git.roojs.org Git - gnome.introspection-doc-generator/blob - JSDOC/TokenReader.js

   1 //<script type="text/javascript">
   2
   3 //imports['Object.js'].load(Object);
   4 XObject = imports.XObject.XObject;
   5 console = imports['console.js'].console;
   6
   7 JSDOC   = imports['JSDOC.js'].JSDOC;
   8 Token   = imports['JSDOC/Token.js'].Token;
   9 Lang    = imports['JSDOC/Token.js'].Lang;
  10
  11 /**
  12         @class Search a {@link JSDOC.TextStream} for language tokens.
  13 */
  14 TokenReader = XObject.define(
  15     function(o) {
  16
  17         this.keepDocs = true;
  18         this.keepWhite = false;
  19         this.keepComments = false;
  20         XObject.extend(this, o || {});
  21
  22     },
  23     Object,
  24     {
  25
  26
  27         /**
  28             @type {JSDOC.Token[]}
  29          */
  30
  31
  32         tokenize : function(/**JSDOC.TextStream*/stream) {
  33             var tokens = [];
  34             /**@ignore*/ tokens.last    = function() { return tokens[tokens.length-1]; }
  35             /**@ignore*/ tokens.lastSym = function() {
  36                 for (var i = tokens.length-1; i >= 0; i--) {
  37                     if (!(tokens[i].is("WHIT") || tokens[i].is("COMM"))) return tokens[i];
  38                 }
  39             }
  40
  41             while (!stream.look().eof) {
  42                 if (this.read_mlcomment(stream, tokens)) continue;
  43                 if (this.read_slcomment(stream, tokens)) continue;
  44                 if (this.read_dbquote(stream, tokens))   continue;
  45                 if (this.read_snquote(stream, tokens))   continue;
  46                 if (this.read_regx(stream, tokens))      continue;
  47                 if (this.read_numb(stream, tokens))      continue;
  48                 if (this.read_punc(stream, tokens))      continue;
  49                 if (this.read_newline(stream, tokens))   continue;
  50                 if (this.read_space(stream, tokens))     continue;
  51                 if (this.read_word(stream, tokens))      continue;
  52
  53                 // if execution reaches here then an error has happened
  54                 tokens.push(new Token(stream.next(), "TOKN", "UNKNOWN_TOKEN"));
  55             }
  56
  57
  58
  59             return tokens;
  60         },
  61
  62         /**
  63             @returns {Boolean} Was the token found?
  64          */
  65         read_word : function(/**JSDOC.TokenStream*/stream, tokens) {
  66             var found = "";
  67             while (!stream.look().eof && Lang.isWordChar(stream.look())) {
  68                 found += stream.next();
  69             }
  70
  71             if (found === "") {
  72                 return false;
  73             }
  74             else {
  75                 var name;
  76                 if ((name = Lang.keyword(found))) tokens.push(new Token(found, "KEYW", name));
  77                 else tokens.push(new Token(found, "NAME", "NAME"));
  78                 return true;
  79             }
  80         },
  81
  82         /**
  83             @returns {Boolean} Was the token found?
  84          */
  85         read_punc : function(/**JSDOC.TokenStream*/stream, tokens) {
  86             var found = "";
  87             var name;
  88             while (!stream.look().eof && Lang.punc(found+stream.look())) {
  89                 found += stream.next();
  90             }
  91
  92             if (found === "") {
  93                 return false;
  94             }
  95             else {
  96                 tokens.push(new Token(found, "PUNC", Lang.punc(found)));
  97                 return true;
  98             }
  99         },
 100
 101         /**
 102             @returns {Boolean} Was the token found?
 103          */
 104         read_space : function(/**JSDOC.TokenStream*/stream, tokens) {
 105             var found = "";
 106
 107             while (!stream.look().eof && Lang.isSpace(stream.look())) {
 108                 found += stream.next();
 109             }
 110
 111             if (found === "") {
 112                 return false;
 113             }
 114             else {
 115                 if (this.collapseWhite) found = " ";
 116                 if (this.keepWhite) tokens.push(new Token(found, "WHIT", "SPACE"));
 117                 return true;
 118             }
 119         },
 120
 121         /**
 122             @returns {Boolean} Was the token found?
 123          */
 124         read_newline : function(/**JSDOC.TokenStream*/stream, tokens) {
 125             var found = "";
 126
 127             while (!stream.look().eof && Lang.isNewline(stream.look())) {
 128                 found += stream.next();
 129             }
 130
 131             if (found === "") {
 132                 return false;
 133             }
 134             else {
 135                 if (this.collapseWhite) found = "\n";
 136                 if (this.keepWhite) tokens.push(new Token(found, "WHIT", "NEWLINE"));
 137                 return true;
 138             }
 139         },
 140
 141         /**
 142             @returns {Boolean} Was the token found?
 143          */
 144         read_mlcomment : function(/**JSDOC.TokenStream*/stream, tokens) {
 145             if (stream.look() == "/" && stream.look(1) == "*") {
 146                 var found = stream.next(2);
 147
 148                 while (!stream.look().eof && !(stream.look(-1) == "/" && stream.look(-2) == "*")) {
 149                     found += stream.next();
 150                 }
 151
 152                 // to start doclet we allow /** or /*** but not /**/ or /****
 153                 if (/^\/\*\*([^\/]|\*[^*])/.test(found) && this.keepDocs) tokens.push(new Token(found, "COMM", "JSDOC"));
 154                 else if (this.keepComments) tokens.push(new Token(found, "COMM", "MULTI_LINE_COMM"));
 155                 return true;
 156             }
 157             return false;
 158         },
 159
 160         /**
 161             @returns {Boolean} Was the token found?
 162          */
 163         read_slcomment : function(/**JSDOC.TokenStream*/stream, tokens) {
 164             var found;
 165             if (
 166                 (stream.look() == "/" && stream.look(1) == "/" && (found=stream.next(2)))
 167                 ||
 168                 (stream.look() == "<" && stream.look(1) == "!" && stream.look(2) == "-" && stream.look(3) == "-" && (found=stream.next(4)))
 169             ) {
 170
 171                 while (!stream.look().eof && !Lang.isNewline(stream.look())) {
 172                     found += stream.next();
 173                 }
 174
 175                 if (this.keepComments) {
 176                     tokens.push(new Token(found, "COMM", "SINGLE_LINE_COMM"));
 177                 }
 178                 return true;
 179             }
 180             return false;
 181         },
 182
 183         /**
 184             @returns {Boolean} Was the token found?
 185          */
 186         read_dbquote : function(/**JSDOC.TokenStream*/stream, tokens) {
 187             if (stream.look() == "\"") {
 188                 // find terminator
 189                 var string = stream.next();
 190
 191                 while (!stream.look().eof) {
 192                     if (stream.look() == "\\") {
 193                         if (Lang.isNewline(stream.look(1))) {
 194                             do {
 195                                 stream.next();
 196                             } while (!stream.look().eof && Lang.isNewline(stream.look()));
 197                             string += "\\\n";
 198                         }
 199                         else {
 200                             string += stream.next(2);
 201                         }
 202                     }
 203                     else if (stream.look() == "\"") {
 204                         string += stream.next();
 205                         tokens.push(new Token(string, "STRN", "DOUBLE_QUOTE"));
 206                         return true;
 207                     }
 208                     else {
 209                         string += stream.next();
 210                     }
 211                 }
 212             }
 213             return false; // error! unterminated string
 214         },
 215
 216         /**
 217             @returns {Boolean} Was the token found?
 218          */
 219         read_snquote : function(/**JSDOC.TokenStream*/stream, tokens) {
 220             if (stream.look() == "'") {
 221                 // find terminator
 222                 var string = stream.next();
 223
 224                 while (!stream.look().eof) {
 225                     if (stream.look() == "\\") { // escape sequence
 226                         string += stream.next(2);
 227                     }
 228                     else if (stream.look() == "'") {
 229                         string += stream.next();
 230                         tokens.push(new Token(string, "STRN", "SINGLE_QUOTE"));
 231                         return true;
 232                     }
 233                     else {
 234                         string += stream.next();
 235                     }
 236                 }
 237             }
 238             return false; // error! unterminated string
 239         },
 240
 241         /**
 242             @returns {Boolean} Was the token found?
 243          */
 244         read_numb : function(/**JSDOC.TokenStream*/stream, tokens) {
 245             if (stream.look() === "0" && stream.look(1) == "x") {
 246                 return this.read_hex(stream, tokens);
 247             }
 248
 249             var found = "";
 250
 251             while (!stream.look().eof && Lang.isNumber(found+stream.look())){
 252                 found += stream.next();
 253             }
 254
 255             if (found === "") {
 256                 return false;
 257             }
 258             else {
 259                 if (/^0[0-7]/.test(found)) tokens.push(new Token(found, "NUMB", "OCTAL"));
 260                 else tokens.push(new Token(found, "NUMB", "DECIMAL"));
 261                 return true;
 262             }
 263         },
 264         /*t:
 265             requires("../lib/JSDOC/TextStream.js");
 266             requires("../lib/JSDOC/Token.js");
 267             requires("../lib/JSDOC/Lang.js");
 268
 269             plan(3, "testing read_numb");
 270
 271             //// setup
 272             var src = "function foo(num){while (num+8.0 >= 0x20 && num < 0777){}}";
 273             var tr = new TokenReader();
 274             var tokens = tr.tokenize(new TextStream(src));
 275
 276             var hexToken, octToken, decToken;
 277             for (var i = 0; i < tokens.length; i++) {
 278                 if (tokens[i].name == "HEX_DEC") hexToken = tokens[i];
 279                 if (tokens[i].name == "OCTAL") octToken = tokens[i];
 280                 if (tokens[i].name == "DECIMAL") decToken = tokens[i];
 281             }
 282             ////
 283
 284             is(decToken.data, "8.0", "decimal number is found in source.");
 285             is(hexToken.data, "0x20", "hexdec number is found in source (issue #99).");
 286             is(octToken.data, "0777", "octal number is found in source.");
 287         */
 288
 289         /**
 290             @returns {Boolean} Was the token found?
 291          */
 292         read_hex : function(/**JSDOC.TokenStream*/stream, tokens) {
 293             var found = stream.next(2);
 294
 295             while (!stream.look().eof) {
 296                 if (Lang.isHexDec(found) && !Lang.isHexDec(found+stream.look())) { // done
 297                     tokens.push(new Token(found, "NUMB", "HEX_DEC"));
 298                     return true;
 299                 }
 300                 else {
 301                     found += stream.next();
 302                 }
 303             }
 304             return false;
 305         },
 306
 307         /**
 308             @returns {Boolean} Was the token found?
 309          */
 310         read_regx : function(/**JSDOC.TokenStream*/stream, tokens) {
 311             var last;
 312             if (
 313                 stream.look() == "/"
 314                 &&
 315                 (
 316
 317                     (
 318                         !(last = tokens.lastSym()) // there is no last, the regex is the first symbol
 319                         ||
 320                         (
 321                                !last.is("NUMB")
 322                             && !last.is("NAME")
 323                             && !last.is("RIGHT_PAREN")
 324                             && !last.is("RIGHT_BRACKET")
 325                         )
 326                     )
 327                 )
 328             ) {
 329                 var regex = stream.next();
 330
 331                 while (!stream.look().eof) {
 332                     if (stream.look() == "\\") { // escape sequence
 333                         regex += stream.next(2);
 334                     }
 335                     else if (stream.look() == "/") {
 336                         regex += stream.next();
 337
 338                         while (/[gmi]/.test(stream.look())) {
 339                             regex += stream.next();
 340                         }
 341
 342                         tokens.push(new Token(regex, "REGX", "REGX"));
 343                         return true;
 344                     }
 345                     else {
 346                         regex += stream.next();
 347                     }
 348                 }
 349                 // error: unterminated regex
 350             }
 351             return false;
 352         }
 353 });