JavaフィルタHTMLラベル
1731 ワード
public static String delHTMLTag( String htmlStr )
{
String regEx_script = "<script[^>]*?>[\\s\\S]*?<\\/script>"; // script
String regEx_style = "<style[^>]*?>[\\s\\S]*?<\\/style>"; // style
String regEx_html = "<[^>]+>"; // HTML
String regEx_blank = "\\s*|\t|\r|
"; // , ,
String regEx_nbsp = " ";
Pattern p_script = Pattern.compile( regEx_script, Pattern.CASE_INSENSITIVE );
Matcher m_script = p_script.matcher( htmlStr );
htmlStr = m_script.replaceAll( "" ); // script
Pattern p_style = Pattern.compile( regEx_style, Pattern.CASE_INSENSITIVE );
Matcher m_style = p_style.matcher( htmlStr );
htmlStr = m_style.replaceAll( "" ); // style
Pattern p_html = Pattern.compile( regEx_html, Pattern.CASE_INSENSITIVE );
Matcher m_html = p_html.matcher( htmlStr );
htmlStr = m_html.replaceAll( "" ); // html
System.out.println("before :");
System.out.println(htmlStr);
Pattern p_nbsp = Pattern.compile( regEx_nbsp, Pattern.CASE_INSENSITIVE );
Matcher m_nbsp = p_nbsp.matcher( htmlStr );
htmlStr = m_nbsp.replaceAll(""); //
System.out.println("after :");
System.out.println(htmlStr);
Pattern p_blank = Pattern.compile( regEx_nbsp, Pattern.CASE_INSENSITIVE );
Matcher m_blank = p_blank.matcher( htmlStr );
htmlStr = m_blank.replaceAll(""); //
return htmlStr.trim(); //
}