HttpRequestBodyToParamMap uses java.net.URLDecoder. But URLDecoder sometimes can't decode query string which encoded by URLCodec.

• User impact:

  Medium
• Similar Issues:

  None

Hi.

At first, please check out MULE-4911.

I found a one more problem in HttpRequestBodyToParamMap class.

HttpRequestBodyToParamMap uses java.net.URLDecoder.
But when multibytes string is used, URLDecoder sometimes can't decode query string which encoded by URLCodec.

For example, Japanese Hiragana characher "\u30a8" is encoded to...
=> "%83%47" (by URLEncoder)
=> "%83G" (by URLCodec)

===== sample code =====
import java.net.URLDecoder;
import java.net.URLEncoder;

import org.apache.commons.codec.net.URLCodec;
import org.junit.Assert;
import org.junit.Test;

public class URLDecoderTest {

private static final String encoding = "Windows-31J";
private static final String value = "\u30a8"; // a Japanese hiragana character

@Test
public void testDecode() throws Exception { System.out.println(URLEncoder.encode(value, encoding)); System.out.println(new URLCodec().encode(value, encoding)); Assert.assertEquals(value, URLDecoder.decode(new URLCodec().encode(value, encoding), encoding)); }

}
===== sample code =====

And according to RFC3986 (Sec 2.3)...
> For consistency, percent-encoded octets in the ranges of ALPHA
> (%41-%5A and %61-%7A), DIGIT (%30-%39), hyphen (%2D), period (%2E),
> underscore (%5F), or tilde (%7E) should not be created by URI
> producers

In the case of "\u30a8", "%83G"(by URLCodec) is preferred result. (correct?)

But URLDecoder can't decode "%83G" to "\u30a8". and URLCodec#decode() can.
So I think that URLCodec#docode() should be used here instead of URLDecoder.

Here is a patch for this problem and functional tests.
(this patch includes all the content of the patch for MULE-4911)

best regards.