Created
July 7, 2018 15:00
-
-
Save holly-hacker/52ebf0a55230bc1b2a0b5b5cd7a10baf to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
namespace SMAZ { | |
const codebook: { [idx: string]: number } = { | |
// tslint:disable-next-line:object-literal-sort-keys | |
" ": 0, "the": 1, "e": 2, "t": 3, "a": 4, "of": 5, "o": 6, "and": 7, "i": 8, "n": 9, "s": 10, "e ": 11, "r": 12, " th": 13, " t": 14, | |
"in": 15, "he": 16, "th": 17, "h": 18, "he ": 19, "to": 20, "\r\n": 21, "l": 22, "s ": 23, "d": 24, " a": 25, "an": 26, "er": 27, "c": 28, | |
" o": 29, "d ": 30, "on": 31, " of": 32, "re": 33, "of ": 34, "t ": 35, ", ": 36, "is": 37, "u": 38, "at": 39, " ": 40, "n ": 41, "or": 42, | |
"which": 43, "f": 44, "m": 45, "as": 46, "it": 47, "that": 48, "\n": 49, "was": 50, "en": 51, " ": 52, " w": 53, "es": 54, " an": 55, | |
" i": 56, "\r": 57, "f ": 58, "g": 59, "p": 60, "nd": 61, " s": 62, "nd ": 63, "ed ": 64, "w": 65, "ed": 66, "http://": 67, "for": 68, | |
"te": 69, "ing": 70, "y ": 71, "The": 72, " c": 73, "ti": 74, "r ": 75, "his": 76, "st": 77, " in": 78, "ar": 79, "nt": 80, ",": 81, | |
" to": 82, "y": 83, "ng": 84, " h": 85, "with": 86, "le": 87, "al": 88, "to ": 89, "b": 90, "ou": 91, "be": 92, "were": 93, " b": 94, | |
"se": 95, "o ": 96, "ent": 97, "ha": 98, "ng ": 99, "their": 100, "\"": 101, "hi": 102, "from": 103, " f": 104, "in ": 105, "de": 106, | |
"ion": 107, "me": 108, "v": 109, ".": 110, "ve": 111, "all": 112, "re ": 113, "ri": 114, "ro": 115, "is ": 116, "co": 117, "f t": 118, | |
"are": 119, "ea": 120, ". ": 121, "her": 122, " m": 123, "er ": 124, " p": 125, "es ": 126, "by": 127, "they": 128, "di": 129, "ra": 130, | |
"ic": 131, "not": 132, "s,": 133, "d t": 134, "at ": 135, "ce": 136, "la": 137, "h ": 138, "ne": 139, "as ": 140, "tio": 141, "on ": 142, | |
"n t": 143, "io": 144, "we": 145, " a ": 146, "om": 147, ", a": 148, "s o": 149, "ur": 150, "li": 151, "ll": 152, "ch": 153, "had": 154, | |
"this": 155, "e t": 156, "g ": 157, "e\r\n": 158, " wh": 159, "ere": 160, " co": 161, "e o": 162, "a ": 163, "us": 164, " d": 165, "ss": 166, | |
"\n\r\n": 167, "\r\n\r": 168, "=\"": 169, " be": 170, " e": 171, "s a": 172, "ma": 173, "one": 174, "t t": 175, "or ": 176, "but": 177, | |
"el": 178, "so": 179, "l ": 180, "e s": 181, /*"s,": 182,*/ "no": 183, "ter": 184, " wa": 185, "iv": 186, "ho": 187, "e a": 188, " r": 189, | |
"hat": 190, "s t": 191, "ns": 192, "ch ": 193, "wh": 194, "tr": 195, "ut": 196, "/": 197, "have": 198, "ly ": 199, "ta": 200, " ha": 201, | |
" on": 202, "tha": 203, "-": 204, " l": 205, "ati": 206, "en ": 207, "pe": 208, " re": 209, "there": 210, "ass": 211, "si": 212, " fo": 213, | |
"wa": 214, "ec": 215, "our": 216, "who": 217, "its": 218, "z": 219, "fo": 220, "rs": 221, ">": 222, "ot": 223, "un": 224, "<": 225, "im": 226, | |
"th ": 227, "nc": 228, "ate": 229, "><": 230, "ver": 231, "ad": 232, " we": 233, "ly": 234, "ee": 235, " n": 236, "id": 237, " cl": 238, | |
"ac": 239, "il": 240, "</": 241, "rt": 242, " wi": 243, "div": 244, "e, ": 245, " it": 246, "whi": 247, " ma": 248, "ge": 249, "x": 250, | |
"e c": 251, "men": 252, ".com": 253, | |
}; | |
const reverseCodebook = [ | |
" ", "the", "e", "t", "a", "of", "o", "and", "i", "n", "s", "e ", "r", " th", " t", "in", "he", "th", "h", "he ", "to", "\r\n", "l", "s ", | |
"d", " a", "an", "er", "c", " o", "d ", "on", " of", "re", "of ", "t ", ", ", "is", "u", "at", " ", "n ", "or", "which", "f", "m", "as", | |
"it", "that", "\n", "was", "en", " ", " w", "es", " an", " i", "\r", "f ", "g", "p", "nd", " s", "nd ", "ed ", "w", "ed", "http://", "for", | |
"te", "ing", "y ", "The", " c", "ti", "r ", "his", "st", " in", "ar", "nt", ",", " to", "y", "ng", " h", "with", "le", "al", "to ", "b", "ou", | |
"be", "were", " b", "se", "o ", "ent", "ha", "ng ", "their", "\"", "hi", "from", " f", "in ", "de", "ion", "me", "v", ".", "ve", "all", "re ", | |
"ri", "ro", "is ", "co", "f t", "are", "ea", ". ", "her", " m", "er ", " p", "es ", "by", "they", "di", "ra", "ic", "not", "s, ", "d t", | |
"at ", "ce", "la", "h ", "ne", "as ", "tio", "on ", "n t", "io", "we", " a ", "om", ", a", "s o", "ur", "li", "ll", "ch", "had", "this", | |
"e t", "g ", "e\r\n", " wh", "ere", " co", "e o", "a ", "us", " d", "ss", "\n\r\n", "\r\n\r", "=\"", " be", " e", "s a", "ma", "one", "t t", | |
"or ", "but", "el", "so", "l ", "e s", "s,", "no", "ter", " wa", "iv", "ho", "e a", " r", "hat", "s t", "ns", "ch ", "wh", "tr", "ut", "/", | |
"have", "ly ", "ta", " ha", " on", "tha", "-", " l", "ati", "en ", "pe", " re", "there", "ass", "si", " fo", "wa", "ec", "our", "who", "its", | |
"z", "fo", "rs", ">", "ot", "un", "<", "im", "th ", "nc", "ate", "><", "ver", "ad", " we", "ly", "ee", " n", "id", " cl", "ac", "il", "</", | |
"rt", " wi", "div", "e, ", " it", "whi", " ma", "ge", "x", "e c", "men", ".com", | |
]; | |
export function compress(input: string) { | |
let verbatim = ""; | |
let output: string[] = []; | |
let inputIndex = 0; | |
while (inputIndex < input.length) { | |
let encoded = false; | |
let j = 7; | |
if (input.length - inputIndex < 7) | |
j = input.length - inputIndex; | |
for (; j >= 0; j--) { | |
let code = codebook[input.substr(inputIndex, j)]; | |
if (code != null) { | |
if (verbatim) { | |
output = output.concat(flush_verbatim(verbatim)); | |
verbatim = ""; | |
} | |
output.push(String.fromCharCode(code)); | |
inputIndex += j; | |
encoded = true; | |
break; | |
} | |
} | |
if (!encoded) { | |
verbatim += input[inputIndex]; | |
inputIndex++; | |
if (verbatim.length === 256) { | |
output = output.concat(flush_verbatim(verbatim)); | |
verbatim = ""; | |
} | |
} | |
} | |
if (verbatim) | |
output = output.concat(flush_verbatim(verbatim)); | |
return output.join(""); | |
} | |
export function decompress(inputString: string) { | |
let output = ""; | |
// Apparently the fastest way: https://jsperf.com/strings-to-number-arrays | |
let input: number[] = []; | |
for (let x = 0; x < inputString.length; x++) | |
input.push(inputString.charCodeAt(x)); | |
let i = 0; | |
while (i < input.length) { | |
if (input[i] === 254) { | |
if (i + 1 > input.length) | |
throw new Error("Malformed SMAZ"); | |
output += inputString[i + 1]; | |
i += 2; | |
} else if (input[i] === 255) { | |
if (i + input[i + 1] + 2 >= input.length) | |
throw new Error("Malformed SMAZ"); | |
for (let j = 0; j < input[i + 1] + 1; j++) | |
output += inputString[i + 2 + j]; | |
i += 3 + input[i + 1]; | |
} else { | |
output += reverseCodebook[input[i]]; | |
i++; | |
} | |
} | |
return output; | |
} | |
function flush_verbatim(verbatim: string) { | |
let output: string[] = []; | |
if (verbatim.length > 1) { | |
output.push(String.fromCharCode(255)); | |
output.push(String.fromCharCode(verbatim.length - 1)); | |
} else { | |
output.push(String.fromCharCode(254)); | |
} | |
for (const x of verbatim) | |
output.push(x); | |
return output; | |
} | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment