java字符串利用正則表達式分割

平時用到，整理的，總感覺缺點什么：

private String getKeywordByContent(String content) {
		StringBuffer sbAllKwyword = new StringBuffer();
		content = content.replaceAll(" ", "");//去掉所有空格
		if(content.length()>10){
			String regex = "[\\u4E00-\u9FA5\\s\\d\\w]{10}";
			Pattern p = Pattern.compile(regex);
			Matcher m = p.matcher(content);
			int i=0;
			while (m.find()&&i<10) {
				sbAllKwyword.append(m.group());
				sbAllKwyword.append("^");		
				i++;
			}
			logger.info("sbAllKwyword={}",sbAllKwyword.toString());
			if(sbAllKwyword.length()>1){				
				sbAllKwyword.deleteCharAt(sbAllKwyword.length() - 1);
			}
		}else{
			sbAllKwyword.append(content);
		}
		return sbAllKwyword.toString();
	}

　需要補充的歡迎隨時指出.