•     
            
Delaware       VIP® Diversified Income Series       (Service Class): Maximum long-term total return consistent with reasonable       risk. 
       
         
            
•     
            
Delaware       VIP® Emerging Markets Series (Service       Class): Long-term capital appreciation. 
       
         
            
•     
            
Delaware       VIP® Limited-Term Diversified Income       Series (Service Class): Maximum total return, consistent with reasonable       risk. 
       
         
            
•     
            
Delaware       VIP® REIT Series (Service Class):       Maximum long-term total return, with capital appreciation as a secondary       objective. 
       
         
            
•     
            
Delaware       VIP® Small Cap Value Series (Service       Class): Capital appreciation. 
       
         
            
•     
            
Delaware       VIP® Smid Cap Core Series (Service       Class): Long-term capital appreciation. 
       
         
            
•     
            
Delaware       VIP® U.S. Growth Series (Service       Class): Long-term capital appreciation. 
       
         
            
•     
            
Delaware       VIP® Value Series (Service Class):       Long-term capital appreciation. 
       
      
     

在上面的Html表格中，我们要抓出每个tr标签中第2个td的文本内容，一开始想到的XPATH语句是这么写的：

//td[contains(text(),':') and contains(text(),'(') and contains(text(),')') and (contains(text(),'Class') or contains(text(),'Shares'))]

结果发现提不出来，将text()函数改为string()函数，就可以提出来了：

//td[contains(string(),':') and contains(string(),'(') and contains(string(),')') and (contains(string(),'Class') or contains(string(),'Shares'))]

原文档中有些td标签文本有换行，而且可能还夹杂着其他子标签，这时候可能用text()提取不出来，可以改用string()，string()可以将所有子标签中的文本串成一起提出来，可以满足绝大部分时候的需求。