1 //<script type="text/javascript">
4 XObject = imports.XObject.XObject;
5 console = imports.console.console;
8 Token = imports.Token.Token;
9 Lang = imports.Lang.Lang;
12 @class Search a {@link JSDOC.TextStream} for language tokens.
14 TokenReader = XObject.define(
18 this.keepWhite = false;
19 this.keepComments = false;
20 this.sepIdents = false; // seperate '.' in identifiers..
21 XObject.extend(this, o || {});
26 collapseWhite : false, // only reduces white space...
30 * @return {Array} of tokens
32 * ts = new TextStream(File.read(str));
33 * tr = TokenReader({ keepComments : true, keepWhite : true });
40 tokenize : function(/**JSDOC.TextStream*/stream) {
43 /**@ignore*/ tokens.last = function() { return tokens[tokens.length-1]; }
44 /**@ignore*/ tokens.lastSym = function() {
45 for (var i = tokens.length-1; i >= 0; i--) {
46 if (!(tokens[i].is("WHIT") || tokens[i].is("COMM"))) return tokens[i];
50 while (!stream.look().eof) {
51 if (this.read_mlcomment(stream, tokens)) continue;
52 if (this.read_slcomment(stream, tokens)) continue;
53 if (this.read_dbquote(stream, tokens)) continue;
54 if (this.read_snquote(stream, tokens)) continue;
55 if (this.read_regx(stream, tokens)) continue;
56 if (this.read_numb(stream, tokens)) continue;
57 if (this.read_punc(stream, tokens)) continue;
58 if (this.read_newline(stream, tokens)) continue;
59 if (this.read_space(stream, tokens)) continue;
60 if (this.read_word(stream, tokens)) continue;
62 // if execution reaches here then an error has happened
63 tokens.push(new Token(stream.next(), "TOKN", "UNKNOWN_TOKEN", this.line));
72 @returns {Boolean} Was the token found?
74 read_word : function(/**JSDOC.TokenStream*/stream, tokens) {
76 while (!stream.look().eof && Lang.isWordChar(stream.look())) {
77 found += stream.next();
85 if ((name = Lang.keyword(found))) {
86 tokens.push(new Token(found, "KEYW", name, this.line));
89 if (!this.sepIdents || found.indexOf('.') < 0 ) {
90 tokens.push(new Token(found, "NAME", "NAME", this.line));
93 var n = found.split('.');
95 n.forEach(function(nm) {
97 tokens.push(new Token('.', "PUNC", "DOT", this.line));
100 tokens.push(new Token(nm, "NAME", "NAME", this.line));
108 @returns {Boolean} Was the token found?
110 read_punc : function(/**JSDOC.TokenStream*/stream, tokens) {
113 while (!stream.look().eof && Lang.punc(found+stream.look())) {
114 found += stream.next();
121 tokens.push(new Token(found, "PUNC", Lang.punc(found), this.line));
127 @returns {Boolean} Was the token found?
129 read_space : function(/**JSDOC.TokenStream*/stream, tokens) {
132 while (!stream.look().eof && Lang.isSpace(stream.look())) {
133 found += stream.next();
140 if (this.collapseWhite) found = " ";
141 if (this.keepWhite) tokens.push(new Token(found, "WHIT", "SPACE", this.line));
147 @returns {Boolean} Was the token found?
149 read_newline : function(/**JSDOC.TokenStream*/stream, tokens) {
152 while (!stream.look().eof && Lang.isNewline(stream.look())) {
154 found += stream.next();
161 if (this.collapseWhite) {
164 if (this.keepWhite) {
165 var last = tokens.pop();
166 if (last.name != "WHIT") {
171 tokens.push(new Token(found, "WHIT", "NEWLINE", this.line));
178 @returns {Boolean} Was the token found?
180 read_mlcomment : function(/**JSDOC.TokenStream*/stream, tokens) {
181 if (stream.look() == "/" && stream.look(1) == "*") {
182 var found = stream.next(2);
184 while (!stream.look().eof && !(stream.look(-1) == "/" && stream.look(-2) == "*")) {
186 if (c == "\n") this.line++;
190 // to start doclet we allow /** or /*** but not /**/ or /****
191 if (/^\/\*\*([^\/]|\*[^*])/.test(found) && this.keepDocs) tokens.push(new Token(found, "COMM", "JSDOC", this.line));
192 else if (this.keepComments) tokens.push(new Token(found, "COMM", "MULTI_LINE_COMM", this.line));
199 @returns {Boolean} Was the token found?
201 read_slcomment : function(/**JSDOC.TokenStream*/stream, tokens) {
204 (stream.look() == "/" && stream.look(1) == "/" && (found=stream.next(2)))
206 (stream.look() == "<" && stream.look(1) == "!" && stream.look(2) == "-" && stream.look(3) == "-" && (found=stream.next(4)))
209 while (!stream.look().eof && !Lang.isNewline(stream.look())) {
210 found += stream.next();
213 if (this.keepComments) {
214 tokens.push(new Token(found, "COMM", "SINGLE_LINE_COMM", this.line));
223 @returns {Boolean} Was the token found?
225 read_dbquote : function(/**JSDOC.TokenStream*/stream, tokens) {
226 if (stream.look() == "\"") {
228 var string = stream.next();
230 while (!stream.look().eof) {
231 if (stream.look() == "\\") {
232 if (Lang.isNewline(stream.look(1))) {
235 } while (!stream.look().eof && Lang.isNewline(stream.look()));
239 string += stream.next(2);
242 else if (stream.look() == "\"") {
243 string += stream.next();
244 tokens.push(new Token(string, "STRN", "DOUBLE_QUOTE", this.line));
248 string += stream.next();
252 return false; // error! unterminated string
256 @returns {Boolean} Was the token found?
258 read_snquote : function(/**JSDOC.TokenStream*/stream, tokens) {
259 if (stream.look() == "'") {
261 var string = stream.next();
263 while (!stream.look().eof) {
264 if (stream.look() == "\\") { // escape sequence
265 string += stream.next(2);
267 else if (stream.look() == "'") {
268 string += stream.next();
269 tokens.push(new Token(string, "STRN", "SINGLE_QUOTE", this.line));
273 string += stream.next();
277 return false; // error! unterminated string
281 @returns {Boolean} Was the token found?
283 read_numb : function(/**JSDOC.TokenStream*/stream, tokens) {
284 if (stream.look() === "0" && stream.look(1) == "x") {
285 return this.read_hex(stream, tokens);
290 while (!stream.look().eof && Lang.isNumber(found+stream.look())){
291 found += stream.next();
298 if (/^0[0-7]/.test(found)) tokens.push(new Token(found, "NUMB", "OCTAL", this.line));
299 else tokens.push(new Token(found, "NUMB", "DECIMAL", this.line));
304 requires("../lib/JSDOC/TextStream.js");
305 requires("../lib/JSDOC/Token.js");
306 requires("../lib/JSDOC/Lang.js");
308 plan(3, "testing read_numb");
311 var src = "function foo(num){while (num+8.0 >= 0x20 && num < 0777){}}";
312 var tr = new TokenReader();
313 var tokens = tr.tokenize(new TextStream(src));
315 var hexToken, octToken, decToken;
316 for (var i = 0; i < tokens.length; i++) {
317 if (tokens[i].name == "HEX_DEC") hexToken = tokens[i];
318 if (tokens[i].name == "OCTAL") octToken = tokens[i];
319 if (tokens[i].name == "DECIMAL") decToken = tokens[i];
323 is(decToken.data, "8.0", "decimal number is found in source.");
324 is(hexToken.data, "0x20", "hexdec number is found in source (issue #99).");
325 is(octToken.data, "0777", "octal number is found in source.");
329 @returns {Boolean} Was the token found?
331 read_hex : function(/**JSDOC.TokenStream*/stream, tokens) {
332 var found = stream.next(2);
334 while (!stream.look().eof) {
335 if (Lang.isHexDec(found) && !Lang.isHexDec(found+stream.look())) { // done
336 tokens.push(new Token(found, "NUMB", "HEX_DEC", this.line));
340 found += stream.next();
347 @returns {Boolean} Was the token found?
349 read_regx : function(/**JSDOC.TokenStream*/stream, tokens) {
357 !(last = tokens.lastSym()) // there is no last, the regex is the first symbol
362 && !last.is("RIGHT_PAREN")
363 && !last.is("RIGHT_BRACKET")
368 var regex = stream.next();
370 while (!stream.look().eof) {
371 if (stream.look() == "\\") { // escape sequence
372 regex += stream.next(2);
374 else if (stream.look() == "/") {
375 regex += stream.next();
377 while (/[gmi]/.test(stream.look())) {
378 regex += stream.next();
381 tokens.push(new Token(regex, "REGX", "REGX", this.line));
385 regex += stream.next();
388 // error: unterminated regex