在Java中剥离HTML标记

Question

Is there an existing Java library which provides a method to strip all HTML tags from a String? I'm looking for something equivalent to the strip_tags function in PHP.

I know that I can use a regex as described in this Stackoverflow question , however I was curious if there may already be a stripTags() method floating around somewhere in the Apache Commons library that can be used.

Answer 1

Use JSoup , it's well documented, available on Maven and after a day of spending time with several libraries, for me, it is the best one i can imagine.. My own opinion is, that a job like that, parsing html into plain-text, should be possible in one line of code -> otherwise the library has failed somehow... just saying ^^ So here it is, the one-liner of JSoup - in Markdown4J, something like that is not possible, in Markdownj too, in htmlCleaner this is pain in the ass with somewhat about 50 lines of code...

String plain = new HtmlToPlainText().getPlainText(Jsoup.parse(html));

And what you got is real plain-text (not just the html-source-code as a String, like in other libs lol) -> he really does a great job on that. It is more or less the same quality as Markdownify for PHP....

Answer 2

This is what I found on google on it. For me it worked fine.

String noHTMLString = htmlString.replaceAll("\\<.*?\\>", "");

Answer 3

Whatever you do, make sure you normalize the data before you start trying to strip tags. I recently attended a web app security workshop that covered XSS filter evasion. One would normally think that searching for < or < or its hex equivalent would be sufficient. I was blown away after seeing a slide with 70 ways that < can be encoded to beat filters.

Update:

Below is the presentation I was referring to, see slide 26 for the 70 ways to encode < .

Filter Evasion: Houdini on the Wire

Answer 4

There may be some, but the most robust thing is to use an actual HTML parser. There's one here , and if it's reasonably well formed, you can also use SAX or another XML parser.

Answer 5

After having this question open for almost a week, I can say with some certainty that there is no method available in the Java API or Apache libaries which strips HTML tags from a String. You would either have to use an HTML parser as described in the previous answers, or write a simple regular expression to strip out the tags.

Answer 6

When using Jsoup it's even easier than described in above answers:

String html = "bla <b>hehe</b> <br> this is awesome simple";

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

Answer 7

I've used nekoHtml to do that. It can strip all tags but it can just as easily keep or strip a subset of tags.

Answer 8

I know that this question is quite old, but I have been looking for this too and it seems that it is still not easy to find a good and easy solution in java.

Today I came across this little functions lib. It actually attempts to imitate the php strip_tags function.

http://jmelo.lyncode.com/java-strip_tags-php-function/

It works like this (copied from their site):

    import static com.lyncode.jtwig.functions.util.HtmlUtils.stripTags;

    public class StripTagsExample {
      public static void main(String... args) {
        String result = stripTags("<!-- <a href='test'></a>--><a>Test</a>", "");
        // Produced result: Test
      }
    }

Answer 9

Hi I know this thread is old but it still came out tops on Google, and I was looking for a quick fix to the same problem. Couldn't find anything useful so I came up with this code snippet -- hope it helps someone. It just loops over the string and skips all the tags. Plain & simple.

boolean intag = false;
String inp = "<H1>Some <b>HTML</b> <span style=blablabla>text</span>";
String outp = "";

for (int i=0; i < inp.length(); ++i)
{
    if (!intag && inp.charAt(i) == '<')
        {
            intag = true;
            continue;
        }
        if (intag && inp.charAt(i) == '>')
        {
            intag = false;
            continue;
        }
        if (!intag)
        {
            outp = outp + inp.charAt(i);
        }
}   
return outp;

Answer 10

With pure iterative approach and no regex :

public String stripTags(final String html) {

    final StringBuilder sbText = new StringBuilder(1000);
    final StringBuilder sbHtml = new StringBuilder(1000);

    boolean isText = true;

    for (char ch : html.toCharArray()) {
        if (isText) { // outside html
            if (ch != '<') {
                sbText.append(ch);
                continue;
            } else {   // switch mode             
                isText = false;      
                sbHtml.append(ch); 
                continue;
            }
        }else { // inside html
            if (ch != '>') {
                sbHtml.append(ch);
                continue;
            } else {      // switch mode    
                isText = true;     
                sbHtml.append(ch); 
                continue;
            }
        }
    }

    return sbText.toString();
}

Answer 11

Because of abbreviation (string truncation) of html fragment, I had also the problem of unclosed html tags that regex can't detect. Eg:

Lorem ipsum dolor sit amet, <b>consectetur</b> adipiscing elit. <a href="abc"

So, referring to the 2 best answers (JSoup and regex), I preferred solution using JSoup:

Jsoup.parse(html).text()

Answer 12

Wicket uses the following method to escape html, located in: org.apache.wicket.util.string.Strings

public static CharSequence escapeMarkup(final String s, final boolean escapeSpaces,
    final boolean convertToHtmlUnicodeEscapes)
{
    if (s == null)
    {
        return null;
    }
    else
    {
        int len = s.length();
        final AppendingStringBuffer buffer = new AppendingStringBuffer((int)(len * 1.1));

        for (int i = 0; i < len; i++)
        {
            final char c = s.charAt(i);

            switch (c)
            {
                case '\t' :
                    if (escapeSpaces)
                    {
                        // Assumption is four space tabs (sorry, but that's
                        // just how it is!)
                        buffer.append("&nbsp;&nbsp;&nbsp;&nbsp;");
                    }
                    else
                    {
                        buffer.append(c);
                    }
                    break;

                case ' ' :
                    if (escapeSpaces)
                    {
                        buffer.append("&nbsp;");
                    }
                    else
                    {
                        buffer.append(c);
                    }
                    break;

                case '<' :
                    buffer.append("&lt;");
                    break;

                case '>' :
                    buffer.append("&gt;");
                    break;

                case '&' :

                    buffer.append("&amp;");
                    break;

                case '"' :
                    buffer.append("&quot;");
                    break;

                case '\'' :
                    buffer.append("&#039;");
                    break;

                default :

                    if (convertToHtmlUnicodeEscapes)
                    {
                        int ci = 0xffff & c;
                        if (ci < 160)
                        {
                            // nothing special only 7 Bit
                            buffer.append(c);
                        }
                        else
                        {
                            // Not 7 Bit use the unicode system
                            buffer.append("&#");
                            buffer.append(new Integer(ci).toString());
                            buffer.append(';');
                        }
                    }
                    else
                    {
                        buffer.append(c);
                    }

                    break;
            }
        }

        return buffer;
    }
}

Answer 13

public static String stripTags(String str) {
    int startPosition = str.indexOf('<');
    int endPosition;
    while (startPosition != -1) {
        endPosition = str.indexOf('>', startPosition);
        str = str.substring(0, startPosition) + (endPosition != -1 ? str.substring(endPosition + 1) : "");
        startPosition = str.indexOf('<');
    }
    return str;
}