CODEWITHSUNDEEP
javaurldecoding
Daniel Rotter
Daniel Rotter

Reputation: 2056

Java URLDecoder throws exception when used with a string containing a %

I have a problem with the URLDecoder of Java. I am escaping a String in JavaScript, and send it to a java servlet. Then I decode the escaped String with the following line:

URLDecoder.decode(request.getParameter("text"), "UTF-8");

This works fine for every special characters I have tried, the only one making problems is the '%'. Everytime I use this character in the string, I get the following exception:

java.lang.IllegalArgumentException: URLDecoder: Incomplete trailing escape (%) pattern
    java.net.URLDecoder.decode(URLDecoder.java:187)
    at.fhv.students.rotter.ajax.count.CountServlet.doGet(CountServlet.java:31)
    javax.servlet.http.HttpServlet.service(HttpServlet.java:621)
    javax.servlet.http.HttpServlet.service(HttpServlet.java:722)

Is this a known bug? Or is it really my mistake?

Upvotes: 11

Views: 30210

Answers (4)

dubet
dubet

Reputation: 1

Even I faced similar issue and it was solved. Following is the example code you can simply run to reproduce and resolve this issue.

public class TestPercentage {

    public static void main(String[] args) {
        // TODO Auto-generated method stub
        String transResult = "Se si utilizza DHCP%2C i valori validi sono S%C3%AC o No.%24%23%24%23%24%23%25NICyUSEWINS%25%24%23%24%23%24%23Se si utilizza WINS%2C i valori validi sono S%C3%AC o No.%24%23%24%23%24%23%25NODEFULL%25%24%23%24%23%24%23Nome completo del computer%24%23%24%23%24%23%25NODENAME%25%24%23%24%23%24%23I primi 8 caratteri del nome effettivo del computer%24%23%24%23%24%23%25NWCONTEXT%25%24%23%24%23%24%23Nome contesto NetWare%24%23%24%23%24%23%25NWSERVER%25%";
        String decode = null;
        try {
            decode = URLDecoder.decode(transResult, "UTF-8");
        } catch (UnsupportedEncodingException ue) {
            System.out.println("UnsupportedEncodingException ! = " + ue);
        } catch (IllegalArgumentException ile) {
            System.out.println("IllegalArgumentException ! = " + ile);
            if (transResult.endsWith("%")) {
                transResult = transResult.substring(0, transResult.lastIndexOf("%"));
                System.out.println("transResult2 = " + transResult);
                try {
                decode = URLDecoder.decode(transResult, "UTF-8");
                } catch (UnsupportedEncodingException ue2) {
                    System.out.println("UnsupportedEncodingException 2 = " + ue2);
                } catch (IllegalArgumentException ile2) {
                    System.out.println("IllegalArgumentException ! = " + ile2);
                }
            }
        }
        System.out.println("decode = " + decode);
    }

}

Upvotes: 0

ravindra nath
ravindra nath

Reputation: 31

We had a similar issue in our angular application where we were encoding % sign once in client side code. When we received the value in servlet it was already decoded due to request.getParameter(). Since we already had URL decoder in our sever side code, decoding the % sign twice was causing a "URLDecoder: Incomplete trailing escape (%) pattern" exception. We figured out that we we should not encode and decode % as a value at all to get face this issue.

Upvotes: 2

Dipendra Shrestha
Dipendra Shrestha

Reputation: 1

In order to get parameter I have written

String requestURL=request.getQueryString();

so that It will give us parameters. From that we can use String.substring() to get prefered parameter in case of fix length or single parameter. Then 

String decodeValue = URLDecoder.decode(value,"UTF-8");

will get preferred string encoded % sign too.

Upvotes: 0

Christian Kuetbach
Christian Kuetbach

Reputation: 16060

It is not a bug. You send a wrong encoded String. The %-sign has to be encoded as %25

If you call request.getParameter(), I think you get a decoded String.

Upvotes: 10

Related Questions