mirror of
https://github.com/scratchfoundation/paper.js.git
synced 2025-01-03 19:45:44 -05:00
1997 lines
54 KiB
JavaScript
1997 lines
54 KiB
JavaScript
/**
|
|
* A JavaScript tokenizer / parser / generator, originally written in Lisp.
|
|
* Copyright (c) Marijn Haverbeke <marijnh@gmail.com>
|
|
* http://marijn.haverbeke.nl/parse-js/
|
|
*
|
|
* Ported by to JavaScript by Mihai Bazon
|
|
* Copyright (c) 2010, Mihai Bazon <mihai.bazon@gmail.com>
|
|
* http://mihai.bazon.net/blog/
|
|
*
|
|
* Modifications and adaptions to browser (c) 2011, Juerg Lehni
|
|
* http://lehni.org/
|
|
*
|
|
* Distributed under the BSD license.
|
|
*/
|
|
|
|
var parse_js = new function() {
|
|
|
|
/* -----[ Tokenizer (constants) ]----- */
|
|
|
|
var KEYWORDS = array_to_hash([
|
|
"break",
|
|
"case",
|
|
"catch",
|
|
"const",
|
|
"continue",
|
|
"default",
|
|
"delete",
|
|
"do",
|
|
"else",
|
|
"finally",
|
|
"for",
|
|
"function",
|
|
"if",
|
|
"in",
|
|
"instanceof",
|
|
"new",
|
|
"return",
|
|
"switch",
|
|
"throw",
|
|
"try",
|
|
"typeof",
|
|
"var",
|
|
"void",
|
|
"while",
|
|
"with"
|
|
]);
|
|
|
|
var RESERVED_WORDS = array_to_hash([
|
|
"abstract",
|
|
"boolean",
|
|
"byte",
|
|
"char",
|
|
"class",
|
|
"debugger",
|
|
"double",
|
|
"enum",
|
|
"export",
|
|
"extends",
|
|
"final",
|
|
"float",
|
|
"goto",
|
|
"implements",
|
|
"import",
|
|
"int",
|
|
"interface",
|
|
"long",
|
|
"native",
|
|
"package",
|
|
"private",
|
|
"protected",
|
|
"public",
|
|
"short",
|
|
"static",
|
|
"super",
|
|
"synchronized",
|
|
"throws",
|
|
"transient",
|
|
"volatile"
|
|
]);
|
|
|
|
var KEYWORDS_BEFORE_EXPRESSION = array_to_hash([
|
|
"return",
|
|
"new",
|
|
"delete",
|
|
"throw",
|
|
"else",
|
|
"case"
|
|
]);
|
|
|
|
var KEYWORDS_ATOM = array_to_hash([
|
|
"false",
|
|
"null",
|
|
"true",
|
|
"undefined"
|
|
]);
|
|
|
|
var OPERATOR_CHARS = array_to_hash(characters("+-*&%=<>!?|~^"));
|
|
|
|
var RE_HEX_NUMBER = /^0x[0-9a-f]+$/i;
|
|
var RE_OCT_NUMBER = /^0[0-7]+$/;
|
|
var RE_DEC_NUMBER = /^\d*\.?\d*(?:e[+-]?\d*(?:\d\.?|\.?\d)\d*)?$/i;
|
|
|
|
var OPERATORS = array_to_hash([
|
|
"in",
|
|
"instanceof",
|
|
"typeof",
|
|
"new",
|
|
"void",
|
|
"delete",
|
|
"++",
|
|
"--",
|
|
"+",
|
|
"-",
|
|
"!",
|
|
"~",
|
|
"&",
|
|
"|",
|
|
"^",
|
|
"*",
|
|
"/",
|
|
"%",
|
|
">>",
|
|
"<<",
|
|
">>>",
|
|
"<",
|
|
">",
|
|
"<=",
|
|
">=",
|
|
"==",
|
|
"===",
|
|
"!=",
|
|
"!==",
|
|
"?",
|
|
"=",
|
|
"+=",
|
|
"-=",
|
|
"/=",
|
|
"*=",
|
|
"%=",
|
|
">>=",
|
|
"<<=",
|
|
">>>=",
|
|
"|=",
|
|
"^=",
|
|
"&=",
|
|
"&&",
|
|
"||"
|
|
]);
|
|
|
|
var WHITESPACE_CHARS = array_to_hash(characters(" \u00a0\n\r\t\f\v\u200b"));
|
|
|
|
var PUNC_BEFORE_EXPRESSION = array_to_hash(characters("[{}(,.;:"));
|
|
|
|
var PUNC_CHARS = array_to_hash(characters("[]{}(),;:"));
|
|
|
|
var REGEXP_MODIFIERS = array_to_hash(characters("gmsiy"));
|
|
|
|
/* -----[ Tokenizer ]----- */
|
|
|
|
// regexps adapted from http://xregexp.com/plugins/#unicode
|
|
var UNICODE = {
|
|
letter: new RegExp("[\\u0041-\\u005A\\u0061-\\u007A\\u00AA\\u00B5\\u00BA\\u00C0-\\u00D6\\u00D8-\\u00F6\\u00F8-\\u02C1\\u02C6-\\u02D1\\u02E0-\\u02E4\\u02EC\\u02EE\\u0370-\\u0374\\u0376\\u0377\\u037A-\\u037D\\u0386\\u0388-\\u038A\\u038C\\u038E-\\u03A1\\u03A3-\\u03F5\\u03F7-\\u0481\\u048A-\\u0523\\u0531-\\u0556\\u0559\\u0561-\\u0587\\u05D0-\\u05EA\\u05F0-\\u05F2\\u0621-\\u064A\\u066E\\u066F\\u0671-\\u06D3\\u06D5\\u06E5\\u06E6\\u06EE\\u06EF\\u06FA-\\u06FC\\u06FF\\u0710\\u0712-\\u072F\\u074D-\\u07A5\\u07B1\\u07CA-\\u07EA\\u07F4\\u07F5\\u07FA\\u0904-\\u0939\\u093D\\u0950\\u0958-\\u0961\\u0971\\u0972\\u097B-\\u097F\\u0985-\\u098C\\u098F\\u0990\\u0993-\\u09A8\\u09AA-\\u09B0\\u09B2\\u09B6-\\u09B9\\u09BD\\u09CE\\u09DC\\u09DD\\u09DF-\\u09E1\\u09F0\\u09F1\\u0A05-\\u0A0A\\u0A0F\\u0A10\\u0A13-\\u0A28\\u0A2A-\\u0A30\\u0A32\\u0A33\\u0A35\\u0A36\\u0A38\\u0A39\\u0A59-\\u0A5C\\u0A5E\\u0A72-\\u0A74\\u0A85-\\u0A8D\\u0A8F-\\u0A91\\u0A93-\\u0AA8\\u0AAA-\\u0AB0\\u0AB2\\u0AB3\\u0AB5-\\u0AB9\\u0ABD\\u0AD0\\u0AE0\\u0AE1\\u0B05-\\u0B0C\\u0B0F\\u0B10\\u0B13-\\u0B28\\u0B2A-\\u0B30\\u0B32\\u0B33\\u0B35-\\u0B39\\u0B3D\\u0B5C\\u0B5D\\u0B5F-\\u0B61\\u0B71\\u0B83\\u0B85-\\u0B8A\\u0B8E-\\u0B90\\u0B92-\\u0B95\\u0B99\\u0B9A\\u0B9C\\u0B9E\\u0B9F\\u0BA3\\u0BA4\\u0BA8-\\u0BAA\\u0BAE-\\u0BB9\\u0BD0\\u0C05-\\u0C0C\\u0C0E-\\u0C10\\u0C12-\\u0C28\\u0C2A-\\u0C33\\u0C35-\\u0C39\\u0C3D\\u0C58\\u0C59\\u0C60\\u0C61\\u0C85-\\u0C8C\\u0C8E-\\u0C90\\u0C92-\\u0CA8\\u0CAA-\\u0CB3\\u0CB5-\\u0CB9\\u0CBD\\u0CDE\\u0CE0\\u0CE1\\u0D05-\\u0D0C\\u0D0E-\\u0D10\\u0D12-\\u0D28\\u0D2A-\\u0D39\\u0D3D\\u0D60\\u0D61\\u0D7A-\\u0D7F\\u0D85-\\u0D96\\u0D9A-\\u0DB1\\u0DB3-\\u0DBB\\u0DBD\\u0DC0-\\u0DC6\\u0E01-\\u0E30\\u0E32\\u0E33\\u0E40-\\u0E46\\u0E81\\u0E82\\u0E84\\u0E87\\u0E88\\u0E8A\\u0E8D\\u0E94-\\u0E97\\u0E99-\\u0E9F\\u0EA1-\\u0EA3\\u0EA5\\u0EA7\\u0EAA\\u0EAB\\u0EAD-\\u0EB0\\u0EB2\\u0EB3\\u0EBD\\u0EC0-\\u0EC4\\u0EC6\\u0EDC\\u0EDD\\u0F00\\u0F40-\\u0F47\\u0F49-\\u0F6C\\u0F88-\\u0F8B\\u1000-\\u102A\\u103F\\u1050-\\u1055\\u105A-\\u105D\\u1061\\u1065\\u1066\\u106E-\\u1070\\u1075-\\u1081\\u108E\\u10A0-\\u10C5\\u10D0-\\u10FA\\u10FC\\u1100-\\u1159\\u115F-\\u11A2\\u11A8-\\u11F9\\u1200-\\u1248\\u124A-\\u124D\\u1250-\\u1256\\u1258\\u125A-\\u125D\\u1260-\\u1288\\u128A-\\u128D\\u1290-\\u12B0\\u12B2-\\u12B5\\u12B8-\\u12BE\\u12C0\\u12C2-\\u12C5\\u12C8-\\u12D6\\u12D8-\\u1310\\u1312-\\u1315\\u1318-\\u135A\\u1380-\\u138F\\u13A0-\\u13F4\\u1401-\\u166C\\u166F-\\u1676\\u1681-\\u169A\\u16A0-\\u16EA\\u1700-\\u170C\\u170E-\\u1711\\u1720-\\u1731\\u1740-\\u1751\\u1760-\\u176C\\u176E-\\u1770\\u1780-\\u17B3\\u17D7\\u17DC\\u1820-\\u1877\\u1880-\\u18A8\\u18AA\\u1900-\\u191C\\u1950-\\u196D\\u1970-\\u1974\\u1980-\\u19A9\\u19C1-\\u19C7\\u1A00-\\u1A16\\u1B05-\\u1B33\\u1B45-\\u1B4B\\u1B83-\\u1BA0\\u1BAE\\u1BAF\\u1C00-\\u1C23\\u1C4D-\\u1C4F\\u1C5A-\\u1C7D\\u1D00-\\u1DBF\\u1E00-\\u1F15\\u1F18-\\u1F1D\\u1F20-\\u1F45\\u1F48-\\u1F4D\\u1F50-\\u1F57\\u1F59\\u1F5B\\u1F5D\\u1F5F-\\u1F7D\\u1F80-\\u1FB4\\u1FB6-\\u1FBC\\u1FBE\\u1FC2-\\u1FC4\\u1FC6-\\u1FCC\\u1FD0-\\u1FD3\\u1FD6-\\u1FDB\\u1FE0-\\u1FEC\\u1FF2-\\u1FF4\\u1FF6-\\u1FFC\\u2071\\u207F\\u2090-\\u2094\\u2102\\u2107\\u210A-\\u2113\\u2115\\u2119-\\u211D\\u2124\\u2126\\u2128\\u212A-\\u212D\\u212F-\\u2139\\u213C-\\u213F\\u2145-\\u2149\\u214E\\u2183\\u2184\\u2C00-\\u2C2E\\u2C30-\\u2C5E\\u2C60-\\u2C6F\\u2C71-\\u2C7D\\u2C80-\\u2CE4\\u2D00-\\u2D25\\u2D30-\\u2D65\\u2D6F\\u2D80-\\u2D96\\u2DA0-\\u2DA6\\u2DA8-\\u2DAE\\u2DB0-\\u2DB6\\u2DB8-\\u2DBE\\u2DC0-\\u2DC6\\u2DC8-\\u2DCE\\u2DD0-\\u2DD6\\u2DD8-\\u2DDE\\u2E2F\\u3005\\u3006\\u3031-\\u3035\\u303B\\u303C\\u3041-\\u3096\\u309D-\\u309F\\u30A1-\\u30FA\\u30FC-\\u30FF\\u3105-\\u312D\\u3131-\\u318E\\u31A0-\\u31B7\\u31F0-\\u31FF\\u3400\\u4DB5\\u4E00\\u9FC3\\uA000-\\uA48C\\uA500-\\uA60C\\uA610-\\uA61F\\uA62A\\uA62B\\uA640-\\uA65F\\uA662-\\uA66E\\uA67F-\\uA697\\uA717-\\uA71F\\uA722-\\uA788\\uA78B\\uA78C\\uA7FB-\\uA801\\uA803-\\uA805\\uA807-\\uA80A\\uA80C-\\uA822\\uA840-\\uA873\\uA882-\\uA8B3\\uA90A-\\uA925\\uA930-\\uA946\\uAA00-\\uAA28\\uAA40-\\uAA42\\uAA44-\\uAA4B\\uAC00\\uD7A3\\uF900-\\uFA2D\\uFA30-\\uFA6A\\uFA70-\\uFAD9\\uFB00-\\uFB06\\uFB13-\\uFB17\\uFB1D\\uFB1F-\\uFB28\\uFB2A-\\uFB36\\uFB38-\\uFB3C\\uFB3E\\uFB40\\uFB41\\uFB43\\uFB44\\uFB46-\\uFBB1\\uFBD3-\\uFD3D\\uFD50-\\uFD8F\\uFD92-\\uFDC7\\uFDF0-\\uFDFB\\uFE70-\\uFE74\\uFE76-\\uFEFC\\uFF21-\\uFF3A\\uFF41-\\uFF5A\\uFF66-\\uFFBE\\uFFC2-\\uFFC7\\uFFCA-\\uFFCF\\uFFD2-\\uFFD7\\uFFDA-\\uFFDC]"),
|
|
non_spacing_mark: new RegExp("[\\u0300-\\u036F\\u0483-\\u0487\\u0591-\\u05BD\\u05BF\\u05C1\\u05C2\\u05C4\\u05C5\\u05C7\\u0610-\\u061A\\u064B-\\u065E\\u0670\\u06D6-\\u06DC\\u06DF-\\u06E4\\u06E7\\u06E8\\u06EA-\\u06ED\\u0711\\u0730-\\u074A\\u07A6-\\u07B0\\u07EB-\\u07F3\\u0816-\\u0819\\u081B-\\u0823\\u0825-\\u0827\\u0829-\\u082D\\u0900-\\u0902\\u093C\\u0941-\\u0948\\u094D\\u0951-\\u0955\\u0962\\u0963\\u0981\\u09BC\\u09C1-\\u09C4\\u09CD\\u09E2\\u09E3\\u0A01\\u0A02\\u0A3C\\u0A41\\u0A42\\u0A47\\u0A48\\u0A4B-\\u0A4D\\u0A51\\u0A70\\u0A71\\u0A75\\u0A81\\u0A82\\u0ABC\\u0AC1-\\u0AC5\\u0AC7\\u0AC8\\u0ACD\\u0AE2\\u0AE3\\u0B01\\u0B3C\\u0B3F\\u0B41-\\u0B44\\u0B4D\\u0B56\\u0B62\\u0B63\\u0B82\\u0BC0\\u0BCD\\u0C3E-\\u0C40\\u0C46-\\u0C48\\u0C4A-\\u0C4D\\u0C55\\u0C56\\u0C62\\u0C63\\u0CBC\\u0CBF\\u0CC6\\u0CCC\\u0CCD\\u0CE2\\u0CE3\\u0D41-\\u0D44\\u0D4D\\u0D62\\u0D63\\u0DCA\\u0DD2-\\u0DD4\\u0DD6\\u0E31\\u0E34-\\u0E3A\\u0E47-\\u0E4E\\u0EB1\\u0EB4-\\u0EB9\\u0EBB\\u0EBC\\u0EC8-\\u0ECD\\u0F18\\u0F19\\u0F35\\u0F37\\u0F39\\u0F71-\\u0F7E\\u0F80-\\u0F84\\u0F86\\u0F87\\u0F90-\\u0F97\\u0F99-\\u0FBC\\u0FC6\\u102D-\\u1030\\u1032-\\u1037\\u1039\\u103A\\u103D\\u103E\\u1058\\u1059\\u105E-\\u1060\\u1071-\\u1074\\u1082\\u1085\\u1086\\u108D\\u109D\\u135F\\u1712-\\u1714\\u1732-\\u1734\\u1752\\u1753\\u1772\\u1773\\u17B7-\\u17BD\\u17C6\\u17C9-\\u17D3\\u17DD\\u180B-\\u180D\\u18A9\\u1920-\\u1922\\u1927\\u1928\\u1932\\u1939-\\u193B\\u1A17\\u1A18\\u1A56\\u1A58-\\u1A5E\\u1A60\\u1A62\\u1A65-\\u1A6C\\u1A73-\\u1A7C\\u1A7F\\u1B00-\\u1B03\\u1B34\\u1B36-\\u1B3A\\u1B3C\\u1B42\\u1B6B-\\u1B73\\u1B80\\u1B81\\u1BA2-\\u1BA5\\u1BA8\\u1BA9\\u1C2C-\\u1C33\\u1C36\\u1C37\\u1CD0-\\u1CD2\\u1CD4-\\u1CE0\\u1CE2-\\u1CE8\\u1CED\\u1DC0-\\u1DE6\\u1DFD-\\u1DFF\\u20D0-\\u20DC\\u20E1\\u20E5-\\u20F0\\u2CEF-\\u2CF1\\u2DE0-\\u2DFF\\u302A-\\u302F\\u3099\\u309A\\uA66F\\uA67C\\uA67D\\uA6F0\\uA6F1\\uA802\\uA806\\uA80B\\uA825\\uA826\\uA8C4\\uA8E0-\\uA8F1\\uA926-\\uA92D\\uA947-\\uA951\\uA980-\\uA982\\uA9B3\\uA9B6-\\uA9B9\\uA9BC\\uAA29-\\uAA2E\\uAA31\\uAA32\\uAA35\\uAA36\\uAA43\\uAA4C\\uAAB0\\uAAB2-\\uAAB4\\uAAB7\\uAAB8\\uAABE\\uAABF\\uAAC1\\uABE5\\uABE8\\uABED\\uFB1E\\uFE00-\\uFE0F\\uFE20-\\uFE26]"),
|
|
space_combining_mark: new RegExp("[\\u0903\\u093E-\\u0940\\u0949-\\u094C\\u094E\\u0982\\u0983\\u09BE-\\u09C0\\u09C7\\u09C8\\u09CB\\u09CC\\u09D7\\u0A03\\u0A3E-\\u0A40\\u0A83\\u0ABE-\\u0AC0\\u0AC9\\u0ACB\\u0ACC\\u0B02\\u0B03\\u0B3E\\u0B40\\u0B47\\u0B48\\u0B4B\\u0B4C\\u0B57\\u0BBE\\u0BBF\\u0BC1\\u0BC2\\u0BC6-\\u0BC8\\u0BCA-\\u0BCC\\u0BD7\\u0C01-\\u0C03\\u0C41-\\u0C44\\u0C82\\u0C83\\u0CBE\\u0CC0-\\u0CC4\\u0CC7\\u0CC8\\u0CCA\\u0CCB\\u0CD5\\u0CD6\\u0D02\\u0D03\\u0D3E-\\u0D40\\u0D46-\\u0D48\\u0D4A-\\u0D4C\\u0D57\\u0D82\\u0D83\\u0DCF-\\u0DD1\\u0DD8-\\u0DDF\\u0DF2\\u0DF3\\u0F3E\\u0F3F\\u0F7F\\u102B\\u102C\\u1031\\u1038\\u103B\\u103C\\u1056\\u1057\\u1062-\\u1064\\u1067-\\u106D\\u1083\\u1084\\u1087-\\u108C\\u108F\\u109A-\\u109C\\u17B6\\u17BE-\\u17C5\\u17C7\\u17C8\\u1923-\\u1926\\u1929-\\u192B\\u1930\\u1931\\u1933-\\u1938\\u19B0-\\u19C0\\u19C8\\u19C9\\u1A19-\\u1A1B\\u1A55\\u1A57\\u1A61\\u1A63\\u1A64\\u1A6D-\\u1A72\\u1B04\\u1B35\\u1B3B\\u1B3D-\\u1B41\\u1B43\\u1B44\\u1B82\\u1BA1\\u1BA6\\u1BA7\\u1BAA\\u1C24-\\u1C2B\\u1C34\\u1C35\\u1CE1\\u1CF2\\uA823\\uA824\\uA827\\uA880\\uA881\\uA8B4-\\uA8C3\\uA952\\uA953\\uA983\\uA9B4\\uA9B5\\uA9BA\\uA9BB\\uA9BD-\\uA9C0\\uAA2F\\uAA30\\uAA33\\uAA34\\uAA4D\\uAA7B\\uABE3\\uABE4\\uABE6\\uABE7\\uABE9\\uABEA\\uABEC]"),
|
|
connector_punctuation: new RegExp("[\\u005F\\u203F\\u2040\\u2054\\uFE33\\uFE34\\uFE4D-\\uFE4F\\uFF3F]")
|
|
};
|
|
|
|
function is_letter(ch) {
|
|
return UNICODE.letter.test(ch);
|
|
};
|
|
|
|
function is_digit(ch) {
|
|
ch = ch.charCodeAt(0);
|
|
return ch >= 48 && ch <= 57; //XXX: find out if "UnicodeDigit" means something else than 0..9
|
|
};
|
|
|
|
function is_alphanumeric_char(ch) {
|
|
return is_digit(ch) || is_letter(ch);
|
|
};
|
|
|
|
function is_unicode_combining_mark(ch) {
|
|
return UNICODE.non_spacing_mark.test(ch) || UNICODE.space_combining_mark.test(ch);
|
|
};
|
|
|
|
function is_unicode_connector_punctuation(ch) {
|
|
return UNICODE.connector_punctuation.test(ch);
|
|
};
|
|
|
|
function is_identifier_start(ch) {
|
|
return ch == "$" || ch == "_" || is_letter(ch);
|
|
};
|
|
|
|
function is_identifier_char(ch) {
|
|
return is_identifier_start(ch)
|
|
|| is_unicode_combining_mark(ch)
|
|
|| is_digit(ch)
|
|
|| is_unicode_connector_punctuation(ch)
|
|
|| ch == "\u200c" // zero-width non-joiner <ZWNJ>
|
|
|| ch == "\u200d" // zero-width joiner <ZWJ> (in my ECMA-262 PDF, this is also 200c)
|
|
;
|
|
};
|
|
|
|
function parse_js_number(num) {
|
|
if (RE_HEX_NUMBER.test(num)) {
|
|
return parseInt(num.substr(2), 16);
|
|
} else if (RE_OCT_NUMBER.test(num)) {
|
|
return parseInt(num.substr(1), 8);
|
|
} else if (RE_DEC_NUMBER.test(num)) {
|
|
return parseFloat(num);
|
|
}
|
|
};
|
|
|
|
function JS_Parse_Error(message, line, col, pos) {
|
|
this.message = message;
|
|
this.line = line;
|
|
this.col = col;
|
|
this.pos = pos;
|
|
};
|
|
|
|
JS_Parse_Error.prototype.toString = function() {
|
|
return this.message + " (line: " + this.line + ", col: " + this.col + ", pos: " + this.pos + ")";
|
|
};
|
|
|
|
function js_error(message, line, col, pos) {
|
|
throw new JS_Parse_Error(message, line, col, pos);
|
|
};
|
|
|
|
function is_token(token, type, val) {
|
|
return token.type == type && (val == null || token.value == val);
|
|
};
|
|
|
|
var EX_EOF = {};
|
|
|
|
function tokenizer($TEXT) {
|
|
|
|
var S = {
|
|
text : $TEXT.replace(/\r\n?|[\n\u2028\u2029]/g, "\n").replace(/^\uFEFF/, ''),
|
|
pos : 0,
|
|
tokpos : 0,
|
|
line : 0,
|
|
tokline : 0,
|
|
col : 0,
|
|
tokcol : 0,
|
|
newline_before : false,
|
|
regex_allowed : false,
|
|
comments_before : []
|
|
};
|
|
|
|
function peek() { return S.text.charAt(S.pos); };
|
|
|
|
function next(signal_eof) {
|
|
var ch = S.text.charAt(S.pos++);
|
|
if (signal_eof && !ch)
|
|
throw EX_EOF;
|
|
if (ch == "\n") {
|
|
S.newline_before = true;
|
|
++S.line;
|
|
S.col = 0;
|
|
} else {
|
|
++S.col;
|
|
}
|
|
return ch;
|
|
};
|
|
|
|
function eof() {
|
|
return !S.peek();
|
|
};
|
|
|
|
function find(what, signal_eof) {
|
|
var pos = S.text.indexOf(what, S.pos);
|
|
if (signal_eof && pos == -1) throw EX_EOF;
|
|
return pos;
|
|
};
|
|
|
|
function start_token() {
|
|
S.tokline = S.line;
|
|
S.tokcol = S.col;
|
|
S.tokpos = S.pos;
|
|
};
|
|
|
|
function token(type, value, is_comment) {
|
|
S.regex_allowed = ((type == "operator" && !HOP(UNARY_POSTFIX, value)) ||
|
|
(type == "keyword" && HOP(KEYWORDS_BEFORE_EXPRESSION, value)) ||
|
|
(type == "punc" && HOP(PUNC_BEFORE_EXPRESSION, value)));
|
|
var ret = {
|
|
type : type,
|
|
value : value,
|
|
line : S.tokline,
|
|
col : S.tokcol,
|
|
pos : S.tokpos,
|
|
nlb : S.newline_before
|
|
};
|
|
if (!is_comment) {
|
|
ret.comments_before = S.comments_before;
|
|
S.comments_before = [];
|
|
}
|
|
S.newline_before = false;
|
|
return ret;
|
|
};
|
|
|
|
function skip_whitespace() {
|
|
while (HOP(WHITESPACE_CHARS, peek()))
|
|
next();
|
|
};
|
|
|
|
function read_while(pred) {
|
|
var ret = "", ch = peek(), i = 0;
|
|
while (ch && pred(ch, i++)) {
|
|
ret += next();
|
|
ch = peek();
|
|
}
|
|
return ret;
|
|
};
|
|
|
|
function parse_error(err) {
|
|
js_error(err, S.tokline, S.tokcol, S.tokpos);
|
|
};
|
|
|
|
function read_num(prefix) {
|
|
var has_e = false, after_e = false, has_x = false, has_dot = prefix == ".";
|
|
var num = read_while(function(ch, i){
|
|
if (ch == "x" || ch == "X") {
|
|
if (has_x) return false;
|
|
return has_x = true;
|
|
}
|
|
if (!has_x && (ch == "E" || ch == "e")) {
|
|
if (has_e) return false;
|
|
return has_e = after_e = true;
|
|
}
|
|
if (ch == "-") {
|
|
if (after_e || (i == 0 && !prefix)) return true;
|
|
return false;
|
|
}
|
|
if (ch == "+") return after_e;
|
|
after_e = false;
|
|
if (ch == ".") {
|
|
if (!has_dot && !has_x)
|
|
return has_dot = true;
|
|
return false;
|
|
}
|
|
return is_alphanumeric_char(ch);
|
|
});
|
|
if (prefix)
|
|
num = prefix + num;
|
|
var valid = parse_js_number(num);
|
|
if (!isNaN(valid)) {
|
|
return token("num", valid);
|
|
} else {
|
|
parse_error("Invalid syntax: " + num);
|
|
}
|
|
};
|
|
|
|
function read_escaped_char() {
|
|
var ch = next(true);
|
|
switch (ch) {
|
|
case "n" : return "\n";
|
|
case "r" : return "\r";
|
|
case "t" : return "\t";
|
|
case "b" : return "\b";
|
|
case "v" : return "\v";
|
|
case "f" : return "\f";
|
|
case "0" : return "\0";
|
|
case "x" : return String.fromCharCode(hex_bytes(2));
|
|
case "u" : return String.fromCharCode(hex_bytes(4));
|
|
case "\n": return "";
|
|
default : return ch;
|
|
}
|
|
};
|
|
|
|
function hex_bytes(n) {
|
|
var num = 0;
|
|
for (; n > 0; --n) {
|
|
var digit = parseInt(next(true), 16);
|
|
if (isNaN(digit))
|
|
parse_error("Invalid hex-character pattern in string");
|
|
num = (num << 4) | digit;
|
|
}
|
|
return num;
|
|
};
|
|
|
|
function read_string() {
|
|
return with_eof_error("Unterminated string constant", function(){
|
|
var quote = next(), ret = "";
|
|
for (;;) {
|
|
var ch = next(true);
|
|
if (ch == "\\") {
|
|
// read OctalEscapeSequence (XXX: deprecated if "strict mode")
|
|
// https://github.com/mishoo/UglifyJS/issues/178
|
|
var octal_len = 0, first = null;
|
|
ch = read_while(function(ch){
|
|
if (ch >= "0" && ch <= "7") {
|
|
if (!first) {
|
|
first = ch;
|
|
return ++octal_len;
|
|
}
|
|
else if (first <= "3" && octal_len <= 2) return ++octal_len;
|
|
else if (first >= "4" && octal_len <= 1) return ++octal_len;
|
|
}
|
|
return false;
|
|
});
|
|
if (octal_len > 0) ch = String.fromCharCode(parseInt(ch, 8));
|
|
else ch = read_escaped_char();
|
|
}
|
|
else if (ch == quote) break;
|
|
ret += ch;
|
|
}
|
|
return token("string", ret);
|
|
});
|
|
};
|
|
|
|
function read_line_comment() {
|
|
next();
|
|
var i = find("\n"), ret;
|
|
if (i == -1) {
|
|
ret = S.text.substr(S.pos);
|
|
S.pos = S.text.length;
|
|
} else {
|
|
ret = S.text.substring(S.pos, i);
|
|
S.pos = i;
|
|
}
|
|
return token("comment1", ret, true);
|
|
};
|
|
|
|
function read_multiline_comment() {
|
|
next();
|
|
return with_eof_error("Unterminated multiline comment", function(){
|
|
var i = find("*/", true),
|
|
text = S.text.substring(S.pos, i),
|
|
tok = token("comment2", text, true);
|
|
S.pos = i + 2;
|
|
S.line += text.split("\n").length - 1;
|
|
S.newline_before = text.indexOf("\n") >= 0;
|
|
return tok;
|
|
});
|
|
};
|
|
|
|
function read_name() {
|
|
var backslash = false, name = "", ch;
|
|
while ((ch = peek()) != null) {
|
|
if (!backslash) {
|
|
if (ch == "\\") backslash = true, next();
|
|
else if (is_identifier_char(ch)) name += next();
|
|
else break;
|
|
}
|
|
else {
|
|
if (ch != "u") parse_error("Expecting UnicodeEscapeSequence -- uXXXX");
|
|
ch = read_escaped_char();
|
|
if (!is_identifier_char(ch)) parse_error("Unicode char: " + ch.charCodeAt(0) + " is not valid in identifier");
|
|
name += ch;
|
|
backslash = false;
|
|
}
|
|
}
|
|
return name;
|
|
};
|
|
|
|
function read_regexp() {
|
|
return with_eof_error("Unterminated regular expression", function(){
|
|
var prev_backslash = false, regexp = "", ch, in_class = false;
|
|
while ((ch = next(true))) if (prev_backslash) {
|
|
regexp += "\\" + ch;
|
|
prev_backslash = false;
|
|
} else if (ch == "[") {
|
|
in_class = true;
|
|
regexp += ch;
|
|
} else if (ch == "]" && in_class) {
|
|
in_class = false;
|
|
regexp += ch;
|
|
} else if (ch == "/" && !in_class) {
|
|
break;
|
|
} else if (ch == "\\") {
|
|
prev_backslash = true;
|
|
} else {
|
|
regexp += ch;
|
|
}
|
|
var mods = read_name();
|
|
return token("regexp", [ regexp, mods ]);
|
|
});
|
|
};
|
|
|
|
function read_operator(prefix) {
|
|
function grow(op) {
|
|
if (!peek()) return op;
|
|
var bigger = op + peek();
|
|
if (HOP(OPERATORS, bigger)) {
|
|
next();
|
|
return grow(bigger);
|
|
} else {
|
|
return op;
|
|
}
|
|
};
|
|
return token("operator", grow(prefix || next()));
|
|
};
|
|
|
|
function handle_slash() {
|
|
next();
|
|
var regex_allowed = S.regex_allowed;
|
|
switch (peek()) {
|
|
case "/":
|
|
S.comments_before.push(read_line_comment());
|
|
S.regex_allowed = regex_allowed;
|
|
return next_token();
|
|
case "*":
|
|
S.comments_before.push(read_multiline_comment());
|
|
S.regex_allowed = regex_allowed;
|
|
return next_token();
|
|
}
|
|
return S.regex_allowed ? read_regexp() : read_operator("/");
|
|
};
|
|
|
|
function handle_dot() {
|
|
next();
|
|
return is_digit(peek())
|
|
? read_num(".")
|
|
: token("punc", ".");
|
|
};
|
|
|
|
function read_word() {
|
|
var word = read_name();
|
|
return !HOP(KEYWORDS, word)
|
|
? token("name", word)
|
|
: HOP(OPERATORS, word)
|
|
? token("operator", word)
|
|
: HOP(KEYWORDS_ATOM, word)
|
|
? token("atom", word)
|
|
: token("keyword", word);
|
|
};
|
|
|
|
function with_eof_error(eof_error, cont) {
|
|
try {
|
|
return cont();
|
|
} catch(ex) {
|
|
if (ex === EX_EOF) parse_error(eof_error);
|
|
else throw ex;
|
|
}
|
|
};
|
|
|
|
function next_token(force_regexp) {
|
|
if (force_regexp)
|
|
return read_regexp();
|
|
skip_whitespace();
|
|
start_token();
|
|
var ch = peek();
|
|
if (!ch) return token("eof");
|
|
if (is_digit(ch)) return read_num();
|
|
if (ch == '"' || ch == "'") return read_string();
|
|
if (HOP(PUNC_CHARS, ch)) return token("punc", next());
|
|
if (ch == ".") return handle_dot();
|
|
if (ch == "/") return handle_slash();
|
|
if (HOP(OPERATOR_CHARS, ch)) return read_operator();
|
|
if (ch == "\\" || is_identifier_start(ch)) return read_word();
|
|
parse_error("Unexpected character '" + ch + "'");
|
|
};
|
|
|
|
next_token.context = function(nc) {
|
|
if (nc) S = nc;
|
|
return S;
|
|
};
|
|
|
|
return next_token;
|
|
|
|
};
|
|
|
|
/* -----[ Parser (constants) ]----- */
|
|
|
|
var UNARY_PREFIX = array_to_hash([
|
|
"typeof",
|
|
"void",
|
|
"delete",
|
|
"--",
|
|
"++",
|
|
"!",
|
|
"~",
|
|
"-",
|
|
"+"
|
|
]);
|
|
|
|
var UNARY_POSTFIX = array_to_hash([ "--", "++" ]);
|
|
|
|
var ASSIGNMENT = (function(a, ret, i){
|
|
while (i < a.length) {
|
|
ret[a[i]] = a[i].substr(0, a[i].length - 1);
|
|
i++;
|
|
}
|
|
return ret;
|
|
})(
|
|
["+=", "-=", "/=", "*=", "%=", ">>=", "<<=", ">>>=", "|=", "^=", "&="],
|
|
{ "=": true },
|
|
0
|
|
);
|
|
|
|
var PRECEDENCE = (function(a, ret){
|
|
for (var i = 0, n = 1; i < a.length; ++i, ++n) {
|
|
var b = a[i];
|
|
for (var j = 0; j < b.length; ++j) {
|
|
ret[b[j]] = n;
|
|
}
|
|
}
|
|
return ret;
|
|
})(
|
|
[
|
|
["||"],
|
|
["&&"],
|
|
["|"],
|
|
["^"],
|
|
["&"],
|
|
["==", "===", "!=", "!=="],
|
|
["<", ">", "<=", ">=", "in", "instanceof"],
|
|
[">>", "<<", ">>>"],
|
|
["+", "-"],
|
|
["*", "/", "%"]
|
|
],
|
|
{}
|
|
);
|
|
|
|
var STATEMENTS_WITH_LABELS = array_to_hash([ "for", "do", "while", "switch" ]);
|
|
|
|
var ATOMIC_START_TOKEN = array_to_hash([ "atom", "num", "string", "regexp", "name" ]);
|
|
|
|
/* -----[ Parser ]----- */
|
|
|
|
function NodeWithToken(str, start, end) {
|
|
this.name = str;
|
|
this.start = start;
|
|
this.end = end;
|
|
};
|
|
|
|
NodeWithToken.prototype.toString = function() { return this.name; };
|
|
|
|
function parse($TEXT, exigent_mode, embed_tokens) {
|
|
|
|
var S = {
|
|
input : typeof $TEXT == "string" ? tokenizer($TEXT, true) : $TEXT,
|
|
token : null,
|
|
prev : null,
|
|
peeked : null,
|
|
in_function : 0,
|
|
in_loop : 0,
|
|
labels : []
|
|
};
|
|
|
|
S.token = next();
|
|
|
|
function is(type, value) {
|
|
return is_token(S.token, type, value);
|
|
};
|
|
|
|
function peek() { return S.peeked || (S.peeked = S.input()); };
|
|
|
|
function next() {
|
|
S.prev = S.token;
|
|
if (S.peeked) {
|
|
S.token = S.peeked;
|
|
S.peeked = null;
|
|
} else {
|
|
S.token = S.input();
|
|
}
|
|
return S.token;
|
|
};
|
|
|
|
function prev() {
|
|
return S.prev;
|
|
};
|
|
|
|
function croak(msg, line, col, pos) {
|
|
var ctx = S.input.context();
|
|
js_error(msg,
|
|
line != null ? line : ctx.tokline,
|
|
col != null ? col : ctx.tokcol,
|
|
pos != null ? pos : ctx.tokpos);
|
|
};
|
|
|
|
function token_error(token, msg) {
|
|
croak(msg, token.line, token.col);
|
|
};
|
|
|
|
function unexpected(token) {
|
|
if (token == null)
|
|
token = S.token;
|
|
token_error(token, "Unexpected token: " + token.type + " (" + token.value + ")");
|
|
};
|
|
|
|
function expect_token(type, val) {
|
|
if (is(type, val)) {
|
|
return next();
|
|
}
|
|
token_error(S.token, "Unexpected token " + S.token.type + ", expected " + type);
|
|
};
|
|
|
|
function expect(punc) { return expect_token("punc", punc); };
|
|
|
|
function can_insert_semicolon() {
|
|
return !exigent_mode && (
|
|
S.token.nlb || is("eof") || is("punc", "}")
|
|
);
|
|
};
|
|
|
|
function semicolon() {
|
|
if (is("punc", ";")) next();
|
|
else if (!can_insert_semicolon()) unexpected();
|
|
};
|
|
|
|
function as() {
|
|
return slice(arguments);
|
|
};
|
|
|
|
function parenthesised() {
|
|
expect("(");
|
|
var ex = expression();
|
|
expect(")");
|
|
return ex;
|
|
};
|
|
|
|
function add_tokens(str, start, end) {
|
|
return str instanceof NodeWithToken ? str : new NodeWithToken(str, start, end);
|
|
};
|
|
|
|
function maybe_embed_tokens(parser) {
|
|
if (embed_tokens) return function() {
|
|
var start = S.token;
|
|
var ast = parser.apply(this, arguments);
|
|
ast[0] = add_tokens(ast[0], start, prev());
|
|
return ast;
|
|
};
|
|
else return parser;
|
|
};
|
|
|
|
var statement = maybe_embed_tokens(function() {
|
|
if (is("operator", "/")) {
|
|
S.peeked = null;
|
|
S.token = S.input(true); // force regexp
|
|
}
|
|
switch (S.token.type) {
|
|
case "num":
|
|
case "string":
|
|
case "regexp":
|
|
case "operator":
|
|
case "atom":
|
|
return simple_statement();
|
|
|
|
case "name":
|
|
return is_token(peek(), "punc", ":")
|
|
? labeled_statement(prog1(S.token.value, next, next))
|
|
: simple_statement();
|
|
|
|
case "punc":
|
|
switch (S.token.value) {
|
|
case "{":
|
|
return as("block", block_());
|
|
case "[":
|
|
case "(":
|
|
return simple_statement();
|
|
case ";":
|
|
next();
|
|
return as("block");
|
|
default:
|
|
unexpected();
|
|
}
|
|
|
|
case "keyword":
|
|
switch (prog1(S.token.value, next)) {
|
|
case "break":
|
|
return break_cont("break");
|
|
|
|
case "continue":
|
|
return break_cont("continue");
|
|
|
|
case "debugger":
|
|
semicolon();
|
|
return as("debugger");
|
|
|
|
case "do":
|
|
return (function(body){
|
|
expect_token("keyword", "while");
|
|
return as("do", prog1(parenthesised, semicolon), body);
|
|
})(in_loop(statement));
|
|
|
|
case "for":
|
|
return for_();
|
|
|
|
case "function":
|
|
return function_(true);
|
|
|
|
case "if":
|
|
return if_();
|
|
|
|
case "return":
|
|
if (S.in_function == 0)
|
|
croak("'return' outside of function");
|
|
return as("return",
|
|
is("punc", ";")
|
|
? (next(), null)
|
|
: can_insert_semicolon()
|
|
? null
|
|
: prog1(expression, semicolon));
|
|
|
|
case "switch":
|
|
return as("switch", parenthesised(), switch_block_());
|
|
|
|
case "throw":
|
|
return as("throw", prog1(expression, semicolon));
|
|
|
|
case "try":
|
|
return try_();
|
|
|
|
case "var":
|
|
return prog1(var_, semicolon);
|
|
|
|
case "const":
|
|
return prog1(const_, semicolon);
|
|
|
|
case "while":
|
|
return as("while", parenthesised(), in_loop(statement));
|
|
|
|
case "with":
|
|
return as("with", parenthesised(), statement());
|
|
|
|
default:
|
|
unexpected();
|
|
}
|
|
}
|
|
});
|
|
|
|
function labeled_statement(label) {
|
|
S.labels.push(label);
|
|
var start = S.token, stat = statement();
|
|
if (exigent_mode && !HOP(STATEMENTS_WITH_LABELS, stat[0]))
|
|
unexpected(start);
|
|
S.labels.pop();
|
|
return as("label", label, stat);
|
|
};
|
|
|
|
function simple_statement() {
|
|
return as("stat", prog1(expression, semicolon));
|
|
};
|
|
|
|
function break_cont(type) {
|
|
var name;
|
|
if (!can_insert_semicolon()) {
|
|
name = is("name") ? S.token.value : null;
|
|
}
|
|
if (name != null) {
|
|
next();
|
|
if (!member(name, S.labels))
|
|
croak("Label " + name + " without matching loop or statement");
|
|
}
|
|
else if (S.in_loop == 0)
|
|
croak(type + " not inside a loop or switch");
|
|
semicolon();
|
|
return as(type, name);
|
|
};
|
|
|
|
function for_() {
|
|
expect("(");
|
|
var init = null;
|
|
if (!is("punc", ";")) {
|
|
init = is("keyword", "var")
|
|
? (next(), var_(true))
|
|
: expression(true, true);
|
|
if (is("operator", "in"))
|
|
return for_in(init);
|
|
}
|
|
return regular_for(init);
|
|
};
|
|
|
|
function regular_for(init) {
|
|
expect(";");
|
|
var test = is("punc", ";") ? null : expression();
|
|
expect(";");
|
|
var step = is("punc", ")") ? null : expression();
|
|
expect(")");
|
|
return as("for", init, test, step, in_loop(statement));
|
|
};
|
|
|
|
function for_in(init) {
|
|
var lhs = init[0] == "var" ? as("name", init[1][0]) : init;
|
|
next();
|
|
var obj = expression();
|
|
expect(")");
|
|
return as("for-in", init, lhs, obj, in_loop(statement));
|
|
};
|
|
|
|
var function_ = maybe_embed_tokens(function(in_statement) {
|
|
var name = is("name") ? prog1(S.token.value, next) : null;
|
|
if (in_statement && !name)
|
|
unexpected();
|
|
expect("(");
|
|
return as(in_statement ? "defun" : "function",
|
|
name,
|
|
// arguments
|
|
(function(first, a){
|
|
while (!is("punc", ")")) {
|
|
if (first) first = false; else expect(",");
|
|
if (!is("name")) unexpected();
|
|
a.push(S.token.value);
|
|
next();
|
|
}
|
|
next();
|
|
return a;
|
|
})(true, []),
|
|
// body
|
|
(function(){
|
|
++S.in_function;
|
|
var loop = S.in_loop;
|
|
S.in_loop = 0;
|
|
var a = block_();
|
|
--S.in_function;
|
|
S.in_loop = loop;
|
|
return a;
|
|
})());
|
|
});
|
|
|
|
function if_() {
|
|
var cond = parenthesised(), body = statement(), belse;
|
|
if (is("keyword", "else")) {
|
|
next();
|
|
belse = statement();
|
|
}
|
|
return as("if", cond, body, belse);
|
|
};
|
|
|
|
function block_() {
|
|
expect("{");
|
|
var a = [];
|
|
while (!is("punc", "}")) {
|
|
if (is("eof")) unexpected();
|
|
a.push(statement());
|
|
}
|
|
next();
|
|
return a;
|
|
};
|
|
|
|
var switch_block_ = curry(in_loop, function(){
|
|
expect("{");
|
|
var a = [], cur = null;
|
|
while (!is("punc", "}")) {
|
|
if (is("eof")) unexpected();
|
|
if (is("keyword", "case")) {
|
|
next();
|
|
cur = [];
|
|
a.push([ expression(), cur ]);
|
|
expect(":");
|
|
}
|
|
else if (is("keyword", "default")) {
|
|
next();
|
|
expect(":");
|
|
cur = [];
|
|
a.push([ null, cur ]);
|
|
}
|
|
else {
|
|
if (!cur) unexpected();
|
|
cur.push(statement());
|
|
}
|
|
}
|
|
next();
|
|
return a;
|
|
});
|
|
|
|
function try_() {
|
|
var body = block_(), bcatch, bfinally;
|
|
if (is("keyword", "catch")) {
|
|
next();
|
|
expect("(");
|
|
if (!is("name"))
|
|
croak("Name expected");
|
|
var name = S.token.value;
|
|
next();
|
|
expect(")");
|
|
bcatch = [ name, block_() ];
|
|
}
|
|
if (is("keyword", "finally")) {
|
|
next();
|
|
bfinally = block_();
|
|
}
|
|
if (!bcatch && !bfinally)
|
|
croak("Missing catch/finally blocks");
|
|
return as("try", body, bcatch, bfinally);
|
|
};
|
|
|
|
function vardefs(no_in) {
|
|
var a = [];
|
|
for (;;) {
|
|
if (!is("name"))
|
|
unexpected();
|
|
var name = S.token.value;
|
|
next();
|
|
if (is("operator", "=")) {
|
|
next();
|
|
a.push([ name, expression(false, no_in) ]);
|
|
} else {
|
|
a.push([ name ]);
|
|
}
|
|
if (!is("punc", ","))
|
|
break;
|
|
next();
|
|
}
|
|
return a;
|
|
};
|
|
|
|
function var_(no_in) {
|
|
return as("var", vardefs(no_in));
|
|
};
|
|
|
|
function const_() {
|
|
return as("const", vardefs());
|
|
};
|
|
|
|
function new_() {
|
|
var newexp = expr_atom(false), args;
|
|
if (is("punc", "(")) {
|
|
next();
|
|
args = expr_list(")");
|
|
} else {
|
|
args = [];
|
|
}
|
|
return subscripts(as("new", newexp, args), true);
|
|
};
|
|
|
|
var expr_atom = maybe_embed_tokens(function(allow_calls) {
|
|
if (is("operator", "new")) {
|
|
next();
|
|
return new_();
|
|
}
|
|
if (is("operator") && HOP(UNARY_PREFIX, S.token.value)) {
|
|
return make_unary("unary-prefix",
|
|
prog1(S.token.value, next),
|
|
expr_atom(allow_calls));
|
|
}
|
|
if (is("punc")) {
|
|
switch (S.token.value) {
|
|
case "(":
|
|
next();
|
|
return subscripts(prog1(expression, curry(expect, ")")), allow_calls);
|
|
case "[":
|
|
next();
|
|
return subscripts(array_(), allow_calls);
|
|
case "{":
|
|
next();
|
|
return subscripts(object_(), allow_calls);
|
|
}
|
|
unexpected();
|
|
}
|
|
if (is("keyword", "function")) {
|
|
next();
|
|
return subscripts(function_(false), allow_calls);
|
|
}
|
|
if (HOP(ATOMIC_START_TOKEN, S.token.type)) {
|
|
var atom = S.token.type == "regexp"
|
|
? as("regexp", S.token.value[0], S.token.value[1])
|
|
: as(S.token.type, S.token.value);
|
|
return subscripts(prog1(atom, next), allow_calls);
|
|
}
|
|
unexpected();
|
|
});
|
|
|
|
function expr_list(closing, allow_trailing_comma, allow_empty) {
|
|
var first = true, a = [];
|
|
while (!is("punc", closing)) {
|
|
if (first) first = false; else expect(",");
|
|
if (allow_trailing_comma && is("punc", closing)) break;
|
|
if (is("punc", ",") && allow_empty) {
|
|
a.push([ "atom", "undefined" ]);
|
|
} else {
|
|
a.push(expression(false));
|
|
}
|
|
}
|
|
next();
|
|
return a;
|
|
};
|
|
|
|
function array_() {
|
|
return as("array", expr_list("]", !exigent_mode, true));
|
|
};
|
|
|
|
function object_() {
|
|
var first = true, a = [];
|
|
while (!is("punc", "}")) {
|
|
if (first) first = false; else expect(",");
|
|
if (!exigent_mode && is("punc", "}"))
|
|
// allow trailing comma
|
|
break;
|
|
var type = S.token.type;
|
|
var name = as_property_name();
|
|
if (type == "name" && (name == "get" || name == "set") && !is("punc", ":")) {
|
|
a.push([ as_name(), function_(false), name ]);
|
|
} else {
|
|
expect(":");
|
|
a.push([ name, expression(false) ]);
|
|
}
|
|
}
|
|
next();
|
|
return as("object", a);
|
|
};
|
|
|
|
function as_property_name() {
|
|
switch (S.token.type) {
|
|
case "num":
|
|
case "string":
|
|
return prog1(S.token.value, next);
|
|
}
|
|
return as_name();
|
|
};
|
|
|
|
function as_name() {
|
|
switch (S.token.type) {
|
|
case "name":
|
|
case "operator":
|
|
case "keyword":
|
|
case "atom":
|
|
return prog1(S.token.value, next);
|
|
default:
|
|
unexpected();
|
|
}
|
|
};
|
|
|
|
function subscripts(expr, allow_calls) {
|
|
if (is("punc", ".")) {
|
|
next();
|
|
return subscripts(as("dot", expr, as_name()), allow_calls);
|
|
}
|
|
if (is("punc", "[")) {
|
|
next();
|
|
return subscripts(as("sub", expr, prog1(expression, curry(expect, "]"))), allow_calls);
|
|
}
|
|
if (allow_calls && is("punc", "(")) {
|
|
next();
|
|
return subscripts(as("call", expr, expr_list(")")), true);
|
|
}
|
|
if (allow_calls && is("operator") && HOP(UNARY_POSTFIX, S.token.value)) {
|
|
return prog1(curry(make_unary, "unary-postfix", S.token.value, expr),
|
|
next);
|
|
}
|
|
return expr;
|
|
};
|
|
|
|
function make_unary(tag, op, expr) {
|
|
if ((op == "++" || op == "--") && !is_assignable(expr))
|
|
croak("Invalid use of " + op + " operator");
|
|
return as(tag, op, expr);
|
|
};
|
|
|
|
function expr_op(left, min_prec, no_in) {
|
|
var op = is("operator") ? S.token.value : null;
|
|
if (op && op == "in" && no_in) op = null;
|
|
var prec = op != null ? PRECEDENCE[op] : null;
|
|
if (prec != null && prec > min_prec) {
|
|
next();
|
|
var right = expr_op(expr_atom(true), prec, no_in);
|
|
return expr_op(as("binary", op, left, right), min_prec, no_in);
|
|
}
|
|
return left;
|
|
};
|
|
|
|
function expr_ops(no_in) {
|
|
return expr_op(expr_atom(true), 0, no_in);
|
|
};
|
|
|
|
function maybe_conditional(no_in) {
|
|
var expr = expr_ops(no_in);
|
|
if (is("operator", "?")) {
|
|
next();
|
|
var yes = expression(false);
|
|
expect(":");
|
|
return as("conditional", expr, yes, expression(false, no_in));
|
|
}
|
|
return expr;
|
|
};
|
|
|
|
function is_assignable(expr) {
|
|
if (!exigent_mode) return true;
|
|
switch (expr[0]) {
|
|
case "dot":
|
|
case "sub":
|
|
case "new":
|
|
case "call":
|
|
return true;
|
|
case "name":
|
|
return expr[1] != "this";
|
|
}
|
|
};
|
|
|
|
function maybe_assign(no_in) {
|
|
var left = maybe_conditional(no_in), val = S.token.value;
|
|
if (is("operator") && HOP(ASSIGNMENT, val)) {
|
|
if (is_assignable(left)) {
|
|
next();
|
|
return as("assign", ASSIGNMENT[val], left, maybe_assign(no_in));
|
|
}
|
|
croak("Invalid assignment");
|
|
}
|
|
return left;
|
|
};
|
|
|
|
var expression = maybe_embed_tokens(function(commas, no_in) {
|
|
if (arguments.length == 0)
|
|
commas = true;
|
|
var expr = maybe_assign(no_in);
|
|
if (commas && is("punc", ",")) {
|
|
next();
|
|
return as("seq", expr, expression(true, no_in));
|
|
}
|
|
return expr;
|
|
});
|
|
|
|
function in_loop(cont) {
|
|
try {
|
|
++S.in_loop;
|
|
return cont();
|
|
} finally {
|
|
--S.in_loop;
|
|
}
|
|
};
|
|
|
|
return as("toplevel", (function(a){
|
|
while (!is("eof"))
|
|
a.push(statement());
|
|
return a;
|
|
})([]));
|
|
|
|
};
|
|
|
|
/* -----[ Utilities ]----- */
|
|
|
|
function curry(f) {
|
|
var args = slice(arguments, 1);
|
|
return function() { return f.apply(this, args.concat(slice(arguments))); };
|
|
};
|
|
|
|
function prog1(ret) {
|
|
if (ret instanceof Function)
|
|
ret = ret();
|
|
for (var i = 1, n = arguments.length; --n > 0; ++i)
|
|
arguments[i]();
|
|
return ret;
|
|
};
|
|
|
|
function array_to_hash(a) {
|
|
var ret = {};
|
|
for (var i = 0; i < a.length; ++i)
|
|
ret[a[i]] = true;
|
|
return ret;
|
|
};
|
|
|
|
function slice(a, start) {
|
|
return Array.prototype.slice.call(a, start || 0);
|
|
};
|
|
|
|
function characters(str) {
|
|
return str.split("");
|
|
};
|
|
|
|
function member(name, array) {
|
|
for (var i = array.length; --i >= 0;)
|
|
if (array[i] === name)
|
|
return true;
|
|
return false;
|
|
};
|
|
|
|
function HOP(obj, prop) {
|
|
return Object.prototype.hasOwnProperty.call(obj, prop);
|
|
};
|
|
|
|
/* -----[ helper for AST traversal ]----- */
|
|
|
|
function ast_walker() {
|
|
function _vardefs(defs) {
|
|
return [ this[0], MAP(defs, function(def){
|
|
var a = [ def[0] ];
|
|
if (def.length > 1)
|
|
a[1] = walk(def[1]);
|
|
return a;
|
|
}) ];
|
|
};
|
|
function _block(statements) {
|
|
var out = [ this[0] ];
|
|
if (statements != null)
|
|
out.push(MAP(statements, walk));
|
|
return out;
|
|
};
|
|
var walkers = {
|
|
"string": function(str) {
|
|
return [ this[0], str ];
|
|
},
|
|
"num": function(num) {
|
|
return [ this[0], num ];
|
|
},
|
|
"name": function(name) {
|
|
return [ this[0], name ];
|
|
},
|
|
"toplevel": function(statements) {
|
|
return [ this[0], MAP(statements, walk) ];
|
|
},
|
|
"block": _block,
|
|
"splice": _block,
|
|
"var": _vardefs,
|
|
"const": _vardefs,
|
|
"try": function(t, c, f) {
|
|
return [
|
|
this[0],
|
|
MAP(t, walk),
|
|
c != null ? [ c[0], MAP(c[1], walk) ] : null,
|
|
f != null ? MAP(f, walk) : null
|
|
];
|
|
},
|
|
"throw": function(expr) {
|
|
return [ this[0], walk(expr) ];
|
|
},
|
|
"new": function(ctor, args) {
|
|
return [ this[0], walk(ctor), MAP(args, walk) ];
|
|
},
|
|
"switch": function(expr, body) {
|
|
return [ this[0], walk(expr), MAP(body, function(branch){
|
|
return [ branch[0] ? walk(branch[0]) : null,
|
|
MAP(branch[1], walk) ];
|
|
}) ];
|
|
},
|
|
"break": function(label) {
|
|
return [ this[0], label ];
|
|
},
|
|
"continue": function(label) {
|
|
return [ this[0], label ];
|
|
},
|
|
"conditional": function(cond, t, e) {
|
|
return [ this[0], walk(cond), walk(t), walk(e) ];
|
|
},
|
|
"assign": function(op, lvalue, rvalue) {
|
|
return [ this[0], op, walk(lvalue), walk(rvalue) ];
|
|
},
|
|
"dot": function(expr) {
|
|
return [ this[0], walk(expr) ].concat(slice(arguments, 1));
|
|
},
|
|
"call": function(expr, args) {
|
|
return [ this[0], walk(expr), MAP(args, walk) ];
|
|
},
|
|
"function": function(name, args, body) {
|
|
return [ this[0], name, args.slice(), MAP(body, walk) ];
|
|
},
|
|
"defun": function(name, args, body) {
|
|
return [ this[0], name, args.slice(), MAP(body, walk) ];
|
|
},
|
|
"if": function(conditional, t, e) {
|
|
return [ this[0], walk(conditional), walk(t), walk(e) ];
|
|
},
|
|
"for": function(init, cond, step, block) {
|
|
return [ this[0], walk(init), walk(cond), walk(step), walk(block) ];
|
|
},
|
|
"for-in": function(vvar, key, hash, block) {
|
|
return [ this[0], walk(vvar), walk(key), walk(hash), walk(block) ];
|
|
},
|
|
"while": function(cond, block) {
|
|
return [ this[0], walk(cond), walk(block) ];
|
|
},
|
|
"do": function(cond, block) {
|
|
return [ this[0], walk(cond), walk(block) ];
|
|
},
|
|
"return": function(expr) {
|
|
return [ this[0], walk(expr) ];
|
|
},
|
|
"binary": function(op, left, right) {
|
|
return [ this[0], op, walk(left), walk(right) ];
|
|
},
|
|
"unary-prefix": function(op, expr) {
|
|
return [ this[0], op, walk(expr) ];
|
|
},
|
|
"unary-postfix": function(op, expr) {
|
|
return [ this[0], op, walk(expr) ];
|
|
},
|
|
"sub": function(expr, subscript) {
|
|
return [ this[0], walk(expr), walk(subscript) ];
|
|
},
|
|
"object": function(props) {
|
|
return [ this[0], MAP(props, function(p){
|
|
return p.length == 2
|
|
? [ p[0], walk(p[1]) ]
|
|
: [ p[0], walk(p[1]), p[2] ]; // get/set-ter
|
|
}) ];
|
|
},
|
|
"regexp": function(rx, mods) {
|
|
return [ this[0], rx, mods ];
|
|
},
|
|
"array": function(elements) {
|
|
return [ this[0], MAP(elements, walk) ];
|
|
},
|
|
"stat": function(stat) {
|
|
return [ this[0], walk(stat) ];
|
|
},
|
|
"seq": function() {
|
|
return [ this[0] ].concat(MAP(slice(arguments), walk));
|
|
},
|
|
"label": function(name, block) {
|
|
return [ this[0], name, walk(block) ];
|
|
},
|
|
"with": function(expr, block) {
|
|
return [ this[0], walk(expr), walk(block) ];
|
|
},
|
|
"atom": function(name) {
|
|
return [ this[0], name ];
|
|
}
|
|
};
|
|
|
|
var user = {};
|
|
var stack = [];
|
|
function walk(ast) {
|
|
if (ast == null)
|
|
return null;
|
|
try {
|
|
stack.push(ast);
|
|
var type = ast[0];
|
|
var gen = user[type];
|
|
if (gen) {
|
|
var ret = gen.apply(ast, ast.slice(1));
|
|
if (ret != null)
|
|
return ret;
|
|
}
|
|
gen = walkers[type];
|
|
return gen.apply(ast, ast.slice(1));
|
|
} finally {
|
|
stack.pop();
|
|
}
|
|
};
|
|
|
|
function with_walkers(walkers, cont){
|
|
var save = {}, i;
|
|
for (i in walkers) if (HOP(walkers, i)) {
|
|
save[i] = user[i];
|
|
user[i] = walkers[i];
|
|
}
|
|
var ret = cont();
|
|
for (i in save) if (HOP(save, i)) {
|
|
if (!save[i]) delete user[i];
|
|
else user[i] = save[i];
|
|
}
|
|
return ret;
|
|
};
|
|
|
|
return {
|
|
walk: walk,
|
|
with_walkers: with_walkers,
|
|
parent: function() {
|
|
return stack[stack.length - 2]; // last one is current node
|
|
},
|
|
stack: function() {
|
|
return stack;
|
|
}
|
|
};
|
|
};
|
|
|
|
function empty(b) {
|
|
return !b || (b[0] == "block" && (!b[1] || b[1].length == 0));
|
|
};
|
|
|
|
/* -----[ re-generate code from the AST ]----- */
|
|
|
|
var DOT_CALL_NO_PARENS = array_to_hash([
|
|
"name",
|
|
"array",
|
|
"object",
|
|
"string",
|
|
"dot",
|
|
"sub",
|
|
"call",
|
|
"regexp"
|
|
]);
|
|
|
|
function make_string(str, ascii_only) {
|
|
var dq = 0, sq = 0;
|
|
str = str.replace(/[\\\b\f\n\r\t\x22\x27\u2028\u2029]/g, function(s){
|
|
switch (s) {
|
|
case "\\": return "\\\\";
|
|
case "\b": return "\\b";
|
|
case "\f": return "\\f";
|
|
case "\n": return "\\n";
|
|
case "\r": return "\\r";
|
|
case "\t": return "\\t";
|
|
case "\u2028": return "\\u2028";
|
|
case "\u2029": return "\\u2029";
|
|
case '"': ++dq; return '"';
|
|
case "'": ++sq; return "'";
|
|
}
|
|
return s;
|
|
});
|
|
if (ascii_only) str = to_ascii(str);
|
|
if (dq > sq) return "'" + str.replace(/\x27/g, "\\'") + "'";
|
|
else return '"' + str.replace(/\x22/g, '\\"') + '"';
|
|
};
|
|
|
|
function to_ascii(str) {
|
|
return str.replace(/[\u0080-\uffff]/g, function(ch) {
|
|
var code = ch.charCodeAt(0).toString(16);
|
|
while (code.length < 4) code = "0" + code;
|
|
return "\\u" + code;
|
|
});
|
|
};
|
|
|
|
var SPLICE_NEEDS_BRACKETS = array_to_hash([ "if", "while", "do", "for", "for-in", "with" ]);
|
|
|
|
function gen_code(ast, options) {
|
|
options = defaults(options, {
|
|
indent_start : 0,
|
|
indent_level : 4,
|
|
quote_keys : false,
|
|
space_colon : false,
|
|
beautify : false,
|
|
ascii_only : false,
|
|
inline_script: false
|
|
});
|
|
var beautify = !!options.beautify;
|
|
var indentation = 0,
|
|
newline = beautify ? "\n" : "",
|
|
space = beautify ? " " : "";
|
|
|
|
function encode_string(str) {
|
|
var ret = make_string(str, options.ascii_only);
|
|
if (options.inline_script)
|
|
ret = ret.replace(/<\x2fscript([>/\t\n\f\r ])/gi, "<\\/script$1");
|
|
return ret;
|
|
};
|
|
|
|
function make_name(name) {
|
|
name = name.toString();
|
|
if (options.ascii_only)
|
|
name = to_ascii(name);
|
|
return name;
|
|
};
|
|
|
|
function indent(line) {
|
|
if (line == null)
|
|
line = "";
|
|
if (beautify)
|
|
line = repeat_string(" ", options.indent_start + indentation * options.indent_level) + line;
|
|
return line;
|
|
};
|
|
|
|
function with_indent(cont, incr) {
|
|
if (incr == null) incr = 1;
|
|
indentation += incr;
|
|
try { return cont.apply(null, slice(arguments, 1)); }
|
|
finally { indentation -= incr; }
|
|
};
|
|
|
|
function add_spaces(a) {
|
|
if (beautify)
|
|
return a.join(" ");
|
|
var b = [];
|
|
for (var i = 0; i < a.length; ++i) {
|
|
var next = a[i + 1];
|
|
b.push(a[i]);
|
|
if (next &&
|
|
((/[a-z0-9_\x24]$/i.test(a[i].toString()) && /^[a-z0-9_\x24]/i.test(next.toString())) ||
|
|
(/[\+\-]$/.test(a[i].toString()) && /^[\+\-]/.test(next.toString())))) {
|
|
b.push(" ");
|
|
}
|
|
}
|
|
return b.join("");
|
|
};
|
|
|
|
function add_commas(a) {
|
|
return a.join("," + space);
|
|
};
|
|
|
|
function parenthesize(expr) {
|
|
var gen = make(expr);
|
|
for (var i = 1; i < arguments.length; ++i) {
|
|
var el = arguments[i];
|
|
if ((el instanceof Function && el(expr)) || expr[0] == el)
|
|
return "(" + gen + ")";
|
|
}
|
|
return gen;
|
|
};
|
|
|
|
function best_of(a) {
|
|
if (a.length == 1) {
|
|
return a[0];
|
|
}
|
|
if (a.length == 2) {
|
|
var b = a[1];
|
|
a = a[0];
|
|
return a.length <= b.length ? a : b;
|
|
}
|
|
return best_of([ a[0], best_of(a.slice(1)) ]);
|
|
};
|
|
|
|
function needs_parens(expr) {
|
|
if (expr[0] == "function" || expr[0] == "object") {
|
|
// dot/call on a literal function requires the
|
|
// function literal itself to be parenthesized
|
|
// only if it's the first "thing" in a
|
|
// statement. This means that the parent is
|
|
// "stat", but it could also be a "seq" and
|
|
// we're the first in this "seq" and the
|
|
// parent is "stat", and so on. Messy stuff,
|
|
// but it worths the trouble.
|
|
var a = slice($stack), self = a.pop(), p = a.pop();
|
|
while (p) {
|
|
if (p[0] == "stat") return true;
|
|
if (((p[0] == "seq" || p[0] == "call" || p[0] == "dot" || p[0] == "sub" || p[0] == "conditional") && p[1] === self) ||
|
|
((p[0] == "binary" || p[0] == "assign" || p[0] == "unary-postfix") && p[2] === self)) {
|
|
self = p;
|
|
p = a.pop();
|
|
} else {
|
|
return false;
|
|
}
|
|
}
|
|
}
|
|
return !HOP(DOT_CALL_NO_PARENS, expr[0]);
|
|
};
|
|
|
|
function make_num(num) {
|
|
var str = num.toString(10), a = [ str.replace(/^0\./, ".") ], m;
|
|
if (Math.floor(num) === num) {
|
|
a.push("0x" + num.toString(16).toLowerCase(), // probably pointless
|
|
"0" + num.toString(8)); // same.
|
|
if ((m = /^(.*?)(0+)$/.exec(num))) {
|
|
a.push(m[1] + "e" + m[2].length);
|
|
}
|
|
} else if ((m = /^0?\.(0+)(.*)$/.exec(num))) {
|
|
a.push(m[2] + "e-" + (m[1].length + m[2].length),
|
|
str.substr(str.indexOf(".")));
|
|
}
|
|
return best_of(a);
|
|
};
|
|
|
|
var generators = {
|
|
"string": encode_string,
|
|
"num": make_num,
|
|
"name": make_name,
|
|
"toplevel": function(statements) {
|
|
return make_block_statements(statements)
|
|
.join(newline);
|
|
},
|
|
"splice": function(statements) {
|
|
var parent = $stack[$stack.length - 2][0];
|
|
if (HOP(SPLICE_NEEDS_BRACKETS, parent)) {
|
|
// we need block brackets in this case
|
|
return make_block.apply(this, arguments);
|
|
} else {
|
|
return MAP(make_block_statements(statements, true),
|
|
function(line, i) {
|
|
// the first line is already indented
|
|
return i > 0 ? indent(line) : line;
|
|
}).join(newline);
|
|
}
|
|
},
|
|
"block": make_block,
|
|
"var": function(defs) {
|
|
return "var " + add_commas(MAP(defs, make_1vardef)) + ";";
|
|
},
|
|
"const": function(defs) {
|
|
return "const " + add_commas(MAP(defs, make_1vardef)) + ";";
|
|
},
|
|
"try": function(tr, ca, fi) {
|
|
var out = [ "try", make_block(tr) ];
|
|
if (ca) out.push("catch", "(" + ca[0] + ")", make_block(ca[1]));
|
|
if (fi) out.push("finally", make_block(fi));
|
|
return add_spaces(out);
|
|
},
|
|
"throw": function(expr) {
|
|
return add_spaces([ "throw", make(expr) ]) + ";";
|
|
},
|
|
"new": function(ctor, args) {
|
|
args = args.length > 0 ? "(" + add_commas(MAP(args, make)) + ")" : "";
|
|
return add_spaces([ "new", parenthesize(ctor, "seq", "binary", "conditional", "assign", function(expr){
|
|
var w = ast_walker(), has_call = {};
|
|
try {
|
|
w.with_walkers({
|
|
"call": function() { throw has_call },
|
|
"function": function() { return this }
|
|
}, function(){
|
|
w.walk(expr);
|
|
});
|
|
} catch(ex) {
|
|
if (ex === has_call)
|
|
return true;
|
|
throw ex;
|
|
}
|
|
}) + args ]);
|
|
},
|
|
"switch": function(expr, body) {
|
|
return add_spaces([ "switch", "(" + make(expr) + ")", make_switch_block(body) ]);
|
|
},
|
|
"break": function(label) {
|
|
var out = "break";
|
|
if (label != null)
|
|
out += " " + make_name(label);
|
|
return out + ";";
|
|
},
|
|
"continue": function(label) {
|
|
var out = "continue";
|
|
if (label != null)
|
|
out += " " + make_name(label);
|
|
return out + ";";
|
|
},
|
|
"conditional": function(co, th, el) {
|
|
return add_spaces([ parenthesize(co, "assign", "seq", "conditional"), "?",
|
|
parenthesize(th, "seq"), ":",
|
|
parenthesize(el, "seq") ]);
|
|
},
|
|
"assign": function(op, lvalue, rvalue) {
|
|
if (op && op !== true) op += "=";
|
|
else op = "=";
|
|
return add_spaces([ make(lvalue), op, parenthesize(rvalue, "seq") ]);
|
|
},
|
|
"dot": function(expr) {
|
|
var out = make(expr), i = 1;
|
|
if (expr[0] == "num") {
|
|
if (!/\./.test(expr[1]))
|
|
out += ".";
|
|
} else if (needs_parens(expr))
|
|
out = "(" + out + ")";
|
|
while (i < arguments.length)
|
|
out += "." + make_name(arguments[i++]);
|
|
return out;
|
|
},
|
|
"call": function(func, args) {
|
|
var f = make(func);
|
|
if (needs_parens(func))
|
|
f = "(" + f + ")";
|
|
return f + "(" + add_commas(MAP(args, function(expr){
|
|
return parenthesize(expr, "seq");
|
|
})) + ")";
|
|
},
|
|
"function": make_function,
|
|
"defun": make_function,
|
|
"if": function(co, th, el) {
|
|
var out = [ "if", "(" + make(co) + ")", el ? make_then(th) : make(th) ];
|
|
if (el) {
|
|
out.push("else", make(el));
|
|
}
|
|
return add_spaces(out);
|
|
},
|
|
"for": function(init, cond, step, block) {
|
|
var out = [ "for" ];
|
|
init = (init != null ? make(init) : "").replace(/;*\s*$/, ";" + space);
|
|
cond = (cond != null ? make(cond) : "").replace(/;*\s*$/, ";" + space);
|
|
step = (step != null ? make(step) : "").replace(/;*\s*$/, "");
|
|
var args = init + cond + step;
|
|
if (args == "; ; ") args = ";;";
|
|
out.push("(" + args + ")", make(block));
|
|
return add_spaces(out);
|
|
},
|
|
"for-in": function(vvar, key, hash, block) {
|
|
return add_spaces([ "for", "(" +
|
|
(vvar ? make(vvar).replace(/;+$/, "") : make(key)),
|
|
"in",
|
|
make(hash) + ")", make(block) ]);
|
|
},
|
|
"while": function(condition, block) {
|
|
return add_spaces([ "while", "(" + make(condition) + ")", make(block) ]);
|
|
},
|
|
"do": function(condition, block) {
|
|
return add_spaces([ "do", make(block), "while", "(" + make(condition) + ")" ]) + ";";
|
|
},
|
|
"return": function(expr) {
|
|
var out = [ "return" ];
|
|
if (expr != null) out.push(make(expr));
|
|
return add_spaces(out) + ";";
|
|
},
|
|
"binary": function(operator, lvalue, rvalue) {
|
|
var left = make(lvalue), right = make(rvalue);
|
|
// XXX: I'm pretty sure other cases will bite here.
|
|
// we need to be smarter.
|
|
// adding parens all the time is the safest bet.
|
|
if (member(lvalue[0], [ "assign", "conditional", "seq" ]) ||
|
|
lvalue[0] == "binary" && PRECEDENCE[operator] > PRECEDENCE[lvalue[1]]) {
|
|
left = "(" + left + ")";
|
|
}
|
|
if (member(rvalue[0], [ "assign", "conditional", "seq" ]) ||
|
|
rvalue[0] == "binary" && PRECEDENCE[operator] >= PRECEDENCE[rvalue[1]] &&
|
|
!(rvalue[1] == operator && member(operator, [ "&&", "||", "*" ]))) {
|
|
right = "(" + right + ")";
|
|
}
|
|
else if (!beautify && options.inline_script && (operator == "<" || operator == "<<")
|
|
&& rvalue[0] == "regexp" && /^script/i.test(rvalue[1])) {
|
|
right = " " + right;
|
|
}
|
|
return add_spaces([ left, operator, right ]);
|
|
},
|
|
"unary-prefix": function(operator, expr) {
|
|
var val = make(expr);
|
|
if (!(expr[0] == "num" || (expr[0] == "unary-prefix" && !HOP(OPERATORS, operator + expr[1])) || !needs_parens(expr)))
|
|
val = "(" + val + ")";
|
|
return operator + (is_alphanumeric_char(operator.charAt(0)) ? " " : "") + val;
|
|
},
|
|
"unary-postfix": function(operator, expr) {
|
|
var val = make(expr);
|
|
if (!(expr[0] == "num" || (expr[0] == "unary-postfix" && !HOP(OPERATORS, operator + expr[1])) || !needs_parens(expr)))
|
|
val = "(" + val + ")";
|
|
return val + operator;
|
|
},
|
|
"sub": function(expr, subscript) {
|
|
var hash = make(expr);
|
|
if (needs_parens(expr))
|
|
hash = "(" + hash + ")";
|
|
return hash + "[" + make(subscript) + "]";
|
|
},
|
|
"object": function(props) {
|
|
if (props.length == 0)
|
|
return "{}";
|
|
return "{" + newline + with_indent(function(){
|
|
return MAP(props, function(p){
|
|
if (p.length == 3) {
|
|
// getter/setter. The name is in p[0], the arg.list in p[1][2], the
|
|
// body in p[1][3] and type ("get" / "set") in p[2].
|
|
return indent(make_function(p[0], p[1][2], p[1][3], p[2]));
|
|
}
|
|
var key = p[0], val = parenthesize(p[1], "seq");
|
|
if (options.quote_keys) {
|
|
key = encode_string(key);
|
|
} else if ((typeof key == "number" || !beautify && +key + "" == key)
|
|
&& parseFloat(key) >= 0) {
|
|
key = make_num(+key);
|
|
} else if (!is_identifier(key)) {
|
|
key = encode_string(key);
|
|
}
|
|
return indent(add_spaces(beautify && options.space_colon
|
|
? [ key, ":", val ]
|
|
: [ key + ":", val ]));
|
|
}).join("," + newline);
|
|
}) + newline + indent("}");
|
|
},
|
|
"regexp": function(rx, mods) {
|
|
return "/" + rx + "/" + mods;
|
|
},
|
|
"array": function(elements) {
|
|
if (elements.length == 0) return "[]";
|
|
return add_spaces([ "[", add_commas(MAP(elements, function(el){
|
|
if (!beautify && el[0] == "atom" && el[1] == "undefined") return "";
|
|
return parenthesize(el, "seq");
|
|
})), "]" ]);
|
|
},
|
|
"stat": function(stmt) {
|
|
return make(stmt).replace(/;*\s*$/, ";");
|
|
},
|
|
"seq": function() {
|
|
return add_commas(MAP(slice(arguments), make));
|
|
},
|
|
"label": function(name, block) {
|
|
return add_spaces([ make_name(name), ":", make(block) ]);
|
|
},
|
|
"with": function(expr, block) {
|
|
return add_spaces([ "with", "(" + make(expr) + ")", make(block) ]);
|
|
},
|
|
"atom": function(name) {
|
|
return make_name(name);
|
|
}
|
|
};
|
|
|
|
// The squeezer replaces "block"-s that contain only a single
|
|
// statement with the statement itself; technically, the AST
|
|
// is correct, but this can create problems when we output an
|
|
// IF having an ELSE clause where the THEN clause ends in an
|
|
// IF *without* an ELSE block (then the outer ELSE would refer
|
|
// to the inner IF). This function checks for this case and
|
|
// adds the block brackets if needed.
|
|
function make_then(th) {
|
|
if (th[0] == "do") {
|
|
// https://github.com/mishoo/UglifyJS/issues/#issue/57
|
|
// IE croaks with "syntax error" on code like this:
|
|
// if (foo) do ... while(cond); else ...
|
|
// we need block brackets around do/while
|
|
return make([ "block", [ th ]]);
|
|
}
|
|
var b = th;
|
|
while (true) {
|
|
var type = b[0];
|
|
if (type == "if") {
|
|
if (!b[3])
|
|
// no else, we must add the block
|
|
return make([ "block", [ th ]]);
|
|
b = b[3];
|
|
}
|
|
else if (type == "while" || type == "do") b = b[2];
|
|
else if (type == "for" || type == "for-in") b = b[4];
|
|
else break;
|
|
}
|
|
return make(th);
|
|
};
|
|
|
|
function make_function(name, args, body, keyword) {
|
|
var out = keyword || "function";
|
|
if (name) {
|
|
out += " " + make_name(name);
|
|
}
|
|
out += "(" + add_commas(MAP(args, make_name)) + ")";
|
|
return add_spaces([ out, make_block(body) ]);
|
|
};
|
|
|
|
function make_block_statements(statements, noindent) {
|
|
for (var a = [], last = statements.length - 1, i = 0; i <= last; ++i) {
|
|
var stat = statements[i];
|
|
var code = make(stat);
|
|
if (code != ";") {
|
|
if (!beautify && i == last) {
|
|
if ((stat[0] == "while" && empty(stat[2])) ||
|
|
(member(stat[0], [ "for", "for-in"] ) && empty(stat[4])) ||
|
|
(stat[0] == "if" && empty(stat[2]) && !stat[3]) ||
|
|
(stat[0] == "if" && stat[3] && empty(stat[3]))) {
|
|
code = code.replace(/;*\s*$/, ";");
|
|
} else {
|
|
code = code.replace(/;+\s*$/, "");
|
|
}
|
|
}
|
|
a.push(code);
|
|
}
|
|
}
|
|
return noindent ? a : MAP(a, indent);
|
|
};
|
|
|
|
function make_switch_block(body) {
|
|
var n = body.length;
|
|
if (n == 0) return "{}";
|
|
return "{" + newline + MAP(body, function(branch, i){
|
|
var has_body = branch[1].length > 0, code = with_indent(function(){
|
|
return indent(branch[0]
|
|
? add_spaces([ "case", make(branch[0]) + ":" ])
|
|
: "default:");
|
|
}, 0.5) + (has_body ? newline + with_indent(function(){
|
|
return make_block_statements(branch[1]).join(newline);
|
|
}) : "");
|
|
if (!beautify && has_body && i < n - 1)
|
|
code += ";";
|
|
return code;
|
|
}).join(newline) + newline + indent("}");
|
|
};
|
|
|
|
function make_block(statements) {
|
|
if (!statements) return ";";
|
|
if (statements.length == 0) return "{}";
|
|
return "{" + newline + with_indent(function(){
|
|
return make_block_statements(statements).join(newline);
|
|
}) + newline + indent("}");
|
|
};
|
|
|
|
function make_1vardef(def) {
|
|
var name = def[0], val = def[1];
|
|
if (val != null)
|
|
name = add_spaces([ make_name(name), "=", parenthesize(val, "seq") ]);
|
|
return name;
|
|
};
|
|
|
|
var $stack = [];
|
|
|
|
function make(node) {
|
|
var type = node[0];
|
|
var gen = generators[type];
|
|
if (!gen)
|
|
throw new Error("Can't find generator for \"" + type + "\"");
|
|
$stack.push(node);
|
|
var ret = gen.apply(type, node.slice(1));
|
|
$stack.pop();
|
|
return ret;
|
|
};
|
|
|
|
return make(ast);
|
|
};
|
|
|
|
/* -----[ Utilities ]----- */
|
|
|
|
function repeat_string(str, i) {
|
|
return i < 1 ? "" : new Array(i + 1).join(str);
|
|
};
|
|
|
|
function defaults(args, defs) {
|
|
var ret = {};
|
|
if (args === true)
|
|
args = {};
|
|
for (var i in defs) if (HOP(defs, i)) {
|
|
ret[i] = (args && HOP(args, i)) ? args[i] : defs[i];
|
|
}
|
|
return ret;
|
|
};
|
|
|
|
function is_identifier(name) {
|
|
return /^[a-z_$][a-z0-9_$]*$/i.test(name)
|
|
&& name != "this"
|
|
&& !HOP(KEYWORDS_ATOM, name)
|
|
&& !HOP(RESERVED_WORDS, name)
|
|
&& !HOP(KEYWORDS, name);
|
|
};
|
|
|
|
function MAP(a, f, o) {
|
|
var ret = [];
|
|
for (var i = 0; i < a.length; ++i) {
|
|
ret.push(f.call(o, a[i], i));
|
|
}
|
|
return ret;
|
|
};
|
|
|
|
/* -----[ Exports ]----- */
|
|
|
|
return {
|
|
parse: parse,
|
|
gen_code: gen_code,
|
|
tokenizer: tokenizer,
|
|
ast_walker: ast_walker
|
|
};
|
|
|
|
};
|