这里有一个稍微更充实的更新，试图处理一些格式的休息和列表。我用Amaya的输出作为指导。

import java.io.IOException;
import java.io.Reader;
import java.io.StringReader;
import java.util.Stack;
import java.util.logging.Logger;

import javax.swing.text.MutableAttributeSet;
import javax.swing.text.html.HTML;
import javax.swing.text.html.HTMLEditorKit;
import javax.swing.text.html.parser.ParserDelegator;

public class HTML2Text extends HTMLEditorKit.ParserCallback {
    private static final Logger log = Logger
            .getLogger(Logger.GLOBAL_LOGGER_NAME);

    private StringBuffer stringBuffer;

    private Stack<IndexType> indentStack;

    public static class IndexType {
        public String type;
        public int counter; // used for ordered lists

        public IndexType(String type) {
            this.type = type;
            counter = 0;
        }
    }

    public HTML2Text() {
        stringBuffer = new StringBuffer();
        indentStack = new Stack<IndexType>();
    }

    public static String convert(String html) {
        HTML2Text parser = new HTML2Text();
        Reader in = new StringReader(html);
        try {
            // the HTML to convert
            parser.parse(in);
        } catch (Exception e) {
            log.severe(e.getMessage());
        } finally {
            try {
                in.close();
            } catch (IOException ioe) {
                // this should never happen
            }
        }
        return parser.getText();
    }

    public void parse(Reader in) throws IOException {
        ParserDelegator delegator = new ParserDelegator();
        // the third parameter is TRUE to ignore charset directive
        delegator.parse(in, this, Boolean.TRUE);
    }

    public void handleStartTag(HTML.Tag t, MutableAttributeSet a, int pos) {
        log.info("StartTag:" + t.toString());
        if (t.toString().equals("p")) {
            if (stringBuffer.length() > 0
                    && !stringBuffer.substring(stringBuffer.length() - 1)
                            .equals("\n")) {
                newLine();
            }
            newLine();
        } else if (t.toString().equals("ol")) {
            indentStack.push(new IndexType("ol"));
            newLine();
        } else if (t.toString().equals("ul")) {
            indentStack.push(new IndexType("ul"));
            newLine();
        } else if (t.toString().equals("li")) {
            IndexType parent = indentStack.peek();
            if (parent.type.equals("ol")) {
                String numberString = "" + (++parent.counter) + ".";
                stringBuffer.append(numberString);
                for (int i = 0; i < (4 - numberString.length()); i++) {
                    stringBuffer.append(" ");
                }
            } else {
                stringBuffer.append("*   ");
            }
            indentStack.push(new IndexType("li"));
        } else if (t.toString().equals("dl")) {
            newLine();
        } else if (t.toString().equals("dt")) {
            newLine();
        } else if (t.toString().equals("dd")) {
            indentStack.push(new IndexType("dd"));
            newLine();
        }
    }

    private void newLine() {
        stringBuffer.append("\n");
        for (int i = 0; i < indentStack.size(); i++) {
            stringBuffer.append("    ");
        }
    }

    public void handleEndTag(HTML.Tag t, int pos) {
        log.info("EndTag:" + t.toString());
        if (t.toString().equals("p")) {
            newLine();
        } else if (t.toString().equals("ol")) {
            indentStack.pop();
            ;
            newLine();
        } else if (t.toString().equals("ul")) {
            indentStack.pop();
            ;
            newLine();
        } else if (t.toString().equals("li")) {
            indentStack.pop();
            ;
            newLine();
        } else if (t.toString().equals("dd")) {
            indentStack.pop();
            ;
        }
    }

    public void handleSimpleTag(HTML.Tag t, MutableAttributeSet a, int pos) {
        log.info("SimpleTag:" + t.toString());
        if (t.toString().equals("br")) {
            newLine();
        }
    }

    public void handleText(char[] text, int pos) {
        log.info("Text:" + new String(text));
        stringBuffer.append(text);
    }

    public String getText() {
        return stringBuffer.toString();
    }

    public static void main(String args[]) {
        String html = "<html><body><p>paragraph at start</p>hello<br />What is happening?<p>this is a<br />mutiline paragraph</p><ol>  <li>This</li>  <li>is</li>  <li>an</li>  <li>ordered</li>  <li>list    <p>with</p>    <ul>      <li>another</li>      <li>list        <dl>          <dt>This</dt>          <dt>is</dt>            <dd>sdasd</dd>            <dd>sdasda</dd>            <dd>asda              <p>aasdas</p>            </dd>            <dd>sdada</dd>          <dt>fsdfsdfsd</dt>        </dl>        <dl>          <dt>vbcvcvbcvb</dt>          <dt>cvbcvbc</dt>            <dd>vbcbcvbcvb</dd>          <dt>cvbcv</dt>          <dt></dt>        </dl>        <dl>          <dt></dt>        </dl></li>      <li>cool</li>    </ul>    <p>stuff</p>  </li>  <li>cool</li></ol><p></p></body></html>";
        System.out.println(convert(html));
    }
}

使用HTML解析器而不是正则表达式。这对于Jsoup来说非常简单。

public static String html2text(String html) {
    return Jsoup.parse(html).text();
}

Jsoup还支持根据可定制的白名单删除HTML标记，如果您只想允许<b>， <i>和<u>，这是非常有用的。

参见:

RegEx匹配打开标记，但XHTML自包含标记除外 主要的Java HTML解析器的优缺点是什么? JSP/Servlet web应用中的跨站防护

使用JSoup保留新行信息的一种方法是在所有新行标记之前使用一些虚拟字符串，执行JSoup并将虚拟字符串替换为“\n”。

String html = "<p>Line one</p><p>Line two</p>Line three<br/>etc.";
String NEW_LINE_MARK = "NEWLINESTART1234567890NEWLINEEND";
for (String tag: new String[]{"</p>","<br/>","</h1>","</h2>","</h3>","</h4>","</h5>","</h6>","</li>"}) {
    html = html.replace(tag, NEW_LINE_MARK+tag);
}

String text = Jsoup.parse(html).text();

text = text.replace(NEW_LINE_MARK + " ", "\n\n");
text = text.replace(NEW_LINE_MARK, "\n\n");

您可能希望在剥离HTML之前将<br/>和</p>标记替换为换行符，以防止它像Tim建议的那样变得难以识别。

我能想到的唯一方法是删除HTML标记，但在尖括号之间留下非HTML标记，将检查HTML标记列表。沿着这条线……

replaceAll("\\<[\s]*tag[^>]*>","")

然后html解码特殊字符，如&结果不应该被认为是消毒的。

也可以使用Apache Tika来实现这个目的。默认情况下，它保留了被剥离的html中的空白，这在某些情况下可能是需要的:

InputStream htmlInputStream = ..
HtmlParser htmlParser = new HtmlParser();
HtmlContentHandler htmlContentHandler = new HtmlContentHandler();
htmlParser.parse(htmlInputStream, htmlContentHandler, new Metadata())
System.out.println(htmlContentHandler.getBodyText().trim())

试试javascript:

const strippedString = htmlString.replace(/(<([^>]+)>)/gi, "");
console.log(strippedString);