github/coffeescript
1
1
Fork 0
mirror of https://github.com/jashkenas/coffeescript.git synced 2026-01-15 01:38:13 -05:00
Code Issues Packages Projects Releases Wiki Activity

lexer: improved logics

Browse Source
This commit is contained in:
satyr
2010-09-23 14:14:18 +09:00
parent ed501ea37e
commit f051d0880e
2 changed files with 142 additions and 166 deletions
Download Patch File Download Diff File Expand all files Collapse all files

256
lib/lexer.js
View File

@@ -1,5 +1,5 @@
(function() { (function() {
var ASSIGNED, CALLABLE, CODE, COFFEE_ALIASES, COFFEE_KEYWORDS, COMMENT, COMPARE, COMPOUND_ASSIGN, CONVERSIONS, HEREDOC, HEREDOC_INDENT, IDENTIFIER, JS_CLEANER, JS_FORBIDDEN, JS_KEYWORDS, LAST_DENT, LAST_DENTS, LINE_BREAK, LOGIC, Lexer, MATH, MULTILINER, MULTI_DENT, NEXT_CHARACTER, NOT_REGEX, NO_NEWLINE, NUMBER, OPERATOR, REGEX_END, REGEX_ESCAPE, REGEX_INTERPOLATION, REGEX_START, RESERVED, Rewriter, SHIFT, UNARY, WHITESPACE, _ref, compact, count, include, starts; var ASSIGNED, CALLABLE, CODE, COFFEE_ALIASES, COFFEE_KEYWORDS, COMMENT, COMPARE, COMPOUND_ASSIGN, CONVERSIONS, HEREDOC, HEREDOC_INDENT, IDENTIFIER, JS_FORBIDDEN, JS_KEYWORDS, LINE_BREAK, LOGIC, Lexer, MATH, MULTILINER, MULTI_DENT, NEXT_CHARACTER, NOT_REGEX, NO_NEWLINE, NUMBER, OPERATOR, REGEX_END, REGEX_ESCAPE, REGEX_INTERPOLATION, REGEX_START, RESERVED, Rewriter, SHIFT, UNARY, WHITESPACE, _ref, compact, count, include, starts;
var __slice = Array.prototype.slice; var __slice = Array.prototype.slice;
_ref = require('./rewriter'); _ref = require('./rewriter');
Rewriter = _ref.Rewriter; Rewriter = _ref.Rewriter;
@@ -12,7 +12,7 @@
Lexer = function() {}; Lexer = function() {};
Lexer.prototype.tokenize = function(code, options) { Lexer.prototype.tokenize = function(code, options) {
var o; var o;
code = code.replace(/(\r|\s+$)/g, ''); code = code.replace(/\r/g, '').replace(/\s+$/, '');
o = options || {}; o = options || {};
this.code = code; this.code = code;
this.i = 0; this.i = 0;
@@ -22,8 +22,7 @@
this.outdebt = 0; this.outdebt = 0;
this.indents = []; this.indents = [];
this.tokens = []; this.tokens = [];
while (this.i < this.code.length) { while ((this.chunk = code.slice(this.i))) {
this.chunk = this.code.slice(this.i);
this.extractNextToken(); this.extractNextToken();
} }
this.closeIndentation(); this.closeIndentation();
@@ -33,56 +32,27 @@
return (new Rewriter()).rewrite(this.tokens); return (new Rewriter()).rewrite(this.tokens);
}; };
Lexer.prototype.extractNextToken = function() { Lexer.prototype.extractNextToken = function() {
if (this.identifierToken()) { return this.identifierToken() || this.commentToken() || this.whitespaceToken() || this.lineToken() || this.heredocToken() || this.stringToken() || this.numberToken() || this.regexToken() || this.jsToken() || this.literalToken();
return null;
}
if (this.commentToken()) {
return null;
}
if (this.whitespaceToken()) {
return null;
}
if (this.lineToken()) {
return null;
}
if (this.heredocToken()) {
return null;
}
if (this.stringToken()) {
return null;
}
if (this.numberToken()) {
return null;
}
if (this.regexToken()) {
return null;
}
if (this.jsToken()) {
return null;
}
return this.literalToken();
}; };
Lexer.prototype.identifierToken = function() { Lexer.prototype.identifierToken = function() {
var close_index, forcedIdentifier, id, tag; var close_index, forcedIdentifier, id, tag;
if (!(id = this.match(IDENTIFIER, 1))) { if (!(id = this.match(IDENTIFIER))) {
return false; return false;
} }
this.i += id.length; this.i += id.length;
forcedIdentifier = this.tagAccessor() || this.match(ASSIGNED, 1); forcedIdentifier = this.tagAccessor() || this.match(ASSIGNED, 1);
tag = 'IDENTIFIER'; tag = 'IDENTIFIER';
if (include(JS_KEYWORDS, id) || (!forcedIdentifier && include(COFFEE_KEYWORDS, id))) { if (include(JS_KEYWORDS, id) || !forcedIdentifier && include(COFFEE_KEYWORDS, id)) {
tag = id.toUpperCase(); tag = id.toUpperCase();
} if (tag === 'WHEN' && include(LINE_BREAK, this.tag())) {
if (tag === 'WHEN' && include(LINE_BREAK, this.tag())) { tag = 'LEADING_WHEN';
tag = 'LEADING_WHEN'; }
} } else if (id === 'all' && this.tag() === 'FOR') {
if (id === 'all' && this.tag() === 'FOR') {
tag = 'ALL'; tag = 'ALL';
} }
if (include(UNARY, tag)) { if (include(UNARY, tag)) {
tag = 'UNARY'; tag = 'UNARY';
} } else if (include(JS_FORBIDDEN, id)) {
if (include(JS_FORBIDDEN, id)) {
if (forcedIdentifier) { if (forcedIdentifier) {
tag = 'STRING'; tag = 'STRING';
id = ("\"" + (id) + "\""); id = ("\"" + (id) + "\"");
@@ -101,11 +71,10 @@
if (include(COFFEE_ALIASES, id)) { if (include(COFFEE_ALIASES, id)) {
tag = (id = CONVERSIONS[id]); tag = (id = CONVERSIONS[id]);
} }
if (include(LOGIC, id)) {
tag = 'LOGIC';
}
if (id === '!') { if (id === '!') {
tag = 'UNARY'; tag = 'UNARY';
} else if (include(LOGIC, id)) {
tag = 'LOGIC';
} }
} }
this.token(tag, id); this.token(tag, id);
@@ -116,10 +85,10 @@
}; };
Lexer.prototype.numberToken = function() { Lexer.prototype.numberToken = function() {
var number; var number;
if (!(number = this.match(NUMBER, 1))) { if (!(number = this.match(NUMBER))) {
return false; return false;
} }
if (this.tag() === '.' && starts(number, '.')) { if (this.tag() === '.' && number.charAt(0) === '.') {
return false; return false;
} }
this.i += number.length; this.i += number.length;
@@ -127,43 +96,47 @@
return true; return true;
}; };
Lexer.prototype.stringToken = function() { Lexer.prototype.stringToken = function() {
var string; var _ref2, string;
if (!(starts(this.chunk, '"') || starts(this.chunk, "'"))) { if (!(("'" === (_ref2 = this.chunk.charAt(0)) || '"' === _ref2))) {
return false; return false;
} }
if (!(string = this.balancedToken(['"', '"'], ['#{', '}']) || this.balancedToken(["'", "'"]))) { if (!(string = this.balancedToken(['"', '"'], ['#{', '}']) || this.balancedToken(["'", "'"]))) {
return false; return false;
} }
this.interpolateString(string.replace(/\n/g, '\\\n')); this.interpolateString(string.replace(MULTILINER, '\\\n'));
this.line += count(string, "\n"); this.line += count(string, '\n');
this.i += string.length; this.i += string.length;
return true; return true;
}; };
Lexer.prototype.heredocToken = function() { Lexer.prototype.heredocToken = function() {
var doc, match, quote; var doc, heredoc, match, quote;
if (!(match = this.chunk.match(HEREDOC))) { if (!(match = this.chunk.match(HEREDOC))) {
return false; return false;
} }
quote = match[1].substr(0, 1); heredoc = match[0];
doc = this.sanitizeHeredoc(match[2] || match[4] || '', { quote = heredoc.charAt(0);
doc = this.sanitizeHeredoc(match[2], {
quote: quote quote: quote
}); });
this.interpolateString(quote + doc + quote, { this.interpolateString(quote + doc + quote, {
heredoc: true heredoc: true
}); });
this.line += count(match[1], "\n"); this.line += count(heredoc, '\n');
this.i += match[1].length; this.i += heredoc.length;
return true; return true;
}; };
Lexer.prototype.commentToken = function() { Lexer.prototype.commentToken = function() {
var match; var _ref2, comment, here, match;
if (!(match = this.chunk.match(COMMENT))) { if (!(match = this.chunk.match(COMMENT))) {
return false; return false;
} }
this.line += count(match[1], "\n"); _ref2 = match;
this.i += match[1].length; comment = _ref2[0];
if (match[2]) { here = _ref2[1];
this.token('HERECOMMENT', this.sanitizeHeredoc(match[2], { this.line += count(comment, '\n');
this.i += comment.length;
if (here) {
this.token('HERECOMMENT', this.sanitizeHeredoc(here, {
herecomment: true, herecomment: true,
indent: Array(this.indent + 1).join(' ') indent: Array(this.indent + 1).join(' ')
})); }));
@@ -173,13 +146,13 @@
}; };
Lexer.prototype.jsToken = function() { Lexer.prototype.jsToken = function() {
var script; var script;
if (!(starts(this.chunk, '`'))) { if (this.chunk.charAt(0) !== '`') {
return false; return false;
} }
if (!(script = this.balancedToken(['`', '`']))) { if (!(script = this.balancedToken(['`', '`']))) {
return false; return false;
} }
this.token('JS', script.replace(JS_CLEANER, '')); this.token('JS', script.slice(1, -1));
this.i += script.length; this.i += script.length;
return true; return true;
}; };
@@ -197,29 +170,25 @@
if (!(regex = this.balancedToken(['/', '/']))) { if (!(regex = this.balancedToken(['/', '/']))) {
return false; return false;
} }
if (!(end = this.chunk.substr(regex.length).match(REGEX_END))) { if (!(end = this.chunk.slice(regex.length).match(REGEX_END))) {
return false; return false;
} }
if (end[2]) { flags = end[0];
regex += (flags = end[2]); if (REGEX_INTERPOLATION.test(regex)) {
} str = regex.slice(1, -1);
if (regex.match(REGEX_INTERPOLATION)) { str = str.replace(REGEX_ESCAPE, '\\$&');
str = regex.substring(1).split('/')[0]; this.tokens.push(['(', '('], ['NEW', 'new'], ['IDENTIFIER', 'RegExp'], ['CALL_START', '(']);
str = str.replace(REGEX_ESCAPE, function(escaped) {
return '\\' + escaped;
});
this.tokens = this.tokens.concat([['(', '('], ['NEW', 'new'], ['IDENTIFIER', 'RegExp'], ['CALL_START', '(']]);
this.interpolateString("\"" + (str) + "\"", { this.interpolateString("\"" + (str) + "\"", {
escapeQuotes: true escapeQuotes: true
}); });
if (flags) { if (flags) {
this.tokens.splice(this.tokens.length, 0, [',', ','], ['STRING', ("\"" + (flags) + "\"")]); this.tokens.push([',', ','], ['STRING', ("\"" + (flags) + "\"")]);
} }
this.tokens.splice(this.tokens.length, 0, [')', ')'], [')', ')']); this.tokens.push([')', ')'], [')', ')']);
} else { } else {
this.token('REGEX', regex); this.token('REGEX', regex + flags);
} }
this.i += regex.length; this.i += regex.length + flags.length;
return true; return true;
}; };
Lexer.prototype.balancedToken = function() { Lexer.prototype.balancedToken = function() {
@@ -229,13 +198,13 @@
}; };
Lexer.prototype.lineToken = function() { Lexer.prototype.lineToken = function() {
var diff, indent, nextCharacter, noNewlines, prev, size; var diff, indent, nextCharacter, noNewlines, prev, size;
if (!(indent = this.match(MULTI_DENT, 1))) { if (!(indent = this.match(MULTI_DENT))) {
return false; return false;
} }
this.line += count(indent, "\n"); this.line += count(indent, '\n');
this.i += indent.length; this.i += indent.length;
prev = this.prev(2); prev = this.prev(2);
size = indent.match(LAST_DENTS).reverse()[0].match(LAST_DENT)[1].length; size = indent.length - 1 - indent.lastIndexOf('\n');
nextCharacter = this.match(NEXT_CHARACTER, 1); nextCharacter = this.match(NEXT_CHARACTER, 1);
noNewlines = nextCharacter === '.' || nextCharacter === ',' || this.unfinished(); noNewlines = nextCharacter === '.' || nextCharacter === ',' || this.unfinished();
if (size - this.indebt === this.indent) { if (size - this.indebt === this.indent) {
@@ -283,13 +252,13 @@
this.outdebt -= moveOut; this.outdebt -= moveOut;
} }
if (!(this.tag() === 'TERMINATOR' || noNewlines)) { if (!(this.tag() === 'TERMINATOR' || noNewlines)) {
this.token('TERMINATOR', "\n"); this.token('TERMINATOR', '\n');
} }
return true; return true;
}; };
Lexer.prototype.whitespaceToken = function() { Lexer.prototype.whitespaceToken = function() {
var prev, space; var prev, space;
if (!(space = this.match(WHITESPACE, 1))) { if (!(space = this.match(WHITESPACE))) {
return false; return false;
} }
prev = this.prev(); prev = this.prev();
@@ -301,25 +270,28 @@
}; };
Lexer.prototype.newlineToken = function(newlines) { Lexer.prototype.newlineToken = function(newlines) {
if (this.tag() !== 'TERMINATOR') { if (this.tag() !== 'TERMINATOR') {
this.token('TERMINATOR', "\n"); this.token('TERMINATOR', '\n');
} }
return true; return true;
}; };
Lexer.prototype.suppressNewlines = function() { Lexer.prototype.suppressNewlines = function() {
if (this.value() === "\\") { if (this.value() === '\\') {
this.tokens.pop(); this.tokens.pop();
} }
return true; return true;
}; };
Lexer.prototype.literalToken = function() { Lexer.prototype.literalToken = function() {
var _ref2, match, prev, space, spaced, tag, value; var _ref2, match, prev, space, spaced, tag, value;
match = this.chunk.match(OPERATOR); if (match = this.chunk.match(OPERATOR)) {
value = match && match[1]; _ref2 = match;
space = match && match[2]; value = _ref2[0];
if (value && value.match(CODE)) { space = _ref2[1];
this.tagParameters(); if (CODE.test(value)) {
this.tagParameters();
}
} else {
value = this.chunk.charAt(0);
} }
value || (value = this.chunk.substr(0, 1));
this.i += value.length; this.i += value.length;
spaced = (prev = this.prev()) && prev.spaced; spaced = (prev = this.prev()) && prev.spaced;
tag = value; tag = value;
@@ -354,11 +326,13 @@
tag = 'CALL_START'; tag = 'CALL_START';
} else if (value === '[') { } else if (value === '[') {
tag = 'INDEX_START'; tag = 'INDEX_START';
if (this.tag() === '?') { switch (this.tag()) {
this.tag(1, 'INDEX_SOAK'); case '?':
} this.tag(1, 'INDEX_SOAK');
if (this.tag() === '::') { break;
this.tag(1, 'INDEX_PROTO'); case '::':
this.tag(1, 'INDEX_PROTO');
break;
} }
} }
} }
@@ -373,7 +347,7 @@
accessor = (function() { accessor = (function() {
if (prev[1] === '::') { if (prev[1] === '::') {
return this.tag(1, 'PROTOTYPE_ACCESS'); return this.tag(1, 'PROTOTYPE_ACCESS');
} else if (prev[1] === '.' && !(this.value(2) === '.')) { } else if (prev[1] === '.' && this.value(2) !== '.') {
if (this.tag(2) === '?') { if (this.tag(2) === '?') {
this.tag(1, 'SOAK_ACCESS'); this.tag(1, 'SOAK_ACCESS');
return this.tokens.splice(-2, 1); return this.tokens.splice(-2, 1);
@@ -393,20 +367,19 @@
return doc; return doc;
} }
if (!(options.herecomment)) { if (!(options.herecomment)) {
while ((match = HEREDOC_INDENT.exec(doc)) !== null) { while ((match = HEREDOC_INDENT.exec(doc))) {
attempt = (typeof (_ref2 = match[2]) !== "undefined" && _ref2 !== null) ? match[2] : match[3]; attempt = (typeof (_ref2 = match[1]) !== "undefined" && _ref2 !== null) ? match[1] : match[2];
if (!(typeof indent !== "undefined" && indent !== null) || (0 < attempt.length) && (attempt.length < indent.length)) { if (!(typeof indent !== "undefined" && indent !== null) || (0 < attempt.length) && (attempt.length < indent.length)) {
indent = attempt; indent = attempt;
} }
} }
} }
indent || (indent = ''); indent || (indent = '');
doc = doc.replace(new RegExp("^" + indent, 'gm'), ''); doc = doc.replace(new RegExp('^' + indent, 'gm'), '');
if (options.herecomment) { if (options.herecomment) {
return doc; return doc;
} }
doc = doc.replace(/^\n/, ''); return doc.replace(/^\n/, '').replace(MULTILINER, '\\n').replace(new RegExp(options.quote, 'g'), "\\" + (options.quote));
return doc.replace(MULTILINER, "\\n").replace(new RegExp(options.quote, 'g'), "\\" + (options.quote));
}; };
Lexer.prototype.tagParameters = function() { Lexer.prototype.tagParameters = function() {
var i, tok; var i, tok;
@@ -444,13 +417,14 @@
throw new Error("SyntaxError: Reserved word \"" + (this.value()) + "\" on line " + (this.line + 1) + " can't be assigned"); throw new Error("SyntaxError: Reserved word \"" + (this.value()) + "\" on line " + (this.line + 1) + " can't be assigned");
}; };
Lexer.prototype.balancedString = function(str, delimited, options) { Lexer.prototype.balancedString = function(str, delimited, options) {
var _i, _len, _ref2, _ref3, close, i, levels, open, pair, slash; var _i, _len, _ref2, _ref3, close, i, levels, open, pair, slash, slen;
options || (options = {}); options || (options = {});
slash = delimited[0][0] === '/'; slash = delimited[0][0] === '/';
levels = []; levels = [];
i = 0; i = 0;
while (i < str.length) { slen = str.length;
if (levels.length && starts(str, '\\', i)) { while (i < slen) {
if (levels.length && str.charAt(i) === '\\') {
i += 1; i += 1;
} else { } else {
_ref2 = delimited; _ref2 = delimited;
@@ -473,7 +447,7 @@
} }
} }
} }
if (!levels.length || slash && starts(str, '\n', i)) { if (!levels.length || slash && str.charAt(i) === '\n') {
break; break;
} }
i += 1; i += 1;
@@ -484,28 +458,29 @@
} }
throw new Error("SyntaxError: Unterminated " + (levels.pop()[0]) + " starting on line " + (this.line + 1)); throw new Error("SyntaxError: Unterminated " + (levels.pop()[0]) + " starting on line " + (this.line + 1));
} }
return !i ? false : str.substring(0, i); return !i ? false : str.slice(0, i);
}; };
Lexer.prototype.interpolateString = function(str, options) { Lexer.prototype.interpolateString = function(str, options) {
var _len, _ref2, _ref3, escaped, expr, i, idx, inner, interpolated, lexer, nested, pi, quote, tag, tok, token, tokens, value; var _len, _ref2, _ref3, end, escaped, expr, i, idx, inner, interpolated, lexer, nested, pi, quote, tag, tok, token, tokens, value;
options || (options = {}); options || (options = {});
if (str.length < 3 || !starts(str, '"')) { if (str.length < 3 || str.charAt(0) !== '"') {
return this.token('STRING', str); return this.token('STRING', str);
} else { } else {
lexer = new Lexer(); lexer = new Lexer();
tokens = []; tokens = [];
quote = str.substring(0, 1); quote = str.charAt(0);
_ref2 = [1, 1]; _ref2 = [1, 1];
i = _ref2[0]; i = _ref2[0];
pi = _ref2[1]; pi = _ref2[1];
while (i < str.length - 1) { end = str.length - 1;
if (starts(str, '\\', i)) { while (i < end) {
if (str.charAt(i) === '\\') {
i += 1; i += 1;
} else if (expr = this.balancedString(str.substring(i), [['#{', '}']])) { } else if (expr = this.balancedString(str.slice(i), [['#{', '}']])) {
if (pi < i) { if (pi < i) {
tokens.push(['STRING', quote + str.substring(pi, i) + quote]); tokens.push(['STRING', quote + str.slice(pi, i) + quote]);
} }
inner = expr.substring(2, expr.length - 1); inner = expr.slice(2, -1);
if (inner.length) { if (inner.length) {
if (options.heredoc) { if (options.heredoc) {
inner = inner.replace(new RegExp('\\\\' + quote, 'g'), quote); inner = inner.replace(new RegExp('\\\\' + quote, 'g'), quote);
@@ -530,8 +505,8 @@
} }
i += 1; i += 1;
} }
if (pi < i && pi < str.length - 1) { if ((i > pi) && (pi < str.length - 1)) {
tokens.push(['STRING', quote + str.substring(pi, i) + quote]); tokens.push(['STRING', quote + str.slice(pi, i) + quote]);
} }
if (tokens[0][0] !== 'STRING') { if (tokens[0][0] !== 'STRING') {
tokens.unshift(['STRING', '""']); tokens.unshift(['STRING', '""']);
@@ -549,7 +524,7 @@
if (tag === 'TOKENS') { if (tag === 'TOKENS') {
this.tokens = this.tokens.concat(value); this.tokens = this.tokens.concat(value);
} else if (tag === 'STRING' && options.escapeQuotes) { } else if (tag === 'STRING' && options.escapeQuotes) {
escaped = value.substring(1, value.length - 1).replace(/"/g, '\\"'); escaped = value.slice(1, -1).replace(/"/g, '\\"');
this.token(tag, "\"" + (escaped) + "\""); this.token(tag, "\"" + (escaped) + "\"");
} else { } else {
this.token(tag, value); this.token(tag, value);
@@ -592,42 +567,37 @@
}; };
Lexer.prototype.match = function(regex, index) { Lexer.prototype.match = function(regex, index) {
var m; var m;
if (!(m = this.chunk.match(regex))) { return (m = this.chunk.match(regex)) ? m[index || 0] : false;
return false;
}
return m ? m[index] : false;
}; };
Lexer.prototype.unfinished = function() { Lexer.prototype.unfinished = function() {
var prev; var prev, value;
prev = this.prev(2); prev = this.prev(2);
return this.value() && this.value().match && this.value().match(NO_NEWLINE) && prev && (prev[0] !== '.') && !this.value().match(CODE) && !this.chunk.match(ASSIGNED); value = this.value();
return value && NO_NEWLINE.test(value) && prev && prev[0] !== '.' && !CODE.test(value) && !ASSIGNED.test(this.chunk);
}; };
return Lexer; return Lexer;
})(); })();
JS_KEYWORDS = ["if", "else", "true", "false", "new", "return", "try", "catch", "finally", "throw", "break", "continue", "for", "in", "while", "delete", "instanceof", "typeof", "switch", "super", "extends", "class", "this", "null", "debugger"]; JS_KEYWORDS = ['if', 'else', 'true', 'false', 'new', 'return', 'try', 'catch', 'finally', 'throw', 'break', 'continue', 'for', 'in', 'while', 'delete', 'instanceof', 'typeof', 'switch', 'super', 'extends', 'class', 'this', 'null', 'debugger'];
COFFEE_ALIASES = ["and", "or", "is", "isnt", "not"]; COFFEE_ALIASES = ['and', 'or', 'is', 'isnt', 'not'];
COFFEE_KEYWORDS = COFFEE_ALIASES.concat(["then", "unless", "until", "loop", "yes", "no", "on", "off", "of", "by", "where", "when"]); COFFEE_KEYWORDS = COFFEE_ALIASES.concat(['then', 'unless', 'until', 'loop', 'yes', 'no', 'on', 'off', 'of', 'by', 'where', 'when']);
RESERVED = ["case", "default", "do", "function", "var", "void", "with", "const", "let", "enum", "export", "import", "native", "__hasProp", "__extends", "__slice"]; RESERVED = ['case', 'default', 'do', 'function', 'var', 'void', 'with', 'const', 'let', 'enum', 'export', 'import', 'native', '__hasProp', '__extends', '__slice'];
JS_FORBIDDEN = JS_KEYWORDS.concat(RESERVED); JS_FORBIDDEN = JS_KEYWORDS.concat(RESERVED);
IDENTIFIER = /^([a-zA-Z\$_](\w|\$)*)/; IDENTIFIER = /^[a-zA-Z_$][\w$]*/;
NUMBER = /^(((\b0(x|X)[0-9a-fA-F]+)|((\b[0-9]+(\.[0-9]+)?|\.[0-9]+)(e[+\-]?[0-9]+)?)))\b/i; NUMBER = /^(?:0x[\da-f]+)|^(?:\d+(\.\d+)?|\.\d+)(?:e[+-]?\d+)?/i;
HEREDOC = /^("{6}|'{6}|"{3}([\s\S]*?)\n?([ \t]*)"{3}|'{3}([\s\S]*?)\n?([ \t]*)'{3})/; HEREDOC = /^("""|''')([\s\S]*?)\n?[ \t]*\1/;
OPERATOR = /^(-[\-=>]?|\+[+=]?|[*&|\/%=<>^:!?]+)([ \t]*)/; OPERATOR = /^(?:-[-=>]?|\+[+=]?|[*&|\/%=<>^:!?]+)(?=([ \t]*))/;
WHITESPACE = /^([ \t]+)/; WHITESPACE = /^[ \t]+/;
COMMENT = /^(###([^#][\s\S]*?)(###[ \t]*\n|(###)?$)|(\s*#(?!##[^#])[^\n]*)+)/; COMMENT = /^###([^#][\s\S]*?)(?:###[ \t]*\n|(?:###)?$)|^(?:\s*#(?!##[^#])[^\n]*)+/;
CODE = /^((-|=)>)/; CODE = /^[-=]>/;
MULTI_DENT = /^((\n([ \t]*))+)(\.)?/; MULTI_DENT = /^(?:\n[ \t]*)+/;
LAST_DENTS = /\n([ \t]*)/g;
LAST_DENT = /\n([ \t]*)/;
REGEX_START = /^\/([^\/])/; REGEX_START = /^\/([^\/])/;
REGEX_INTERPOLATION = /([^\\]#\{.*[^\\]\})/; REGEX_INTERPOLATION = /[^\\]#\{.*[^\\]\}/;
REGEX_END = /^(([imgy]{1,4})\b|\W|$)/; REGEX_END = /^[imgy]{0,4}(?![a-zA-Z])/;
REGEX_ESCAPE = /\\[^\$]/g; REGEX_ESCAPE = /\\[^#]/g;
JS_CLEANER = /(^`|`$)/g;
MULTILINER = /\n/g; MULTILINER = /\n/g;
NO_NEWLINE = /^([+\*&|\/\-%=<>!.\\][<>=&|]*|and|or|is|isnt|not|delete|typeof|instanceof)$/; NO_NEWLINE = /^(?:[-+*&|\/%=<>!.\\][<>=&|]*|and|or|is(?:nt)?|not|delete|typeof|instanceof)$/;
HEREDOC_INDENT = /(\n+([ \t]*)|^([ \t]+))/g; HEREDOC_INDENT = /\n+([ \t]*)|^([ \t]+)/g;
ASSIGNED = /^\s*(([a-zA-Z\$_@]\w*|["'][^\r\n]+?["']|\d+)[ \t]*?[:=][^:=])/; ASSIGNED = /^\s*((?:[a-zA-Z$_@]\w*|["'][^\n]+?["']|\d+)[ \t]*?[:=][^:=])/;
NEXT_CHARACTER = /^\s*(\S)/; NEXT_CHARACTER = /^\s*(\S)/;
COMPOUND_ASSIGN = ['-=', '+=', '/=', '*=', '%=', '||=', '&&=', '?=', '<<=', '>>=', '>>>=', '&=', '^=', '|=']; COMPOUND_ASSIGN = ['-=', '+=', '/=', '*=', '%=', '||=', '&&=', '?=', '<<=', '>>=', '>>>=', '&=', '^=', '|='];
UNARY = ['UMINUS', 'UPLUS', '!', '!!', '~', 'TYPEOF', 'DELETE']; UNARY = ['UMINUS', 'UPLUS', '!', '!!', '~', 'TYPEOF', 'DELETE'];

52
src/lexer.coffee
View File

@@ -43,8 +43,7 @@ exports.Lexer = class Lexer
@outdebt = 0 # The under-outdentation at the current level. @outdebt = 0 # The under-outdentation at the current level.
@indents = [] # The stack of all current indentation levels. @indents = [] # The stack of all current indentation levels.
@tokens = [] # Stream of parsed tokens in the form ['TYPE', value, line] @tokens = [] # Stream of parsed tokens in the form ['TYPE', value, line]
while @i < @code.length while (@chunk = code[@i..])
@chunk = @code[@i..]
@extractNextToken() @extractNextToken()
@closeIndentation() @closeIndentation()
return @tokens if o.rewrite is off return @tokens if o.rewrite is off
@@ -54,16 +53,16 @@ exports.Lexer = class Lexer
# short-circuiting if any of them succeed. Their order determines precedence: # short-circuiting if any of them succeed. Their order determines precedence:
# `@literalToken` is the fallback catch-all. # `@literalToken` is the fallback catch-all.
extractNextToken: -> extractNextToken: ->
return if @identifierToken() @identifierToken() or
return if @commentToken() @commentToken() or
return if @whitespaceToken() @whitespaceToken() or
return if @lineToken() @lineToken() or
return if @heredocToken() @heredocToken() or
return if @stringToken() @stringToken() or
return if @numberToken() @numberToken() or
return if @regexToken() @regexToken() or
return if @jsToken() @jsToken() or
return @literalToken() @literalToken()
# Tokenizers # Tokenizers
# ---------- # ----------
@@ -79,11 +78,15 @@ exports.Lexer = class Lexer
@i += id.length @i += id.length
forcedIdentifier = @tagAccessor() or @match ASSIGNED, 1 forcedIdentifier = @tagAccessor() or @match ASSIGNED, 1
tag = 'IDENTIFIER' tag = 'IDENTIFIER'
tag = id.toUpperCase() if include(JS_KEYWORDS, id) or (not forcedIdentifier and include(COFFEE_KEYWORDS, id)) if include(JS_KEYWORDS, id) or
tag = 'LEADING_WHEN' if tag is 'WHEN' and include LINE_BREAK, @tag() not forcedIdentifier and include(COFFEE_KEYWORDS, id)
tag = 'ALL' if id is 'all' and @tag() is 'FOR' tag = id.toUpperCase()
tag = 'UNARY' if include UNARY, tag tag = 'LEADING_WHEN' if tag is 'WHEN' and include LINE_BREAK, @tag()
if include(JS_FORBIDDEN, id) else if id is 'all' and @tag() is 'FOR'
tag = 'ALL'
if include UNARY, tag
tag = 'UNARY'
else if include JS_FORBIDDEN, id
if forcedIdentifier if forcedIdentifier
tag = 'STRING' tag = 'STRING'
id = "\"#{id}\"" id = "\"#{id}\""
@@ -95,8 +98,10 @@ exports.Lexer = class Lexer
@identifierError id @identifierError id
unless forcedIdentifier unless forcedIdentifier
tag = id = CONVERSIONS[id] if include COFFEE_ALIASES, id tag = id = CONVERSIONS[id] if include COFFEE_ALIASES, id
tag = 'LOGIC' if include LOGIC, id if id is '!'
tag = 'UNARY' if id is '!' tag = 'UNARY'
else if include LOGIC, id
tag = 'LOGIC'
@token tag, id @token tag, id
@token ']', ']' if close_index @token ']', ']' if close_index
true true
@@ -177,7 +182,7 @@ exports.Lexer = class Lexer
@i += regex.length + flags.length @i += regex.length + flags.length
true true
# Matches a token in which which the passed delimiter pairs must be correctly # Matches a token in which the passed delimiter pairs must be correctly
# balanced (ie. strings, JS literals). # balanced (ie. strings, JS literals).
balancedToken: (delimited...) -> balancedToken: (delimited...) ->
@balancedString @chunk, delimited @balancedString @chunk, delimited
@@ -292,8 +297,9 @@ exports.Lexer = class Lexer
tag = 'CALL_START' tag = 'CALL_START'
else if value is '[' else if value is '['
tag = 'INDEX_START' tag = 'INDEX_START'
@tag 1, 'INDEX_SOAK' if @tag() is '?' switch @tag()
@tag 1, 'INDEX_PROTO' if @tag() is '::' when '?' then @tag 1, 'INDEX_SOAK'
when '::' then @tag 1, 'INDEX_PROTO'
@token tag, value @token tag, value
true true
@@ -307,7 +313,7 @@ exports.Lexer = class Lexer
return false if (not prev = @prev()) or (prev and prev.spaced) return false if (not prev = @prev()) or (prev and prev.spaced)
accessor = if prev[1] is '::' accessor = if prev[1] is '::'
@tag 1, 'PROTOTYPE_ACCESS' @tag 1, 'PROTOTYPE_ACCESS'
else if prev[1] is '.' and not (@value(2) is '.') else if prev[1] is '.' and @value(2) isnt '.'
if @tag(2) is '?' if @tag(2) is '?'
@tag(1, 'SOAK_ACCESS') @tag(1, 'SOAK_ACCESS')
@tokens.splice(-2, 1) @tokens.splice(-2, 1)