使用PHP缩短文本后，字符解码失败

我在缩短文本时遇到了一些问题。我使用PHP类来缩短文本并防止HTML标记被破坏。缩短文本的代码行：

Html::trim(html_entity_decode($item["content"]), 500)

我从中得到的输出为我提供了一些奇怪的字符，其中应该放置特殊字符。例如：

ï

用于缩短文本的Html类：

<?php
class Html{

  protected
    $reachedLimit = false,
    $totalLen     = 0,
    $maxLen       = 25,
    $toRemove     = array();

  public static function trim($html, $maxLen = 25){

    $dom = new DomDocument();
    $dom->loadHTML($html);

    $html = new static();
    $toRemove = $html->walk($dom, $maxLen);

    // remove any nodes that passed our limit
    foreach($toRemove as $child) 
      $child->parentNode->removeChild($child);

    // remove wrapper tags added by DD (doctype, html...)
    if(version_compare(PHP_VERSION, '5.3.6') < 0){
      // http://stackoverflow.com/a/6953808/1058140
      $dom->removeChild($dom->firstChild);            
      $dom->replaceChild($dom->firstChild->firstChild->firstChild, $dom->firstChild);
      return $dom->saveHTML();
    }

    return $dom->saveHTML($dom->getElementsByTagName('body')->item(0));   
  }

  protected function walk(DomNode $node, $maxLen){

    if($this->reachedLimit){
      $this->toRemove[] = $node;

    }else{
      // only text nodes should have text,
      // so do the splitting here
      if($node instanceof DomText){
        $this->totalLen += $nodeLen = strlen($node->nodeValue);

        // use mb_strlen / mb_substr for UTF-8 support
        if($this->totalLen > $maxLen){
          $node->nodeValue = substr($node->nodeValue, 0, $nodeLen - ($this->totalLen - $maxLen)) . '...';
          $this->reachedLimit = true;
        }

      }

      // if node has children, walk its child elements 
      if(isset($node->childNodes))
        foreach($node->childNodes as $child)
          $this->walk($child, $maxLen);
    }  

    return $this->toRemove;
  }  
}
?>

---

如何正确显示特殊字符？我为字符解码设置了UTF-8。

尝试使用多字节安全功能

mb_substr

而不是

substr