What is the encoding used by AppEngine TaskQueue when they encode byte[] arrays as Strings?

What is the encoding of a HTTP POST body called from the AppEngine TaskQueue service?

If I create a task via TaskOptions#payload(byte[], String), what will the encoding of the HTTP request body be?

Similarly, what will be the encoding of the String created via TaskOptions#param(String, byte[]) and retrieved via ServletRequest#getParameter(String)?

UPDATE: What is the charset name I have to use in

req.getParameter("myParam").getBytes(charset)

to get back the binary data I've开发者_高级运维 submitted via TaskOptions#param(String, byte[]) ?

It seems to be a servlet-container specific default value which is not defined in the definition for 'application/x-www-form-urlencoded' at http://www.w3.org/TR/html4/interact/forms.html#h-17.13.4.1 -- because all that is abstracted away in the servlet API already.

If I create a task via TaskOptions#payload(byte[], String), what will the encoding of the HTTP request body be?

There is no encoding - the byte array you pass in becomes the literal body of the HTTP request.

Similarly, what will be the encoding of the String created via TaskOptions#param(String, byte[]) and retrieved via ServletRequest#getParameter(String)?

Parameters are encoded using formencoding, as in a regular GET or POST request.

On the first one, I have no idea. I'll however do a bet on UTF-8 since the Javadoc mentions UTF-8 everywhere. You could debug the request body by a HTTP debugger tool like Fiddler2. You could test around with strings with UTF-8 specific characters which are transformed to byte array by string.getBytes("UTF-8") and then read it in the servlet side. If it returns the same characters, then the chance is definitely big that it is using UTF-8.

On the second one, that depends on the charset attribute in the Content-Type request header. This is however more than often absent (at least, when a normal webbrowser is used). You can however set it yourself by ServletRequest#setCharacterEncoding() before you access any data from the request body.

if (request.getCharacterEncoding() == null) {
    request.setCharacterEncoding("UTF-8");
}

Otherwise the platform default one will be used, as specified by Charset#defaultCharset().