Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

Library and examples of parsing multipart/form-data from inputstream

The response to one kind of HTTP request I send is a multipart/form-data looks something like:

--------boundary123
Content-Disposition: form-data; name="json"
Content-Type: application/json

{"some":"json"}
--------boundary123
Content-Disposition: form-data; name="bin"
Content-Type: application/octet-stream

<file data>

--------boundary123

I've been using apache to send and receive the HTTP requests, but I can't seem to find an easy way to use it to parse the above for easy access of the form fields.

I would prefer not to reinvent the wheel, so I'm looking for a library that allows me to do something similar to:

MultipartEntity multipart = new MultipartEntity(inputStream);
InputStream bin = multipart.get("bin");

Any suggestions?

like image 531
Linus Avatar asked Nov 19 '12 16:11

Linus


3 Answers

I wanted a solution that worked from Azure functions with everything being in-memory and no reliance on the servlet or portal apis. So I wrote one:

package my.package;

import java.io.ByteArrayInputStream;
import java.io.ByteArrayOutputStream;
import java.io.File;
import java.io.IOException;
import java.io.InputStream;
import java.io.OutputStream;
import java.io.UnsupportedEncodingException;
import java.nio.charset.StandardCharsets;
import java.util.List;
import java.util.Map;

import org.apache.commons.fileupload.FileItem;
import org.apache.commons.fileupload.FileItemFactory;
import org.apache.commons.fileupload.FileItemHeaders;
import org.apache.commons.fileupload.FileUpload;
import org.apache.commons.fileupload.FileUploadException;
import org.apache.commons.fileupload.ParameterParser;
import org.apache.commons.fileupload.UploadContext;

public class MultipartParser {
  private static class SimpleContext implements UploadContext {
    private final byte[] request;
    private final String contentType;

    private SimpleContext(byte[] requestBody, String contentTypeHeader) {
        this.request = requestBody;
        this.contentType = contentTypeHeader;
    }

    @Override
    public long contentLength() {
        return request.length;
    }

    @Override
    public String getCharacterEncoding() {
        // The 'Content-Type' header may look like:
        // multipart/form-data; charset=UTF-8; boundary="xxxx"
        // in which case we can extract the charset, otherwise,
        // just default to UTF-8.
        ParameterParser parser = new ParameterParser();
        parser.setLowerCaseNames(true);
        String charset = parser.parse(contentType, ';').get("charset");
        return charset != null ? charset : "UTF-8";
    }

    @Override
    public int getContentLength() {
        return request.length;
    }

    @Override
    public String getContentType() {
        return contentType;
    }

    @Override
    public InputStream getInputStream() throws IOException {
        return new ByteArrayInputStream(request);
    }
  }

  /**
   * A form field which stores the field or file data completely in
   * memory. Will be limited by the maximum size of
   * a byte array (about 2GB).
   */
  private static class MemoryFileItem implements FileItem {
    private String fieldName;
    private String fileName;
    private String contentType;
    private boolean isFormField;
    private FileItemHeaders headers;
    private final ByteArrayOutputStream os = new ByteArrayOutputStream();

    public MemoryFileItem(String fieldName, String contentType, boolean isFormField, String fileName) {
        this.fieldName = fieldName;
        this.contentType = contentType;
        this.isFormField = isFormField;
        this.fileName = fileName;
    }

    @Override
    public void delete() {
    }

    /**
     * Not cached, so only call once.
     */
    @Override
    public byte[] get() {
        return os.toByteArray();
    }

    @Override
    public String getContentType() {
        return contentType;
    }

    @Override
    public String getFieldName() {
        return fieldName;
    }

    @Override
    public InputStream getInputStream() throws IOException {
        return new ByteArrayInputStream(get());
    }

    @Override
    public String getName() {
        return fileName;
    }

    @Override
    public OutputStream getOutputStream() throws IOException {
        return os;
    }

    @Override
    public long getSize() {
        return os.size();
    }

    @Override
    public String getString() {
        return new String(get(), StandardCharsets.UTF_8);
    }

    @Override
    public String getString(String encoding) throws UnsupportedEncodingException {
        return new String(get(), encoding);
    }

    @Override
    public boolean isFormField() {
        return isFormField;
    }

    @Override
    public boolean isInMemory() {
        return true;
    }

    @Override
    public void setFieldName(String name) {
        fieldName = name;
    }

    @Override
    public void setFormField(boolean state) {
        isFormField = state;
    }

    @Override
    public void write(File file) throws Exception {
    }

    @Override
    public FileItemHeaders getHeaders() {
        return headers;
    }

    @Override
    public void setHeaders(FileItemHeaders headers) {
        this.headers = headers;
    }
  }

  private static class MemoryFileItemFactory implements FileItemFactory {
    @Override
    public FileItem createItem(String fieldName, String contentType, boolean isFormField, String fileName) {
        return new MemoryFileItem(fieldName, contentType, isFormField, fileName);
    }
  }

  /**
   * Assumes the request body really is multipart/form-data.
   * Given the binary request body and the Content-Type header value,
   * attempts to parse fields into a map from field name to list
   * of FileItem objects.
   * 
   * Everything is stored in memory and an individual item will only be limited
   * by the maximum size of a byte array (about 2GB). It is recommended that the
   * user sets a limit on maximum upload request size. Doing this will obviously
   * differ by environment.
   *
   * Example:
   *   <code>
   *   var fields = MultipartParser.parseRequest(requestBody, contentTypeHeader);
   *   String firstName = fields.get("firstname").get(0).getString();
   *   byte[] profilePic = fields.get("picture").get(0).get();
   *   </code>
   *
   * @param requestBody The binary request body
   * @param contentTypeHeader The string value of the Content-Type header.
   * @return a map, with each entry having one or more values for that named field.
   * @throws FileUploadException
   */
  public static Map<String, List<FileItem>> parseRequest(byte[] requestBody, String contentTypeHeader)
        throws FileUploadException {
    FileUpload fileUpload = new FileUpload(new MemoryFileItemFactory());
    return fileUpload.parseParameterMap(new SimpleContext(requestBody, contentTypeHeader));
  }
}

And a sample use with Azure functions:

public class Function {
  @FunctionName("doSomethingWithBinaryFile")
  public HttpResponseMessage run(
        @HttpTrigger(
            name = "req",
            methods = {HttpMethod.POST},
            authLevel = AuthorizationLevel.ANONYMOUS)
            HttpRequestMessage<byte[]> request,
        final ExecutionContext context) throws IOException, FileUploadException {
    Logger logger = context.getLogger();
    byte[] body = request.getBody();
    String contentTypeHeader = request.getHeaders().get("content-type");
    byte[] fileBytes = MultipartParser.parseRequest(body, contentTypeHeader).get("file").get(0).get();

    return request.createResponseBuilder(HttpStatus.OK).body(
        transformBytesForExample(fileBytes)).build();
  }
}
like image 152
Daniel F Avatar answered Nov 17 '22 19:11

Daniel F


Example code using deprecated constructor:

import java.io.ByteArrayInputStream;

import org.apache.commons.fileupload.MultipartStream;

public class MultipartTest {

    // Lines should end with CRLF
    public static final String MULTIPART_BODY =
            "Content-Type: multipart/form-data; boundary=--AaB03x\r\n"
            + "\r\n"
            + "----AaB03x\r\n"
            + "Content-Disposition: form-data; name=\"submit-name\"\r\n"
            + "\r\n"
            + "Larry\r\n"
            + "----AaB03x\r\n"
            + "Content-Disposition: form-data; name=\"files\"; filename=\"file1.txt\"\r\n"
            + "Content-Type: text/plain\r\n"
            + "\r\n"
            + "HELLO WORLD!\r\n"
            + "----AaB03x--\r\n";

    public static void main(String[] args) throws Exception {

        byte[] boundary = "--AaB03x".getBytes();

        ByteArrayInputStream content = new ByteArrayInputStream(MULTIPART_BODY.getBytes());

        @SuppressWarnings("deprecation")
        MultipartStream multipartStream =
                new MultipartStream(content, boundary);

        boolean nextPart = multipartStream.skipPreamble();
        while (nextPart) {
            String header = multipartStream.readHeaders();
            System.out.println("");
            System.out.println("Headers:");
            System.out.println(header);
            System.out.println("Body:");
            multipartStream.readBodyData(System.out);
            System.out.println("");
            nextPart = multipartStream.readBoundary();
        }
    }
}
like image 14
Linus Avatar answered Nov 17 '22 21:11

Linus


Example code without using deprecated methods.

import com.google.common.net.MediaType;
import org.apache.commons.fileupload.RequestContext;

import java.io.ByteArrayInputStream;
import java.io.IOException;
import java.io.InputStream;
import java.nio.charset.Charset;

public class SimpleRequestContext implements RequestContext {
    private final Charset charset;
    private final MediaType contentType;
    private final byte[] content;

    public SimpleRequestContext(Charset charset, MediaType contentType, byte[] content) {
        this.charset = charset;
        this.contentType = contentType;
        this.content = content;
    }

    public String getCharacterEncoding() {
        return charset.displayName();
    }

    public String getContentType() {
        return contentType.toString();
    }

    @Deprecated
    public int getContentLength() {
        return content.length;
    }

    public InputStream getInputStream() throws IOException {
        return new ByteArrayInputStream(content);
    }
}

{
  ...
  Charset encoding = UTF_8;
  RequestContext requestContext = new SimpleRequestContext(encoding, contentType, body.getBytes());
  FileUploadBase fileUploadBase = new PortletFileUpload();
  FileItemFactory fileItemFactory = new DiskFileItemFactory();
  fileUploadBase.setFileItemFactory(fileItemFactory);
  fileUploadBase.setHeaderEncoding(encoding.displayName());
  List<FileItem> fileItems = fileUploadBase.parseRequest(requestContext);
  ...
}
like image 1
user1853665 Avatar answered Nov 17 '22 21:11

user1853665