/* vim: set sw=4 ts=4 et tw=78: */
/* ***** BEGIN LICENSE BLOCK *****
 * Version: MPL 1.1/GPL 2.0/LGPL 2.1
 *
 * The contents of this file are subject to the Mozilla Public License Version
 * 1.1 (the "License"); you may not use this file except in compliance with
 * the License. You may obtain a copy of the License at
 * http://www.mozilla.org/MPL/
 *
 * Software distributed under the License is distributed on an "AS IS" basis,
 * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
 * for the specific language governing rights and limitations under the
 * License.
 *
 * The Original Code is the Narcissus JavaScript engine.
 *
 * The Initial Developer of the Original Code is
 * Brendan Eich <brendan@mozilla.org>.
 * Portions created by the Initial Developer are Copyright (C) 2004
 * the Initial Developer. All Rights Reserved.
 *
 * Contributor(s):
 *   Tom Austin <taustin@ucsc.edu>
 *   Brendan Eich <brendan@mozilla.org>
 *   Shu-Yu Guo <shu@rfrn.org>
 *   Dave Herman <dherman@mozilla.com>
 *   Dimitris Vardoulakis <dimvar@ccs.neu.edu>
 *   Patrick Walton <pcwalton@mozilla.com>
 *
 * Alternatively, the contents of this file may be used under the terms of
 * either the GNU General Public License Version 2 or later (the "GPL"), or
 * the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
 * in which case the provisions of the GPL or the LGPL are applicable instead
 * of those above. If you wish to allow use of your version of this file only
 * under the terms of either the GPL or the LGPL, and not to allow others to
 * use your version of this file under the terms of the MPL, indicate your
 * decision by deleting the provisions above and replace them with the notice
 * and other provisions required by the GPL or the LGPL. If you do not delete
 * the provisions above, a recipient may use your version of this file under
 * the terms of any one of the MPL, the GPL or the LGPL.
 *
 * ***** END LICENSE BLOCK ***** */

/*
 * Narcissus - JS implemented in JS.
 *
 * Lexical scanner.
 */

Narcissus.lexer = (function() {

    var definitions = Narcissus.definitions;

    // Set constants in the local scope.
    eval(definitions.consts);

    // Build up a trie of operator tokens.
    var opTokens = {};
    for (var op in definitions.opTypeNames) {
        if (op === '\n' || op === '.')
            continue;

        var node = opTokens;
        for (var i = 0; i < op.length; i++) {
            var ch = op[i];
            if (!(ch in node))
                node[ch] = {};
            node = node[ch];
            node.op = op;
        }
    }

    /*
     * Tokenizer :: (source, filename, line number) -> Tokenizer
     */
    function Tokenizer(s, f, l) {
        this.cursor = 0;
        this.source = String(s);
        this.tokens = [];
        this.tokenIndex = 0;
        this.lookahead = 0;
        this.scanNewlines = false;
        this.unexpectedEOF = false;
        this.filename = f || "";
        this.lineno = l || 1;
    }

    Tokenizer.prototype = {
        get done() {
            // We need to set scanOperand to true here because the first thing
            // might be a regexp.
            return this.peek(true) === END;
        },

        get token() {
            return this.tokens[this.tokenIndex];
        },

        match: function (tt, scanOperand) {
            return this.get(scanOperand) === tt || this.unget();
        },

        mustMatch: function (tt) {
            if (!this.match(tt)) {
                throw this.newSyntaxError("Missing " +
                                          definitions.tokens[tt].toLowerCase());
            }
            return this.token;
        },

        forceIdentifier: function() {
        	if (!this.match(IDENTIFIER)) {
        		// keywords are valid property names in ES 5
        		if (this.get() >= definitions.keywords[0] || this.unget) {
        			this.token.type = IDENTIFIER;
        		}
        		else {
        			throw this.newSyntaxError("Missing identifier");
        		}
        	}
        	return this.token;
        },

        peek: function (scanOperand) {
            var tt, next;
            if (this.lookahead) {
                next = this.tokens[(this.tokenIndex + this.lookahead) & 3];
                tt = (this.scanNewlines && next.lineno !== this.lineno)
                     ? NEWLINE
                     : next.type;
            } else {
                tt = this.get(scanOperand);
                this.unget();
            }
            return tt;
        },

        peekOnSameLine: function (scanOperand) {
            this.scanNewlines = true;
            var tt = this.peek(scanOperand);
            this.scanNewlines = false;
            return tt;
        },

        // Eat comments and whitespace.
        skip: function () {
            var input = this.source;
            for (;;) {
                var ch = input[this.cursor++];
                var next = input[this.cursor];
                if (ch === '\n' && !this.scanNewlines) {
                    this.lineno++;
                } else if (ch === '/' && next === '*') {
                    this.cursor++;
                    for (;;) {
                        ch = input[this.cursor++];
                        if (ch === undefined)
                            throw this.newSyntaxError("Unterminated comment");

                        if (ch === '*') {
                            next = input[this.cursor];
                            if (next === '/') {
                                this.cursor++;
                                break;
                            }
                        } else if (ch === '\n') {
                            this.lineno++;
                        }
                    }
                } else if (ch === '/' && next === '/') {
                    this.cursor++;
                    for (;;) {
                        ch = input[this.cursor++];
                        if (ch === undefined)
                            return;

                        if (ch === '\n') {
                            this.lineno++;
                            break;
                        }
                    }
                } else if (ch !== ' ' && ch !== '\t') {
                    this.cursor--;
                    return;
                }
            }
        },

        // Lex the exponential part of a number, if present. Return true iff an
        // exponential part was found.
        lexExponent: function() {
            var input = this.source;
            var next = input[this.cursor];
            if (next === 'e' || next === 'E') {
                this.cursor++;
                ch = input[this.cursor++];
                if (ch === '+' || ch === '-')
                    ch = input[this.cursor++];

                if (ch < '0' || ch > '9')
                    throw this.newSyntaxError("Missing exponent");

                do {
                    ch = input[this.cursor++];
                } while (ch >= '0' && ch <= '9');
                this.cursor--;

                return true;
            }

            return false;
        },

        lexZeroNumber: function (ch) {
            var token = this.token, input = this.source;
            token.type = NUMBER;

            ch = input[this.cursor++];
            if (ch === '.') {
                do {
                    ch = input[this.cursor++];
                } while (ch >= '0' && ch <= '9');
                this.cursor--;

                this.lexExponent();
                token.value = parseFloat(input.substring(token.start, this.cursor));
            } else if (ch === 'x' || ch === 'X') {
                do {
                    ch = input[this.cursor++];
                } while ((ch >= '0' && ch <= '9') || (ch >= 'a' && ch <= 'f') ||
                         (ch >= 'A' && ch <= 'F'));
                this.cursor--;

                token.value = parseInt(input.substring(token.start, this.cursor));
            } else if (ch >= '0' && ch <= '7') {
                do {
                    ch = input[this.cursor++];
                } while (ch >= '0' && ch <= '7');
                this.cursor--;

                token.value = parseInt(input.substring(token.start, this.cursor));
            } else {
                this.cursor--;
                this.lexExponent();     // 0E1, &c.
                token.value = 0;
            }
        },

        lexNumber: function (ch) {
            var token = this.token, input = this.source;
            token.type = NUMBER;

            var floating = false;
            do {
                ch = input[this.cursor++];
                if (ch === '.' && !floating) {
                    floating = true;
                    ch = input[this.cursor++];
                }
            } while (ch >= '0' && ch <= '9');

            this.cursor--;

            var exponent = this.lexExponent();
            floating = floating || exponent;

            var str = input.substring(token.start, this.cursor);
            token.value = floating ? parseFloat(str) : parseInt(str);
        },

        lexDot: function (ch) {
            var token = this.token, input = this.source;
            var next = input[this.cursor];
            if (next >= '0' && next <= '9') {
                do {
                    ch = input[this.cursor++];
                } while (ch >= '0' && ch <= '9');
                this.cursor--;

                this.lexExponent();

                token.type = NUMBER;
                token.value = parseFloat(input.substring(token.start, this.cursor));
            } else {
                token.type = DOT;
                token.assignOp = null;
                token.value = '.';
            }
        },

        lexString: function (ch) {
            var token = this.token, input = this.source;
            token.type = STRING;

            var hasEscapes = false;
            var delim = ch;
            while ((ch = input[this.cursor++]) !== delim) {
                if (this.cursor == input.length)
                    throw this.newSyntaxError("Unterminated string literal");
                if (ch === '\\') {
                    hasEscapes = true;
                    if (input[this.cursor] === '\n') this.lineno++; // fix for escaped newline
                    if (++this.cursor == input.length)
                        throw this.newSyntaxError("Unterminated string literal");
                }
            }

            token.value = hasEscapes
                          ? eval(input.substring(token.start, this.cursor))
                          : input.substring(token.start + 1, this.cursor - 1);
        },

        lexRegExp: function (ch) {
            var token = this.token, input = this.source;
            token.type = REGEXP;

            do {
                ch = input[this.cursor++];
                if (ch === '\\') {
                    this.cursor++;
                } else if (ch === '[') {
                    do {
                        if (ch === undefined)
                            throw this.newSyntaxError("Unterminated character class");

                        if (ch === '\\')
                            this.cursor++;

                        ch = input[this.cursor++];
                    } while (ch !== ']');
                } else if (ch === undefined) {
                    throw this.newSyntaxError("Unterminated regex");
                }
            } while (ch !== '/');

            do {
                ch = input[this.cursor++];
            } while (ch >= 'a' && ch <= 'z');

            this.cursor--;

            token.value = eval(input.substring(token.start, this.cursor));
        },

        lexOp: function (ch) {
            var token = this.token, input = this.source;

            // A bit ugly, but it seems wasteful to write a trie lookup routine
            // for only 3 characters...
            var node = opTokens[ch];
            var next = input[this.cursor];
            if (next in node) {
                node = node[next];
                this.cursor++;
                next = input[this.cursor];
                if (next in node) {
                    node = node[next];
                    this.cursor++;
                    next = input[this.cursor];
                }
            }

            var op = node.op;
            if (definitions.assignOps[op] && input[this.cursor] === '=') {
                this.cursor++;
                token.type = ASSIGN;
                token.assignOp = definitions.tokenIds[definitions.opTypeNames[op]];
                op += '=';
            } else {
                token.type = definitions.tokenIds[definitions.opTypeNames[op]];
                token.assignOp = null;
            }

            token.value = op;
        },

        // FIXME: Unicode escape sequences
        // FIXME: Unicode identifiers
        lexIdent: function (ch) {
            var token = this.token, input = this.source;

            do {
                ch = input[this.cursor++];
            } while ((ch >= 'a' && ch <= 'z') || (ch >= 'A' && ch <= 'Z') ||
                     (ch >= '0' && ch <= '9') || ch === '$' || ch === '_');

            this.cursor--;  // Put the non-word character back.

            var id = input.substring(token.start, this.cursor);
            token.type = definitions.keywords[id] || IDENTIFIER;
            token.value = id;
        },

        /*
         * Tokenizer.get :: void -> token type
         *
         * Consume input *only* if there is no lookahead.
         * Dispatch to the appropriate lexing function depending on the input.
         */
        get: function (scanOperand) {
            var token;
            while (this.lookahead) {
                --this.lookahead;
                this.tokenIndex = (this.tokenIndex + 1) & 3;
                token = this.tokens[this.tokenIndex];
                if (token.type !== NEWLINE || this.scanNewlines)
                    return token.type;
            }

            this.skip();

            this.tokenIndex = (this.tokenIndex + 1) & 3;
            token = this.tokens[this.tokenIndex];
            if (!token)
                this.tokens[this.tokenIndex] = token = {};

            var input = this.source;
            if (this.cursor === input.length)
                return token.type = END;

            token.start = this.cursor;
            token.lineno = this.lineno;

            var ch = input[this.cursor++];
            if ((ch >= 'a' && ch <= 'z') || (ch >= 'A' && ch <= 'Z') || ch === '$' || ch === '_') {
                this.lexIdent(ch);
            } else if (scanOperand && ch === '/') {
                this.lexRegExp(ch);
            } else if (ch in opTokens) {
                this.lexOp(ch);
            } else if (ch === '.') {
                this.lexDot(ch);
            } else if (ch >= '1' && ch <= '9') {
                this.lexNumber(ch);
            } else if (ch === '0') {
                this.lexZeroNumber(ch);
            } else if (ch === '"' || ch === "'") {
                this.lexString(ch);
            } else if (this.scanNewlines && ch === '\n') {
                token.type = NEWLINE;
                token.value = '\n';
                this.lineno++;
            } else {
                throw this.newSyntaxError("Illegal token");
            }

            token.end = this.cursor;
            return token.type;
        },

        /*
         * Tokenizer.unget :: void -> undefined
         *
         * Match depends on unget returning undefined.
         */
        unget: function () {
            if (++this.lookahead === 4) throw "PANIC: too much lookahead!";
            this.tokenIndex = (this.tokenIndex - 1) & 3;
        },

        newSyntaxError: function (m) {
            var e = new SyntaxError(this.filename + ":" + this.lineno + ":" + m);
            e.source = this.source;
            e.cursor = this.lookahead
                       ? this.tokens[(this.tokenIndex + this.lookahead) & 3].start
                       : this.cursor;
            return e;
        },
    };

    return { Tokenizer: Tokenizer };

}());