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/Lang.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         this.sepIdents = false; // seperate '.' in identifiers..
  21         XObject.extend(this, o || {});
  22
  23     },
  24     Object,
  25     {
  26
  27
  28         /**
  29             @type {JSDOC.Token[]}
  30          */
  31
  32
  33         tokenize : function(/**JSDOC.TextStream*/stream) {
  34             var tokens = [];
  35             /**@ignore*/ tokens.last    = function() { return tokens[tokens.length-1]; }
  36             /**@ignore*/ tokens.lastSym = function() {
  37                 for (var i = tokens.length-1; i >= 0; i--) {
  38                     if (!(tokens[i].is("WHIT") || tokens[i].is("COMM"))) return tokens[i];
  39                 }
  40             }
  41
  42             while (!stream.look().eof) {
  43                 if (this.read_mlcomment(stream, tokens)) continue;
  44                 if (this.read_slcomment(stream, tokens)) continue;
  45                 if (this.read_dbquote(stream, tokens))   continue;
  46                 if (this.read_snquote(stream, tokens))   continue;
  47                 if (this.read_regx(stream, tokens))      continue;
  48                 if (this.read_numb(stream, tokens))      continue;
  49                 if (this.read_punc(stream, tokens))      continue;
  50                 if (this.read_newline(stream, tokens))   continue;
  51                 if (this.read_space(stream, tokens))     continue;
  52                 if (this.read_word(stream, tokens))      continue;
  53
  54                 // if execution reaches here then an error has happened
  55                 tokens.push(new Token(stream.next(), "TOKN", "UNKNOWN_TOKEN"));
  56             }
  57
  58
  59
  60             return tokens;
  61         },
  62
  63         /**
  64             @returns {Boolean} Was the token found?
  65          */
  66         read_word : function(/**JSDOC.TokenStream*/stream, tokens) {
  67             var found = "";
  68             while (!stream.look().eof && Lang.isWordChar(stream.look())) {
  69                 found += stream.next();
  70             }
  71
  72             if (found === "") {
  73                 return false;
  74             }
  75             else {
  76                 var name;
  77                 if ((name = Lang.keyword(found))) tokens.push(new Token(found, "KEYW", name));
  78                 else tokens.push(new Token(found, "NAME", "NAME"));
  79                 return true;
  80             }
  81         },
  82
  83         /**
  84             @returns {Boolean} Was the token found?
  85          */
  86         read_punc : function(/**JSDOC.TokenStream*/stream, tokens) {
  87             var found = "";
  88             var name;
  89             while (!stream.look().eof && Lang.punc(found+stream.look())) {
  90                 found += stream.next();
  91             }
  92
  93             if (found === "") {
  94                 return false;
  95             }
  96             else {
  97                 tokens.push(new Token(found, "PUNC", Lang.punc(found)));
  98                 return true;
  99             }
 100         },
 101
 102         /**
 103             @returns {Boolean} Was the token found?
 104          */
 105         read_space : function(/**JSDOC.TokenStream*/stream, tokens) {
 106             var found = "";
 107
 108             while (!stream.look().eof && Lang.isSpace(stream.look())) {
 109                 found += stream.next();
 110             }
 111
 112             if (found === "") {
 113                 return false;
 114             }
 115             else {
 116                 if (this.collapseWhite) found = " ";
 117                 if (this.keepWhite) tokens.push(new Token(found, "WHIT", "SPACE"));
 118                 return true;
 119             }
 120         },
 121
 122         /**
 123             @returns {Boolean} Was the token found?
 124          */
 125         read_newline : function(/**JSDOC.TokenStream*/stream, tokens) {
 126             var found = "";
 127
 128             while (!stream.look().eof && Lang.isNewline(stream.look())) {
 129                 found += stream.next();
 130             }
 131
 132             if (found === "") {
 133                 return false;
 134             }
 135             else {
 136                 if (this.collapseWhite) found = "\n";
 137                 if (this.keepWhite) tokens.push(new Token(found, "WHIT", "NEWLINE"));
 138                 return true;
 139             }
 140         },
 141
 142         /**
 143             @returns {Boolean} Was the token found?
 144          */
 145         read_mlcomment : function(/**JSDOC.TokenStream*/stream, tokens) {
 146             if (stream.look() == "/" && stream.look(1) == "*") {
 147                 var found = stream.next(2);
 148
 149                 while (!stream.look().eof && !(stream.look(-1) == "/" && stream.look(-2) == "*")) {
 150                     found += stream.next();
 151                 }
 152
 153                 // to start doclet we allow /** or /*** but not /**/ or /****
 154                 if (/^\/\*\*([^\/]|\*[^*])/.test(found) && this.keepDocs) tokens.push(new Token(found, "COMM", "JSDOC"));
 155                 else if (this.keepComments) tokens.push(new Token(found, "COMM", "MULTI_LINE_COMM"));
 156                 return true;
 157             }
 158             return false;
 159         },
 160
 161         /**
 162             @returns {Boolean} Was the token found?
 163          */
 164         read_slcomment : function(/**JSDOC.TokenStream*/stream, tokens) {
 165             var found;
 166             if (
 167                 (stream.look() == "/" && stream.look(1) == "/" && (found=stream.next(2)))
 168                 ||
 169                 (stream.look() == "<" && stream.look(1) == "!" && stream.look(2) == "-" && stream.look(3) == "-" && (found=stream.next(4)))
 170             ) {
 171
 172                 while (!stream.look().eof && !Lang.isNewline(stream.look())) {
 173                     found += stream.next();
 174                 }
 175
 176                 if (this.keepComments) {
 177                     tokens.push(new Token(found, "COMM", "SINGLE_LINE_COMM"));
 178                 }
 179                 return true;
 180             }
 181             return false;
 182         },
 183
 184         /**
 185             @returns {Boolean} Was the token found?
 186          */
 187         read_dbquote : function(/**JSDOC.TokenStream*/stream, tokens) {
 188             if (stream.look() == "\"") {
 189                 // find terminator
 190                 var string = stream.next();
 191
 192                 while (!stream.look().eof) {
 193                     if (stream.look() == "\\") {
 194                         if (Lang.isNewline(stream.look(1))) {
 195                             do {
 196                                 stream.next();
 197                             } while (!stream.look().eof && Lang.isNewline(stream.look()));
 198                             string += "\\\n";
 199                         }
 200                         else {
 201                             string += stream.next(2);
 202                         }
 203                     }
 204                     else if (stream.look() == "\"") {
 205                         string += stream.next();
 206                         tokens.push(new Token(string, "STRN", "DOUBLE_QUOTE"));
 207                         return true;
 208                     }
 209                     else {
 210                         string += stream.next();
 211                     }
 212                 }
 213             }
 214             return false; // error! unterminated string
 215         },
 216
 217         /**
 218             @returns {Boolean} Was the token found?
 219          */
 220         read_snquote : function(/**JSDOC.TokenStream*/stream, tokens) {
 221             if (stream.look() == "'") {
 222                 // find terminator
 223                 var string = stream.next();
 224
 225                 while (!stream.look().eof) {
 226                     if (stream.look() == "\\") { // escape sequence
 227                         string += stream.next(2);
 228                     }
 229                     else if (stream.look() == "'") {
 230                         string += stream.next();
 231                         tokens.push(new Token(string, "STRN", "SINGLE_QUOTE"));
 232                         return true;
 233                     }
 234                     else {
 235                         string += stream.next();
 236                     }
 237                 }
 238             }
 239             return false; // error! unterminated string
 240         },
 241
 242         /**
 243             @returns {Boolean} Was the token found?
 244          */
 245         read_numb : function(/**JSDOC.TokenStream*/stream, tokens) {
 246             if (stream.look() === "0" && stream.look(1) == "x") {
 247                 return this.read_hex(stream, tokens);
 248             }
 249
 250             var found = "";
 251
 252             while (!stream.look().eof && Lang.isNumber(found+stream.look())){
 253                 found += stream.next();
 254             }
 255
 256             if (found === "") {
 257                 return false;
 258             }
 259             else {
 260                 if (/^0[0-7]/.test(found)) tokens.push(new Token(found, "NUMB", "OCTAL"));
 261                 else tokens.push(new Token(found, "NUMB", "DECIMAL"));
 262                 return true;
 263             }
 264         },
 265         /*t:
 266             requires("../lib/JSDOC/TextStream.js");
 267             requires("../lib/JSDOC/Token.js");
 268             requires("../lib/JSDOC/Lang.js");
 269
 270             plan(3, "testing read_numb");
 271
 272             //// setup
 273             var src = "function foo(num){while (num+8.0 >= 0x20 && num < 0777){}}";
 274             var tr = new TokenReader();
 275             var tokens = tr.tokenize(new TextStream(src));
 276
 277             var hexToken, octToken, decToken;
 278             for (var i = 0; i < tokens.length; i++) {
 279                 if (tokens[i].name == "HEX_DEC") hexToken = tokens[i];
 280                 if (tokens[i].name == "OCTAL") octToken = tokens[i];
 281                 if (tokens[i].name == "DECIMAL") decToken = tokens[i];
 282             }
 283             ////
 284
 285             is(decToken.data, "8.0", "decimal number is found in source.");
 286             is(hexToken.data, "0x20", "hexdec number is found in source (issue #99).");
 287             is(octToken.data, "0777", "octal number is found in source.");
 288         */
 289
 290         /**
 291             @returns {Boolean} Was the token found?
 292          */
 293         read_hex : function(/**JSDOC.TokenStream*/stream, tokens) {
 294             var found = stream.next(2);
 295
 296             while (!stream.look().eof) {
 297                 if (Lang.isHexDec(found) && !Lang.isHexDec(found+stream.look())) { // done
 298                     tokens.push(new Token(found, "NUMB", "HEX_DEC"));
 299                     return true;
 300                 }
 301                 else {
 302                     found += stream.next();
 303                 }
 304             }
 305             return false;
 306         },
 307
 308         /**
 309             @returns {Boolean} Was the token found?
 310          */
 311         read_regx : function(/**JSDOC.TokenStream*/stream, tokens) {
 312             var last;
 313             if (
 314                 stream.look() == "/"
 315                 &&
 316                 (
 317
 318                     (
 319                         !(last = tokens.lastSym()) // there is no last, the regex is the first symbol
 320                         ||
 321                         (
 322                                !last.is("NUMB")
 323                             && !last.is("NAME")
 324                             && !last.is("RIGHT_PAREN")
 325                             && !last.is("RIGHT_BRACKET")
 326                         )
 327                     )
 328                 )
 329             ) {
 330                 var regex = stream.next();
 331
 332                 while (!stream.look().eof) {
 333                     if (stream.look() == "\\") { // escape sequence
 334                         regex += stream.next(2);
 335                     }
 336                     else if (stream.look() == "/") {
 337                         regex += stream.next();
 338
 339                         while (/[gmi]/.test(stream.look())) {
 340                             regex += stream.next();
 341                         }
 342
 343                         tokens.push(new Token(regex, "REGX", "REGX"));
 344                         return true;
 345                     }
 346                     else {
 347                         regex += stream.next();
 348                     }
 349                 }
 350                 // error: unterminated regex
 351             }
 352             return false;
 353         }
 354 });