Java HTML Parse Jsoup getDoctypeName(InputStream s)

Description

get Doctype Name

License

Open Source License

Declaration

static String getDoctypeName(InputStream s) throws IOException 

Method Source Code

//package com.java2s;
/*/*from w w  w  .j av  a  2 s  .  c  o  m*/
   Copyright 2012-2018 Michael Pozhidaev <michael.pozhidaev@gmail.com>
    
   This file is part of LUWRAIN.
    
   LUWRAIN is free software; you can redistribute it and/or
   modify it under the terms of the GNU General Public
   License as published by the Free Software Foundation; either
   version 3 of the License, or (at your option) any later version.
    
   LUWRAIN is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
   General Public License for more details.
*/

import java.util.*;
import java.io.*;

public class Main {
    static String getDoctypeName(InputStream s) throws IOException {
        final org.jsoup.nodes.Document doc = org.jsoup.Jsoup.parse(s, "us-ascii", "",
                org.jsoup.parser.Parser.xmlParser());
        List<org.jsoup.nodes.Node> nods = doc.childNodes();
        for (org.jsoup.nodes.Node node : nods)
            if (node instanceof org.jsoup.nodes.DocumentType) {
                org.jsoup.nodes.DocumentType documentType = (org.jsoup.nodes.DocumentType) node;
                final String res = documentType.attr("name");
                if (res != null)
                    return res;
            }
        for (org.jsoup.nodes.Node node : nods)
            if (node instanceof org.jsoup.nodes.Element) {
                org.jsoup.nodes.Element el = (org.jsoup.nodes.Element) node;
                final String res = el.tagName();
                if (res != null)
                    return res;
            }
        return "";
    }
}

Related

1. getDistinctImageUrls(String htmlContent)
2. getDoc(Connection conn)
3. getDoc(File file)
4. getDoc(String path)
5. getDoc(String url)
6. getErrorMessage(String htmlStr)
7. getExplanation(String html)
8. getFirstImageSrc(String html)
9. getFirstSentence(final String html)