Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add new operation: Shuffle #1472

Merged
merged 1 commit into from
Nov 25, 2022
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions src/core/config/Categories.json
Original file line number Diff line number Diff line change
Expand Up @@ -249,6 +249,7 @@
"To Table",
"Reverse",
"Sort",
"Shuffle",
"Unique",
"Split",
"Filter",
Expand Down
157 changes: 157 additions & 0 deletions src/core/operations/Shuffle.mjs
Original file line number Diff line number Diff line change
@@ -0,0 +1,157 @@
/**
* @author mikecat
* @copyright Crown Copyright 2022
* @license Apache-2.0
*/

import Operation from "../Operation.mjs";

/**
* Shuffle operation
*/
class Shuffle extends Operation {

/**
* Shuffle constructor
*/
constructor() {
super();

this.name = "Shuffle";
this.module = "Default";
this.description = "Randomly reorders input elements.";
this.infoURL = "https://wikipedia.org/wiki/Shuffling";
this.inputType = "ArrayBuffer";
this.outputType = "ArrayBuffer";
this.args = [
{
"name": "By",
"type": "option",
"value": ["Byte", "Character", "Line"],
"defaultIndex": 1
}
];
}

/**
* @param {ArrayBuffer} input
* @param {Object[]} args
* @returns {ArrayBuffer}
*/
run(input, args) {
const type = args[0];
if (input.byteLength === 0) return input;
if (ArrayBuffer.isView(input)) {
if (input.byteOffset === 0 && input.byteLength === input.buffer.byteLength) {
input = input.buffer;
} else {
input = input.buffer.slice(input.byteOffset, input.byteOffset + input.byteLength);
}
}
const inputBytes = new Uint8Array(input);

// return a random number in [0, 1)
const rng = (typeof crypto) !== "undefined" && crypto.getRandomValues ? (function() {
const buf = new Uint32Array(2);
return function() {
// generate 53-bit random integer: 21 + 32 bits
crypto.getRandomValues(buf);
const value = (buf[0] >>> (32 - 21)) * ((1 << 30) * 4) + buf[1];
return value / ((1 << 23) * (1 << 30));
};
})() : Math.random;

// return a random integer in [0, max)
const randint = function(max) {
return Math.floor(rng() * max);
};

const toShuffle = [];
let addLastNewLine = false;
switch (type) {
case "Character":
// split input into UTF-8 code points
for (let i = 0; i < inputBytes.length;) {
const charLength = (function() {
if (inputBytes[i] < 0xc0) return 1;
if (inputBytes[i] < 0xe0) return 2;
if (inputBytes[i] < 0xf0) return 3;
if (inputBytes[i] < 0xf8) return 4;
return 1;
})();
if (i + charLength <= inputBytes.length) {
let elementLength = charLength;
for (let j = 1; j < charLength; j++) {
if ((inputBytes[i + j] & 0xc0) !== 0x80) {
elementLength = 1;
break;
}
}
toShuffle.push([i, elementLength]);
i += elementLength;
} else {
toShuffle.push([i, 1]);
i++;
}
}
break;
case "Line":
{
// split input by newline characters
let lineBegin = 0;
for (let i = 0; i < inputBytes.length; i++) {
if (inputBytes[i] === 0xd || inputBytes[i] === 0xa) {
if (i + 1 < inputBytes.length && inputBytes[i] === 0xd && inputBytes[i + 1] === 0xa) {
i++;
}
toShuffle.push([lineBegin, i - lineBegin + 1]);
lineBegin = i + 1;
}
}
if (lineBegin < inputBytes.length) {
toShuffle.push([lineBegin, inputBytes.length - lineBegin]);
addLastNewLine = true;
}
}
break;
default:
{
// Creating element information for each bytes looks very wasteful.
// Therefore, directly shuffle here.
const outputBytes = new Uint8Array(inputBytes);
for (let i = outputBytes.length - 1; i > 0; i--) {
const idx = randint(i + 1);
const tmp = outputBytes[idx];
outputBytes[idx] = outputBytes[i];
outputBytes[i] = tmp;
}
return outputBytes.buffer;
}
}

// shuffle elements
const lastStart = toShuffle[toShuffle.length - 1][0];
for (let i = toShuffle.length - 1; i > 0; i--) {
const idx = randint(i + 1);
const tmp = toShuffle[idx];
toShuffle[idx] = toShuffle[i];
toShuffle[i] = tmp;
}

// place shuffled elements
const outputBytes = new Uint8Array(inputBytes.length + (addLastNewLine ? 1 : 0));
let outputPos = 0;
for (let i = 0; i < toShuffle.length; i++) {
outputBytes.set(new Uint8Array(input, toShuffle[i][0], toShuffle[i][1]), outputPos);
outputPos += toShuffle[i][1];
if (addLastNewLine && toShuffle[i][0] === lastStart) {
outputBytes[outputPos] = 0xa;
outputPos++;
}
}
return outputBytes.buffer;
}

}

export default Shuffle;
1 change: 1 addition & 0 deletions tests/operations/index.mjs
Original file line number Diff line number Diff line change
Expand Up @@ -124,6 +124,7 @@ import "./tests/UnescapeString.mjs";
import "./tests/LS47.mjs";
import "./tests/LZString.mjs";
import "./tests/NTLM.mjs";
import "./tests/Shuffle.mjs";

// Cannot test operations that use the File type yet
// import "./tests/SplitColourChannels.mjs";
Expand Down
92 changes: 92 additions & 0 deletions tests/operations/tests/Shuffle.mjs
Original file line number Diff line number Diff line change
@@ -0,0 +1,92 @@
/**
* @author mikecat
* @copyright Crown Copyright 2022
* @license Apache-2.0
*/
import TestRegister from "../../lib/TestRegister.mjs";

TestRegister.addTests([
{
"name": "Shuffle empty",
"input": "",
"expectedOutput": "",
"recipeConfig": [
{
"op": "Shuffle",
"args": ["Character"]
}
]
},
{
"name": "Shuffle bytes",
"input": "12345678",
"expectedOutput": "31 32 33 34 35 36 37 38",
"recipeConfig": [
{
"op": "Shuffle",
"args": ["Byte"]
},
{
"op": "To Hex",
"args": ["Space", 0]
},
{
"op": "Sort",
"args": ["Space", false, "Alphabetical (case sensitive)"]
}
]
},
{
"name": "Shuffle characters",
"input": "1234\uff15\uff16\uff17\uff18",
"expectedOutput": " 0031 0032 0033 0034 FF15 FF16 FF17 FF18",
"recipeConfig": [
{
"op": "Shuffle",
"args": ["Character"]
},
{
"op": "Escape Unicode Characters",
"args": ["%u", true, 4, true]
},
{
"op": "Split",
"args": ["%u", " "]
},
{
"op": "Sort",
"args": ["Space", false, "Alphabetical (case sensitive)"]
}
]
},
{
"name": "Shuffle lines",
"input": "1\n2\n3\n4\n5\n6\n7\n8\n9\na\nb\nc\nd\ne\nf\n",
"expectedOutput": "\n1\n2\n3\n4\n5\n6\n7\n8\n9\na\nb\nc\nd\ne\nf",
"recipeConfig": [
{
"op": "Shuffle",
"args": ["Line"]
},
{
"op": "Sort",
"args": ["Line feed", false, "Alphabetical (case sensitive)"]
}
]
},
{
"name": "Shuffle lines (last character is not newline)",
"input": "1\n2\n3\n4\n5\n6\n7\n8\n9\na\nb\nc\nd\ne\nf",
"expectedOutput": "\n1\n2\n3\n4\n5\n6\n7\n8\n9\na\nb\nc\nd\ne\nf",
"recipeConfig": [
{
"op": "Shuffle",
"args": ["Line"]
},
{
"op": "Sort",
"args": ["Line feed", false, "Alphabetical (case sensitive)"]
}
]
},
]);