org.apache.stanbol.entityhub.jersey.utils

Class MessageBodyReaderUtils

java.lang.Object

org.apache.stanbol.entityhub.jersey.utils.MessageBodyReaderUtils

public final class MessageBodyReaderUtils
extends Object

Utilities for implementing MessageBodyReader.

Author:

Rupert Westenthaler

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	MessageBodyReaderUtils.RequestData Simple class that holds the MediaType, Name and the content as InputStream.

Method Summary

Methods

Modifier and Type	Method and Description
static MessageBodyReaderUtils.RequestData	formForm(InputStream formData, String charset, String encodingParam, List<String> contentParams) Returns content parsed as MediaType.APPLICATION_FORM_URLENCODED.
static List<MessageBodyReaderUtils.RequestData>	fromMultipart(InputStream mimeData, javax.ws.rs.core.MediaType mediaType) Returns content parsed from MediaType.MULTIPART_FORM_DATA.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

formForm

public static MessageBodyReaderUtils.RequestData formForm(InputStream formData,
                                          String charset,
                                          String encodingParam,
                                          List<String> contentParams)
                                                   throws IOException,
                                                          UnsupportedEncodingException,
                                                          IllegalArgumentException

Returns content parsed as MediaType.APPLICATION_FORM_URLENCODED. It assumes that the encoding and the content is defined by own parameters. For the content this method allows to parse several parameters. The first existing one is used to get the content. The parameter actually used to retrieve the content will be available via MessageBodyReaderUtils.RequestData.getName().

This Method will load the content several time into memory and should not be used for big contents. However this should be fine in cases data are parsed as MediaType.APPLICATION_FORM_URLENCODED

This Method is necessary because within MessageBodyReader one can not use the FormParam annotations because only the InputStream is parsed to the MessageBodyReader.readFrom(Class, Type, java.lang.annotation.Annotation[], MediaType, javax.ws.rs.core.MultivaluedMap, InputStream) method

To test this Method with curl use:

 curl -v -X POST --data-urlencode "{encodingParam}=application/rdf+xml" 
                 --data-urlencode "{contentParam}@{datafile}" 
                  {serviceURL}
 

Note that between {contentParam} and the datafile MUST NOT be a '='!

Parameters:

formData - the data of the form as stream

charset - the charset used for the form data

encodingParam - the parameter name used to parse the encoding

contentParams - the list of parameters used for the content. The first existing parameter is used to parse the content. Additional ones are ignored.

Returns:

The parsed content (MediaType and InputStream)

Throws:

IOException - On any exception while reading from the parsed stream.

UnsupportedEncodingException - if the parsed charset is not supported by this plattform

IllegalArgumentException - In case of a Response.Status.BAD_REQUEST

fromMultipart

public static List<MessageBodyReaderUtils.RequestData> fromMultipart(InputStream mimeData,
                                                     javax.ws.rs.core.MediaType mediaType)
                                                              throws IOException,
                                                                     IllegalArgumentException

Returns content parsed from MediaType.MULTIPART_FORM_DATA. It iterates over all BodyParts and tries to create MessageBodyReaderUtils.RequestData instances. In case the Part.getContentType() is not present or can not be parsed, the MessageBodyReaderUtils.RequestData.getMediaType() is set to null. If Part.getInputStream() is not defined an IllegalArgumentException is thrown. The Part.getFileName() is used for MessageBodyReaderUtils.RequestData.getName(). The ordering of the returned Content instances is the same as within the MimeMultipart instance parsed from the input stream.

This Method does NOT load the data into memory, but returns directly the InputStreams as returned by the BodyParts. Therefore it is saved to be used with big attachments.

This Method is necessary because within MessageBodyReader one can not use the usual annotations as used within Resources. so this method allows to access the data directly from the parameters available from the MessageBodyReader.readFrom(Class, Type, java.lang.annotation.Annotation[], MediaType, javax.ws.rs.core.MultivaluedMap, InputStream) method

To test this Method with curl use:

 curl -v -X POST -F "content=@{dataFile};type={mimeType}" 
      {serviceURL}
 

Note that between {contentParam} and the datafile MUST NOT be a '='!

Parameters:

mimeData - the mime encoded data

mediaType - the mediaType (parsed to the ByteArrayDataSource constructor)

Returns:

the contents parsed from the BodyParts

Throws:

IOException - an any Exception while reading the stream or MessagingException exceptions other than ParseExceptions

IllegalArgumentException - If a InputStream is not available for any BodyPart or on ParseExceptions while reading the MimeData from the stream.