private function _ignore_stopword($output)
{
$ignore = $this->ignore;
$charset = $this->charset;
$dict = $this->dict;
$rule = $this->rule;
//清理標點符號
if ($ignore == true) {
$lines = explode("\n", file_get_contents($rule));
$start_record = false;
$start_split = false;
$stop_words = "";
for ($i = 0; $i < count($lines); $i++) {
$l = trim($lines[$i]);
if ($l == "[nostats]") {
$start_record = true;
continue;
}
if ($l == "[noname]") {
$start_split = true;
continue;
}
if ($l == '[symbol]') {
continue;
}
if (substr($l, 0, 1) == ":" || substr($l, 0, 1) == ";") {
continue;
}
if ($l == "[pubname]") {
break;
}
if ($start_record == true) {
$data = $lines[$i];
if ($start_split == true) {
$data_ary = Segmentor::utf8_str_split($data);
$data = implode(" ", $data_ary);
}
$stop_words .= $data . " ";
}
}
//把字串切割吧
//$stop_words_ary = utf8_str_split($stop_words_ary);
$stop_words_ary = explode(" ", $stop_words);
//檢查每一個單一的字，是否有符合此字串
$words_array = explode(" ", $output);
$output = '';
for ($i = 0; $i < count($words_array); $i++) {
$match = false;
for ($j = 0; $j < count($stop_words_ary); $j++) {
$word = $words_array[$i];
$index = strrpos($word, Segmentor::$speech_separator);
if (FALSE !== $index) {
$word = substr($word, 0, $index);
}
if ($word == $stop_words_ary[$j]) {
$match = true;
break;
}
}
if ($match == false) {
$output .= $words_array[$i] . " ";
}
}
}
$output = trim($output);
return $output;
}
