can generate jsdoc; private methods marked as such
[dygraphs.git] / jsdoc-toolkit / app / lib / JSDOC / TokenReader.js
diff --git a/jsdoc-toolkit/app/lib/JSDOC/TokenReader.js b/jsdoc-toolkit/app/lib/JSDOC/TokenReader.js
new file mode 100644 (file)
index 0000000..9f658fb
--- /dev/null
@@ -0,0 +1,332 @@
+if (typeof JSDOC == "undefined") JSDOC = {};
+
+/**
+       @class Search a {@link JSDOC.TextStream} for language tokens.
+*/
+JSDOC.TokenReader = function() {
+       this.keepDocs = true;
+       this.keepWhite = false;
+       this.keepComments = false;
+}
+
+/**
+       @type {JSDOC.Token[]}
+ */
+JSDOC.TokenReader.prototype.tokenize = function(/**JSDOC.TextStream*/stream) {
+       var tokens = [];
+       /**@ignore*/ tokens.last    = function() { return tokens[tokens.length-1]; }
+       /**@ignore*/ tokens.lastSym = function() {
+               for (var i = tokens.length-1; i >= 0; i--) {
+                       if (!(tokens[i].is("WHIT") || tokens[i].is("COMM"))) return tokens[i];
+               }
+       }
+
+       while (!stream.look().eof) {
+               if (this.read_mlcomment(stream, tokens)) continue;
+               if (this.read_slcomment(stream, tokens)) continue;
+               if (this.read_dbquote(stream, tokens))   continue;
+               if (this.read_snquote(stream, tokens))   continue;
+               if (this.read_regx(stream, tokens))      continue;
+               if (this.read_numb(stream, tokens))      continue;
+               if (this.read_punc(stream, tokens))      continue;
+               if (this.read_newline(stream, tokens))   continue;
+               if (this.read_space(stream, tokens))     continue;
+               if (this.read_word(stream, tokens))      continue;
+               
+               // if execution reaches here then an error has happened
+               tokens.push(new JSDOC.Token(stream.next(), "TOKN", "UNKNOWN_TOKEN"));
+       }
+       return tokens;
+}
+
+/**
+       @returns {Boolean} Was the token found?
+ */
+JSDOC.TokenReader.prototype.read_word = function(/**JSDOC.TokenStream*/stream, tokens) {
+       var found = "";
+       while (!stream.look().eof && JSDOC.Lang.isWordChar(stream.look())) {
+               found += stream.next();
+       }
+       
+       if (found === "") {
+               return false;
+       }
+       else {
+               var name;
+               if ((name = JSDOC.Lang.keyword(found))) tokens.push(new JSDOC.Token(found, "KEYW", name));
+               else tokens.push(new JSDOC.Token(found, "NAME", "NAME"));
+               return true;
+       }
+}
+
+/**
+       @returns {Boolean} Was the token found?
+ */
+JSDOC.TokenReader.prototype.read_punc = function(/**JSDOC.TokenStream*/stream, tokens) {
+       var found = "";
+       var name;
+       while (!stream.look().eof && JSDOC.Lang.punc(found+stream.look())) {
+               found += stream.next();
+       }
+       
+       if (found === "") {
+               return false;
+       }
+       else {
+               tokens.push(new JSDOC.Token(found, "PUNC", JSDOC.Lang.punc(found)));
+               return true;
+       }
+}
+
+/**
+       @returns {Boolean} Was the token found?
+ */
+JSDOC.TokenReader.prototype.read_space = function(/**JSDOC.TokenStream*/stream, tokens) {
+       var found = "";
+       
+       while (!stream.look().eof && JSDOC.Lang.isSpace(stream.look())) {
+               found += stream.next();
+       }
+       
+       if (found === "") {
+               return false;
+       }
+       else {
+               if (this.collapseWhite) found = " ";
+               if (this.keepWhite) tokens.push(new JSDOC.Token(found, "WHIT", "SPACE"));
+               return true;
+       }
+}
+
+/**
+       @returns {Boolean} Was the token found?
+ */
+JSDOC.TokenReader.prototype.read_newline = function(/**JSDOC.TokenStream*/stream, tokens) {
+       var found = "";
+       
+       while (!stream.look().eof && JSDOC.Lang.isNewline(stream.look())) {
+               found += stream.next();
+       }
+       
+       if (found === "") {
+               return false;
+       }
+       else {
+               if (this.collapseWhite) found = "\n";
+               if (this.keepWhite) tokens.push(new JSDOC.Token(found, "WHIT", "NEWLINE"));
+               return true;
+       }
+}
+
+/**
+       @returns {Boolean} Was the token found?
+ */
+JSDOC.TokenReader.prototype.read_mlcomment = function(/**JSDOC.TokenStream*/stream, tokens) {
+       if (stream.look() == "/" && stream.look(1) == "*") {
+               var found = stream.next(2);
+               
+               while (!stream.look().eof && !(stream.look(-1) == "/" && stream.look(-2) == "*")) {
+                       found += stream.next();
+               }
+               
+               // to start doclet we allow /** or /*** but not /**/ or /****
+               if (/^\/\*\*([^\/]|\*[^*])/.test(found) && this.keepDocs) tokens.push(new JSDOC.Token(found, "COMM", "JSDOC"));
+               else if (this.keepComments) tokens.push(new JSDOC.Token(found, "COMM", "MULTI_LINE_COMM"));
+               return true;
+       }
+       return false;
+}
+
+/**
+       @returns {Boolean} Was the token found?
+ */
+JSDOC.TokenReader.prototype.read_slcomment = function(/**JSDOC.TokenStream*/stream, tokens) {
+       var found;
+       if (
+               (stream.look() == "/" && stream.look(1) == "/" && (found=stream.next(2)))
+               || 
+               (stream.look() == "<" && stream.look(1) == "!" && stream.look(2) == "-" && stream.look(3) == "-" && (found=stream.next(4)))
+       ) {
+               
+               while (!stream.look().eof && !JSDOC.Lang.isNewline(stream.look())) {
+                       found += stream.next();
+               }
+               
+               if (this.keepComments) {
+                       tokens.push(new JSDOC.Token(found, "COMM", "SINGLE_LINE_COMM"));
+               }
+               return true;
+       }
+       return false;
+}
+
+/**
+       @returns {Boolean} Was the token found?
+ */
+JSDOC.TokenReader.prototype.read_dbquote = function(/**JSDOC.TokenStream*/stream, tokens) {
+       if (stream.look() == "\"") {
+               // find terminator
+               var string = stream.next();
+               
+               while (!stream.look().eof) {
+                       if (stream.look() == "\\") {
+                               if (JSDOC.Lang.isNewline(stream.look(1))) {
+                                       do {
+                                               stream.next();
+                                       } while (!stream.look().eof && JSDOC.Lang.isNewline(stream.look()));
+                                       string += "\\\n";
+                               }
+                               else {
+                                       string += stream.next(2);
+                               }
+                       }
+                       else if (stream.look() == "\"") {
+                               string += stream.next();
+                               tokens.push(new JSDOC.Token(string, "STRN", "DOUBLE_QUOTE"));
+                               return true;
+                       }
+                       else {
+                               string += stream.next();
+                       }
+               }
+       }
+       return false; // error! unterminated string
+}
+
+/**
+       @returns {Boolean} Was the token found?
+ */
+JSDOC.TokenReader.prototype.read_snquote = function(/**JSDOC.TokenStream*/stream, tokens) {
+       if (stream.look() == "'") {
+               // find terminator
+               var string = stream.next();
+               
+               while (!stream.look().eof) {
+                       if (stream.look() == "\\") { // escape sequence
+                               string += stream.next(2);
+                       }
+                       else if (stream.look() == "'") {
+                               string += stream.next();
+                               tokens.push(new JSDOC.Token(string, "STRN", "SINGLE_QUOTE"));
+                               return true;
+                       }
+                       else {
+                               string += stream.next();
+                       }
+               }
+       }
+       return false; // error! unterminated string
+}
+
+/**
+       @returns {Boolean} Was the token found?
+ */
+JSDOC.TokenReader.prototype.read_numb = function(/**JSDOC.TokenStream*/stream, tokens) {
+       if (stream.look() === "0" && stream.look(1) == "x") {
+               return this.read_hex(stream, tokens);
+       }
+       
+       var found = "";
+       
+       while (!stream.look().eof && JSDOC.Lang.isNumber(found+stream.look())){
+               found += stream.next();
+       }
+       
+       if (found === "") {
+               return false;
+       }
+       else {
+               if (/^0[0-7]/.test(found)) tokens.push(new JSDOC.Token(found, "NUMB", "OCTAL"));
+               else tokens.push(new JSDOC.Token(found, "NUMB", "DECIMAL"));
+               return true;
+       }
+}
+/*t:
+       requires("../lib/JSDOC/TextStream.js");
+       requires("../lib/JSDOC/Token.js");
+       requires("../lib/JSDOC/Lang.js");
+       
+       plan(3, "testing JSDOC.TokenReader.prototype.read_numb");
+       
+       //// setup
+       var src = "function foo(num){while (num+8.0 >= 0x20 && num < 0777){}}";
+       var tr = new JSDOC.TokenReader();
+       var tokens = tr.tokenize(new JSDOC.TextStream(src));
+       
+       var hexToken, octToken, decToken;
+       for (var i = 0; i < tokens.length; i++) {
+               if (tokens[i].name == "HEX_DEC") hexToken = tokens[i];
+               if (tokens[i].name == "OCTAL") octToken = tokens[i];
+               if (tokens[i].name == "DECIMAL") decToken = tokens[i];
+       }
+       ////
+       
+       is(decToken.data, "8.0", "decimal number is found in source.");
+       is(hexToken.data, "0x20", "hexdec number is found in source (issue #99).");
+       is(octToken.data, "0777", "octal number is found in source.");
+*/
+
+/**
+       @returns {Boolean} Was the token found?
+ */
+JSDOC.TokenReader.prototype.read_hex = function(/**JSDOC.TokenStream*/stream, tokens) {
+       var found = stream.next(2);
+       
+       while (!stream.look().eof) {
+               if (JSDOC.Lang.isHexDec(found) && !JSDOC.Lang.isHexDec(found+stream.look())) { // done
+                       tokens.push(new JSDOC.Token(found, "NUMB", "HEX_DEC"));
+                       return true;
+               }
+               else {
+                       found += stream.next();
+               }
+       }
+       return false;
+}
+
+/**
+       @returns {Boolean} Was the token found?
+ */
+JSDOC.TokenReader.prototype.read_regx = function(/**JSDOC.TokenStream*/stream, tokens) {
+       var last;
+       if (
+               stream.look() == "/"
+               && 
+               (
+                       
+                       (
+                               !(last = tokens.lastSym()) // there is no last, the regex is the first symbol
+                               || 
+                               (
+                                          !last.is("NUMB")
+                                       && !last.is("NAME")
+                                       && !last.is("RIGHT_PAREN")
+                                       && !last.is("RIGHT_BRACKET")
+                               )
+                       )
+               )
+       ) {
+               var regex = stream.next();
+               
+               while (!stream.look().eof) {
+                       if (stream.look() == "\\") { // escape sequence
+                               regex += stream.next(2);
+                       }
+                       else if (stream.look() == "/") {
+                               regex += stream.next();
+                               
+                               while (/[gmi]/.test(stream.look())) {
+                                       regex += stream.next();
+                               }
+                               
+                               tokens.push(new JSDOC.Token(regex, "REGX", "REGX"));
+                               return true;
+                       }
+                       else {
+                               regex += stream.next();
+                       }
+               }
+               // error: unterminated regex
+       }
+       return false;
+}