@speed-highlight/core 1.1.6

package/examples/languages/test.sql

@@ -0,0 +1,6 @@
+-- Comment
+SELECT 'Geeks' || ' ' || @test_hello FROM dual;
+INSERT INTO database (geek_id, geek_name) VALUES (5000, 'abc');
+REPLACE('123geeks123', '123', TRUE);
+/* FIXME Multiline
+	Comment */

package/examples/languages/test.todo

@@ -0,0 +1,18 @@
+TODO ...
+FIXME ...
+CHANGED ...
+CHANGE ...
+FIX ...
+XXX ...
+IDEA ...
+HACK ...
+NOTE ...
+REVIEW ...
+NB ...
+BUG ...
+QUESTION ...
+COMBAK ...
+TEMP ...
+DEBUG ...
+OPTIMIZE ...
+WARNING ...

package/examples/languages/test.toml

@@ -0,0 +1,16 @@
+# CHANGED This is a TOML document
+
+title = "TOML Example"
+
+[owner]
+name = "Tom Preston-Werner"
+dob = 1979-05-27T07:32:00-08:00
+
+[database]
+enabled = true
+ports = [ 8000, inf, 8002 ]
+data = [ ["delta", "phi"], [3.14] ]
+temp_targets = { cpu = 79.5, case = 72.0 }
+re = """
+test
+"""

package/examples/languages/test.ts

@@ -0,0 +1,43 @@
+type Result = "pass" | "fail"
+
+interface User {
+	id: number
+	firstName: string
+	lastName: string
+	role: string
+	result: Result
+}
+
+const test = {
+	test: value
+};
+
+function updateUser(id: number, update: Partial<User>) {
+const user = getUser(id)
+const newUser = { ...user, ...update }
+saveUser(id, newUser)
+}
+
+function compact(arr: string[]) {
+	if (arr.length > 10)
+		return arr.slice(0, 10)
+	return arr
+}
+
+namespace Validation {
+	export interface StringValidator {
+		isAcceptable(s: string): boolean;
+	}
+	const lettersRegexp = /^[A-Za-z]+$/;
+	const numberRegexp = /^[0-9]+$/;
+	export class LettersOnlyValidator implements StringValidator {
+		isAcceptable(s: string) {
+			return lettersRegexp.test(s);
+		}
+	}
+	export class ZipCodeValidator implements StringValidator {
+		isAcceptable(s: string) {
+			return s.length === 5 && numberRegexp.test(s);
+		}
+	}
+}

package/examples/languages/test.uri

@@ -0,0 +1,9 @@
+https://john.doe@www.example.com:123/forum/questions/?tag=networking&order=newest#top
+# TODO more uri
+http://github.matubu.com/shj/examples/index.html
+ldap://[2001:db8::7]/c=GB?objectClass?one
+mailto:John.Doe@example.com
+news:comp.infosystems.www.servers.unix
+tel:+1-816-555-1212
+telnet://192.0.2.16:80/
+urn:oasis:names:specification:docbook:dtd:xml:4.1.2

package/examples/languages/test.xml

@@ -0,0 +1,44 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!--Simple example from w3schools-->
+<breakfast_menu>
+<food>
+	<name>Belgian Waffles</name>
+	<price>$5.95</price>
+	<description>
+	Two of our famous Belgian Waffles with plenty of real maple syrup
+	</description>
+	<calories>650</calories>
+</food>
+<food>
+	<name>Strawberry Belgian Waffles</name>
+	<price>$7.95</price>
+	<description>
+	Light Belgian waffles covered with strawberries and whipped cream
+	</description>
+	<calories>900</calories>
+</food>
+<food>
+	<name>Berry-Berry Belgian Waffles</name>
+	<price>$8.95</price>
+	<description>
+	Belgian waffles covered with assorted fresh berries and whipped cream
+	</description>
+	<calories>900</calories>
+</food>
+<food>
+	<name>French Toast</name>
+	<price>$4.50</price>
+	<description>
+	Thick slices made from our homemade sourdough bread
+	</description>
+	<calories>600</calories>
+</food>
+<food>
+	<name>Homestyle Breakfast</name>
+	<price>$6.95</price>
+	<description>
+	Two eggs, bacon or sausage, toast, and our ever-popular hash browns
+	</description>
+	<calories>950</calories>
+</food>
+</breakfast_menu> 

package/examples/languages/test.yaml

@@ -0,0 +1,18 @@
+--- # The Smiths
+- {name: John Smith, age: 33}
+- name: Mary Smith
+	age: !!str 27
+- [name, age]: [Rae Smith, 4] # sequences as kwds are supported
+--- # TODO People, by gender
+men: ["John Smith", 'Bill Jones', Yes]
+data: >
+	Wrapped text
+	will be folded
+	into a single
+	paragraph
+
+	Blank lines denote
+ 	paragraph breaks
+women:
+	- Mary Smith
+	- Susan Williams

package/examples/node.js

@@ -0,0 +1,11 @@
+const fs = require('fs'),
+	path = require('path'),
+	{ setTheme, printHighlight } = require('../dist/node/term.js');
+
+const lang = process.argv[2] ?? 'js';
+
+setTheme('default').then(_ => {
+	const code = fs.readFileSync(path.resolve(__dirname, `./languages/test.${lang}`));
+
+	printHighlight(code, lang);
+});

package/examples/style.css

@@ -0,0 +1,28 @@
+label, .chrono {font: 14px Arial}
+.radio-list {
+	margin: 10px 0;
+	display: flex;
+	flex-wrap: wrap;
+	gap: 5px
+}
+input {
+	position: absolute;
+	visibility: hidden
+}
+label {
+	cursor: pointer;
+	transition: .2s;
+	background: #c5c5e022;
+	border-radius: 5px;
+	padding: 10px
+}
+label:hover {background: #c5c5e033}
+input:checked + label {
+	color: #58f;
+	background: #58f3
+}
+.chrono {
+	padding: 3px;
+	border-radius: 2px;
+	color: #889
+}

package/examples/worker.html

@@ -0,0 +1,19 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+	<meta charset="UTF-8">
+	<title>Web worker | Speed highlight JS</title>
+	<link rel="stylesheet" href="../src/themes/atom-dark.css">
+</head>
+<body>
+	<div class="shj-lang-js shj-multiline shj-mode-noheader" id="code">this.code = 'highlighted on a web worker';
+this.isAwesome = new Boolean(.5);
+this.doesNotWorkOnFirefox = '💩'</div>
+	<script type="module">
+		const code = document.querySelector('#code');
+		const worker = new Worker('./worker.js', { type: "module" });
+		worker.onmessage = e => code.innerHTML = e.data;
+		worker.postMessage([code.textContent, 'js']);
+	</script>
+</body>
+</html>

package/examples/worker.js

@@ -0,0 +1,6 @@
+import { highlightText } from '../src/index.js'
+
+onmessage = async e => {
+	const result = await highlightText(e.data[0], e.data[1] || 'js');
+	postMessage(result);
+};

package/package.json

@@ -0,0 +1,40 @@
+{
+  "name": "@speed-highlight/core",
+  "version": "1.1.6",
+  "description": "🌈 Light, fast, and easy to use, dependencies free javascript syntax highlighter, with automatic language detection",
+  "main": "dist/node/index.js",
+  "directories": {
+    "example": "examples"
+  },
+  "scripts": {
+    "test": "echo \"Error: no test specified\" && exit 1"
+  },
+  "repository": {
+    "type": "git",
+    "url": "git://github.com/speed-highlight/core.git"
+  },
+  "keywords": [
+    "javascript",
+    "syntax-highlighting",
+    "language",
+    "fast",
+    "js",
+    "simple",
+    "highlighter",
+    "regex",
+    "highlighting",
+    "highlightjs",
+    "small",
+    "deno"
+  ],
+  "author": "matubu",
+  "license": "CC0-1.0",
+  "bugs": {
+    "url": "https://github.com/speed-highlight/core/issues"
+  },
+  "homepage": "https://github.com/speed-highlight/core#readme",
+  "publishConfig": {
+    "registry": "https://npm.pkg.github.com/",
+    "access": "public"
+  }
+}

package/src/common.js

@@ -0,0 +1,18 @@
+/**
+ * Commonly used match pattern
+*/
+
+export default {
+	num: {
+		type: 'num',
+		match: /(\.e?|\b)\d(e-|[\d.oxa-fA-F_])*(\.|\b)/g
+	},
+	str: {
+		type: 'str',
+		match: /(["'])(\\[^]|(?!\1)[^\r\n\\])*\1?/g
+	},
+	strDouble: {
+		type: 'str',
+		match: /"((?!")[^\r\n\\]|\\[^])*"?/g
+	}
+}

package/src/detect.js

@@ -0,0 +1,33 @@
+const languages = [
+	['bash', [/#!(\/usr)?\/bin\/bash/g, 500], [/\b(if|elif|then|fi|echo)\b|\$/g, 10]],
+	['html', [/<\/?[a-z-]+[^\n>]*>/g, 10], [/^\s+<!DOCTYPE\s+html/g, 500]],
+	['http', [/^(GET|HEAD|POST|PUT|DELETE|PATCH|HTTP)\b/g, 500]],
+	['js', [/\b(console|await|async|function|export|import|this|class|for|let|const|map|join|require)\b/g, 10]],
+	['ts', [/\b(console|await|async|function|export|import|this|class|for|let|const|map|join|require|implements|interface|namespace)\b/g, 10]],
+	['py', [/\b(def|print|class|and|or|lambda)\b/g, 10]],
+	['sql', [/\b(SELECT|INSERT|FROM)\b/g, 50]],
+	['pl', [/#!(\/usr)?\/bin\/perl/g, 500], [/\b(use|print)\b|\$/g, 10]],
+	['lua', [/#!(\/usr)?\/bin\/lua/g, 500]],
+	['make', [/\b(ifneq|endif|if|elif|then|fi|echo|.PHONY|^[a-z]+ ?:$)\b|\$/gm, 10]],
+	['uri', [/https?:|mailto:|tel:|ftp:/g, 30]],
+	['css', [/^(@import|@page|@media|(\.|#)[a-z]+)/gm, 20]],
+	['diff', [/^[+><-]/gm, 10], [/^@@ ?[-+,0-9 ]+ ?@@/gm, 25]],
+	['md', [/^(>|\t\*|\t\d+.)/gm, 10], [/\[.*\](.*)/g, 10]],
+	['docker', [/^(FROM|ENTRYPOINT|RUN)/gm, 500]],
+	['xml', [/<\/?[a-z-]+[^\n>]*>/g, 10], [/^<\?xml/g, 500]],
+	['c', [/#include\b|\bprintf\s+\(/g, 100]],
+	['rs', [/^\s+(use|fn|mut|match)\b/gm, 100]],
+	['go', [/\b(func|fmt|package)\b/g, 100]],
+	['java', [/^import\s+java/gm, 500]],
+	['asm', [/^(section|global main|extern|\t(call|mov|ret))/gm, 100]],
+]
+
+export const detectLanguage = code => {
+	return (languages
+		.map(([lang, ...features]) => [
+			features.reduce((acc, [match, score]) => acc + [...code.matchAll(match)].length * score, 0),
+			lang
+		])
+		.filter(([score, lang]) => score > 10)
+		.sort((a, b) => b[0] - a[0])[0]?.[1] || 'plain');
+}

package/src/index.js

@@ -0,0 +1,129 @@
+/**
+ * Base script
+*/
+
+import expandData from './common.js';
+
+const langs = {},
+	sanitize = (str = '') =>
+		str.replaceAll('&', '&#38;').replaceAll?.('<', '&lt;').replaceAll?.('>', '&gt;'),
+	/**
+	 * A function that turn text in HTML to apply a class to it
+	 * @param {String} str need to be sanitize to be safe
+	 * @param {String} [token] token type
+	 * @returns A string with HTML semantics
+	 */
+	toSpan = (str, token) => token ? `<span class="shj-syn-${token}">${str}</span>` : str;
+
+/**
+ * A
+ * @param {String} src
+ * @param {String|Object} lang
+ * @param {Function} token the callback that will be call for every token found.
+ * the arguments pass will be:
+ * * the token as a string (can be an empty string)
+ * * the token type also as a string like 'err', 'cmnt', ...
+ */
+export async function tokenize(src, lang, token) {
+	try {
+		let m,
+			part,
+			first = {},
+			match,
+			cache = [],
+			i = 0,
+			data = typeof lang === 'string' ? (await (langs[lang] ??= import(`./languages/${lang}.js`))) : lang,
+			// make a fast shallow copy to bee able to splice lang without change the original one
+			arr = [...typeof lang === 'string' ? data.default : lang.sub];
+
+		while (i < src.length) {
+			first.index = null;
+			for (m = arr.length; m-- > 0;) {
+				part = arr[m].expand ? expandData[arr[m].expand] : arr[m];
+				// do not call again exec if the previous result is sufficient
+				if (cache[m] === undefined || cache[m].match.index < i) {
+					part.match.lastIndex = i;
+					match = part.match.exec(src);
+					if (match === null) {
+						// no more match with this regex can be disposed
+						arr.splice(m, 1);
+						cache.splice(m, 1);
+						continue;
+					}
+					// save match for later use to decrease performance cost
+					cache[m] = { match, lastIndex: part.match.lastIndex };
+				}
+				// check if it the first match in the string
+				if (cache[m].match[0] && (cache[m].match.index <= first.index || first.index === null))
+					first = {
+						part: part,
+						index: cache[m].match.index,
+						match: cache[m].match[0],
+						end: cache[m].lastIndex
+					}
+			}
+			if (first.index === null)
+				break;
+			token(src.slice(i, first.index), data.type);
+			i = first.end;
+			if (first.part.sub)
+				await tokenize(first.match, typeof first.part.sub === 'string' ? first.part.sub : (typeof first.part.sub === 'function' ? first.part.sub(first.match) : first.part), token);
+			else
+				token(first.match, first.part.type);
+		}
+		token(src.slice(i, src.length), data.type);
+	}
+	catch {
+		token(src);
+	}
+}
+
+/**
+ * @async
+ * @function highlightText A function that highlight a string text and return it
+ * @example
+ * elm.innerHTML = await highlightText(code, 'js');
+ * @param {String} src the text content to be highlighted
+ * @param {String} lang the lang name ex: 'js'
+ * @param {Boolean} [multiline=true] inline mode
+ * @returns {String} the highlighted as String text
+ */
+export async function highlightText(src, lang, multiline = true) {
+	let res = multiline ? `<div><div class="shj-numbers">${'<div></div>'.repeat(src.split('\n').length)}</div><div>` : '';
+
+	await tokenize(src, lang, (str, type) => res += toSpan(sanitize(str), type))
+
+	if (multiline)
+		res += '</div></div>';
+	return res;
+}
+
+/**
+ * @async
+ * @function highlightElement highlight a element code with a 'pre' parent
+ * @param {HTMLElement} elm the code elm
+ * @param {String} [lang] the lang used for syntax highlighting by default is found in the className of the parent or the elm it self
+ * @param {String} [mode] can be set to inline multiline or oneline by default it's auto detected: inline if `code` tag else depends of the number of lines
+ */
+export async function highlightElement(elm, lang = elm.className.match(/shj-lang-([\w-]+)/)?.[1], mode) {
+	let txt = elm.textContent;
+	mode ??= `${elm.tagName == 'CODE' ? 'in' : (txt.split('\n').length < 2 ? 'one' : 'multi')}line`;
+	elm.dataset.lang = lang;
+	elm.className = `${[...elm.classList].filter(className => !className.startsWith('shj-') || className.startsWith('shj-mode-')).join(' ')} shj-lang-${lang} shj-${mode}`;
+	elm.innerHTML = await highlightText(txt, lang, mode == 'multiline');
+}
+
+/**
+ * for all element that have a class name starting with "shj-lang-"
+ * this function will call highlightElement with the html element as argument
+ * The function will select those scheme for example:
+ * ```html
+ * <div class='shj-lang-[code-language]'>[code]</div>
+ * or
+ * <code class='shj-lang-[code-language]'>[inline code]</code>
+ * ```
+ */
+export let highlightAll = async () =>
+	document
+		.querySelectorAll('[class*="shj-lang-"]')
+		.forEach(elm => highlightElement(elm));

package/src/languages/asm.js

@@ -0,0 +1,30 @@
+export default [
+	{
+		type: 'cmnt',
+		match: /;.*/gm
+	},
+	{
+		expand: 'str'
+	},
+	{
+		expand: 'num'
+	},
+	{
+		type: 'kwd',
+		match: /^[a-z]+\s+[a-z.]+/gm,
+		sub: [
+			{
+				type: 'func',
+				match: /^[a-z]+/g
+			}
+		]
+	},
+	{
+		type: 'kwd',
+		match: /^\t*[a-z]+/gm,
+	},
+	{
+		match: /%|\$/g,
+		type: 'oper'
+	}
+]

package/src/languages/bash.js

@@ -0,0 +1,44 @@
+let variable = {
+	type: 'var',
+	match: /\$\w+|\${[^}]*}|\$\([^)]*\)/g
+};
+
+export default [
+	{
+		sub: 'todo',
+		match: /#.*/g
+	},
+	{
+		type: 'str',
+		match: /(["'])((?!\1)[^\r\n\\]|\\[^])*\1?/g,
+		sub: [ variable ]
+	},
+	{
+		type: 'kwd',
+		match: /-[a-zA-Z]+|$<|[&|;]+|\b(unset|readonly|shift|export|if|fi|else|elif|while|do|done|for|until|case|esac|break|continue|exit|return|trap|wait|eval|exec|then|declare|enable|local|select|typeset|time|add|remove|install|update|delete)\b/g
+	},
+	{
+		expand: 'num'
+	},
+	{
+		type: 'func',
+		match: /\b(set|alias|bg|bind|builtin|caller|command|compgen|complete|dirs|disown|echo|enable|eval|exec|exit|fc|fg|getopts|hash|help|history|jobs|kill|let|logout|popd|printf|pushd|pwd|read|set|shift|shopt|source|suspend|test|times|trap|type|ulimit|umask|unalias|unset)\b/g
+	},
+	{
+		type: 'bool',
+		match: /\b(true|false)\b/g
+	},
+	{
+		type: 'func',
+		match: /[a-z_]+(?=\s*\()/g
+	},
+	{
+		type: 'oper',
+		match: /[=(){}<>+*/!?~^-]+/g
+	},
+	{
+		type: 'var',
+		match: /\w+(?=\s*=)/g
+	},
+	variable
+]

package/src/languages/bf.js

@@ -0,0 +1,18 @@
+export default [
+	{
+		match: /[^\[\->+.<\]\s].*/g,
+		sub: 'todo'
+	},
+	{
+		type: 'func',
+		match: /\.+/g
+	},
+	{
+		type: 'kwd',
+		match: /[<>]+/g
+	},
+	{
+		type: 'oper',
+		match: /[+-]+/g
+	}
+]

package/src/languages/c.js

@@ -0,0 +1,52 @@
+export default [
+	{
+		match: /\/\/.*\n?|\/\*((?!\*\/)[^])*(\*\/)?/g,
+		sub: 'todo'
+	},
+	{
+		expand: 'str'
+	},
+	{
+		expand: 'num'
+	},
+	{
+		type: 'kwd',
+		match: /#\s*include (<.*>|".*")/g,
+		sub: [
+			{
+				type: 'str',
+				match: /(<|").*/g
+			}
+		]
+	},
+	{
+		match: /asm\s*{[^}]*}/g,
+		sub: [
+			{
+				type: 'kwd',
+				match: /^asm/g
+			},
+			{
+				//type: 'str',
+				match: /[^{}]*(?=}$)/g,
+				sub: 'asm'
+			}
+		]
+	},
+	{
+		type: 'kwd',
+		match: /\*|&|#[a-z]+\b|\b(asm|auto|double|int|struct|break|else|long|switch|case|enum|register|typedef|char|extern|return|union|const|float|short|unsigned|continue|for|signed|void|default|goto|sizeof|volatile|do|if|static|while)\b/g
+	},
+	{
+		type: 'oper',
+		match: /[/*+:?&|%^~=!,<>.^-]+/g
+	},
+	{
+		type: 'func',
+		match: /[a-zA-Z_][\w_]*(?=\s*\()/g
+	},
+	{
+		type: 'class',
+		match: /[A-Z][\w_]*/g
+	}
+]

package/src/languages/css.js

@@ -0,0 +1,56 @@
+export default [
+	{
+		match: /\/\*((?!\*\/)[^])*(\*\/)?/g,
+		sub: 'todo'
+	},
+	{
+		expand: 'str'
+	},
+	{
+		type: 'kwd',
+		match: /@\w+\b|\b(and|not|only|or)\b|\b[a-z-]+(?=[^{}]*{)/g
+	},
+	{
+		type: 'var',
+		match: /\b[\w-]+(?=\s*:)|(::?|\.)[\w-]+(?=[^{}]*{)/g
+	},
+	{
+		type: 'func',
+		match: /#[\w-]+(?=[^{}]*{)/g
+	},
+	{
+		type: 'num',
+		match:  /#[\da-f]{3,8}/g
+	},
+	{
+		type: 'num',
+		match: /\d+(\.\d+)?(cm|mm|in|px|pt|pc|em|ex|ch|rem|vm|vh|vmin|vmax|%)?/g,
+		sub: [
+			{
+				type: 'var',
+				match: /[a-z]+|%/g
+			}
+		]
+	},
+	{
+		match: /url\([^)]*\)/g,
+		sub: [
+			{
+				type: 'func',
+				match: /url(?=\()/g
+			},
+			{
+				type: 'str',
+				match: /[^()]+/g
+			}
+		]
+	},
+	{
+		type: 'func',
+		match: /\b[a-zA-Z]\w*(?=\s*\()/g
+	},
+	{
+		type: 'num',
+		match: /\b[a-z-]+\b/g
+	}
+]

package/src/languages/csv.js

@@ -0,0 +1,9 @@
+export default [
+	{
+		expand: 'strDouble'
+	},
+	{
+		type: 'oper',
+		match: /,/g
+	}
+]

package/src/languages/diff.js

@@ -0,0 +1,18 @@
+export default [
+	{
+		type: 'deleted',
+		match: /^[-<].*/gm
+	},
+	{
+		type: 'insert',
+		match: /^[+>].*/gm
+	},
+	{
+		type: 'kwd',
+		match: /!.*/gm
+	},
+	{
+		type: 'section',
+		match: /^@@.*@@$|^\d.*|^([*-+])\1\1.*/gm
+	}
+]