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..console;
   6
   7
   8 Token   = imports.Token.Token;
   9 Lang    = imports.Lang.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             this.line =1;
  35             var tokens = [];
  36             /**@ignore*/ tokens.last    = function() { return tokens[tokens.length-1]; }
  37             /**@ignore*/ tokens.lastSym = function() {
  38                 for (var i = tokens.length-1; i >= 0; i--) {
  39                     if (!(tokens[i].is("WHIT") || tokens[i].is("COMM"))) return tokens[i];
  40                 }
  41             }
  42
  43             while (!stream.look().eof) {
  44                 if (this.read_mlcomment(stream, tokens)) continue;
  45                 if (this.read_slcomment(stream, tokens)) continue;
  46                 if (this.read_dbquote(stream, tokens))   continue;
  47                 if (this.read_snquote(stream, tokens))   continue;
  48                 if (this.read_regx(stream, tokens))      continue;
  49                 if (this.read_numb(stream, tokens))      continue;
  50                 if (this.read_punc(stream, tokens))      continue;
  51                 if (this.read_newline(stream, tokens))   continue;
  52                 if (this.read_space(stream, tokens))     continue;
  53                 if (this.read_word(stream, tokens))      continue;
  54
  55                 // if execution reaches here then an error has happened
  56                 tokens.push(new Token(stream.next(), "TOKN", "UNKNOWN_TOKEN", this.line));
  57             }
  58
  59
  60
  61             return tokens;
  62         },
  63
  64         /**
  65             @returns {Boolean} Was the token found?
  66          */
  67         read_word : function(/**JSDOC.TokenStream*/stream, tokens) {
  68             var found = "";
  69             while (!stream.look().eof && Lang.isWordChar(stream.look())) {
  70                 found += stream.next();
  71             }
  72
  73             if (found === "") {
  74                 return false;
  75             }
  76             else {
  77                 var name;
  78                 if ((name = Lang.keyword(found))) {
  79                     tokens.push(new Token(found, "KEYW", name, this.line));
  80                     return true;
  81                 }
  82                 if (!this.sepIdents || found.indexOf('.') < 0 ) {
  83                     tokens.push(new Token(found, "NAME", "NAME", this.line));
  84                     return true;
  85                 }
  86                 var n = found.split('.');
  87                 var p = false;
  88                 n.forEach(function(nm) {
  89                     if (p) {
  90                         tokens.push(new Token('.', "PUNC", "DOT", this.line));
  91                     }
  92                     p=true;
  93                     tokens.push(new Token(nm, "NAME", "NAME", this.line));
  94                 });
  95                 return true;
  96
  97             }
  98         },
  99
 100         /**
 101             @returns {Boolean} Was the token found?
 102          */
 103         read_punc : function(/**JSDOC.TokenStream*/stream, tokens) {
 104             var found = "";
 105             var name;
 106             while (!stream.look().eof && Lang.punc(found+stream.look())) {
 107                 found += stream.next();
 108             }
 109
 110             if (found === "") {
 111                 return false;
 112             }
 113             else {
 114                 tokens.push(new Token(found, "PUNC", Lang.punc(found), this.line));
 115                 return true;
 116             }
 117         },
 118
 119         /**
 120             @returns {Boolean} Was the token found?
 121          */
 122         read_space : function(/**JSDOC.TokenStream*/stream, tokens) {
 123             var found = "";
 124
 125             while (!stream.look().eof && Lang.isSpace(stream.look())) {
 126                 found += stream.next();
 127             }
 128
 129             if (found === "") {
 130                 return false;
 131             }
 132             else {
 133                 if (this.collapseWhite) found = " ";
 134                 if (this.keepWhite) tokens.push(new Token(found, "WHIT", "SPACE", this.line));
 135                 return true;
 136             }
 137         },
 138
 139         /**
 140             @returns {Boolean} Was the token found?
 141          */
 142         read_newline : function(/**JSDOC.TokenStream*/stream, tokens) {
 143             var found = "";
 144
 145             while (!stream.look().eof && Lang.isNewline(stream.look())) {
 146                 this.line++;
 147                 found += stream.next();
 148             }
 149
 150             if (found === "") {
 151                 return false;
 152             }
 153             else {
 154                 if (this.collapseWhite) found = "\n";
 155                 if (this.keepWhite) tokens.push(new Token(found, "WHIT", "NEWLINE", this.line));
 156                 return true;
 157             }
 158         },
 159
 160         /**
 161             @returns {Boolean} Was the token found?
 162          */
 163         read_mlcomment : function(/**JSDOC.TokenStream*/stream, tokens) {
 164             if (stream.look() == "/" && stream.look(1) == "*") {
 165                 var found = stream.next(2);
 166                 var c = '';
 167                 while (!stream.look().eof && !(stream.look(-1) == "/" && stream.look(-2) == "*")) {
 168                     c = stream.next();
 169                     if (c == "\n") this.line++;
 170                     found += c;
 171                 }
 172
 173                 // to start doclet we allow /** or /*** but not /**/ or /****
 174                 if (/^\/\*\*([^\/]|\*[^*])/.test(found) && this.keepDocs) tokens.push(new Token(found, "COMM", "JSDOC", this.line));
 175                 else if (this.keepComments) tokens.push(new Token(found, "COMM", "MULTI_LINE_COMM", this.line));
 176                 return true;
 177             }
 178             return false;
 179         },
 180
 181         /**
 182             @returns {Boolean} Was the token found?
 183          */
 184         read_slcomment : function(/**JSDOC.TokenStream*/stream, tokens) {
 185             var found;
 186             if (
 187                 (stream.look() == "/" && stream.look(1) == "/" && (found=stream.next(2)))
 188                 ||
 189                 (stream.look() == "<" && stream.look(1) == "!" && stream.look(2) == "-" && stream.look(3) == "-" && (found=stream.next(4)))
 190             ) {
 191
 192                 while (!stream.look().eof && !Lang.isNewline(stream.look())) {
 193                     found += stream.next();
 194                 }
 195
 196                 if (this.keepComments) {
 197                     tokens.push(new Token(found, "COMM", "SINGLE_LINE_COMM", this.line));
 198                 }
 199                 this.line++;
 200                 return true;
 201             }
 202             return false;
 203         },
 204
 205         /**
 206             @returns {Boolean} Was the token found?
 207          */
 208         read_dbquote : function(/**JSDOC.TokenStream*/stream, tokens) {
 209             if (stream.look() == "\"") {
 210                 // find terminator
 211                 var string = stream.next();
 212
 213                 while (!stream.look().eof) {
 214                     if (stream.look() == "\\") {
 215                         if (Lang.isNewline(stream.look(1))) {
 216                             do {
 217                                 stream.next();
 218                             } while (!stream.look().eof && Lang.isNewline(stream.look()));
 219                             string += "\\\n";
 220                         }
 221                         else {
 222                             string += stream.next(2);
 223                         }
 224                     }
 225                     else if (stream.look() == "\"") {
 226                         string += stream.next();
 227                         tokens.push(new Token(string, "STRN", "DOUBLE_QUOTE", this.line));
 228                         return true;
 229                     }
 230                     else {
 231                         string += stream.next();
 232                     }
 233                 }
 234             }
 235             return false; // error! unterminated string
 236         },
 237
 238         /**
 239             @returns {Boolean} Was the token found?
 240          */
 241         read_snquote : function(/**JSDOC.TokenStream*/stream, tokens) {
 242             if (stream.look() == "'") {
 243                 // find terminator
 244                 var string = stream.next();
 245
 246                 while (!stream.look().eof) {
 247                     if (stream.look() == "\\") { // escape sequence
 248                         string += stream.next(2);
 249                     }
 250                     else if (stream.look() == "'") {
 251                         string += stream.next();
 252                         tokens.push(new Token(string, "STRN", "SINGLE_QUOTE", this.line));
 253                         return true;
 254                     }
 255                     else {
 256                         string += stream.next();
 257                     }
 258                 }
 259             }
 260             return false; // error! unterminated string
 261         },
 262
 263         /**
 264             @returns {Boolean} Was the token found?
 265          */
 266         read_numb : function(/**JSDOC.TokenStream*/stream, tokens) {
 267             if (stream.look() === "0" && stream.look(1) == "x") {
 268                 return this.read_hex(stream, tokens);
 269             }
 270
 271             var found = "";
 272
 273             while (!stream.look().eof && Lang.isNumber(found+stream.look())){
 274                 found += stream.next();
 275             }
 276
 277             if (found === "") {
 278                 return false;
 279             }
 280             else {
 281                 if (/^0[0-7]/.test(found)) tokens.push(new Token(found, "NUMB", "OCTAL", this.line));
 282                 else tokens.push(new Token(found, "NUMB", "DECIMAL", this.line));
 283                 return true;
 284             }
 285         },
 286         /*t:
 287             requires("../lib/JSDOC/TextStream.js");
 288             requires("../lib/JSDOC/Token.js");
 289             requires("../lib/JSDOC/Lang.js");
 290
 291             plan(3, "testing read_numb");
 292
 293             //// setup
 294             var src = "function foo(num){while (num+8.0 >= 0x20 && num < 0777){}}";
 295             var tr = new TokenReader();
 296             var tokens = tr.tokenize(new TextStream(src));
 297
 298             var hexToken, octToken, decToken;
 299             for (var i = 0; i < tokens.length; i++) {
 300                 if (tokens[i].name == "HEX_DEC") hexToken = tokens[i];
 301                 if (tokens[i].name == "OCTAL") octToken = tokens[i];
 302                 if (tokens[i].name == "DECIMAL") decToken = tokens[i];
 303             }
 304             ////
 305
 306             is(decToken.data, "8.0", "decimal number is found in source.");
 307             is(hexToken.data, "0x20", "hexdec number is found in source (issue #99).");
 308             is(octToken.data, "0777", "octal number is found in source.");
 309         */
 310
 311         /**
 312             @returns {Boolean} Was the token found?
 313          */
 314         read_hex : function(/**JSDOC.TokenStream*/stream, tokens) {
 315             var found = stream.next(2);
 316
 317             while (!stream.look().eof) {
 318                 if (Lang.isHexDec(found) && !Lang.isHexDec(found+stream.look())) { // done
 319                     tokens.push(new Token(found, "NUMB", "HEX_DEC", this.line));
 320                     return true;
 321                 }
 322                 else {
 323                     found += stream.next();
 324                 }
 325             }
 326             return false;
 327         },
 328
 329         /**
 330             @returns {Boolean} Was the token found?
 331          */
 332         read_regx : function(/**JSDOC.TokenStream*/stream, tokens) {
 333             var last;
 334             if (
 335                 stream.look() == "/"
 336                 &&
 337                 (
 338
 339                     (
 340                         !(last = tokens.lastSym()) // there is no last, the regex is the first symbol
 341                         ||
 342                         (
 343                                !last.is("NUMB")
 344                             && !last.is("NAME")
 345                             && !last.is("RIGHT_PAREN")
 346                             && !last.is("RIGHT_BRACKET")
 347                         )
 348                     )
 349                 )
 350             ) {
 351                 var regex = stream.next();
 352
 353                 while (!stream.look().eof) {
 354                     if (stream.look() == "\\") { // escape sequence
 355                         regex += stream.next(2);
 356                     }
 357                     else if (stream.look() == "/") {
 358                         regex += stream.next();
 359
 360                         while (/[gmi]/.test(stream.look())) {
 361                             regex += stream.next();
 362                         }
 363
 364                         tokens.push(new Token(regex, "REGX", "REGX", this.line));
 365                         return true;
 366                     }
 367                     else {
 368                         regex += stream.next();
 369                     }
 370                 }
 371                 // error: unterminated regex
 372             }
 373             return false;
 374         }
 375 });