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