+ This behavior is effectively identical to Ruby's scope for local variables. + Because you don't have direct access to the var keyword, + it's impossible to shadow an outer variable on purpose, you may only refer + to it. So be careful that you're not reusing the name of an external + variable accidentally, if you're writing a deeply nested function. +
Although suppressed within this documentation for clarity, all
CoffeeScript output is wrapped in an anonymous function:
diff --git a/examples/web_server.coffee b/examples/web_server.coffee
index 8bf66d1d..b9927ef1 100644
--- a/examples/web_server.coffee
+++ b/examples/web_server.coffee
@@ -1,5 +1,6 @@
# Contributed by Jason Huggins
+process.mixin require 'sys'
http: require 'http'
server: http.createServer (req, res) ->
diff --git a/extras/CoffeeScript.tmbundle/Preferences/Indent.tmPreferences b/extras/CoffeeScript.tmbundle/Preferences/Indent.tmPreferences
index 08e29612..b331f305 100644
--- a/extras/CoffeeScript.tmbundle/Preferences/Indent.tmPreferences
+++ b/extras/CoffeeScript.tmbundle/Preferences/Indent.tmPreferences
@@ -15,8 +15,8 @@
^\s*
(.*class
|[a-zA-Z\$_](\w|\$|:|\.)*\s*(?=\:(\s*\(.*\))?\s*((=|-)>\s*$)) # function that is not one line
- |[a-zA-Z\$_](\w|\$|\.)*\s*(?=(?!\::)\:(?!(\s*\(.*\))?\s*((=|-)>))):\s*(if(?!.*?then)|while|for|$) # assignment using multiline if/while/for
- |if(?!.*?then)|while|for
+ |[a-zA-Z\$_](\w|\$|\.)*\s*(?=(?!\::)\:(?!(\s*\(.*\))?\s*((=|-)>))):\s*((if|while)(?!.*?then)|for|$) # assignment using multiline if/while/for
+ |(if|while)(?!.*?then)|for
|.*\{$
|.*\[$)
diff --git a/extras/CoffeeScript.tmbundle/Syntaxes/CoffeeScript.tmLanguage b/extras/CoffeeScript.tmbundle/Syntaxes/CoffeeScript.tmLanguage
index 0a9729ee..16a325bb 100644
--- a/extras/CoffeeScript.tmbundle/Syntaxes/CoffeeScript.tmLanguage
+++ b/extras/CoffeeScript.tmbundle/Syntaxes/CoffeeScript.tmLanguage
@@ -140,7 +140,7 @@
+ This behavior is effectively identical to Ruby's scope for local variables. + Because you don't have direct access to the var keyword, + it's impossible to shadow an outer variable on purpose, you may only refer + to it. So be careful that you're not reusing the name of an external + variable accidentally, if you're writing a deeply nested function. +
Although suppressed within this documentation for clarity, all CoffeeScript output is wrapped in an anonymous function: @@ -995,7 +1002,7 @@ numbers_copy = numbers.slice(0, numbers.length);
numbers: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9] -numbers[3..6]: [-3, -4, -5, -6] +numbers[3..6]: [-3, -4, -5, -6]
var numbers;
diff --git a/lib/lexer.js b/lib/lexer.js
index fab26120..ee349cce 100644
--- a/lib/lexer.js
+++ b/lib/lexer.js
@@ -1,5 +1,5 @@
(function(){
- var ACCESSORS, ASSIGNMENT, BEFORE_WHEN, CALLABLE, CODE, COFFEE_KEYWORDS, COMMENT, COMMENT_CLEANER, HEREDOC, HEREDOC_INDENT, IDENTIFIER, INTERPOLATION, JS_CLEANER, JS_FORBIDDEN, JS_KEYWORDS, KEYWORDS, LAST_DENT, LAST_DENTS, Lexer, MULTILINER, MULTI_DENT, NOT_REGEX, NO_NEWLINE, NUMBER, OPERATOR, REGEX, RESERVED, Rewriter, STRING_NEWLINES, WHITESPACE, compact, count, include, starts;
+ var ACCESSORS, ASSIGNMENT, BEFORE_WHEN, CALLABLE, CODE, COFFEE_KEYWORDS, COMMENT, COMMENT_CLEANER, HEREDOC, HEREDOC_INDENT, IDENTIFIER, INTERPOLATION, JS_CLEANER, JS_FORBIDDEN, JS_KEYWORDS, KEYWORDS, LAST_DENT, LAST_DENTS, Lexer, MULTILINER, MULTI_DENT, NOT_REGEX, NO_NEWLINE, NUMBER, OPERATOR, RESERVED, Rewriter, STRING_NEWLINES, WHITESPACE, compact, count, include, starts;
// The CoffeeScript Lexer. Uses a series of token-matching regexes to attempt
// matches against the beginning of the source code. When a match is found,
// a token is produced, we consume the match, and start again. Tokens are in the
@@ -33,7 +33,7 @@
Lexer.prototype.tokenize = function tokenize(code, options) {
var o;
o = options || {};
- this.code = code;
+ this.code = code || '';
// The remainder of the source code.
this.i = 0;
// Current character position we're parsing.
@@ -129,18 +129,14 @@
// Matches strings, including multi-line strings. Ensures that quotation marks
// are balanced within the string's contents, and within nested interpolations.
Lexer.prototype.string_token = function string_token() {
- var string;
+ var merge, string;
if (!(starts(this.chunk, '"') || starts(this.chunk, "'"))) {
return false;
}
- string = this.balanced_token(['"', '"'], ['${', '}']);
- if (!(string)) {
- string = this.balanced_token(["'", "'"]);
- }
- if (!(string)) {
+ if (!((string = this.balanced_token(['"', '"'], ['${', '}']) || this.balanced_token(["'", "'"])))) {
return false;
}
- this.interpolate_string(string.replace(STRING_NEWLINES, " \\\n"));
+ this.interpolate_string(string.replace(STRING_NEWLINES, " \\\n"), (merge = true));
this.line += count(string, "\n");
this.i += string.length;
return true;
@@ -175,14 +171,51 @@
// to distinguish from division, so we borrow some basic heuristics from
// JavaScript and Ruby.
Lexer.prototype.regex_token = function regex_token() {
- var regex;
- if (!((regex = this.match(REGEX, 1)))) {
+ var _a, _b, _c, _d, _e, each, flags, i, index, interp_tokens, merge, regex, str;
+ if (!(starts(this.chunk, '/'))) {
return false;
}
if (include(NOT_REGEX, this.tag())) {
return false;
}
- this.token('REGEX', regex);
+ if (!((regex = this.balanced_token(['/', '/'])))) {
+ return false;
+ }
+ if (regex.length < 3 || regex.match(/^\/\s+/m)) {
+ return false;
+ }
+ flags = ['i', 'm', 'g', 'y'];
+ while (((index = flags.indexOf(this.chunk.substr(regex.length, 1)))) >= 0) {
+ regex += flags[index];
+ flags.splice(index, 1);
+ }
+ if (((0 < (_e = regex.indexOf('${'))) && (_e < regex.indexOf('}'))) || regex.match(/[^\\]\$[a-zA-Z_@]/)) {
+ _a = regex.substring(1).split('/');
+ str = _a[0];
+ flags = _a[1];
+ str = str.replace(/\\[^\$]/g, function(escaped) {
+ return '\\' + escaped;
+ });
+ this.tokens = this.tokens.concat([['(', '('], ['NEW', 'new'], ['IDENTIFIER', 'RegExp'], ['CALL_START', '(']]);
+ interp_tokens = this.interpolate_string("\"" + str + "\"", (merge = false));
+ _b = interp_tokens;
+ for (i = 0, _c = _b.length; i < _c; i++) {
+ each = _b[i];
+ if ((_d = each[0]) === 'TOKENS') {
+ this.tokens = this.tokens.concat(each[1]);
+ } else if (_d === 'STRING') {
+ this.token(each[0], each[1].substring(0, 1) + each[1].substring(1, each[1].length - 1).replace(/"/g, '\\"') + each[1].substring(0, 1));
+ } else {
+ this.token(each[0], each[1]);
+ }
+ if (i < interp_tokens.length - 1) {
+ this.token('+', '+');
+ }
+ }
+ this.tokens = this.tokens.concat([[',', ','], ['STRING', "'" + flags + "'"], [')', ')'], [')', ')']]);
+ } else {
+ this.token('REGEX', regex);
+ }
this.i += regex.length;
return true;
};
@@ -429,6 +462,9 @@
i += 1;
}
if (levels.length) {
+ if (delimited[0][0] === '/') {
+ return false;
+ }
throw new Error("SyntaxError: Unterminated " + (levels.pop()[0]) + " starting on line " + (this.line + 1));
}
if (i === 0) {
@@ -444,8 +480,8 @@
// If it encounters an interpolation, this method will recursively create a
// new Lexer, tokenize the interpolated contents, and merge them into the
// token stream.
- Lexer.prototype.interpolate_string = function interpolate_string(str) {
- var _a, _b, _c, _d, _e, each, expr, group, i, inner, interp, lexer, match, nested, pi, quote, tokens;
+ Lexer.prototype.interpolate_string = function interpolate_string(str, merge) {
+ var _a, _b, _c, _d, _e, _f, _g, each, expr, group, has_string, i, inner, interp, lexer, match, nested, pi, quote, tokens;
if (str.length < 3 || !starts(str, '"')) {
return this.token('STRING', str);
} else {
@@ -466,14 +502,14 @@
interp = "this." + (interp.substring(1));
}
if (pi < i) {
- tokens.push(['STRING', quote + (str.substring(pi, i)) + quote]);
+ tokens.push(['STRING', '' + quote + (str.substring(pi, i)) + quote]);
}
tokens.push(['IDENTIFIER', interp]);
i += group.length - 1;
pi = i + 1;
} else if (((expr = this.balanced_string(str.substring(i), ['${', '}'])))) {
if (pi < i) {
- tokens.push(['STRING', quote + (str.substring(pi, i)) + quote]);
+ tokens.push(['STRING', '' + quote + (str.substring(pi, i)) + quote]);
}
inner = expr.substring(2, expr.length - 1);
if (inner.length) {
@@ -484,7 +520,7 @@
nested.pop();
tokens.push(['TOKENS', nested]);
} else {
- tokens.push(['STRING', quote + quote]);
+ tokens.push(['STRING', '' + quote + quote]);
}
i += expr.length - 1;
pi = i + 1;
@@ -492,19 +528,27 @@
i += 1;
}
if (pi < i && pi < str.length - 1) {
- tokens.push(['STRING', quote + (str.substring(pi, i)) + quote]);
+ tokens.push(['STRING', '' + quote + (str.substring(pi, i)) + quote]);
}
- _c = []; _d = tokens;
- for (i = 0, _e = _d.length; i < _e; i++) {
- each = _d[i];
- _c.push((function() {
+ _c = tokens;
+ for (_d = 0, _e = _c.length; _d < _e; _d++) {
+ each = _c[_d];
+ each[0] === 'STRING' ? ((has_string = true)) : null;
+ }
+ if (!has_string) {
+ tokens.unshift(['STRING', "''"]);
+ }
+ if (((typeof merge !== "undefined" && merge !== null) ? merge : true)) {
+ _f = tokens;
+ for (i = 0, _g = _f.length; i < _g; i++) {
+ each = _f[i];
each[0] === 'TOKENS' ? (this.tokens = this.tokens.concat(each[1])) : this.token(each[0], each[1]);
if (i < tokens.length - 1) {
- return this.token('+', '+');
+ this.token('+', '+');
}
- }).call(this));
+ }
}
- return _c;
+ return tokens;
}
};
// Helpers
@@ -568,7 +612,7 @@
// be used as identifiers or properties.
JS_FORBIDDEN = JS_KEYWORDS.concat(RESERVED);
// Token matching regexes.
- IDENTIFIER = /^([a-zA-Z$_](\w|\$)*)/;
+ IDENTIFIER = /^([a-zA-Z\$_](\w|\$)*)/;
NUMBER = /^(\b((0(x|X)[0-9a-fA-F]+)|([0-9]+(\.[0-9]+)?(e[+\-]?[0-9]+)?)))\b/i;
HEREDOC = /^("{6}|'{6}|"{3}\n?([\s\S]*?)\n?([ \t]*)"{3}|'{3}\n?([\s\S]*?)\n?([ \t]*)'{3})/;
INTERPOLATION = /^\$([a-zA-Z_@]\w*(\.\w+)*)/;
@@ -576,7 +620,6 @@
WHITESPACE = /^([ \t]+)/;
COMMENT = /^(((\n?[ \t]*)?#[^\n]*)+)/;
CODE = /^((-|=)>)/;
- REGEX = /^(\/(\S.*?)?([^\\]|\\\\)\/[imgy]{0,4})/;
MULTI_DENT = /^((\n([ \t]*))+)(\.)?/;
LAST_DENTS = /\n([ \t]*)/g;
LAST_DENT = /\n([ \t]*)/;
diff --git a/lib/nodes.js b/lib/nodes.js
index 3ea376a6..bca3d9a8 100644
--- a/lib/nodes.js
+++ b/lib/nodes.js
@@ -289,7 +289,7 @@ idt += TAB
var end, idt;
idt = this.is_statement() ? this.idt() : '';
end = this.is_statement() ? ';' : '';
- return idt + this.value + end;
+ return '' + idt + this.value + end;
};
LiteralNode.prototype.toString = function toString(idt) {
return " \"" + this.value + "\"";
@@ -762,7 +762,7 @@ idt += TAB
props = props.empty() ? '' : props.compile(o) + '\n';
extension = extension ? this.idt() + extension.compile(o) + ';\n' : '';
returns = ret ? '\n' + this.idt() + 'return ' + this.variable.compile(o) + ';' : '';
- return construct + extension + props + returns;
+ return '' + construct + extension + props + returns;
};
return ClassNode;
}).call(this);
@@ -1540,7 +1540,7 @@ idt += TAB
indent: this.idt(),
chain_child: true
})) : " else {\n" + (Expressions.wrap([this.else_body]).compile(o)) + "\n" + this.tab + "}";
- return if_part + else_part;
+ return '' + if_part + else_part;
};
// Compile the IfNode as a ternary operator.
IfNode.prototype.compile_ternary = function compile_ternary(o) {
@@ -1559,7 +1559,7 @@ idt += TAB
// with Git.
TRAILING_WHITESPACE = /\s+$/gm;
// Keep this identifier regex in sync with the Lexer.
- IDENTIFIER = /^[a-zA-Z$_](\w|\$)*$/;
+ IDENTIFIER = /^[a-zA-Z\$_](\w|\$)*$/;
// Utility Functions
// -----------------
// Merge objects, returning a fresh copy with attributes from both sides.
diff --git a/lib/rewriter.js b/lib/rewriter.js
index cbf63c26..e09299c4 100644
--- a/lib/rewriter.js
+++ b/lib/rewriter.js
@@ -81,7 +81,7 @@
Rewriter.prototype.remove_leading_newlines = function remove_leading_newlines() {
var _a;
_a = [];
- while (this.tokens[0][0] === 'TERMINATOR') {
+ while (this.tokens[0] && this.tokens[0][0] === 'TERMINATOR') {
_a.push(this.tokens.shift());
}
return _a;
diff --git a/src/lexer.coffee b/src/lexer.coffee
index 30ba0d9b..9e71baf3 100644
--- a/src/lexer.coffee
+++ b/src/lexer.coffee
@@ -36,7 +36,7 @@ exports.Lexer: class Lexer
# unless explicitly asked not to.
tokenize: (code, options) ->
o : options or {}
- @code : code # The remainder of the source code.
+ @code : code or '' # The remainder of the source code.
@i : 0 # Current character position we're parsing.
@line : o.line or 0 # The current line.
@indent : 0 # The current indentation level.
@@ -96,10 +96,10 @@ exports.Lexer: class Lexer
# are balanced within the string's contents, and within nested interpolations.
string_token: ->
return false unless starts(@chunk, '"') or starts(@chunk, "'")
- string: @balanced_token ['"', '"'], ['${', '}']
- string: @balanced_token ["'", "'"] unless string
- return false unless string
- @interpolate_string string.replace STRING_NEWLINES, " \\\n"
+ return false unless string:
+ @balanced_token(['"', '"'], ['${', '}']) or
+ @balanced_token ["'", "'"]
+ @interpolate_string string.replace(STRING_NEWLINES, " \\\n"), merge: true
@line += count string, "\n"
@i += string.length
true
@@ -126,9 +126,28 @@ exports.Lexer: class Lexer
# to distinguish from division, so we borrow some basic heuristics from
# JavaScript and Ruby.
regex_token: ->
- return false unless regex: @match REGEX, 1
+ return false unless starts @chunk, '/'
return false if include NOT_REGEX, @tag()
- @token 'REGEX', regex
+ return false unless regex: @balanced_token ['/', '/']
+ return false if regex.length < 3 or regex.match /^\/\s+/m
+ flags: ['i', 'm', 'g', 'y']
+ while (index: flags.indexOf @chunk.substr regex.length, 1) >= 0
+ regex += flags[index]
+ flags.splice index, 1
+ if (0 < regex.indexOf('${') < regex.indexOf('}')) or regex.match /[^\\]\$[a-zA-Z_@]/
+ [str, flags]: regex.substring(1).split('/')
+ str: str.replace /\\[^\$]/g, (escaped) -> '\\' + escaped
+ @tokens: @tokens.concat [['(', '('], ['NEW', 'new'], ['IDENTIFIER', 'RegExp'], ['CALL_START', '(']]
+ interp_tokens: @interpolate_string "\"$str\"", merge: false
+ for each, i in interp_tokens
+ switch each[0]
+ when 'TOKENS' then @tokens: @tokens.concat each[1]
+ when 'STRING' then @token each[0], each[1].substring(0, 1) + each[1].substring(1, each[1].length - 1).replace(/"/g, '\\"') + each[1].substring(0, 1)
+ else @token each[0], each[1]
+ @token '+', '+' if i < interp_tokens.length - 1
+ @tokens: @tokens.concat [[',', ','], ['STRING', "'$flags'"], [')', ')'], [')', ')']]
+ else
+ @token 'REGEX', regex
@i += regex.length
true
@@ -317,7 +336,9 @@ exports.Lexer: class Lexer
break
break unless levels.length
i += 1
- throw new Error "SyntaxError: Unterminated ${levels.pop()[0]} starting on line ${@line + 1}" if levels.length
+ if levels.length
+ return false if delimited[0][0] is '/'
+ throw new Error "SyntaxError: Unterminated ${levels.pop()[0]} starting on line ${@line + 1}"
return false if i is 0
return str.substring(0, i)
@@ -331,7 +352,7 @@ exports.Lexer: class Lexer
# If it encounters an interpolation, this method will recursively create a
# new Lexer, tokenize the interpolated contents, and merge them into the
# token stream.
- interpolate_string: (str) ->
+ interpolate_string: (str, merge) ->
if str.length < 3 or not starts str, '"'
@token 'STRING', str
else
@@ -362,12 +383,16 @@ exports.Lexer: class Lexer
pi: i + 1
i += 1
tokens.push ['STRING', "$quote${ str.substring(pi, i) }$quote"] if pi < i and pi < str.length - 1
- for each, i in tokens
- if each[0] is 'TOKENS'
- @tokens: @tokens.concat each[1]
- else
- @token each[0], each[1]
- @token '+', '+' if i < tokens.length - 1
+ (has_string: yes) for each in tokens when each[0] is 'STRING'
+ tokens.unshift ['STRING', "''"] if not has_string
+ if (merge ? true)
+ for each, i in tokens
+ if each[0] is 'TOKENS'
+ @tokens: @tokens.concat each[1]
+ else
+ @token each[0], each[1]
+ @token '+', '+' if i < tokens.length - 1
+ tokens
# Helpers
# -------
@@ -440,7 +465,7 @@ RESERVED: [
JS_FORBIDDEN: JS_KEYWORDS.concat RESERVED
# Token matching regexes.
-IDENTIFIER : /^([a-zA-Z$_](\w|\$)*)/
+IDENTIFIER : /^([a-zA-Z\$_](\w|\$)*)/
NUMBER : /^(\b((0(x|X)[0-9a-fA-F]+)|([0-9]+(\.[0-9]+)?(e[+\-]?[0-9]+)?)))\b/i
HEREDOC : /^("{6}|'{6}|"{3}\n?([\s\S]*?)\n?([ \t]*)"{3}|'{3}\n?([\s\S]*?)\n?([ \t]*)'{3})/
INTERPOLATION : /^\$([a-zA-Z_@]\w*(\.\w+)*)/
@@ -448,7 +473,6 @@ OPERATOR : /^([+\*&|\/\-%=<>:!?]+)/
WHITESPACE : /^([ \t]+)/
COMMENT : /^(((\n?[ \t]*)?#[^\n]*)+)/
CODE : /^((-|=)>)/
-REGEX : /^(\/(\S.*?)?([^\\]|\\\\)\/[imgy]{0,4})/
MULTI_DENT : /^((\n([ \t]*))+)(\.)?/
LAST_DENTS : /\n([ \t]*)/g
LAST_DENT : /\n([ \t]*)/
diff --git a/src/nodes.coffee b/src/nodes.coffee
index 142928f3..bcd18331 100644
--- a/src/nodes.coffee
+++ b/src/nodes.coffee
@@ -1192,7 +1192,7 @@ TAB: ' '
TRAILING_WHITESPACE: /\s+$/gm
# Keep this identifier regex in sync with the Lexer.
-IDENTIFIER: /^[a-zA-Z$_](\w|\$)*$/
+IDENTIFIER: /^[a-zA-Z\$_](\w|\$)*$/
# Utility Functions
# -----------------
diff --git a/src/rewriter.coffee b/src/rewriter.coffee
index 1fe3a8d8..8848f6db 100644
--- a/src/rewriter.coffee
+++ b/src/rewriter.coffee
@@ -62,7 +62,7 @@ exports.Rewriter: class Rewriter
# Leading newlines would introduce an ambiguity in the grammar, so we
# dispatch them here.
remove_leading_newlines: ->
- @tokens.shift() while @tokens[0][0] is 'TERMINATOR'
+ @tokens.shift() while @tokens[0] and @tokens[0][0] is 'TERMINATOR'
# Some blocks occur in the middle of expressions -- when we're expecting
# this, remove their trailing newlines.
diff --git a/test/test_regexp_interpolation.coffee b/test/test_regexp_interpolation.coffee
new file mode 100644
index 00000000..c5b2c74b
--- /dev/null
+++ b/test/test_regexp_interpolation.coffee
@@ -0,0 +1,17 @@
+name: 'Bob'
+
+ok not not '"Bob"'.match(/^"${name}"$/i)
+ok '"Bobby"'.match(/^"${name}"$/i) is null
+
+ok not not 'Bob'.match(/^$name$/)
+ok 'Bobby'.match(/^$name/)
+
+ok 'Bobby'.match(/${"${"${"$name"}"}"}/imgy)
+
+ok '$a$b$c'.match(/\$A\$B\$C/i)
+
+a: 1
+b: 2
+c: 3
+
+ok '123'.match(/$a$b$c/i)
diff --git a/test/test_string_interpolation.coffee b/test/test_string_interpolation.coffee
index 5c454544..a69c2780 100644
--- a/test/test_string_interpolation.coffee
+++ b/test/test_string_interpolation.coffee
@@ -61,3 +61,8 @@ ok "Where is ${"the nested ${obj["name"]}"}?" is 'Where is the nested Joe?'
ok "Hello ${world ? "$hello"}" is 'Hello World'
ok "Hello ${"${"${obj["name"]}" + '!'}"}" is 'Hello Joe!'
+
+a: 1
+b: 2
+c: 3
+ok "$a$b$c" is '123'