From de98e0b232509d5f40c135d540a70e415272ff85 Mon Sep 17 00:00:00 2001 From: Florian Dold Date: Wed, 3 May 2017 15:35:00 +0200 Subject: node_modules --- node_modules/acorn/src/tokenize.js | 687 +++++++++++++++++++++++++++++++++++++ 1 file changed, 687 insertions(+) create mode 100644 node_modules/acorn/src/tokenize.js (limited to 'node_modules/acorn/src/tokenize.js') diff --git a/node_modules/acorn/src/tokenize.js b/node_modules/acorn/src/tokenize.js new file mode 100644 index 000000000..917a013b8 --- /dev/null +++ b/node_modules/acorn/src/tokenize.js @@ -0,0 +1,687 @@ +import {isIdentifierStart, isIdentifierChar} from "./identifier" +import {types as tt, keywords as keywordTypes} from "./tokentype" +import {Parser} from "./state" +import {SourceLocation} from "./locutil" +import {lineBreak, lineBreakG, isNewLine, nonASCIIwhitespace} from "./whitespace" + +// Object type used to represent tokens. Note that normally, tokens +// simply exist as properties on the parser object. This is only +// used for the onToken callback and the external tokenizer. + +export class Token { + constructor(p) { + this.type = p.type + this.value = p.value + this.start = p.start + this.end = p.end + if (p.options.locations) + this.loc = new SourceLocation(p, p.startLoc, p.endLoc) + if (p.options.ranges) + this.range = [p.start, p.end] + } +} + +// ## Tokenizer + +const pp = Parser.prototype + +// Are we running under Rhino? +const isRhino = typeof Packages == "object" && Object.prototype.toString.call(Packages) == "[object JavaPackage]" + +// Move to the next token + +pp.next = function() { + if (this.options.onToken) + this.options.onToken(new Token(this)) + + this.lastTokEnd = this.end + this.lastTokStart = this.start + this.lastTokEndLoc = this.endLoc + this.lastTokStartLoc = this.startLoc + this.nextToken() +} + +pp.getToken = function() { + this.next() + return new Token(this) +} + +// If we're in an ES6 environment, make parsers iterable +if (typeof Symbol !== "undefined") + pp[Symbol.iterator] = function() { + return { + next: () => { + let token = this.getToken() + return { + done: token.type === tt.eof, + value: token + } + } + } + } + +// Toggle strict mode. Re-reads the next number or string to please +// pedantic tests (`"use strict"; 010;` should fail). + +pp.curContext = function() { + return this.context[this.context.length - 1] +} + +// Read a single token, updating the parser object's token-related +// properties. + +pp.nextToken = function() { + let curContext = this.curContext() + if (!curContext || !curContext.preserveSpace) this.skipSpace() + + this.start = this.pos + if (this.options.locations) this.startLoc = this.curPosition() + if (this.pos >= this.input.length) return this.finishToken(tt.eof) + + if (curContext.override) return curContext.override(this) + else this.readToken(this.fullCharCodeAtPos()) +} + +pp.readToken = function(code) { + // Identifier or keyword. '\uXXXX' sequences are allowed in + // identifiers, so '\' also dispatches to that. + if (isIdentifierStart(code, this.options.ecmaVersion >= 6) || code === 92 /* '\' */) + return this.readWord() + + return this.getTokenFromCode(code) +} + +pp.fullCharCodeAtPos = function() { + let code = this.input.charCodeAt(this.pos) + if (code <= 0xd7ff || code >= 0xe000) return code + let next = this.input.charCodeAt(this.pos + 1) + return (code << 10) + next - 0x35fdc00 +} + +pp.skipBlockComment = function() { + let startLoc = this.options.onComment && this.curPosition() + let start = this.pos, end = this.input.indexOf("*/", this.pos += 2) + if (end === -1) this.raise(this.pos - 2, "Unterminated comment") + this.pos = end + 2 + if (this.options.locations) { + lineBreakG.lastIndex = start + let match + while ((match = lineBreakG.exec(this.input)) && match.index < this.pos) { + ++this.curLine + this.lineStart = match.index + match[0].length + } + } + if (this.options.onComment) + this.options.onComment(true, this.input.slice(start + 2, end), start, this.pos, + startLoc, this.curPosition()) +} + +pp.skipLineComment = function(startSkip) { + let start = this.pos + let startLoc = this.options.onComment && this.curPosition() + let ch = this.input.charCodeAt(this.pos += startSkip) + while (this.pos < this.input.length && ch !== 10 && ch !== 13 && ch !== 8232 && ch !== 8233) { + ++this.pos + ch = this.input.charCodeAt(this.pos) + } + if (this.options.onComment) + this.options.onComment(false, this.input.slice(start + startSkip, this.pos), start, this.pos, + startLoc, this.curPosition()) +} + +// Called at the start of the parse and after every token. Skips +// whitespace and comments, and. + +pp.skipSpace = function() { + loop: while (this.pos < this.input.length) { + let ch = this.input.charCodeAt(this.pos) + switch (ch) { + case 32: case 160: // ' ' + ++this.pos + break + case 13: + if (this.input.charCodeAt(this.pos + 1) === 10) { + ++this.pos + } + case 10: case 8232: case 8233: + ++this.pos + if (this.options.locations) { + ++this.curLine + this.lineStart = this.pos + } + break + case 47: // '/' + switch (this.input.charCodeAt(this.pos + 1)) { + case 42: // '*' + this.skipBlockComment() + break + case 47: + this.skipLineComment(2) + break + default: + break loop + } + break + default: + if (ch > 8 && ch < 14 || ch >= 5760 && nonASCIIwhitespace.test(String.fromCharCode(ch))) { + ++this.pos + } else { + break loop + } + } + } +} + +// Called at the end of every token. Sets `end`, `val`, and +// maintains `context` and `exprAllowed`, and skips the space after +// the token, so that the next one's `start` will point at the +// right position. + +pp.finishToken = function(type, val) { + this.end = this.pos + if (this.options.locations) this.endLoc = this.curPosition() + let prevType = this.type + this.type = type + this.value = val + + this.updateContext(prevType) +} + +// ### Token reading + +// This is the function that is called to fetch the next token. It +// is somewhat obscure, because it works in character codes rather +// than characters, and because operator parsing has been inlined +// into it. +// +// All in the name of speed. +// +pp.readToken_dot = function() { + let next = this.input.charCodeAt(this.pos + 1) + if (next >= 48 && next <= 57) return this.readNumber(true) + let next2 = this.input.charCodeAt(this.pos + 2) + if (this.options.ecmaVersion >= 6 && next === 46 && next2 === 46) { // 46 = dot '.' + this.pos += 3 + return this.finishToken(tt.ellipsis) + } else { + ++this.pos + return this.finishToken(tt.dot) + } +} + +pp.readToken_slash = function() { // '/' + let next = this.input.charCodeAt(this.pos + 1) + if (this.exprAllowed) { ++this.pos; return this.readRegexp() } + if (next === 61) return this.finishOp(tt.assign, 2) + return this.finishOp(tt.slash, 1) +} + +pp.readToken_mult_modulo_exp = function(code) { // '%*' + let next = this.input.charCodeAt(this.pos + 1) + let size = 1 + let tokentype = code === 42 ? tt.star : tt.modulo + + // exponentiation operator ** and **= + if (this.options.ecmaVersion >= 7 && next === 42) { + ++size + tokentype = tt.starstar + next = this.input.charCodeAt(this.pos + 2) + } + + if (next === 61) return this.finishOp(tt.assign, size + 1) + return this.finishOp(tokentype, size) +} + +pp.readToken_pipe_amp = function(code) { // '|&' + let next = this.input.charCodeAt(this.pos + 1) + if (next === code) return this.finishOp(code === 124 ? tt.logicalOR : tt.logicalAND, 2) + if (next === 61) return this.finishOp(tt.assign, 2) + return this.finishOp(code === 124 ? tt.bitwiseOR : tt.bitwiseAND, 1) +} + +pp.readToken_caret = function() { // '^' + let next = this.input.charCodeAt(this.pos + 1) + if (next === 61) return this.finishOp(tt.assign, 2) + return this.finishOp(tt.bitwiseXOR, 1) +} + +pp.readToken_plus_min = function(code) { // '+-' + let next = this.input.charCodeAt(this.pos + 1) + if (next === code) { + if (next == 45 && this.input.charCodeAt(this.pos + 2) == 62 && + lineBreak.test(this.input.slice(this.lastTokEnd, this.pos))) { + // A `-->` line comment + this.skipLineComment(3) + this.skipSpace() + return this.nextToken() + } + return this.finishOp(tt.incDec, 2) + } + if (next === 61) return this.finishOp(tt.assign, 2) + return this.finishOp(tt.plusMin, 1) +} + +pp.readToken_lt_gt = function(code) { // '<>' + let next = this.input.charCodeAt(this.pos + 1) + let size = 1 + if (next === code) { + size = code === 62 && this.input.charCodeAt(this.pos + 2) === 62 ? 3 : 2 + if (this.input.charCodeAt(this.pos + size) === 61) return this.finishOp(tt.assign, size + 1) + return this.finishOp(tt.bitShift, size) + } + if (next == 33 && code == 60 && this.input.charCodeAt(this.pos + 2) == 45 && + this.input.charCodeAt(this.pos + 3) == 45) { + if (this.inModule) this.unexpected() + // `