp@sawp.com.br>
! May, 2011
!
! License: Creative Commons http://creativecommons.org/licenses/by-nc-nd/3.0/
!
MODULE LZW_DECODER
USE LZW_SHARED_PARAMETERS
USE CODECIO
IMPLICIT NONE
!
! Derived Type definitions
!
TYPE :: DECODE_BUFFER_STACK
INTEGER , DIMENSION(0:MAX_DICTIONARY_SIZE) :: DECODERSTACK
INTEGER :: TOP
END TYPE DECODE_BUFFER_STACK
!
! Local variables
!
TYPE(DECODE_BUFFER_STACK) :: stack
CONTAINS
!
! Can be hand optimized - optimization
SUBROUTINE DECOMPRESS()
IMPLICIT NONE
!
! Local variables
!
INTEGER :: newsymbol
INTEGER :: nextsymbol
INTEGER :: oldsymbol
INTEGER :: popedsymbol
INTEGER :: symbol
!
nextsymbol = COMPILER_INTEGER_SIZE*SYMBOL_SIZE
oldsymbol = GETINPUTCODE()
symbol = oldsymbol
CALL SETRAWBYTE(oldsymbol)
DO
newsymbol = GETINPUTCODE()
IF( newsymbol==MAX_VALUE )RETURN
IF( newsymbol>=nextsymbol )THEN
STAck%DECODERSTACK(0) = symbol
CALL DECODESYMBOL(STAck%DECODERSTACK(1:) , oldsymbol)
ELSE
CALL DECODESYMBOL(STAck%DECODERSTACK(:) , newsymbol)
END IF
symbol = STAck%DECODERSTACK(STAck%TOP)
DO WHILE ( STAck%TOP>=0 )
popedsymbol = STAck%DECODERSTACK(STAck%TOP)
CALL SETRAWBYTE(popedsymbol)
STAck%TOP = STAck%TOP - 1
END DO
IF( nextsymbol<=MAX_CODE )THEN
PREfixcodes(nextsymbol) = oldsymbol
CONcatenatedsymbols(nextsymbol) = symbol
nextsymbol = nextsymbol + 1
END IF
oldsymbol = newsymbol
END DO
RETURN
END SUBROUTINE DECOMPRESS
SUBROUTINE DECODESYMBOL(Buffer , Code)
IMPLICIT NONE
!
! Dummy arguments
!
INTEGER :: Code
INTEGER , DIMENSION(:) :: Buffer
INTENT (IN) Code
INTENT (INOUT) Buffer
!
! Local variables
!
INTEGER :: i
INTEGER :: j
INTEGER :: symbol
!
j = 0
symbol = Code
STAck%TOP = 0
DO WHILE ( symbol>=COMPILER_INTEGER_SIZE*SYMBOL_SIZE )
! IF( symbol<COMPILER_INTEGER_SIZE*SYMBOL_SIZE )EXIT
IF( j>=MAX_CODE )THEN
PRINT * , "Decoding error"
STOP
END IF
i = STAck%TOP + 1
Buffer(i) = CONcatenatedsymbols(symbol)
symbol = PREfixcodes(symbol)
STAck%TOP = STAck%TOP + 1
j = j + 1
END DO
i = j + 1
Buffer(i) = symbol
END SUBROUTINE DECODESYMBOL
end module LZW_Decoder
! lzw.f90
!
! LZW Coder and Decoder
!
! Author: Pedro Garcia Freitas <sawp@sawp.com.br>
! May, 2011
!
! License: Creative Commons http://creativecommons.org/licenses/by-nc-nd/3.0/
!
MODULE LZW
USE LZW_SHARED_PARAMETERS
USE LZW_ENCODER
USE LZW_DECODER
IMPLICIT NONE
CONTAINS
SUBROUTINE INIT(Input , Output , Operation , Filename)
IMPLICIT NONE
!
! Dummy arguments
!
CHARACTER(100) :: Filename
CHARACTER(100) :: Input
CHARACTER(1) :: Operation
CHARACTER(100) :: Output
INTENT (IN) Filename , Input , Operation , Output
!
IF( Operation/='d' .AND. Operation/='e' )THEN
PRINT * , "Usage: " // TRIM(Filename) // " <operation> input output"
PRINT * , "Possible operations: "
PRINT * , " e -> encode (compress)"
PRINT * , " d -> decode (inflate)"
STOP
END IF
OPEN(UNIT = FILEIN , FILE = Input , ACTION = "read" , STATUS = "old" , &
&ACCESS = 'stream' , FORM = "formatted")
OPEN(UNIT = FILEOUT , FILE = Output , ACTION = "write" , STATUS = "replace" , &
& ACCESS = 'stream' , FORM = "formatted")
IF( Operation=='d' )THEN
PRINT * , "Decoding..."
CALL DECOMPRESS()
ELSE
PRINT * , "Encoding..."
CALL COMPRESS()
END IF
CLOSE(UNIT = FILEIN)
CLOSE(UNIT = FILEOUT)
END SUBROUTINE INIT
end module LZW
!
PROGRAM MAIN
USE LZW
IMPLICIT NONE
!
! Local variables
!
CHARACTER(100) :: filename
REAL :: finish
CHARACTER(100) :: input
CHARACTER(1) :: operation
CHARACTER(100) :: output
REAL :: start
!
CALL GETARG(0 , filename)
CALL GETARG(1 , operation)
CALL GETARG(2 , input)
CALL GETARG(3 , output)
CALL CPU_TIME(start)
CALL INIT(input , output , operation , filename)
CALL CPU_TIME(finish)
PRINT '("Time = ",f6.3," seconds.")' , finish - start
END PROGRAM MAIN</syntaxhighlight>
=={{header|FreeBASIC}}==
<syntaxhighlight lang="freebasic">' version 22-02-2019
' compile with: fbc -s console
Type dict
prefix As Integer
B As String
End Type
Sub init(dictionary() As dict, ByRef last As ULong)
For i As ULong = 0 To 255
dictionary(i).prefix = -1
dictionary(i).B = Chr(i)
Next
last = 255
End Sub
Function encode_LZW(dictionary() As dict, last_entry As ULong, input_str As String) As String
If Len(input_str) < 2 Then
Print "input string is to short"
Return ""
End If
Dim As String word, output_str, char
Dim As ULong i = 1, index, j, len_input = Len(input_str)
Do
If i > len_input Then
output_str = output_str + " " + Str(index)
Return output_str ' no more chars to process. we are done
End If
char = Mid(Input_str, i, 1)
i += 1
For j = 0 To last_entry
If dictionary(j).B = word + char Then
word += char
index = j
Continue Do
End If
Next
output_str = output_str + " " + Str(index)
last_entry = last_entry +1
dictionary(last_entry).B = word + char
dictionary(last_entry).prefix = index
word = char : index = Asc(char)
Loop
End Function
Function decode_LZW(dictionary() As dict, last_entry As ULong, input_str As String) As String
Dim As String temp, word, output_str
Dim As ULong i, i1 = 1, j, index
input_str = Trim(input_str)
Dim As ULong len_input = Len(input_str)
input_str = input_str + " "
i = InStr(i1, input_str, " ")
index = Val(Mid(Input_str, i1, i - i1))
word = dictionary(index).B
output_str = word
i1 = i +1
Do
i = InStr(i1, input_str, " ")
If i >= len_input Then
index = Val(Mid(input_str, i1))
output_str = output_str + dictionary(index).B
Return output_str
End If
index = Val(Mid(Input_str, i1, i - i1))
i1 = i +1
If index <= last_entry Then
temp = dictionary(index).B
Else
temp = word + Left(word, 1)
End If
output_str = output_str + temp
last_entry = last_entry +1
dictionary(last_entry).B = word + Left(temp, 1)
word = temp
Loop
End Function
' ------=< MAIN >=------
Dim As ULong last_entry, max_bit = 9
Dim As ULong dict_max = 1 Shl max_bit -1
Dim As String output_str, input_str = "TOBEORNOTTOBEORTOBEORNOT"
Dim As dict dictionary()
Print " input str: ";input_str
ReDim dictionary(dict_max +1)
init(dictionary(), last_entry)
output_str = encode_LZW(dictionary(), last_entry, input_str)
Print "encoded str: ";output_str
ReDim dictionary(dict_max +1)
init(dictionary(), last_entry)
output_str = decode_LZW(dictionary(), last_entry, output_str)
Print "decoded str: "; output_str
' empty keyboard buffer
While InKey <> "" : Wend
Print : Print "hit any key to end program"
Sleep
End</syntaxhighlight>
{{out}}
<pre> input str: TOBEORNOTTOBEORTOBEORNOT
encoded str: 84 79 66 69 79 82 78 79 84 256 258 260 265 259 261 263
decoded str: TOBEORNOTTOBEORTOBEORNOT</pre>
=={{header|Go}}==
Go also has
<code>[https://golang.org/pkg/compress/lzw compress/lzw]</code>
package in the standard library.
{{trans|Java}}
This handles any series of bytes in the input string,
not just ASCII or valid UTF8 encoding
(tested with [https://github.com/dvyukov/go-fuzz go-fuzz]).
<syntaxhighlight lang="go">package main
import (
"fmt"
"log"
"strings"
)
// Compress a string to a list of output symbols.
func compress(uncompressed string) []int {
// We actually want a map of []byte -> int but
// slices are not acceptable map key types.
dictionary := make(map[string]int, dictSize)
for i := 0; i < dictSize; i++ {
// Ugly mess to work around not having a []byte key type.
// Using `string(i)` would do utf8 encoding for i>127.
dictionary[string([]byte{byte(i)})] = i
}
var result []int
var w []byte
for i := 0; i < len(uncompressed); i++ {
c := uncompressed[i]
wc := append(w, c)
if _, ok := dictionary[string(wc)]; ok {
w = wc
} else {
result = append(result, dictionary[string(w)])
// Add wc to the dictionary.
dictionary[string(wc)] = dictSize
dictSize++
//w = []byte{c}, but re-using wc
wc[0] = c
w = wc[:1]
}
}
if len(w) > 0 {
// Output the code for w.
result = append(result, dictionary[string(w)])
}
return result
}
type BadSymbolError int
func (e BadSymbolError) Error() string {
return fmt.Sprint("Bad compressed symbol ", int(e))
}
// Decompress a list of output symbols to
func decompress(compressed []int) (string, error) {
// Build the dictionary.
dictSize := 256
dictionary := make(map[int][]byte, dictSize)
for i := 0; i < dictSize; i++ {
dictionary[i] = []byte{byte(i)}
}
var result strings.Builder
var w []byte
for _, k := range
var entry []byte
if x, ok := dictionary[k]; ok {
//entry = x, but ensuring any append will make a copy
entry = x[:len(x):len(x)]
} else if k == dictSize && len(w) > 0 {
entry = append(w, w[0])
} else {
return result.String(), BadSymbolError(k)
}
if len(w) > 0 {
// Add w+entry[0] to the dictionary.
w = append(w, entry[0])
dictionary[dictSize] = w
dictSize++
}
w = entry
}
return result.String(), nil
}
func main() {
decompressed,
if err != nil {
log.Fatal(err)
}
fmt.Println(decompressed)
}</syntaxhighlight>
{{out}}
<pre>
Line 1,534 ⟶ 2,675:
=={{header|Groovy}}==
<
def dictionary = (
def w = '', compressed = []
text.each { ch ->
Line 1,543 ⟶ 2,684:
} else {
compressed << dictionary[w]
dictionary[wc] = dictionary.size()
w = "$ch"
}
Line 1,550 ⟶ 2,691:
compressed
}
def decompress = { compressed ->
def dictionary = (
int dictSize = 128;
String w = "${(char)compressed[0]}"
StringBuffer result = new StringBuffer(w)
compressed.drop(1).each { k ->
String entry = dictionary[k]
if (!entry) {
Line 1,562 ⟶ 2,705:
}
result << entry
dictionary[dictionary.size()] = "$w${entry[0]}"
w = entry
}
result.toString()
}</
Testing:
<
def compressed = compress(plaintext)
def result = decompress(compressed)
Line 1,575 ⟶ 2,720:
Plaintext: '$plaintext'
Compressed: $compressed
Uncompressed: '$result'""".stripIndent()</
{{out}}
<pre>Plaintext: 'TOBEORNOTTOBEORTOBEORNOT'
Compressed: [84, 79, 66, 69, 79, 82, 78, 79, 84, 256, 258, 260, 265, 259, 261, 263]
Uncompressed: 'TOBEORNOTTOBEORTOBEORNOT'</pre>
=={{header|Haskell}}==
<
import Data.
doLZW :: Eq a => [a] -> [a] -> [Int]
doLZW _ [] = []
doLZW as (x:xs) =
where
lzw a w (x:xs)
| w_ `elem` a = lzw a w_
| otherwise = fromJust (elemIndex w a) : lzw (a ++ [w_]) [x] xs
where
w_ = w ++ [x]
undoLZW :: [a] -> [Int] -> [a]
undoLZW _ [] = []
undoLZW a cs =
(!!)
(foldl
((.) <$> (++) <*>
(\x xs -> return (((++) <$> head <*> take 1 . last) ((x !!) <$> xs))))
(return <$> a)
(take2 cs))
take2 :: [a] -> [[a]]
take2 xs = filter ((2 ==) . length) (take 2 <$> tails xs)
main :: IO ()
main = do
print $ doLZW ['\0' .. '\255'] "TOBEORNOTTOBEORTOBEORNOT"
print $
undoLZW
['\0' .. '\255']
[84, 79, 66, 69, 79, 82, 78, 79, 84, 256, 258, 260, 265, 259, 261, 263]
print $
((==) <*> ((.) <$> undoLZW <*> doLZW) ['\NUL' .. '\255'])
"TOBEORNOTTOBEORTOBEORNOT"</syntaxhighlight>
{{Out}}
<pre>[84,79,66,69,79,82,78,79,84,256,258,260,265,259,261,263]
"TOBEORNOTTOBEORTOBEORNOT"
True</pre>
Other (elegant) code can be found at Haskell wiki [http://www.haskell.org/haskellwiki/Toy_compression_implementations Toy compression]
Line 1,619 ⟶ 2,776:
Straightforward implementations of encoding and decoding:
<
d=. ;/x
r=.0$0
Line 1,632 ⟶ 2,789:
end.
r, d i.<w
)</
Test:
<pre> a. encodeLZW 'TOBEORNOTTOBEORTOBEORNOT'
84 79 66 69 79 82 78 79 84 256 258 260 265 259 261 263</pre>
Decoding:
<
d=.;/x
w=.r=. >d{~{.y
Line 1,652 ⟶ 2,809:
end.
;r
)</
Test:
<pre> a. decodeLZW 84 79 66 69 79 82 78 79 84 256 258 260 265 259 261 263
Line 1,670 ⟶ 2,827:
=={{header|Java}}==
{{works with|Java|1.5+}}
<
public class LZW {
Line 1,736 ⟶ 2,893:
System.out.println(decompressed);
}
}</
{{out}} (Command Line direct output):
<
TOBEORNOTTOBEORTOBEORNOT</
=={{header|JavaScript}}==
<
var LZW = {
compress: function (uncompressed) {
Line 1,823 ⟶ 2,980:
comp = LZW.compress("TOBEORNOTTOBEORTOBEORNOT"),
decomp = LZW.decompress(comp);
document.write(comp + '<br>' + decomp);</
=== ES6 Version ===
This is the the same thing, but for ES6. The code has been refactored and cleaned up a bit to look neater.
<syntaxhighlight lang="javascript">'use strict';
/**
Namespace for LZW compression and decompression.
Methods:
LZW.compress(uncompressed)
LZW.decompress(compressed)
*/
class LZW
{
/**
Perform the LZW compression
uncompressed - String. The string on which to perform the compression.
*/
static compress(uncompressed)
{
// Initialize dictionary
let dictionary = {};
for (let i = 0; i < 256; i++)
{
dictionary[String.fromCharCode(i)] = i;
}
let word = '';
let result = [];
let dictSize = 256;
for (let i = 0, len = uncompressed.length; i < len; i++)
{
let curChar = uncompressed[i];
let joinedWord = word + curChar;
// Do not use dictionary[joinedWord] because javascript objects
// will return values for myObject['toString']
if (dictionary.hasOwnProperty(joinedWord))
{
word = joinedWord;
}
else
{
result.push(dictionary[word]);
// Add wc to the dictionary.
dictionary[joinedWord] = dictSize++;
word = curChar;
}
}
if (word !== '')
{
result.push(dictionary[word]);
}
return result;
}
/**
Decompress LZW array generated by LZW.compress()
compressed - Array. The array that holds LZW compressed data.
*/
static decompress(compressed)
{
// Initialize Dictionary (inverse of compress)
let dictionary = {};
for (let i = 0; i < 256; i++)
{
dictionary[i] = String.fromCharCode(i);
}
let word = String.fromCharCode(compressed[0]);
let result = word;
let entry = '';
let dictSize = 256;
for (let i = 1, len = compressed.length; i < len; i++)
{
let curNumber = compressed[i];
if (dictionary[curNumber] !== undefined)
{
entry = dictionary[curNumber];
}
else
{
if (curNumber === dictSize)
{
entry = word + word[0];
}
else
{
throw 'Error in processing';
return null;
}
}
result += entry;
// Add word + entry[0] to dictionary
dictionary[dictSize++] = word + entry[0];
word = entry;
}
return result;
}
}
let comp = LZW.compress('TOBEORNOTTOBEORTOBEORNOT');
let decomp = LZW.decompress(comp);
console.log(`${comp}
${decomp}`);</syntaxhighlight>
{{out}}
Line 1,832 ⟶ 3,105:
{{ works with|jq|1.4}}
{{trans|JavaScript}}
<
def lzw_compress:
def decode: [.] | implode;
Line 1,877 ⟶ 3,150:
| .[2] = $entry # w = entry
) | .[3]
;</
'''Example''':
<
{{Out}}
$ jq -n -f LZW.jq
"TOBEORNOTTOBEORTOBEORNOT"
=={{header|Julia}}==
{{works with|Julia|1.1.1}}
<syntaxhighlight lang="julia">function compressLZW(decompressed::String)
dictsize = 256
dict = Dict{String,Int}(string(Char(i)) => i for i in 0:dictsize)
result = Vector{Int}(undef, 0)
w = ""
for c in decompressed
wc = string(w, c)
if haskey(dict, wc)
w = wc
else
push!(result, dict[w])
dict[wc] = dictsize
dictsize += 1
w = string(c)
end
end
if !isempty(w) push!(result, dict[w]) end
return result
end
function decompressLZW(compressed::Vector{Int})
dictsize = 256
dict = Dict{Int,String}(i => string('\0' + i) for i in 0:dictsize)
result = IOBuffer()
w = string(Char(popfirst!(compressed)))
write(result, w)
for k in compressed
if haskey(dict, k)
entry = dict[k]
elseif k == dictsize
entry = string(w, w[1])
else
error("bad compressed k: $k")
end
write(result, entry)
dict[dictsize] = string(w, entry[1])
dictsize += 1
w = entry
end
return String(take!(result))
end
original = ["0123456789", "TOBEORNOTTOBEORTOBEORNOT", "dudidudidudida"]
compressed = compressLZW.(original)
decompressed = decompressLZW.(compressed)
for (word, comp, decomp) in zip(original, compressed, decompressed)
comprate = (length(word) - length(comp)) / length(word) * 100
println("Original: $word\n-> Compressed: $comp (compr.rate: $(round(comprate, digits=2))%)\n-> Decompressed: $decomp")
end</syntaxhighlight>
{{out}}
<pre>Original: 0123456789
-> Compressed: [49, 50, 51, 52, 53, 54, 55, 56, 57] (compr.rate: 10.0%)
-> Decompressed: 0123456789
Original: TOBEORNOTTOBEORTOBEORNOT
-> Compressed: [79, 66, 69, 79, 82, 78, 79, 84, 256, 258, 260, 265, 259, 261, 263] (compr.rate: 37.5%)
-> Decompressed: TOBEORNOTTOBEORTOBEORNOT
Original: dudidudidudida
-> Compressed: [117, 100, 105, 256, 258, 260, 259, 97] (compr.rate: 42.86%)
-> Decompressed: dudidudidudida</pre>
=={{header|Kotlin}}==
{{trans|Java}}
<syntaxhighlight lang="scala">// version 1.1.2
object Lzw {
/** Compress a string to a list of output symbols. */
fun compress(uncompressed: String): MutableList<Int> {
// Build the dictionary.
var dictSize = 256
val dictionary = mutableMapOf<String, Int>()
(0 until dictSize).forEach { dictionary.put(it.toChar().toString(), it)}
var w = ""
val result = mutableListOf<Int>()
for (c in uncompressed) {
val wc = w + c
if (dictionary.containsKey(wc))
w = wc
else {
result.add(dictionary[w]!!)
// Add wc to the dictionary.
dictionary.put(wc, dictSize++)
w = c.toString()
}
}
// Output the code for w
if (!w.isEmpty()) result.add(dictionary[w]!!)
return result
}
/** Decompress a list of output symbols to a string. */
fun decompress(compressed: MutableList<Int>): String {
// Build the dictionary.
var dictSize = 256
val dictionary = mutableMapOf<Int, String>()
(0 until dictSize).forEach { dictionary.put(it, it.toChar().toString())}
var w = compressed.removeAt(0).toChar().toString()
val result = StringBuilder(w)
for (k in compressed) {
var entry: String
if (dictionary.containsKey(k))
entry = dictionary[k]!!
else if (k == dictSize)
entry = w + w[0]
else
throw IllegalArgumentException("Bad compressed k: $k")
result.append(entry)
// Add w + entry[0] to the dictionary.
dictionary.put(dictSize++, w + entry[0])
w = entry
}
return result.toString()
}
}
fun main(args: Array<String>) {
val compressed = Lzw.compress("TOBEORNOTTOBEORTOBEORNOT")
println(compressed)
val decompressed = Lzw.decompress(compressed)
println(decompressed)
}</syntaxhighlight>
{{out}}
<pre>
[84, 79, 66, 69, 79, 82, 78, 79, 84, 256, 258, 260, 265, 259, 261, 263]
TOBEORNOTTOBEORTOBEORNOT
</pre>
=={{header|Liberty BASIC}}==
Line 1,889 ⟶ 3,297:
It also has the option to write the encoding/decoding dictionaries to file so the encoder can be checked for accuracy.
This code directly follows the methodology described in an excellent web article by Juha Nieminen entitled "An efficient LZW implementation".
<syntaxhighlight lang="text"> DIM LZW(1, 1)
DIM JDlzw(1)
DIM JDch$(1)
Line 2,168 ⟶ 3,576:
fileTag$ = STR$(tagCount) + "_"
RETURN
''''''''''''''''''''''''''''''''''''''''</
=={{header|
<syntaxhighlight lang="lua">local function compress(uncompressed) -- string
local dictionary, result, dictSize, w, c = {}, {}, 255, ""
for i = 0, 255 do
dictionary[string.char(i)] = i
end
for i = 1, #uncompressed do
c = string.sub(uncompressed, i, i)
if dictionary[w .. c] then
w = w .. c
else
table.insert(result, dictionary[w])
dictSize = dictSize + 1
dictionary[w .. c] = dictSize
w = c
end
end
if w ~= "" then
table.insert(result, dictionary[w])
end
return result
end
local function decompress(compressed) -- table
local dictionary, dictSize, entry, w, k = {}, 256, "", string.char(compressed[1])
local result = {w}
for i = 0, 255 do
dictionary[i] = string.char(i)
end
for i = 2, #compressed do
k = compressed[i]
if dictionary[k] then
entry = dictionary[k]
elseif k == dictSize then
entry = w .. string.sub(w, 1, 1)
else
return nil, i
end
table.insert(result, entry)
dictionary[dictSize] = w .. string.sub(entry, 1, 1)
dictSize = dictSize + 1
w = entry
end
return table.concat(result)
end
local example = "TOBEORNOTTOBEORTOBEORNOT"
local com = compress(example)
local dec = decompress(com)
print(table.concat(com, ", "))
print(dec)</syntaxhighlight>
{{Out}}
<pre>
84, 79, 66, 69, 79, 82, 78, 79, 84, 256, 258, 260, 265, 259, 261, 263
TOBEORNOTTOBEORTOBEORNOT
</pre>
=={{header|M2000 Interpreter}}==
{{trans|BBC BASIC}}
<syntaxhighlight lang="m2000 interpreter">
Module BBCtrans {
\\ LZW compression
plaintext$="TOBEORNOTTOBEORTOBEORNOT"
Function encodeLZW$(i$) {
Def long c, d, i, l, o$, w$
DIM dict$(0 to 4095)
FOR i = 0 TO 255 : dict$(i) = CHR$(i) : NEXT i
l = i
i = 1
w$ = LEFT$(i$,1)
REPEAT{
d = 0
REPEAT {
c = d
IF i > LEN(i$) THEN EXIT
FOR d = 1 TO l-1
IF w$ = dict$(d) THEN EXIT
NEXT d
IF d < l Then i += 1 : w$ += MID$(i$, i, 1)
} UNTIL d >= l
dict$(l) = w$ : l += 1 : w$ = RIGHT$(w$, 1)
o$ += CHR$(c MOD 256) + CHR$(c DIV 256)
} UNTIL i > LEN(i$)
= o$
}
encodeLZW$ = encodeLZW$(plaintext$)
FOR i = 1 TO LEN(encodeLZW$) STEP 2
PRINT ASC(MID$(encodeLZW$,i)) + 256*ASC(MID$(encodeLZW$,i+1));" ";
NEXT i
Print
Function decodeLZW$(i$) {
Def c, i, l, o$, t$, w$
DIM dict$(0 to 4095)
FOR i = 0 TO 255 : dict$(i) = CHR$(i) : NEXT i
l = i
c = ASC(i$) + 256*ASC(MID$(i$,2))
w$ = dict$(c)
o$ = w$
IF LEN(i$) < 4 THEN = o$
FOR i = 3 TO LEN(i$) STEP 2
c = ASC(MID$(i$,i)) + 256*ASC(MID$(i$,i+1))
IF c < l Then { t$ = dict$(c) } ELSE t$ = w$ + LEFT$(w$,1)
o$ += t$
dict$(l) = w$ + LEFT$(t$,1)
l += 1
w$ = t$
NEXT i
= o$
}
Print decodeLZW$(encodeLZW$)
}
BBCtrans
</syntaxhighlight>
And here a change for using Inventories, where we have hash function, and we find entry in O(1).
<syntaxhighlight lang="m2000 interpreter">
Module FastM2000 {
plaintext$="TOBEORNOTTOBEORTOBEORNOT"
Function encodeLZW$(i$) {
Def long c, d, i, l, o$, w$
Inventory dict
For i = 0 to 255 {Append dict , Chr$(i):=i}
l = i
i = 1
w$ = LEFT$(i$,1)
REPEAT{
d = 0
Repeat {
c = d
IF i > Len(i$) Then Exit
if exist(dict, w$) Then {
d=eval(dict)
} Else Append dict, w$:=l: Exit
if d<l Then i += 1 : w$ += Mid$(i$, i, 1)
} Until d >= l
l += 1 : w$ = Right$(w$, 1)
o$ += Chr$(c Mod 256) + Chr$(c div 256)
} Until i > Len(i$)
= o$
}
encodeLZW$ = encodeLZW$(plaintext$)
Document Doc$
For i = 1 to Len(encodeLZW$) STEP 2
Doc$= Str$(Asc(Mid$(encodeLZW$,i)) + 256*Asc(Mid$(encodeLZW$,i+1)))
Next i
Doc$={
}
Function decodeLZW$(i$) {
Def c, i, l, o$, t$, w$
Inventory Dict
For i = 0 to 255 {Append dict , i:=chr$(i)}
l = i
c = Asc(i$) + 256*Asc(Mid$(i$,2))
w$ = dict$(c)
o$ = w$
IF Len(i$) < 4 Then = o$
For i = 3 to Len(i$) STEP 2 {
c = Asc(Mid$(i$,i)) + 256*Asc(Mid$(i$,i+1))
IF c < l Then {
t$ = dict$(c)
} Else t$ = w$ + LEFT$(w$,1)
o$ += t$
Append dict, l:=w$ + LEFT$(t$,1)
l += 1 : w$ = t$
}
= o$
}
Doc$=decodeLZW$(encodeLZW$)+{
}
Clipboard Doc$
Report Doc$
}
FastM2000
</syntaxhighlight>
{{out}}
<pre >
84 79 66 69 79 82 78 79 84 256 258 260 265 259 261 263
TOBEORNOTTOBEORTOBEORNOT
</pre>
=={{header|Mathematica}}/{{header|Wolfram Language}}==
{{trans|Ruby}}
<syntaxhighlight lang="text">compress[uncompressed_] :=
Module[{dictsize, dictionary, w, result, wc},
dictsize = 256;
Line 2,208 ⟶ 3,798:
(*How to use:*)
compress["TOBEORNOTTOBEORTOBEORNOT"]
decompress[%]</
{{Out}}
<pre>{"T", "O", "B", "E", "O", "R", "N", "O", "T", 256, 258, 260, 265, 259, 261, 263}
"TOBEORNOTTOBEORTOBEORNOT"</pre>
(* stm, 6 June 2022: I did not edit the above code, but I believe that there is a small error. The Range@dictsize should be replaced by Range[0,dictsize-1] for LZW standard; Mathematica runs from 1 to 256 instead of 0 to 255. The code works as is in String format in Mathematica because Mathematica distinguishes between Strings and numbers. However, this example fails when the code is adapted to byte type input and output, as used in images. Adjusting Range[...] fixes the problem.*)
=={{header|Nim}}==
<syntaxhighlight lang="text">import tables
proc compress*(uncompressed: string): seq[int] =
# Build the dictionary.
var
for i in 0..255: dictionary[$chr(i)] = i
var w = ""
for c in uncompressed:
var wc = w & c
if
w = wc
else:
#
# "wc" is a new sequence
dictionary
w = $c
#
if w.len > 0: result.add dictionary[w]
proc decompress*(compressed: var seq[int]): string =
var dictionary: Table[int, string]
for i in 0..255: dictionary[i] = $chr(i)
var w = dictionary[compressed[0]]
compressed.delete(0)
for k in compressed:
var entry: string
if k in dictionary:
entry = dictionary[k]
elif
entry = w & w[0]
else:
raise newException(ValueError, "Bad compressed k: " & $k)
result.add entry
# New sequence: add it to the dictionary.
dictionary[dictionary.len] = w & entry[0]
w = entry
when isMainModule:
Line 2,278 ⟶ 3,858:
echo compressed
var decompressed = decompress(compressed)
echo decompressed</syntaxhighlight>
{{
<pre>@[84, 79, 66, 69, 79, 82, 78, 79, 84, 256, 258, 260, 265, 259, 261, 263]
TOBEORNOTTOBEORTOBEORNOT</pre>
=={{header|Objeck}}==
{{trans|Java}}
<
class LZW {
Line 2,391 ⟶ 3,969:
"]"->PrintLine();
}
}</
<pre>[84, 79, 66, 69, 79, 82, 78, 79, 84, 256, 258, 260, 265, 259, 261, 263]
Line 2,402 ⟶ 3,980:
The class for the LZW compression algorithm:
<
#import <stdio.h>
Line 2,487 ⟶ 4,065:
}
@end</
Usage example:
<
int main()
Line 2,511 ⟶ 4,089:
}
return EXIT_SUCCESS;
}</
{{out}} (reformatted by hand):
Line 2,521 ⟶ 4,099:
=={{header|OCaml}}==
<
#load "extLib.cma"
open ExtString
Line 2,600 ⟶ 4,178:
in
(List.rev result)
;;</
here is the interface:
<
val decompress : compressed:int list -> string list</
How to use:<br />
Line 2,610 ⟶ 4,188:
So to know how many bits are required, you need to know how many bits are required for the greatest symbol in the list.
<
(** number of bits needed to encode the integer m *)
Line 2,648 ⟶ 4,226:
List.iter (Buffer.add_string buf) result;
(Buffer.contents buf)
;;</
=={{header|Ol}}==
This version use lazy streams which is pair (symbol . function-to-get-next-symbol).
<syntaxhighlight lang="scheme">
(define (compress str)
(let loop ((dc (fold (lambda (f x) ; dictionary (simplest, not optimized), with reversed codes
(cons (list x) (cons x f)))
'() (iota 256)))
(w '()) ; output sequence (reversed)
(s 256) ; maximal dictionary code value + 1
(x '()) ; current sequence
(r (str-iter str))); input stream
(cond
((null? r)
(reverse (cons (cadr (member x dc)) w)))
((pair? r)
(let ((xy (cons (car r) x)))
(if (member xy dc)
(loop dc w s xy (cdr r))
(loop (cons xy (cons s dc)) ; update dictionary with xy . s
(cons (cadr (member x dc)) w) ; add code to output stream
(+ s 1) ; increase code
(list (car r)) ; new current sequence
(cdr r))))) ; next input
(else
(loop dc w s x (r)))))
)
(print (compress "TOBEORNOTTOBEORTOBEORNOT")) ; => (84 79 66 69 79 82 78 79 84 256 258 260 265 259 261 263)
</syntaxhighlight>
And decoder (runes->string used to unify functions - both used string iterators):
<syntaxhighlight lang="scheme">
(define (decompress str)
(let loop ((dc (fold (lambda (f x) ; dictionary (simplest, not optimized), with reversed codes
(cons x (cons (list x) f)))
'() (iota 256)))
(w '()) ; output sequence (reversed)
(s 256) ; maximal dictionary code value + 1
(x '()) ; current symbols sequence
(r (str-iter str))); input stream
(cond
((null? r)
(reverse w))
((pair? r)
(let*((y (cadr (member (car r) dc)))
(xy (append y x)))
(if (member xy dc)
(loop dc (append y w) s xy (cdr r)) ; вряд ли такое будет...
(loop (cons s (cons xy dc)) ; update dictionary with xy . s
(append y w) ; add phrase to output stream
(+ s 1)
y ; new initial code
(cdr r))))) ; next input
(else
(loop dc w s x (r))))))
(print (runes->string
(decompress (runes->string '(84 79 66 69 79 82 78 79 84 256 258 260 265 259 261 263)))))
; => TOBEORNOTTOBEORTOBEEORNOT
</syntaxhighlight>
=={{header|Perl}}==
In this version the hashes contain mixed typed data:
<
sub compress {
my $uncompressed = shift;
Line 2,717 ⟶ 4,356:
print "@compressed\n";
my $decompressed = decompress(@compressed);
print "$decompressed\n";</
{{out}}
Line 2,725 ⟶ 4,364:
</pre>
=={{header|
{{trans|
<!--<syntaxhighlight lang="phix">(phixonline)-->
<span style="color: #008080;">with</span> <span style="color: #008080;">javascript_semantics</span>
<span style="color: #008080;">function</span> <span style="color: #000000;">compress</span><span style="color: #0000FF;">(</span><span style="color: #004080;">string</span> <span style="color: #000000;">uncompressed</span><span style="color: #0000FF;">)</span>
<span style="color: #004080;">integer</span> <span style="color: #000000;">dict</span> <span style="color: #0000FF;">=</span> <span style="color: #7060A8;">new_dict</span><span style="color: #0000FF;">()</span>
<span style="color: #004080;">sequence</span> <span style="color: #000000;">result</span> <span style="color: #0000FF;">=</span> <span style="color: #0000FF;">{}</span>
<span style="color: #004080;">integer</span> <span style="color: #000000;">dictSize</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">255</span><span style="color: #0000FF;">,</span> <span style="color: #000000;">c</span>
<span style="color: #004080;">string</span> <span style="color: #000000;">word</span> <span style="color: #0000FF;">=</span> <span style="color: #008000;">""</span>
<span style="color: #008080;">for</span> <span style="color: #000000;">i</span><span style="color: #0000FF;">=</span><span style="color: #000000;">0</span> <span style="color: #008080;">to</span> <span style="color: #000000;">255</span> <span style="color: #008080;">do</span>
<span style="color: #7060A8;">setd</span><span style="color: #0000FF;">(</span><span style="color: #008000;">""</span><span style="color: #0000FF;">&</span><span style="color: #000000;">i</span><span style="color: #0000FF;">,</span><span style="color: #000000;">i</span><span style="color: #0000FF;">,</span><span style="color: #000000;">dict</span><span style="color: #0000FF;">)</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">for</span>
<span style="color: #008080;">for</span> <span style="color: #000000;">i</span><span style="color: #0000FF;">=</span><span style="color: #000000;">1</span> <span style="color: #008080;">to</span> <span style="color: #7060A8;">length</span><span style="color: #0000FF;">(</span><span style="color: #000000;">uncompressed</span><span style="color: #0000FF;">)</span> <span style="color: #008080;">do</span>
<span style="color: #000000;">c</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">uncompressed</span><span style="color: #0000FF;">[</span><span style="color: #000000;">i</span><span style="color: #0000FF;">]</span>
<span style="color: #008080;">if</span> <span style="color: #7060A8;">getd_index</span><span style="color: #0000FF;">(</span><span style="color: #000000;">word</span><span style="color: #0000FF;">&</span><span style="color: #000000;">c</span><span style="color: #0000FF;">,</span><span style="color: #000000;">dict</span><span style="color: #0000FF;">)</span> <span style="color: #008080;">then</span>
<span style="color: #000000;">word</span> <span style="color: #0000FF;">&=</span> <span style="color: #000000;">c</span>
<span style="color: #008080;">else</span>
<span style="color: #000000;">result</span> <span style="color: #0000FF;">&=</span> <span style="color: #7060A8;">getd</span><span style="color: #0000FF;">(</span><span style="color: #000000;">word</span><span style="color: #0000FF;">,</span><span style="color: #000000;">dict</span><span style="color: #0000FF;">)</span>
<span style="color: #000000;">dictSize</span> <span style="color: #0000FF;">+=</span> <span style="color: #000000;">1</span>
<span style="color: #7060A8;">setd</span><span style="color: #0000FF;">(</span><span style="color: #000000;">word</span><span style="color: #0000FF;">&</span><span style="color: #000000;">c</span><span style="color: #0000FF;">,</span><span style="color: #000000;">dictSize</span><span style="color: #0000FF;">,</span><span style="color: #000000;">dict</span><span style="color: #0000FF;">)</span>
<span style="color: #000000;">word</span> <span style="color: #0000FF;">=</span> <span style="color: #008000;">""</span><span style="color: #0000FF;">&</span><span style="color: #000000;">c</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">if</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">for</span>
<span style="color: #008080;">if</span> <span style="color: #000000;">word</span><span style="color: #0000FF;">!=</span><span style="color: #008000;">""</span> <span style="color: #008080;">then</span>
<span style="color: #000000;">result</span> <span style="color: #0000FF;">&=</span> <span style="color: #7060A8;">getd</span><span style="color: #0000FF;">(</span><span style="color: #000000;">word</span><span style="color: #0000FF;">,</span><span style="color: #000000;">dict</span><span style="color: #0000FF;">)</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">if</span>
<span style="color: #7060A8;">destroy_dict</span><span style="color: #0000FF;">(</span><span style="color: #000000;">dict</span><span style="color: #0000FF;">)</span>
<span style="color: #008080;">return</span> <span style="color: #000000;">result</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">function</span>
<span style="color: #008080;">function</span> <span style="color: #000000;">decompress</span><span style="color: #0000FF;">(</span><span style="color: #004080;">sequence</span> <span style="color: #000000;">compressed</span><span style="color: #0000FF;">)</span>
<span style="color: #004080;">integer</span> <span style="color: #000000;">dict</span> <span style="color: #0000FF;">=</span> <span style="color: #7060A8;">new_dict</span><span style="color: #0000FF;">()</span>
<span style="color: #004080;">integer</span> <span style="color: #000000;">dictSize</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">255</span><span style="color: #0000FF;">,</span> <span style="color: #000000;">k</span><span style="color: #0000FF;">,</span> <span style="color: #000000;">ki</span>
<span style="color: #004080;">string</span> <span style="color: #000000;">dent</span> <span style="color: #0000FF;">=</span> <span style="color: #008000;">""</span><span style="color: #0000FF;">,</span> <span style="color: #000000;">result</span> <span style="color: #0000FF;">=</span> <span style="color: #008000;">""</span><span style="color: #0000FF;">,</span> <span style="color: #000000;">word</span> <span style="color: #0000FF;">=</span> <span style="color: #008000;">""</span>
<span style="color: #008080;">for</span> <span style="color: #000000;">i</span><span style="color: #0000FF;">=</span><span style="color: #000000;">0</span> <span style="color: #008080;">to</span> <span style="color: #000000;">255</span> <span style="color: #008080;">do</span>
<span style="color: #7060A8;">setd</span><span style="color: #0000FF;">(</span><span style="color: #000000;">i</span><span style="color: #0000FF;">,</span><span style="color: #008000;">""</span><span style="color: #0000FF;">&</span><span style="color: #000000;">i</span><span style="color: #0000FF;">,</span><span style="color: #000000;">dict</span><span style="color: #0000FF;">)</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">for</span>
<span style="color: #008080;">for</span> <span style="color: #000000;">i</span><span style="color: #0000FF;">=</span><span style="color: #000000;">1</span> <span style="color: #008080;">to</span> <span style="color: #7060A8;">length</span><span style="color: #0000FF;">(</span><span style="color: #000000;">compressed</span><span style="color: #0000FF;">)</span> <span style="color: #008080;">do</span>
<span style="color: #000000;">k</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">compressed</span><span style="color: #0000FF;">[</span><span style="color: #000000;">i</span><span style="color: #0000FF;">]</span>
<span style="color: #000000;">ki</span> <span style="color: #0000FF;">=</span> <span style="color: #7060A8;">getd_index</span><span style="color: #0000FF;">(</span><span style="color: #000000;">k</span><span style="color: #0000FF;">,</span><span style="color: #000000;">dict</span><span style="color: #0000FF;">)</span>
<span style="color: #008080;">if</span> <span style="color: #000000;">ki</span> <span style="color: #008080;">then</span>
<span style="color: #000000;">dent</span> <span style="color: #0000FF;">=</span> <span style="color: #7060A8;">getd_by_index</span><span style="color: #0000FF;">(</span><span style="color: #000000;">ki</span><span style="color: #0000FF;">,</span><span style="color: #000000;">dict</span><span style="color: #0000FF;">)</span>
<span style="color: #008080;">elsif</span> <span style="color: #000000;">k</span><span style="color: #0000FF;">=</span><span style="color: #000000;">dictSize</span> <span style="color: #008080;">then</span>
<span style="color: #000000;">dent</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">word</span><span style="color: #0000FF;">&</span><span style="color: #000000;">word</span><span style="color: #0000FF;">[</span><span style="color: #000000;">1</span><span style="color: #0000FF;">]</span>
<span style="color: #008080;">else</span>
<span style="color: #008080;">return</span> <span style="color: #0000FF;">{</span><span style="color: #004600;">NULL</span><span style="color: #0000FF;">,</span><span style="color: #000000;">i</span><span style="color: #0000FF;">}</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">if</span>
<span style="color: #000000;">result</span> <span style="color: #0000FF;">&=</span> <span style="color: #000000;">dent</span>
<span style="color: #7060A8;">setd</span><span style="color: #0000FF;">(</span><span style="color: #000000;">dictSize</span><span style="color: #0000FF;">,</span><span style="color: #000000;">word</span><span style="color: #0000FF;">&</span><span style="color: #000000;">dent</span><span style="color: #0000FF;">[</span><span style="color: #000000;">1</span><span style="color: #0000FF;">],</span><span style="color: #000000;">dict</span><span style="color: #0000FF;">)</span>
<span style="color: #000000;">dictSize</span> <span style="color: #0000FF;">+=</span> <span style="color: #000000;">1</span>
<span style="color: #000000;">word</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">dent</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">for</span>
<span style="color: #7060A8;">destroy_dict</span><span style="color: #0000FF;">(</span><span style="color: #000000;">dict</span><span style="color: #0000FF;">)</span>
<span style="color: #008080;">return</span> <span style="color: #000000;">result</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">function</span>
<span style="color: #008080;">constant</span> <span style="color: #000000;">example</span> <span style="color: #0000FF;">=</span> <span style="color: #008000;">"TOBEORNOTTOBEORTOBEORNOT"</span>
<span style="color: #004080;">sequence</span> <span style="color: #000000;">com</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">compress</span><span style="color: #0000FF;">(</span><span style="color: #000000;">example</span><span style="color: #0000FF;">)</span>
<span style="color: #7060A8;">pp</span><span style="color: #0000FF;">(</span><span style="color: #000000;">com</span><span style="color: #0000FF;">,{</span><span style="color: #004600;">pp_IntCh</span><span style="color: #0000FF;">,</span><span style="color: #004600;">true</span><span style="color: #0000FF;">,</span><span style="color: #004600;">pp_Maxlen</span><span style="color: #0000FF;">,</span><span style="color: #000000;">90</span><span style="color: #0000FF;">})</span>
<span style="color: #0000FF;">?</span><span style="color: #000000;">decompress</span><span style="color: #0000FF;">(</span><span style="color: #000000;">com</span><span style="color: #0000FF;">)</span>
<!--</syntaxhighlight>-->
{{out}}
<pre>
{84'T
"TOBEORNOTTOBEORTOBEORNOT"
</pre>
=={{header|PHP}}==
{{trans|
<
{
function compress($unc) {
Line 2,819 ⟶ 4,474:
for ($i = 1; $i < count($com);$i++) {
$k = $com[$i];
if (isset($dictionary[$k])) {
$entry = $dictionary[$k];
} else {
Line 2,829 ⟶ 4,484:
}
$result .= $entry;
$dictionary[$dictSize++] = $w
$w = $entry;
}
Line 2,842 ⟶ 4,497:
$dec = $lzw->decompress($com);
echo $com . "<br>" . $dec;
</syntaxhighlight>
{{out}}
<pre>
84,79,66,69,79,82,78,79,84,
TOBEORNOTTOBEORTOBEORNOT
</pre>
=={{header|Picat}}==
<syntaxhighlight lang="picat">go =>
S = "TOBEORNOTTOBEORTOBEORNOT",
println(s=S),
println(len=S.length),
Compressed = compress(S),
println(compressed=Compressed),
println(len=Compressed.length),
Uncompressed = uncompress(Compressed),
println(uncompressed=Uncompressed),
printf("compressed to %3.3f%%\n", 100*(Compressed.length / S.length)),
if S = Uncompressed then
println("Same!")
else
println("Error: S != Uncompressed!"),
printf("S.length: %d Uncompressed.length: %d\n", S.length, Uncompressed.length),
edit(S,Uncompressed,Distance,Diffs),
println(distance=Distance),
println(diffs=Diffs)
end,
nl.
compress(Uncompressed) = Compressed =>
DictSize = 256,
Dict = new_map([C=C : I in 1..DictSize-1, C=chr(I).to_string()]),
W = "",
Result = [],
foreach(C in Uncompressed)
C := C.to_string(),
WC = W ++ C,
if Dict.has_key(WC) then
W := WC
else
Result := Result ++ [Dict.get(W)],
Dict.put(WC, DictSize),
DictSize := DictSize + 1,
W := C
end
end,
if W.length > 0 then
Result := Result ++ [Dict.get(W)]
end,
Compressed = Result.
uncompress(Compressed) = Uncompressed =>
DictSize = 256,
Dict = new_map([ C=C : I in 1..DictSize-1, C=chr(I).to_string()]),
W = Compressed.first(),
Compressed := Compressed.tail(),
Result = W,
Entry = "",
foreach(K in Compressed)
if Dict.has_key(K) then
Entry := Dict.get(K)
elseif K == DictSize then
Entry := W ++ W[1].to_string()
else
printf("Bad compressed K: %w\n", K)
end,
Result := Result ++ Entry,
Dict.put(DictSize,(W ++ Entry[1].to_string())),
DictSize := DictSize + 1,
W := Entry
end,
Uncompressed = Result.flatten().
%
% Computing the minimal editing distance of two given lists
%
table(+,+,min,-)
edit([],[],D,Diffs) => D=0, Diffs=[].
edit([X|Xs],[X|Ys],D,Diffs) => % copy
edit(Xs,Ys,D,Diffs).
edit(Xs,[Y|Ys],D,Diffs) ?=> % insert
edit(Xs,Ys,D1,Diffs1),
D=D1+1,
Diffs = [insert=Y,xPos=Xs.length,yPos=Ys.length|Diffs1].
edit([X|Xs],Ys,D,Diffs) => % delete
edit(Xs,Ys,D1,Diffs1),
D=D1+1,
Diffs = [[delete=X,xPos=Xs.length,yPos=Ys.length]|Diffs1].</syntaxhighlight>
{{out}}
<pre>s = TOBEORNOTTOBEORTOBEORNOT
len = 24
compressed = [T,O,B,E,O,R,N,O,T,256,258,260,265,259,261,263]
len = 16
uncompressed = TOBEORNOTTOBEORTOBEORNOT
compressed to 66.667%
Same!</pre>
=={{header|PicoLisp}}==
<
(let (Codes 255 Dict)
(balance 'Dict
Line 2,880 ⟶ 4,631:
(when W
(idx 'Dict (cons (inc 'Codes) (cons (last WC) W)) T) )
(setq W WC) ) ) ) ) ) )</
Test:
<pre>: (lzwCompress (chop "TOBEORNOTTOBEORTOBEORNOT"))
Line 2,887 ⟶ 4,638:
: (pack (lzwDecompress @))
-> "TOBEORNOTTOBEORTOBEORNOT"</pre>
=={{header|PL/I}}==
{{trans|REXX}}
The interesting point is the implementation of REXX's associative array (compound variable).
<
lzwt: Proc Options(main);
Line 3,041 ⟶ 4,791:
Return;
End;</
{{out}}
<pre>str=TOBEORNOTTOBEORTOBEORNOT
Line 3,053 ⟶ 4,803:
This is because PureBasic uses these to terminate strings.
Only slight modifications are necessary to handle Null values that would be present for a more generic routine that could be used with a buffer containing any data type.
<
;Compress a string to a list of output symbols
Line 3,149 ⟶ 4,899:
Input()
CloseConsole()
EndIf</
Sample output:
<pre>Type something: TOBEORNOTTOBEORTOBEORNOT
Line 3,156 ⟶ 4,906:
=={{header|Python}}==
{{works with|Python|3.x}}
In this version the dicts contain mixed typed data:
<
"""Compress a string to a list of output symbols."""
# Build the dictionary.
dict_size = 256
dictionary = dict((chr(i),
# in Python 3: dictionary = {chr(i):
w = ""
Line 3,187 ⟶ 4,937:
def decompress(compressed):
"""Decompress a list of output ks to a string."""
from
# Build the dictionary.
dict_size = 256
dictionary = dict(
# in Python 3: dictionary = {
# use StringIO, otherwise this becomes O(N^2)
# due to string concatenation in a loop
result = StringIO()
w = chr(compressed.pop(0))
result.write(w)
for k in compressed:
Line 3,220 ⟶ 4,970:
print (compressed)
decompressed = decompress(compressed)
print (decompressed)</
Output:
<pre>
[
TOBEORNOTTOBEORTOBEORNOT
</pre>
=={{header|Racket}}==
<
#lang racket
; utilities
Line 3,287 ⟶ 5,037:
(def decompressed (decompress compressed))
(displayln decompressed)
</syntaxhighlight>
Output:
Line 3,296 ⟶ 5,046:
</pre>
=={{header|
(formerly Perl 6) I just came across [https://stackoverflow.com/questions/30531078/ this SO question] by chance hence the update. Notably the ancestor Perl entry simply works without any further tweak.
{{trans|Perl}}
<syntaxhighlight lang="raku" line># 20200421 Updated Raku programming solution ; add unicode support
sub compress(Str $uncompressed --> Seq) {
my $dict-size = 256;
my %dictionary = (.chr => .chr for ^$dict-size);
my $w = "";
gather {
for $uncompressed.encode('utf8').list.chrs.comb -> $c {
my $wc = $w ~ $c;
if %dictionary{$wc}:exists { $w = $wc }
else {
take %dictionary{$w};
%dictionary{$wc} = +%dictionary;
$w = $c;
}
}
take %dictionary{$w} if $w.chars;
}
}
sub decompress(@compressed --> Str) {
my $dict-size = 256;
my %dictionary = (.chr => .chr for ^$dict-size);
my $w = shift @compressed;
( Blob.new: flat ( gather {
take $w;
for @compressed -> $k {
my $entry;
if %dictionary{$k}:exists { take $entry = %dictionary{$k} }
elsif $k == $dict-size { take $entry = $w ~ $w.substr(0,1) }
else { die "Bad compressed k: $k" }
%dictionary{$dict-size++} = $w ~ $entry.substr(0,1);
$w = $entry;
}
} )».ords ).decode('utf-8')
}
say my @compressed = compress('TOBEORNOTTOBEORTOBEORNOT');
say decompress(@compressed);
@compressed = compress('こんにちは𝒳𝒴𝒵こんにちは𝒳𝒴𝒵こんにちは𝒳𝒴𝒵');
say decompress(@compressed);</syntaxhighlight>
{{out}}
<pre>
[T O B E O R N O T 256 258 260 265 259 261 263]
TOBEORNOTTOBEORTOBEORNOT
こんにちは𝒳𝒴𝒵こんにちは𝒳𝒴𝒵こんにちは𝒳𝒴𝒵
</pre>
=={{header|REXX}}==
===version 1===
{{trans|Java}}
<
* 20.07.2014 Walter Pachl translated from Java
* 21.07.2014 WP allow for blanks in the string
Line 3,338 ⟶ 5,145:
res=res||dict.w', '
dict.wc=dict_size
dict_size
w=c
End
Line 3,361 ⟶ 5,168:
k=word(compressed,i)
Select
When d.k<>'' | d.k=='
entry=d.k
When k=dict_size Then
Line 3,370 ⟶ 5,177:
res=res||entry
d.dict_size=w||substr(entry,1,1)
dict_size
w=entry
End
Return res</
'''Output:'''
<pre>str=TOBEORNOTTOBEORTOBEORNOT
Line 3,385 ⟶ 5,192:
dec=To be or not to be that is the question!
decompression ok
</pre>
===version 2===
(Based on the '''Java''' program.)
This REXX version can execute on '''ASCII''' or '''EBCDIC''' systems.
<syntaxhighlight lang="rexx">/*REXX program compresses text using the LZW (Lempel─Ziv─Welch), and reconstitutes it.*/
$$$= '"There is nothing permanent except change." ─── Heraclitus [540 ── 475 BCE]'
parse arg text; if text='' then text= $$$ /*get an optional argument from the CL.*/
say 'original text=' text /* [↑] Not specified? Then use default*/
cypher= LZWc(text) /*compress text using the LZW algorithm*/
say 'reconstituted=' LZWd(cypher) /*display the reconstituted string. */
say; say ' LZW integers=' cypher /* " " LZW integers used. */
exit 0 /*stick a fork in it, we're all done. */
/*──────────────────────────────────────────────────────────────────────────────────────*/
LZWi: arg i,@.; #=256; do j=0 for #; _=d2c(j); if i then @.j=_; else @._=j; end; return
/*──────────────────────────────────────────────────────────────────────────────────────*/
LZWc: procedure; parse arg y,,$; call LZWi 0; w= /*LZW compress.*/
do k=1 for length(y)+1; z= w || substr(y, k, 1)
if @.z=='' then do; $= $ @.w; @.z= #; #= # + 1; w= substr(y, k, 1); end
else w= z /*#: the dictionary size.*/
end /*k*/; return substr($, 2) /*elide a leading blank. */
/*──────────────────────────────────────────────────────────────────────────────────────*/
LZWd: procedure; parse arg x y; call LZWi 1; $= @.x; w= $ /*LZW decompress.*/
do k=1 for words(y); z= word(y, k)
if @.z\=='' | @.k==" " then ?= @.z
else if z==# then ?= w || left(w, 1)
$= $ || ?
@.#= w || left(?, 1); w= ?; #= # + 1 /*bump dict. size*/
end /*k*/; return $</syntaxhighlight>
{{out|output|text= when using the default input:}}
<pre>
original text= "There is nothing permanent except change." ─── Heraclitus [540 ── 475 BCE]
reconstituted= "There is nothing permanent except change." ─── Heraclitus [540 ── 475 BCE]
LZW integers= 34 84 104 101 114 101 32 105 115 32 110 111 116 104 105 110 103 32 112 259 109 97 110 101 110 116 32 101 120 99 101 112 281 99 104 277 103 101 46 34 32 296 196 298 296 32 72 259 97 99 108 105 116 117 264 32 91 53 52 48 32 299 52 55 53 32 66 67 69 93
</pre>
=={{header|Ring}}==
<syntaxhighlight lang="ring">
# Project : LZW compression
plaintext = "TOBEORNOTTOBEORTOBEORNOT"
result = []
encode = encodelzw(plaintext)
for i = 1 to len(encode) step 2
add(result,ascii(substr(encode,i,1)) + 256*ascii(substr(encode,i+1,1)))
next
showarray(result)
see decodelzw(encode)
func encodelzw(text)
o = ""
dict = list(4096)
for i = 1 to 255
dict[i] = char(i)
next
l = i
i = 1
w = left(text,1)
while i < len(text)
d = 0
while d < l
c = d
if i > len(text)
exit
ok
for d = 1 to l
if w = dict[d]
exit
ok
next
if d < l
i = i + 1
w = w + substr(text,i,1)
ok
end
dict[l] = w
l = l + 1
w = right(w,1)
o = o + char(c % 256) + char(floor(c / 256))
end
return o
func decodelzw(text)
o = ""
dict = list(4096)
for i = 1 to 255
dict[i] = char(i)
next
l = i
c = ascii(left(text,1)) + 256*ascii(substr(text,2,1))
w = dict[c]
o = w
if len(text) < 4
return o
ok
for i = 3 to len(text) step 2
c = ascii(substr(text,i,1)) + 256*ascii(substr(text,i+1,1))
if c < l
t = dict[c]
else
t = w + left(w,1)
ok
o = o + t
dict[l] = w + left(t,1)
l = l + 1
w = t
next
return o
func showarray(vect)
svect = ""
for n = 1 to len(vect)
svect = svect + vect[n] + " "
next
svect = left(svect, len(svect) - 1)
see svect + nl
</syntaxhighlight>
Output:
<pre>
84 79 66 69 79 82 78 79 84 256 258 260 265 259 261 263
TOBEORNOTTOBEORTOBEORNOT
</pre>
Line 3,390 ⟶ 5,320:
In this version the hashes contain mixed typed data:
<
def compress(uncompressed)
# Build the dictionary.
Line 3,446 ⟶ 5,376:
p compressed
decompressed = decompress(compressed)
puts decompressed</
Output:
<pre>
["T", "O", "B", "E", "O", "R", "N", "O", "T", 256, 258, 260, 265, 259, 261, 263]
TOBEORNOTTOBEORTOBEORNOT
</pre>
=={{header|Rust}}==
{{trans|C Sharp}}
Handles arbitrary byte sequences.
<syntaxhighlight lang="rust">use std::collections::HashMap;
fn compress(data: &[u8]) -> Vec<u32> {
// Build initial dictionary.
let mut dictionary: HashMap<Vec<u8>, u32> = (0u32..=255)
.map(|i| (vec![i as u8], i))
.collect();
let mut w = Vec::new();
let mut compressed = Vec::new();
for &b in data {
let mut wc = w.clone();
wc.push(b);
if dictionary.contains_key(&wc) {
w = wc;
} else {
// Write w to output.
compressed.push(dictionary[&w]);
// wc is a new sequence; add it to the dictionary.
dictionary.insert(wc, dictionary.len() as u32);
w.clear();
w.push(b);
}
}
// Write remaining output if necessary.
if !w.is_empty() {
compressed.push(dictionary[&w]);
}
compressed
}
fn decompress(mut data: &[u32]) -> Vec<u8> {
// Build the dictionary.
let mut dictionary: HashMap::<u32, Vec<u8>> = (0u32..=255)
.map(|i| (i, vec![i as u8]))
.collect();
let mut w = dictionary[&data[0]].clone();
data = &data[1..];
let mut decompressed = w.clone();
for &k in data {
let entry = if dictionary.contains_key(&k) {
dictionary[&k].clone()
} else if k == dictionary.len() as u32 {
let mut entry = w.clone();
entry.push(w[0]);
entry
} else {
panic!("Invalid dictionary!");
};
decompressed.extend_from_slice(&entry);
// New sequence; add it to the dictionary.
w.push(entry[0]);
dictionary.insert(dictionary.len() as u32, w);
w = entry;
}
decompressed
}
fn main() {
let compressed = compress("TOBEORNOTTOBEORTOBEORNOT".as_bytes());
println!("{:?}", compressed);
let decompressed = decompress(&compressed);
let decompressed = String::from_utf8(decompressed).unwrap();
println!("{}", decompressed);
}</syntaxhighlight>
Output:
<pre>
[84, 79, 66, 69, 79, 82, 78, 79, 84, 256, 258, 260, 265, 259, 261, 263]
TOBEORNOTTOBEORTOBEORNOT
</pre>
=={{header|Scala}}==
<
def compress(tc:String) = {
//initial dictionary
Line 3,501 ⟶ 5,518:
val result = decompress(compressed)
println(result)
</syntaxhighlight>
=={{header|Scheme}}==
<
(define (member-string-ref m l)
(define r #f)
Line 3,575 ⟶ 5,592:
(append dictionary
(list (string-append
;; Build the resulting string
Line 3,588 ⟶ 5,605:
(display compressed) (newline)
(define decompressed (lzw-decompress compressed))
(display decompressed) (newline)</
Output:<pre>(84 79 66 69 79 82 78 79 84 256 258 260 265 259 261 263)
TOBEORNOTTOBEORTOBEORNOT</pre>
=={{header|Seed7}}==
<
const func string: lzwCompress (in string: uncompressed) is func
result
var string:
local
var char: ch is ' ';
Line 3,612 ⟶ 5,629:
buffer &:= str(ch)
else
mydict @:= [xstr] chr(length(mydict));
buffer := str(ch);
Line 3,618 ⟶ 5,635:
end for;
if buffer <> "" then
end if;
end func;
const func string: lzwDecompress (in string: compressed) is func
result
var string:
local
var char: ch is ' ';
Line 3,638 ⟶ 5,655:
if buffer = "" then
buffer := mydict[ch];
elsif ch <= chr(255) then
current := mydict[ch];
chain := buffer & current;
mydict @:= [chr(length(mydict))] chain;
Line 3,651 ⟶ 5,668:
chain := buffer & str(buffer[1]);
end if;
mydict @:= [chr(length(mydict))] buffer & str(chain[1]);
buffer := chain;
Line 3,657 ⟶ 5,674:
end for;
end func;
const proc: main is func
local
Line 3,667 ⟶ 5,684:
uncompressed := lzwDecompress(compressed);
writeln(uncompressed);
end func;</
Output:
<pre>
"TOBEORNOT\256
TOBEORNOTTOBEORTOBEORNOT
</pre>
Line 3,679 ⟶ 5,696:
=={{header|Sidef}}==
{{trans|Perl}}
<
func compress(String uncompressed
# Build the dictionary.
var dict_size = 256
var dictionary = Hash
dictionary
}
var w = ''
var result = []
uncompressed.each { |c|
var wc = w+c
if (dictionary.exists(wc)) {
w = wc
} else {
result
# Add wc to the dictionary.
dictionary
dict_size++
w = c
}
}
# Output the code for w.
if (w != '') {
result
}
return result
}
# Decompress a list of output ks to a string.
func decompress(Array compressed
# Build the dictionary.
var dict_size = 256
var dictionary = Hash
dictionary
}
var w = compressed.shift
var result = w
compressed.each { |k|
var entry
if (dictionary.exists(k)) {
entry = dictionary
} elsif (k == dict_size) {
entry = w+w.
} else {
die "Bad compressed k: #{k}"
}
result += entry
# Add w+entry[0] to the dictionary.
dictionary
dict_size++
w = entry
}
return result
}
# How to use:
var compressed = compress('TOBEORNOTTOBEORTOBEORNOT')
say compressed.join(' ')
var decompressed = decompress(compressed)
say decompressed
{{out}}
<pre>T O B E O R N O T 256 258 260 265 259 261 263
Line 3,757 ⟶ 5,774:
=={{header|Swift}}==
{{trans|JavaScript}}
<
class func compress(_ uncompressed:String) -> [Int] {
var dict = [String : Int]()
for i in 0 ..< 256 {
let s = String(
dict[s] = i
}
var dictSize = 256
var w = ""
var result = [Int]()
for c in uncompressed {
let wc = w + String(c)
if dict[wc] != nil {
w = wc
} else {
result.append(dict[w]!)
dict[wc] = dictSize
dictSize += 1
w = String(c)
}
}
if w != "" {
result.append(dict[w]!)
}
return result
}
class func decompress(_ compressed:[Int]) -> String? {
var
for
for k in compressed[1 ..< compressed.count] {
let entry : String
if let x = dict[k] {
entry = x
} else if k == dictSize {
entry = w + String(w[w.startIndex])
} else {
return nil
}
result += entry
dict[dictSize] = w + String(entry[entry.startIndex])
dictSize += 1
w = entry
}
return result
}
}
let comp = LZW.compress("TOBEORNOTTOBEORTOBEORNOT")
if let decomp = LZW.decompress(comp) {
}</
{{out}}
<pre>
Line 3,826 ⟶ 5,846:
=={{header|Tcl}}==
<
variable char2int
variable chars
Line 3,885 ⟶ 5,905:
# or
if {$s eq [LZW::decode [LZW::encode $s]]} then {puts success} else {puts fail} ;# ==> success</
=={{header|VBScript}}==
<syntaxhighlight lang="vb">
Option Explicit
Const numchars=127 'plain ASCII
Function LZWCompress(si)
Dim oDict, intMaxCode, i,z,ii,ss,strCurrent,strNext,j
Set oDict = CreateObject("Scripting.Dictionary")
ReDim a(Len(si))
intMaxCode = numchars
For i = 0 To numchars
oDict.Add Chr(i), i
Next
'strCurrent = ofread.ReadText(1)
strCurrent = Left(si,1)
j=0
For ii=2 To Len(si)
strNext = Mid(si,ii,1)
ss=strCurrent & strNext
If oDict.Exists(ss) Then
strCurrent = ss
Else
a(j)=oDict.Item(strCurrent) :j=j+1
intMaxCode = intMaxCode + 1
oDict.Add ss, intMaxCode
strCurrent = strNext
End If
Next
a(j)=oDict.Item(strCurrent)
ReDim preserve a(j)
LZWCompress=a
Set oDict = Nothing
End Function
Function lzwUncompress(sc)
Dim intNext, intCurrent, intMaxCode, i,ss,istr,s,j
s=""
reDim dict(1000)
intMaxCode = numchars
For i = 0 To numchars : dict(i)= Chr(i) : Next
intCurrent=sc(0)
For j=1 To UBound(sc)
ss=dict(intCurrent)
s= s & ss
intMaxCode = intMaxCode + 1
intnext=sc(j)
If intNext<intMaxCode Then
dict(intMaxCode)=ss & Left(dict(intNext), 1)
Else
dict(intMaxCode)=ss & Left(ss, 1)
End If
intCurrent = intNext
Next
s= s & dict(intCurrent)
lzwUncompress=s
End function
Sub printvec(a)
Dim s,i,x
s="("
For i=0 To UBound (a)
s=s & x & a(i)
x=", "
Next
WScript.echo s &")"
End sub
Dim a,b
b="TOBEORNOTTOBEORTOBEORNOT"
WScript.Echo b
a=LZWCompress (b)
printvec(a)
WScript.echo lzwUncompress (a )
wscript.quit 1
</syntaxhighlight>
{{out}}
<small>
<pre>
TOBEORNOTTOBEORTOBEORNOT
(84, 79, 66, 69, 79, 82, 78, 79, 84, 128, 130, 132, 137, 131, 133, 135)
TOBEORNOTTOBEORTOBEORNOT
</pre>
</small>
=={{header|Wren}}==
{{trans|Kotlin}}
<syntaxhighlight lang="wren">class LZW {
/* Compress a string to a list of output symbols. */
static compress(uncompressed) {
// Build the dictionary.
var dictSize = 256
var dictionary = {}
for (i in 0...dictSize) dictionary[String.fromByte(i)] = i
var w = ""
var result = []
for (c in uncompressed.bytes) {
var cs = String.fromByte(c)
var wc = w + cs
if (dictionary.containsKey(wc)) {
w = wc
} else {
result.add(dictionary[w])
// Add wc to the dictionary.
dictionary[wc] = dictSize
dictSize = dictSize + 1
w = cs
}
}
// Output the code for w
if (w != "") result.add(dictionary[w])
return result
}
/* Decompress a list of output symbols to a string. */
static decompress(compressed) {
// Build the dictionary.
var dictSize = 256
var dictionary = {}
for (i in 0...dictSize) dictionary[i] = String.fromByte(i)
var w = String.fromByte(compressed[0])
var result = w
for (k in compressed.skip(1)) {
var entry
if (dictionary.containsKey(k)) {
entry = dictionary[k]
} else if (k == dictSize) {
entry = w + String.fromByte(w.bytes[0])
} else {
Fiber.abort("Bad compressed k: %(k)")
}
result = result + entry
// Add w + entry[0] to the dictionary.
dictionary[dictSize] = w + String.fromByte(entry.bytes[0])
dictSize = dictSize + 1
w = entry
}
return result
}
}
var compressed = LZW.compress("TOBEORNOTTOBEORTOBEORNOT")
System.print(compressed)
var decompressed = LZW.decompress(compressed)
System.print(decompressed)</syntaxhighlight>
{{out}}
<pre>
[84, 79, 66, 69, 79, 82, 78, 79, 84, 256, 258, 260, 265, 259, 261, 263]
TOBEORNOTTOBEORTOBEORNOT
</pre>
=={{header|Xojo}}==
{{trans|PHP}}
<syntaxhighlight lang="vb">
Function compress(str as String) As String
Dim i as integer
Dim w as String = ""
Dim c as String
Dim strLen as Integer
Dim wc as string
Dim dic() as string
Dim result() as string
Dim lookup as integer
Dim startPos as integer = 0
Dim combined as String
strLen = len(str)
dic = populateDictionary
for i = 1 to strLen
c = str.mid(i, 1)
wc = w + c
startPos = getStartPos(wc)
lookup = findArrayPos(dic, wc, startPos)
if (lookup <> -1) then
w = wc
Else
startPos = getStartPos(w)
lookup = findArrayPos(dic, w, startPos)
if lookup <> -1 then
result.Append(lookup.ToText)
end if
dic.append(wc)
w = c
end if
next i
if (w <> "") then
startPos = getStartPos(w)
lookup = findArrayPos(dic, w, startPos)
result.Append(lookup.ToText)
end if
return join(result, ",")
End Function
Function decompress(str as string) As string
dim comStr() as string
dim w as string
dim result as string
dim comStrLen as integer
dim entry as string
dim dic() as string
dim i as integer
comStr = str.Split(",")
comStrLen = comStr.Ubound
dic = populateDictionary
w = chr(val(comStr(0)))
result = w
for i = 1 to comStrLen
entry = dic(val(comStr(i)))
result = result + entry
dic.append(w + entry.mid(1,1))
w = entry
next i
return result
End Function
Private Function findArrayPos(arr() as String, search as String, start as integer) As Integer
dim arraySize as Integer
dim arrayPosition as Integer = -1
dim i as Integer
arraySize = UBound(arr)
for i = start to arraySize
if (strcomp(arr(i), search, 0) = 0) then
arrayPosition = i
exit
end if
next i
return arrayPosition
End Function
Private Function getStartPos(str as String) As integer
if (len(str) = 1) then
return 0
else
return 255
end if
End Function
Private Function populateDictionary() As string()
dim dic() as string
dim i as integer
for i = 0 to 255
dic.append(Chr(i))
next i
return dic
End Function
</syntaxhighlight>
Test:
<pre>
Dim str as String = "TOBEORNOTTOBEORTOBEORNOT"
Dim com as string = LZW.compress(str)
print com
print LZW.decompress(com)
</pre>
Output:
<pre>
84,79,66,69,79,82,78,79,84,256,258,260,265,259,261,263
TOBEORNOTTOBEORTOBEORNOT
</pre>
=={{header|zkl}}==
<syntaxhighlight lang="zkl">fcn lzwCompress(uncompressed){ // text-->list of 12 bit ints
dictionary:=(256).pump(Dictionary(),fcn(n){ return(n.toChar(),n) });
w,compressed:="",List();
foreach c in (uncompressed){
wc:=w+c;
if(dictionary.holds(wc)) w=wc;
else{
compressed.append(dictionary[w]); // 12 bits
dictionary[wc]=dictionary.len();
w=c;
}
}
if(w) compressed.append(dictionary[w]);
compressed
}
fcn lzwUncompress(compressed){ // compressed data-->text
dictionary:=(256).pump(Dictionary(),fcn(n){ return(n,n.toChar()) });
w,decommpressed:=dictionary[compressed[0]],Data(Void,w);
foreach k in (compressed[1,*]){
if(dictionary.holds(k)) entry:=dictionary[k];
else if(k==dictionary.len()) entry:=w+w[0];
else throw(Exception.ValueError("Invalid compressed data"));
decommpressed.append(entry);
dictionary.add(dictionary.len(),w+entry[0]);
w=entry;
}
decommpressed.text
}</syntaxhighlight>
<syntaxhighlight lang="zkl">compressed:=lzwCompress("TOBEORNOTTOBEORTOBEORNOT");
compressed.toString(*).println();
lzwUncompress(compressed).println();</syntaxhighlight>
{{out}}
<pre>
L(84,79,66,69,79,82,78,79,84,256,258,260,265,259,261,263)
TOBEORNOTTOBEORTOBEORNOT
</pre>
{{omit from|Lilypond}}
|