<?php
/***********************************************************************
Class:        Gender Guesser
Version:      v0.02.0
Date:         11/12/2005
Author:       Wudi <hide@address.com>
License:      GPL (GNU General Public License)
PHP Classes:  http://www.phpclasses.org/browse/package/2701.html
Description:  This class can guess the gender by name.
***********************************************************************/

/***********************************************************************
This program is free software; you can redistribute it and/or
modify it under the terms of the GNU General Public License
as published by the Free Software Foundation; either version 2
of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

The GNU General Public License can be found at
http://www.gnu.org/copyleft/gpl.html
***********************************************************************/

class GenderGuesser {
    var $options = array('s' => TRUE);
    var $lexicon_name = 'Unknown';
    var $lexicon_argv = 0;
    var $lexicon_chars = array();

    function setOptions($options) {
        if (is_string($options)) {
            $options = explode(' ', $options);
        }
        if (is_array($options)) {
            foreach ($options as $option) {
                if ($option{0} == '-') {
                    $this->options[$option{1}] = FALSE;
                } else {
                    $this->options[$option{0}] = TRUE;
                }
            }
        } else {
            return FALSE;
        }
    }

    function loadLexicon($lexicon_file) {
        if (!file_exists($lexicon_file)) {
            return FALSE;
        }
        $fp = fopen($lexicon_file, 'r');
        $temp = fgets($fp, 512);
        if ($temp === FALSE) {
            return FALSE;
        } else {
            if (strpos($temp, "\t") === FALSE) {
                return FALSE;
            } else {
                list ($lexicon_type, $lexicon_ver, $lexicon_name) = explode("\t", trim($temp));
                if (($lexicon_type!='LEXICON_CHAR_CS') || ($lexicon_ver!='001')) {
                    return FALSE;
                }
                $this->lexicon_name = $lexicon_name;
            }
        }
        while (!feof($fp)) {
            list ($char, $offset) = split("\t", rtrim(fgets($fp, 32)));
            $this->lexicon_chars[$char] = $offset;
        }
        fclose($fp);
        $this->lexicon_argv = max($this->lexicon_chars);
        return TRUE;
    }

    function getLexiconName() {
        return $this->lexicon_name;
    }

    function getOffset($name) {
        if (count($this->lexicon_chars) == 0) {
            return FALSE;
        }
        $name_length = strlen($name);
        $chars = array();
        for ($i=0; $i<$name_length; $i++) {
            if ((ord($name{$i})>=224) && (ord($name{$i})<=239)) {
                $chars[] = $name{$i} . $name{$i+1} . $name{$i+2};
                $i += 2;
            }
        }
        if ($this->options['s']) {
            if (count($chars) == 2) {
                $chars = array($chars[1]);
            } elseif (count($chars) == 4) {
                $chars = array($chars[2], $chars[3]);
            } else {
                $chars = array($chars[1], $chars[2]);
            }
        } else {
            if (count($chars) > 2) {
                $chars = array($chars[0], $chars[1]);
            }
        }
        $offset = 0;
        if (count($chars) == 1) {
            if (array_key_exists($chars[0], $this->lexicon_chars)) {
                $offset += $this->lexicon_chars[$chars[0]] * 6;
            }
        } else {
            if (array_key_exists($chars[0], $this->lexicon_chars)) {
                $offset += $this->lexicon_chars[$chars[0]] * 4;
            }
            if (array_key_exists($chars[1], $this->lexicon_chars)) {
                $offset += $this->lexicon_chars[$chars[1]] * 5;
            }
        }
        return $offset;
    }

    function getTendencyByOffset($offset) {
        if ($this->lexicon_argv == 0) {
            $tendency = FALSE;
        } else {
            $tendency = $offset / ($this->lexicon_argv * 9);
        }
        return $tendency;
    }

    function getTendency($name) {
        $tendency = $this->getOffset($name);
        return $this->getTendencyByOffset($tendency);
    }
}
?>