ESM: Ported case converters, generic beautifier and syntax highlighting

This commit is contained in:
Matt C 2018-05-26 18:04:53 +01:00
parent 176e83a79f
commit 905bc6699e
6 changed files with 428 additions and 0 deletions

31
src/core/lib/Code.mjs Normal file
View File

@ -0,0 +1,31 @@
/**
* Code functions.
*
* @author n1474335 [n1474335@gmail.com]
*
* @copyright Crown Copyright 2018
* @license Apache-2.0
*
*/
/**
* This tries to rename variable names in a code snippet according to a function.
*
* @param {string} input
* @param {function} replacer - this function will be fed the token which should be renamed.
* @returns {string}
*/
export function replaceVariableNames(input, replacer) {
const tokenRegex = /\\"|"(?:\\"|[^"])*"|(\b[a-z0-9\-_]+\b)/ig;
return input.replace(tokenRegex, (...args) => {
const match = args[0],
quotes = args[1];
if (!quotes) {
return match;
} else {
return replacer(match);
}
});
}

View File

@ -0,0 +1,161 @@
/**
* @author n1474335 [n1474335@gmail.com]
* @copyright Crown Copyright 2016
* @license Apache-2.0
*/
import Operation from "../Operation";
/**
* Generic Code Beautify operation
*/
class GenericCodeBeautify extends Operation {
/**
* GenericCodeBeautify constructor
*/
constructor() {
super();
this.name = "Generic Code Beautify";
this.module = "Code";
this.description = "Attempts to pretty print C-style languages such as C, C++, C#, Java, PHP, JavaScript etc.<br><br>This will not do a perfect job, and the resulting code may not work any more. This operation is designed purely to make obfuscated or minified code more easy to read and understand.<br><br>Things which will not work properly:<ul><li>For loop formatting</li><li>Do-While loop formatting</li><li>Switch/Case indentation</li><li>Certain bit shift operators</li></ul>";
this.inputType = "string";
this.outputType = "string";
this.args = [];
}
/**
* @param {string} input
* @param {Object[]} args
* @returns {string}
*/
run(input, args) {
const preservedTokens = [];
let code = input,
t = 0,
m;
// Remove strings
const sstrings = /'([^'\\]|\\.)*'/g;
while ((m = sstrings.exec(code))) {
code = preserveToken(code, m, t++);
sstrings.lastIndex = m.index;
}
const dstrings = /"([^"\\]|\\.)*"/g;
while ((m = dstrings.exec(code))) {
code = preserveToken(code, m, t++);
dstrings.lastIndex = m.index;
}
// Remove comments
const scomments = /\/\/[^\n\r]*/g;
while ((m = scomments.exec(code))) {
code = preserveToken(code, m, t++);
scomments.lastIndex = m.index;
}
const mcomments = /\/\*[\s\S]*?\*\//gm;
while ((m = mcomments.exec(code))) {
code = preserveToken(code, m, t++);
mcomments.lastIndex = m.index;
}
const hcomments = /(^|\n)#[^\n\r#]+/g;
while ((m = hcomments.exec(code))) {
code = preserveToken(code, m, t++);
hcomments.lastIndex = m.index;
}
// Remove regexes
const regexes = /\/.*?[^\\]\/[gim]{0,3}/gi;
while ((m = regexes.exec(code))) {
code = preserveToken(code, m, t++);
regexes.lastIndex = m.index;
}
code = code
// Create newlines after ;
.replace(/;/g, ";\n")
// Create newlines after { and around }
.replace(/{/g, "{\n")
.replace(/}/g, "\n}\n")
// Remove carriage returns
.replace(/\r/g, "")
// Remove all indentation
.replace(/^\s+/g, "")
.replace(/\n\s+/g, "\n")
// Remove trailing spaces
.replace(/\s*$/g, "")
.replace(/\n{/g, "{");
// Indent
let i = 0,
level = 0,
indent;
while (i < code.length) {
switch (code[i]) {
case "{":
level++;
break;
case "\n":
if (i+1 >= code.length) break;
if (code[i+1] === "}") level--;
indent = (level >= 0) ? Array(level*4+1).join(" ") : "";
code = code.substring(0, i+1) + indent + code.substring(i+1);
if (level > 0) i += level*4;
break;
}
i++;
}
code = code
// Add strategic spaces
.replace(/\s*([!<>=+-/*]?)=\s*/g, " $1= ")
.replace(/\s*<([=]?)\s*/g, " <$1 ")
.replace(/\s*>([=]?)\s*/g, " >$1 ")
.replace(/([^+])\+([^+=])/g, "$1 + $2")
.replace(/([^-])-([^-=])/g, "$1 - $2")
.replace(/([^*])\*([^*=])/g, "$1 * $2")
.replace(/([^/])\/([^/=])/g, "$1 / $2")
.replace(/\s*,\s*/g, ", ")
.replace(/\s*{/g, " {")
.replace(/}\n/g, "}\n\n")
// Hacky horribleness
.replace(/(if|for|while|with|elif|elseif)\s*\(([^\n]*)\)\s*\n([^{])/gim, "$1 ($2)\n $3")
.replace(/(if|for|while|with|elif|elseif)\s*\(([^\n]*)\)([^{])/gim, "$1 ($2) $3")
.replace(/else\s*\n([^{])/gim, "else\n $1")
.replace(/else\s+([^{])/gim, "else $1")
// Remove strategic spaces
.replace(/\s+;/g, ";")
.replace(/\{\s+\}/g, "{}")
.replace(/\[\s+\]/g, "[]")
.replace(/}\s*(else|catch|except|finally|elif|elseif|else if)/gi, "} $1");
// Replace preserved tokens
const ptokens = /###preservedToken(\d+)###/g;
while ((m = ptokens.exec(code))) {
const ti = parseInt(m[1], 10);
code = code.substring(0, m.index) + preservedTokens[ti] + code.substring(m.index + m[0].length);
ptokens.lastIndex = m.index;
}
return code;
/**
* Replaces a matched token with a placeholder value.
*/
function preserveToken(str, match, t) {
preservedTokens[t] = match[0];
return str.substring(0, match.index) +
"###preservedToken" + t + "###" +
str.substring(match.index + match[0].length);
}
}
}
export default GenericCodeBeautify;

View File

@ -0,0 +1,78 @@
/**
* @author n1474335 [n1474335@gmail.com]
* @copyright Crown Copyright 2016
* @license Apache-2.0
*/
import Operation from "../Operation";
import hljs from "highlight.js";
/**
* Syntax highlighter operation
*/
class SyntaxHighlighter extends Operation {
/**
* SyntaxHighlighter constructor
*/
constructor() {
super();
this.name = "Syntax highlighter";
this.module = "Code";
this.description = "Adds syntax highlighting to a range of source code languages. Note that this will not indent the code. Use one of the 'Beautify' operations for that.";
this.inputType = "string";
this.outputType = "html";
this.args = [
{
"name": "Language",
"type": "option",
"value": ["auto detect"].concat(hljs.listLanguages())
}
];
}
/**
* @param {string} input
* @param {Object[]} args
* @returns {html}
*/
run(input, args) {
const language = args[0];
if (language === "auto detect") {
return hljs.highlightAuto(input).value;
}
return hljs.highlight(language, input, true).value;
}
/**
* Highlight Syntax highlighter
*
* @param {Object[]} pos
* @param {number} pos[].start
* @param {number} pos[].end
* @param {Object[]} args
* @returns {Object[]} pos
*/
highlight(pos, args) {
return pos;
}
/**
* Highlight Syntax highlighter in reverse
*
* @param {Object[]} pos
* @param {number} pos[].start
* @param {number} pos[].end
* @param {Object[]} args
* @returns {Object[]} pos
*/
highlightReverse(pos, args) {
return pos;
}
}
export default SyntaxHighlighter;

View File

@ -0,0 +1,53 @@
/**
* @author tlwr [toby@toby.codes]
* @copyright Crown Copyright 2017
* @license Apache-2.0
*/
import { camelCase } from "lodash";
import Operation from "../Operation";
import { replaceVariableNames } from "../lib/Code";
/**
* To Camel case operation
*/
class ToCamelCase extends Operation {
/**
* ToCamelCase constructor
*/
constructor() {
super();
this.name = "To Camel case";
this.module = "Code";
this.description = "Converts the input string to camel case.\n<br><br>\nCamel case is all lower case except letters after word boundaries which are uppercase.\n<br><br>\ne.g. thisIsCamelCase\n<br><br>\n'Attempt to be context aware' will make the operation attempt to nicely transform variable and function names.";
this.inputType = "string";
this.outputType = "string";
this.args = [
{
"name": "Attempt to be context aware",
"type": "boolean",
"value": false
}
];
}
/**
* @param {string} input
* @param {Object[]} args
* @returns {string}
*/
run(input, args) {
const smart = args[0];
if (smart) {
return replaceVariableNames(input, camelCase);
} else {
return camelCase(input);
}
}
}
export default ToCamelCase;

View File

@ -0,0 +1,53 @@
/**
* @author n1474335 [n1474335@gmail.com]
* @copyright Crown Copyright 2016
* @license Apache-2.0
*/
import { kebabCase } from "lodash";
import Operation from "../Operation";
import { replaceVariableNames } from "../lib/Code";
/**
* To Kebab case operation
*/
class ToKebabCase extends Operation {
/**
* ToKebabCase constructor
*/
constructor() {
super();
this.name = "To Kebab case";
this.module = "Code";
this.description = "Converts the input string to kebab case.\n<br><br>\nKebab case is all lower case with dashes as word boundaries.\n<br><br>\ne.g. this-is-kebab-case\n<br><br>\n'Attempt to be context aware' will make the operation attempt to nicely transform variable and function names.";
this.inputType = "string";
this.outputType = "string";
this.args = [
{
"name": "Attempt to be context aware",
"type": "boolean",
"value": false
}
];
}
/**
* @param {string} input
* @param {Object[]} args
* @returns {string}
*/
run(input, args) {
const smart = args[0];
if (smart) {
return replaceVariableNames(input, kebabCase);
} else {
return kebabCase(input);
}
}
}
export default ToKebabCase;

View File

@ -0,0 +1,52 @@
/**
* @author tlwr [toby@toby.codes]
* @copyright Crown Copyright 2017
* @license Apache-2.0
*/
import { snakeCase } from "lodash";
import Operation from "../Operation";
import { replaceVariableNames } from "../lib/Code";
/**
* To Snake case operation
*/
class ToSnakeCase extends Operation {
/**
* ToSnakeCase constructor
*/
constructor() {
super();
this.name = "To Snake case";
this.module = "Code";
this.description = "Converts the input string to snake case.\n<br><br>\nSnake case is all lower case with underscores as word boundaries.\n<br><br>\ne.g. this_is_snake_case\n<br><br>\n'Attempt to be context aware' will make the operation attempt to nicely transform variable and function names.";
this.inputType = "string";
this.outputType = "string";
this.args = [
{
"name": "Attempt to be context aware",
"type": "boolean",
"value": false
}
];
}
/**
* @param {string} input
* @param {Object[]} args
* @returns {string}
*/
run(input, args) {
const smart = args[0];
if (smart) {
return replaceVariableNames(input, snakeCase);
} else {
return snakeCase(input);
}
}
}
export default ToSnakeCase;