Description
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

602 lines
14 KiB

4 years ago
const IDENT_RE = '[A-Za-z$_][0-9A-Za-z$_]*';
const KEYWORDS = [
"as", // for exports
"in",
"of",
"if",
"for",
"while",
"finally",
"var",
"new",
"function",
"do",
"return",
"void",
"else",
"break",
"catch",
"instanceof",
"with",
"throw",
"case",
"default",
"try",
"switch",
"continue",
"typeof",
"delete",
"let",
"yield",
"const",
"class",
// JS handles these with a special rule
// "get",
// "set",
"debugger",
"async",
"await",
"static",
"import",
"from",
"export",
"extends"
];
const LITERALS = [
"true",
"false",
"null",
"undefined",
"NaN",
"Infinity"
];
const TYPES = [
"Intl",
"DataView",
"Number",
"Math",
"Date",
"String",
"RegExp",
"Object",
"Function",
"Boolean",
"Error",
"Symbol",
"Set",
"Map",
"WeakSet",
"WeakMap",
"Proxy",
"Reflect",
"JSON",
"Promise",
"Float64Array",
"Int16Array",
"Int32Array",
"Int8Array",
"Uint16Array",
"Uint32Array",
"Float32Array",
"Array",
"Uint8Array",
"Uint8ClampedArray",
"ArrayBuffer"
];
const ERROR_TYPES = [
"EvalError",
"InternalError",
"RangeError",
"ReferenceError",
"SyntaxError",
"TypeError",
"URIError"
];
const BUILT_IN_GLOBALS = [
"setInterval",
"setTimeout",
"clearInterval",
"clearTimeout",
"require",
"exports",
"eval",
"isFinite",
"isNaN",
"parseFloat",
"parseInt",
"decodeURI",
"decodeURIComponent",
"encodeURI",
"encodeURIComponent",
"escape",
"unescape"
];
const BUILT_IN_VARIABLES = [
"arguments",
"this",
"super",
"console",
"window",
"document",
"localStorage",
"module",
"global" // Node.js
];
const BUILT_INS = [].concat(
BUILT_IN_GLOBALS,
BUILT_IN_VARIABLES,
TYPES,
ERROR_TYPES
);
/**
* @param {string} value
* @returns {RegExp}
* */
/**
* @param {RegExp | string } re
* @returns {string}
*/
function source(re) {
if (!re) return null;
if (typeof re === "string") return re;
return re.source;
}
/**
* @param {RegExp | string } re
* @returns {string}
*/
function lookahead(re) {
return concat('(?=', re, ')');
}
/**
* @param {...(RegExp | string) } args
* @returns {string}
*/
function concat(...args) {
const joined = args.map((x) => source(x)).join("");
return joined;
}
/*
Language: JavaScript
Description: JavaScript (JS) is a lightweight, interpreted, or just-in-time compiled programming language with first-class functions.
Category: common, scripting
Website: https://developer.mozilla.org/en-US/docs/Web/JavaScript
*/
/** @type LanguageFn */
function javascript(hljs) {
/**
* Takes a string like "<Booger" and checks to see
* if we can find a matching "</Booger" later in the
* content.
* @param {RegExpMatchArray} match
* @param {{after:number}} param1
*/
const hasClosingTag = (match, { after }) => {
const tag = "</" + match[0].slice(1);
const pos = match.input.indexOf(tag, after);
return pos !== -1;
};
const IDENT_RE$1 = IDENT_RE;
const FRAGMENT = {
begin: '<>',
end: '</>'
};
const XML_TAG = {
begin: /<[A-Za-z0-9\\._:-]+/,
end: /\/[A-Za-z0-9\\._:-]+>|\/>/,
/**
* @param {RegExpMatchArray} match
* @param {CallbackResponse} response
*/
isTrulyOpeningTag: (match, response) => {
const afterMatchIndex = match[0].length + match.index;
const nextChar = match.input[afterMatchIndex];
// nested type?
// HTML should not include another raw `<` inside a tag
// But a type might: `<Array<Array<number>>`, etc.
if (nextChar === "<") {
response.ignoreMatch();
return;
}
// <something>
// This is now either a tag or a type.
if (nextChar === ">") {
// if we cannot find a matching closing tag, then we
// will ignore it
if (!hasClosingTag(match, { after: afterMatchIndex })) {
response.ignoreMatch();
}
}
}
};
const KEYWORDS$1 = {
$pattern: IDENT_RE,
keyword: KEYWORDS.join(" "),
literal: LITERALS.join(" "),
built_in: BUILT_INS.join(" ")
};
// https://tc39.es/ecma262/#sec-literals-numeric-literals
const decimalDigits = '[0-9](_?[0-9])*';
const frac = `\\.(${decimalDigits})`;
// DecimalIntegerLiteral, including Annex B NonOctalDecimalIntegerLiteral
// https://tc39.es/ecma262/#sec-additional-syntax-numeric-literals
const decimalInteger = `0|[1-9](_?[0-9])*|0[0-7]*[89][0-9]*`;
const NUMBER = {
className: 'number',
variants: [
// DecimalLiteral
{ begin: `(\\b(${decimalInteger})((${frac})|\\.)?|(${frac}))` +
`[eE][+-]?(${decimalDigits})\\b` },
{ begin: `\\b(${decimalInteger})\\b((${frac})\\b|\\.)?|(${frac})\\b` },
// DecimalBigIntegerLiteral
{ begin: `\\b(0|[1-9](_?[0-9])*)n\\b` },
// NonDecimalIntegerLiteral
{ begin: "\\b0[xX][0-9a-fA-F](_?[0-9a-fA-F])*n?\\b" },
{ begin: "\\b0[bB][0-1](_?[0-1])*n?\\b" },
{ begin: "\\b0[oO][0-7](_?[0-7])*n?\\b" },
// LegacyOctalIntegerLiteral (does not include underscore separators)
// https://tc39.es/ecma262/#sec-additional-syntax-numeric-literals
{ begin: "\\b0[0-7]+n?\\b" },
],
relevance: 0
};
const SUBST = {
className: 'subst',
begin: '\\$\\{',
end: '\\}',
keywords: KEYWORDS$1,
contains: [] // defined later
};
const HTML_TEMPLATE = {
begin: 'html`',
end: '',
starts: {
end: '`',
returnEnd: false,
contains: [
hljs.BACKSLASH_ESCAPE,
SUBST
],
subLanguage: 'xml'
}
};
const CSS_TEMPLATE = {
begin: 'css`',
end: '',
starts: {
end: '`',
returnEnd: false,
contains: [
hljs.BACKSLASH_ESCAPE,
SUBST
],
subLanguage: 'css'
}
};
const TEMPLATE_STRING = {
className: 'string',
begin: '`',
end: '`',
contains: [
hljs.BACKSLASH_ESCAPE,
SUBST
]
};
const JSDOC_COMMENT = hljs.COMMENT(
'/\\*\\*',
'\\*/',
{
relevance: 0,
contains: [
{
className: 'doctag',
begin: '@[A-Za-z]+',
contains: [
{
className: 'type',
begin: '\\{',
end: '\\}',
relevance: 0
},
{
className: 'variable',
begin: IDENT_RE$1 + '(?=\\s*(-)|$)',
endsParent: true,
relevance: 0
},
// eat spaces (not newlines) so we can find
// types or variables
{
begin: /(?=[^\n])\s/,
relevance: 0
}
]
}
]
}
);
const COMMENT = {
className: "comment",
variants: [
JSDOC_COMMENT,
hljs.C_BLOCK_COMMENT_MODE,
hljs.C_LINE_COMMENT_MODE
]
};
const SUBST_INTERNALS = [
hljs.APOS_STRING_MODE,
hljs.QUOTE_STRING_MODE,
HTML_TEMPLATE,
CSS_TEMPLATE,
TEMPLATE_STRING,
NUMBER,
hljs.REGEXP_MODE
];
SUBST.contains = SUBST_INTERNALS
.concat({
// we need to pair up {} inside our subst to prevent
// it from ending too early by matching another }
begin: /\{/,
end: /\}/,
keywords: KEYWORDS$1,
contains: [
"self"
].concat(SUBST_INTERNALS)
});
const SUBST_AND_COMMENTS = [].concat(COMMENT, SUBST.contains);
const PARAMS_CONTAINS = SUBST_AND_COMMENTS.concat([
// eat recursive parens in sub expressions
{
begin: /\(/,
end: /\)/,
keywords: KEYWORDS$1,
contains: ["self"].concat(SUBST_AND_COMMENTS)
}
]);
const PARAMS = {
className: 'params',
begin: /\(/,
end: /\)/,
excludeBegin: true,
excludeEnd: true,
keywords: KEYWORDS$1,
contains: PARAMS_CONTAINS
};
return {
name: 'Javascript',
aliases: ['js', 'jsx', 'mjs', 'cjs'],
keywords: KEYWORDS$1,
// this will be extended by TypeScript
exports: { PARAMS_CONTAINS },
illegal: /#(?![$_A-z])/,
contains: [
hljs.SHEBANG({
label: "shebang",
binary: "node",
relevance: 5
}),
{
label: "use_strict",
className: 'meta',
relevance: 10,
begin: /^\s*['"]use (strict|asm)['"]/
},
hljs.APOS_STRING_MODE,
hljs.QUOTE_STRING_MODE,
HTML_TEMPLATE,
CSS_TEMPLATE,
TEMPLATE_STRING,
COMMENT,
NUMBER,
{ // object attr container
begin: concat(/[{,\n]\s*/,
// we need to look ahead to make sure that we actually have an
// attribute coming up so we don't steal a comma from a potential
// "value" container
//
// NOTE: this might not work how you think. We don't actually always
// enter this mode and stay. Instead it might merely match `,
// <comments up next>` and then immediately end after the , because it
// fails to find any actual attrs. But this still does the job because
// it prevents the value contain rule from grabbing this instead and
// prevening this rule from firing when we actually DO have keys.
lookahead(concat(
// we also need to allow for multiple possible comments inbetween
// the first key:value pairing
/(((\/\/.*$)|(\/\*(\*[^/]|[^*])*\*\/))\s*)*/,
IDENT_RE$1 + '\\s*:'))),
relevance: 0,
contains: [
{
className: 'attr',
begin: IDENT_RE$1 + lookahead('\\s*:'),
relevance: 0
}
]
},
{ // "value" container
begin: '(' + hljs.RE_STARTERS_RE + '|\\b(case|return|throw)\\b)\\s*',
keywords: 'return throw case',
contains: [
COMMENT,
hljs.REGEXP_MODE,
{
className: 'function',
// we have to count the parens to make sure we actually have the
// correct bounding ( ) before the =>. There could be any number of
// sub-expressions inside also surrounded by parens.
begin: '(\\(' +
'[^()]*(\\(' +
'[^()]*(\\(' +
'[^()]*' +
'\\))*[^()]*' +
'\\))*[^()]*' +
'\\)|' + hljs.UNDERSCORE_IDENT_RE + ')\\s*=>',
returnBegin: true,
end: '\\s*=>',
contains: [
{
className: 'params',
variants: [
{
begin: hljs.UNDERSCORE_IDENT_RE,
relevance: 0
},
{
className: null,
begin: /\(\s*\)/,
skip: true
},
{
begin: /\(/,
end: /\)/,
excludeBegin: true,
excludeEnd: true,
keywords: KEYWORDS$1,
contains: PARAMS_CONTAINS
}
]
}
]
},
{ // could be a comma delimited list of params to a function call
begin: /,/, relevance: 0
},
{
className: '',
begin: /\s/,
end: /\s*/,
skip: true
},
{ // JSX
variants: [
{ begin: FRAGMENT.begin, end: FRAGMENT.end },
{
begin: XML_TAG.begin,
// we carefully check the opening tag to see if it truly
// is a tag and not a false positive
'on:begin': XML_TAG.isTrulyOpeningTag,
end: XML_TAG.end
}
],
subLanguage: 'xml',
contains: [
{
begin: XML_TAG.begin,
end: XML_TAG.end,
skip: true,
contains: ['self']
}
]
}
],
relevance: 0
},
{
className: 'function',
beginKeywords: 'function',
end: /[{;]/,
excludeEnd: true,
keywords: KEYWORDS$1,
contains: [
'self',
hljs.inherit(hljs.TITLE_MODE, { begin: IDENT_RE$1 }),
PARAMS
],
illegal: /%/
},
{
// prevent this from getting swallowed up by function
// since they appear "function like"
beginKeywords: "while if switch catch for"
},
{
className: 'function',
// we have to count the parens to make sure we actually have the correct
// bounding ( ). There could be any number of sub-expressions inside
// also surrounded by parens.
begin: hljs.UNDERSCORE_IDENT_RE +
'\\(' + // first parens
'[^()]*(\\(' +
'[^()]*(\\(' +
'[^()]*' +
'\\))*[^()]*' +
'\\))*[^()]*' +
'\\)\\s*\\{', // end parens
returnBegin:true,
contains: [
PARAMS,
hljs.inherit(hljs.TITLE_MODE, { begin: IDENT_RE$1 }),
]
},
// hack: prevents detection of keywords in some circumstances
// .keyword()
// $keyword = x
{
variants: [
{ begin: '\\.' + IDENT_RE$1 },
{ begin: '\\$' + IDENT_RE$1 }
],
relevance: 0
},
{ // ES6 class
className: 'class',
beginKeywords: 'class',
end: /[{;=]/,
excludeEnd: true,
illegal: /[:"[\]]/,
contains: [
{ beginKeywords: 'extends' },
hljs.UNDERSCORE_TITLE_MODE
]
},
{
begin: /\b(?=constructor)/,
end: /[{;]/,
excludeEnd: true,
contains: [
hljs.inherit(hljs.TITLE_MODE, { begin: IDENT_RE$1 }),
'self',
PARAMS
]
},
{
begin: '(get|set)\\s+(?=' + IDENT_RE$1 + '\\()',
end: /\{/,
keywords: "get set",
contains: [
hljs.inherit(hljs.TITLE_MODE, { begin: IDENT_RE$1 }),
{ begin: /\(\)/ }, // eat to avoid empty params
PARAMS
]
},
{
begin: /\$[(.]/ // relevance booster for a pattern common to JS libs: `$(something)` and `$.something`
}
]
};
}
module.exports = javascript;