HTTP Model

Apache Pekko HTTP model contains a deeply structured, fully immutable, case-class based model of all the major HTTP data structures, like HTTP requests, responses and common headers. It lives in the pekko-http-core module and forms the basis for most of Apache Pekko HTTP’s APIs.

Overview

Since pekko-http-core provides the central HTTP data structures you will find the following import in quite a few places around the code base (and probably your own code as well):

Scala

copysourceimport org.apache.pekko.http.scaladsl.model._

Java

copysourceimport org.apache.pekko.http.javadsl.model.*;
import org.apache.pekko.http.javadsl.model.headers.*;

import java.util.Optional;

This brings all of the most relevant types in scope, mainly:

• HttpRequestHttpRequest and HttpResponseHttpResponse, the central message model
• headers, the package containing all the predefined HTTP header models and supporting types
• Supporting types like UriUri, HttpMethodsHttpMethods, MediaTypesMediaTypes, StatusCodesStatusCodes, etc.

A common pattern is that the model of a certain entity is represented by an immutable type (class or trait), while the actual instances of the entity defined by the HTTP spec live in an accompanying object carrying the name of the type plus a trailing plural ‘s’.

For example:

• Defined HttpMethodHttpMethod instances live inare defined as static fields of the HttpMethodsHttpMethods objectclass.
• Defined HttpCharsetHttpCharset instances live inare defined as static fields of the HttpCharsetsHttpCharsets objectclass.
• Defined HttpEncodingHttpEncoding instances live inare defined as static fields of the HttpEncodingsHttpEncodings objectclass.
• Defined HttpProtocolHttpProtocol instances live inare defined as static fields of the HttpProtocolsHttpProtocols objectclass.
• Defined MediaTypeMediaType instances live inare defined as static fields of the MediaTypesMediaTypes objectclass.
• Defined StatusCodeStatusCode instances live inare defined as static fields of the StatusCodesStatusCodes objectclass.

HttpRequest

HttpRequestHttpRequest and HttpResponseHttpResponse are the basic caseimmutable classes representing HTTP messages.

An HttpRequestHttpRequest consists of

• a method (GET, POST, etc.)
• a URI (see URI model for more information)
• a seq of headers
• an entity (body data)
• a protocol

Here are some examples how to construct an HttpRequestHttpRequest:

Scala

copysourceimport org.apache.pekko.http.scaladsl.model._
import HttpMethods._

// construct a simple GET request to `homeUri`
val homeUri = Uri("/abc")
HttpRequest(GET, uri = homeUri)

// construct simple GET request to "/index" (implicit string to Uri conversion)
HttpRequest(GET, uri = "/index")

// construct simple POST request containing entity
val data = ByteString("abc")
HttpRequest(POST, uri = "/receive", entity = data)

// customize every detail of HTTP request
import HttpProtocols._
import MediaTypes._
import HttpCharsets._
val userData = ByteString("abc")
val authorization = headers.Authorization(BasicHttpCredentials("user", "pass"))
HttpRequest(
  PUT,
  uri = "/user",
  entity = HttpEntity(`text/plain`.withCharset(`UTF-8`), userData),
  headers = List(authorization),
  protocol = `HTTP/1.0`)

Java

copysourceimport org.apache.pekko.http.javadsl.model.*;
import org.apache.pekko.http.javadsl.model.headers.*;

import java.util.Optional;

// construct a simple GET request to `homeUri`
Uri homeUri = Uri.create("/home");
HttpRequest request1 = HttpRequest.create().withUri(homeUri);

// construct simple GET request to "/index" using helper methods
HttpRequest request2 = HttpRequest.GET("/index");

// construct simple POST request containing entity
ByteString data = ByteString.fromString("abc");
HttpRequest postRequest1 = HttpRequest.POST("/receive").withEntity(data);

// customize every detail of HTTP request
// import HttpProtocols.*
// import MediaTypes.*
Authorization authorization = Authorization.basic("user", "pass");
HttpRequest complexRequest =
    HttpRequest.PUT("/user")
        .withEntity(HttpEntities.create(ContentTypes.TEXT_PLAIN_UTF8, "abc"))
        .addHeader(authorization)
        .withProtocol(HttpProtocols.HTTP_1_0);

Synthetic Headers

In some cases it may be necessary to deviate from fully RFC-Compliant behavior. For instance, Amazon S3 treats the + character in the path part of the URL as a space, even though the RFC specifies that this behavior should be limited exclusively to the query portion of the URI.

In order to work around these types of edge cases, Apache Pekko HTTP provides for the ability to provide extra, non-standard information to the request via synthetic headers. These headers are not passed to the client but are instead consumed by the request engine and used to override default behavior.

For instance, in order to provide a raw request URI, bypassing the default URL normalization, you could do the following:

Scala

copysourceimport org.apache.pekko.http.scaladsl.model.headers.`Raw-Request-URI`
val req = HttpRequest(uri = "/ignored", headers = List(`Raw-Request-URI`("/a/b%2Bc")))

Java

copysource// imports org.apache.pekko.http.javadsl.model.headers.RawRequestURI
HttpRequest.create("/ignored").addHeader(RawRequestURI.create("/a/b%2Bc"));

HttpResponse

An HttpResponseHttpResponse consists of

• a status code
• a Seqlist of headers
• an entity (body data)
• a protocol

Here are some examples how to construct an HttpResponseHttpResponse:

Scala

copysourceimport StatusCodes._

// simple OK response without data created using the integer status code
HttpResponse(200)

// 404 response created using the named StatusCode constant
HttpResponse(NotFound)

// 404 response with a body explaining the error
HttpResponse(404, entity = "Unfortunately, the resource couldn't be found.")

// A redirecting response containing an extra header
val locationHeader = headers.Location("http://example.com/other")
HttpResponse(Found, headers = List(locationHeader))

Java

copysource// simple OK response without data created using the integer status code
HttpResponse ok = HttpResponse.create().withStatus(200);

// 404 response created using the named StatusCode constant
HttpResponse notFound = HttpResponse.create().withStatus(StatusCodes.NOT_FOUND);

// 404 response with a body explaining the error
HttpResponse notFoundCustom =
    HttpResponse.create()
        .withStatus(404)
        .withEntity("Unfortunately, the resource couldn't be found.");

// A redirecting response containing an extra header
Location locationHeader = Location.create("http://example.com/other");
HttpResponse redirectResponse =
    HttpResponse.create().withStatus(StatusCodes.FOUND).addHeader(locationHeader);

In addition to the simple HttpEntityHttpEntity constructorsHttpEntities.create methods which create an entity from a fixed String or ByteStringByteString as shown here the Apache Pekko HTTP model defines a number of subclasses of HttpEntityHttpEntity which allow body data to be specified as a stream of bytes. All of these types can be created using the method on HttpEntites.

HttpEntity

An HttpEntityHttpEntity carries the data bytes of a message together with its Content-Type and, if known, its Content-Length. In Apache Pekko HTTP, there are five different kinds of entities which model the various ways that message content can be received or sent:

HttpEntity.StrictHttpEntityStrict

The simplest entity, which is used when all the entity are already available in memory. It wraps a plain ByteStringByteString and represents a standard, unchunked entity with a known Content-Length.

HttpEntity.DefaultHttpEntityDefault

The general, unchunked HTTP/1.1 message entity. It has a known length and presents its data as a Source<ByteString, ?>Source[ByteString, _] which can be only materialized once. It is an error if the provided source doesn’t produce exactly as many bytes as specified. The distinction of StrictHttpEntityStrict and DefaultHttpEntityDefault is an API-only one. On the wire, both kinds of entities look the same.

HttpEntity.ChunkedHttpEntityChunked

The model for HTTP/1.1 chunked content (i.e. sent with Transfer-Encoding: chunked). The content length is unknown and the individual chunks are presented as a Source[HttpEntity.ChunkStreamPart]Source<ChunkStreamPart, ?>Source[ChunkStreamPart, ?]. A ChunkStreamPart is either a non-empty Chunkchunk or a LastChunkthe empty last chunk containing optional trailer headers. The stream consists of zero or more Chunkednon-empty chunks parts and can be terminated by an optional LastChunk partlast chunk.

HttpEntity.CloseDelimitedHttpEntityCloseDelimited

An unchunked entity of unknown length that is implicitly delimited by closing the connection (Connection: close). The content data are presented as a Source<ByteString, ?>Source[ByteString, _]. Since the connection must be closed after sending an entity of this type it can only be used on the server-side for sending a response. Also, the main purpose of CloseDelimited entities is compatibility with HTTP/1.0 peers, which do not support chunked transfer encoding. If you are building a new application and are not constrained by legacy requirements you shouldn’t rely on CloseDelimited entities, since implicit terminate-by-connection-close is not a robust way of signaling response end, especially in the presence of proxies. Additionally this type of entity prevents connection reuse which can seriously degrade performance. Use HttpEntity.ChunkedHttpEntityChunked instead!

HttpEntity.IndefiniteLengthHttpEntityIndefiniteLength

A streaming entity of unspecified length for use in a Multipart.BodyPart.

Entity types StrictHttpEntityStrict, DefaultHttpEntityDefault, and ChunkedHttpEntityChunked are a subtype of HttpEntity.RegularRequestEntityRequestEntity which allows to use them for requests and responses. In contrast, HttpEntity.CloseDelimitedHttpEntityCloseDelimited can only be used for responses.

Streaming entity types (i.e. all but StrictHttpEntityStrict) cannot be shared or serialized. To create a strict, shareable copy of an entity or message use HttpEntity.toStrict or HttpMessage.toStrict which returns a FutureCompletionStage of the object with the body data collected into a ByteStringByteString.

The HttpEntityHttpEntity companion objectclass HttpEntities contains several helper constructorsstatic methods to create entities from common types easily.

You can pattern match overuse the subtypesisX methods of HttpEntityHttpEntity to find out of which subclass an entity is if you want to provide special handling for each of the subtypes. However, in many cases a recipient of an HttpEntityHttpEntity doesn’t care about of which subtype an entity is (and how data is transported exactly on the HTTP layer). Therefore, the general method HttpEntity.dataBytesHttpEntity.getDataBytes() is provided which returns a Source<ByteString, ?>Source[ByteString, _] that allows access to the data of an entity regardless of its concrete subtype.

Limiting message entity length

All message entities that Apache Pekko HTTP reads from the network automatically get a length verification check attached to them. This check makes sure that the total entity size is less than or equal to the configured max-content-length [1], which is an important defense against certain Denial-of-Service attacks. However, a single global limit for all requests (or responses) is often too inflexible for applications that need to allow large limits for some requests (or responses) but want to clamp down on all messages not belonging into that group.

In order to give you maximum flexibility in defining entity size limits according to your needs the HttpEntityHttpEntity features a withSizeLimit method, which lets you adjust the globally configured maximum size for this particular entity, be it to increase or decrease any previously set value. This means that your application will receive all requests (or responses) from the HTTP layer, even the ones whose Content-Length exceeds the configured limit (because you might want to increase the limit yourself). Only when the actual data stream SourceSource contained in the entity is materialized will the boundary checks be actually applied. In case the length verification fails the respective stream will be terminated with an EntityStreamSizeException either directly at materialization time (if the Content-Length is known) or whenever more data bytes than allowed have been read.

When called on Strict entities the withSizeLimit method will return the entity itself if the length is within the bound, otherwise a Default entity with a single element data stream. This allows for potential refinement of the entity size limit at a later point (before materialization of the data stream).

By default all message entities produced by the HTTP layer automatically carry the limit that is defined in the application’s max-content-length config setting. If the entity is transformed in a way that changes the content-length and then another limit is applied then this new limit will be evaluated against the new content-length. If the entity is transformed in a way that changes the content-length and no new limit is applied then the previous limit will be applied against the previous content-length. Generally this behavior should be in line with your expectations.

[1] pekko.http.parsing.max-content-length (applying to server- as well as client-side), pekko.http.server.parsing.max-content-length (server-side only), pekko.http.client.parsing.max-content-length (client-side only) or pekko.http.host-connection-pool.client.parsing.max-content-length (only host-connection-pools)

Special processing for HEAD requests

RFC 7230 defines very clear rules for the entity length of HTTP messages.

Especially this rule requires special treatment in Apache Pekko HTTP:

Any response to a HEAD request and any response with a 1xx (Informational), 204 (No Content), or 304 (Not Modified) status code is always terminated by the first empty line after the header fields, regardless of the header fields present in the message, and thus cannot contain a message body.

Responses to HEAD requests introduce the complexity that Content-Length or Transfer-Encoding headers can be present but the entity is empty. This is modeled by allowing HttpEntity.DefaultHttpEntityDefault and HttpEntity.ChunkedHttpEntityChunked to be used for HEAD responses with an empty data stream.

Also, when a HEAD response has an HttpEntity.CloseDelimitedHttpEntityCloseDelimited entity the Apache Pekko HTTP implementation will not close the connection after the response has been sent. This allows the sending of HEAD responses without Content-Length header across persistent HTTP connections.

Header Model

Apache Pekko HTTP contains a rich model of the most common HTTP headers. Parsing and rendering is done automatically so that applications don’t need to care for the actual syntax of headers. Headers not modelled explicitly are represented as a RawHeaderRawHeader, which is essentially a String/String name/value pair.

See these examples of how to deal with headers:

Scala

copysourceimport org.apache.pekko.http.scaladsl.model.headers._

// create a ``Location`` header
val loc = Location("http://example.com/other")

// create an ``Authorization`` header with HTTP Basic authentication data
val auth = Authorization(BasicHttpCredentials("joe", "josepp"))

// custom type
case class User(name: String, pass: String)

// a method that extracts basic HTTP credentials from a request
def credentialsOfRequest(req: HttpRequest): Option[User] =
  for {
    case Authorization(BasicHttpCredentials(user, pass)) <- req.header[Authorization]
  } yield User(user, pass)

Java

copysource  // create a ``Location`` header
  Location locationHeader = Location.create("http://example.com/other");

  // create an ``Authorization`` header with HTTP Basic authentication data
  Authorization authorization = Authorization.basic("user", "pass");

// a method that extracts basic HTTP credentials from a request
private Optional<BasicHttpCredentials> getCredentialsOfRequest(HttpRequest request) {
  Optional<Authorization> auth = request.getHeader(Authorization.class);
  if (auth.isPresent() && auth.get().credentials() instanceof BasicHttpCredentials)
    return Optional.of((BasicHttpCredentials) auth.get().credentials());
  else return Optional.empty();
}

HTTP Headers

When the Apache Pekko HTTP server receives an HTTP request it tries to parse all its headers into their respective model classes. Independently of whether this succeeds or not, the HTTP layer will always pass on all received headers to the application. Unknown headers as well as ones with invalid syntax (according to the header parser) will be made available as RawHeaderRawHeader instances. For the ones exhibiting parsing errors a warning message is logged depending on the value of the illegal-header-warnings config setting.

Some headers have special status in HTTP and are therefore treated differently from “regular” headers:

Content-Type

The Content-Type of an HTTP message is modeled as the contentType field of the HttpEntityHttpEntity. The Content-Type header therefore doesn’t appear in the headers sequence of a message. Also, a Content-Type header instance that is explicitly added to the headers of a request or response will not be rendered onto the wire and trigger a warning being logged instead!

Transfer-Encoding

Messages with Transfer-Encoding: chunked are represented via the HttpEntity.Chunkedas a HttpEntityChunked entity. As such chunked messages that do not have another deeper nested transfer encoding will not have a Transfer-Encoding header in their headers sequencelist. Similarly, a Transfer-Encoding header instance that is explicitly added to the headers of a request or response will not be rendered onto the wire and trigger a warning being logged instead!

Content-Length

The content length of a message is modelled via its HttpEntity. As such no Content-Length header will ever be part of a message’s header sequence. Similarly, a Content-Length header instance that is explicitly added to the headers of a request or response will not be rendered onto the wire and trigger a warning being logged instead!

Server

A Server header is usually added automatically to any response and its value can be configured via the pekko.http.server.server-header setting. Additionally an application can override the configured header with a custom one by adding it to the response’s header sequence.

User-Agent

A User-Agent header is usually added automatically to any request and its value can be configured via the pekko.http.client.user-agent-header setting. Additionally an application can override the configured header with a custom one by adding it to the request’s header sequence.

Date

The DateDate response header is added automatically but can be overridden by supplying it manually.

Connection

On the server-side Apache Pekko HTTP watches for explicitly added Connection: close response headers and as such honors the potential wish of the application to close the connection after the respective response has been sent out. The actual logic for determining whether to close the connection is quite involved. It takes into account the request’s method, protocol and potential ConnectionConnection header as well as the response’s protocol, entity and potential ConnectionConnection header. See this test for a full table of what happens when.

Strict-Transport-Security

HTTP Strict Transport Security (HSTS) is a web security policy mechanism which is communicated by the Strict-Transport-Security header. The most important security vulnerability that HSTS can fix is SSL-stripping man-in-the-middle attacks. The SSL-stripping attack works by transparently converting a secure HTTPS connection into a plain HTTP connection. The user can see that the connection is insecure, but crucially there is no way of knowing whether the connection should be secure. HSTS addresses this problem by informing the browser that connections to the site should always use TLS/SSL. See also RFC 6797.

Custom Headers

Sometimes you may need to model a custom header type which is not part of HTTP and still be able to use it as convenient as is possible with the built-in types.

Because of the number of ways one may interact with headers (i.e. try to matchconvert a CustomHeaderCustomHeader againstto a RawHeaderRawHeader or the other way around etc), a helper traitclasses for custom Header types and their companions classes are provided by Apache Pekko HTTP. Thanks to extending ModeledCustomHeaderModeledCustomHeader instead of the plain CustomHeaderCustomHeader such header can be matchedthe following methods are at your disposal:

Scala

copysourcefinal class ApiTokenHeader(token: String) extends ModeledCustomHeader[ApiTokenHeader] {
  override def renderInRequests = true
  override def renderInResponses = true
  override val companion = ApiTokenHeader
  override def value: String = token
}
object ApiTokenHeader extends ModeledCustomHeaderCompanion[ApiTokenHeader] {
  override val name = "apiKey"
  override def parse(value: String) = Try(new ApiTokenHeader(value))
}

Java

copysourcepublic static class ApiTokenHeader extends ModeledCustomHeader {

  ApiTokenHeader(String name, String value) {
    super(name, value);
  }

  public boolean renderInResponses() {
    return false;
  }

  public boolean renderInRequests() {
    return false;
  }
}

static class ApiTokenHeaderFactory extends ModeledCustomHeaderFactory<ApiTokenHeader> {

  public String name() {
    return "apiKey";
  }

  @Override
  public ApiTokenHeader parse(String value) {
    return new ApiTokenHeader(name(), value);
  }
}

Which allows this CustomHeadermodeled custom header to be used in the following scenarios:

Scala

copysourceval ApiTokenHeader(t1) = ApiTokenHeader("token")
t1 should ===("token")

val RawHeader(k2, v2) = ApiTokenHeader("token")
k2 should ===("apiKey")
v2 should ===("token")

// will match, header keys are case insensitive
val ApiTokenHeader(v3) = RawHeader("APIKEY", "token")
v3 should ===("token")

intercept[MatchError] {
  // won't match, different header name
  val ApiTokenHeader(v4) = DifferentHeader("token")
}

intercept[MatchError] {
  // won't match, different header name
  val RawHeader("something", v5) = DifferentHeader("token")
}

intercept[MatchError] {
  // won't match, different header name
  val ApiTokenHeader(v6) = RawHeader("different", "token")
}

Java

copysourcefinal ApiTokenHeaderFactory apiTokenHeaderFactory = new ApiTokenHeaderFactory();
final ApiTokenHeader token = apiTokenHeaderFactory.create("token");
assertEquals("token", token.value());

final HttpHeader header = apiTokenHeaderFactory.create("token");
assertEquals("apiKey", header.name());
assertEquals("token", header.value());

final Optional<ApiTokenHeader> fromRaw =
    apiTokenHeaderFactory.from(RawHeader.create("apiKey", "token"));
assertTrue("Expected a header", fromRaw.isPresent());
assertEquals("apiKey", fromRaw.get().name());
assertEquals("token", fromRaw.get().value());

// will match, header keys are case insensitive
final Optional<ApiTokenHeader> fromRawUpper =
    apiTokenHeaderFactory.from(RawHeader.create("APIKEY", "token"));
assertTrue("Expected a header", fromRawUpper.isPresent());
assertEquals("apiKey", fromRawUpper.get().name());
assertEquals("token", fromRawUpper.get().value());

// won't match, different header name
final Optional<ApiTokenHeader> wrong =
    apiTokenHeaderFactory.from(RawHeader.create("different", "token"));
assertFalse(wrong.isPresent());

Including usage within the header directives like in the following headerValuePF example:

Scala

copysourcedef extractFromCustomHeader = headerValuePF {
  case t @ ApiTokenHeader(token) => s"extracted> $t"
  case raw: RawHeader            => s"raw> $raw"
}

val routes = extractFromCustomHeader { s =>
  complete(s)
}

Get().withHeaders(RawHeader("apiKey", "TheKey")) ~> routes ~> check {
  status should ===(StatusCodes.OK)
  responseAs[String] should ===("extracted> apiKey: TheKey")
}

Get().withHeaders(RawHeader("somethingElse", "TheKey")) ~> routes ~> check {
  status should ===(StatusCodes.OK)
  responseAs[String] should ===("raw> somethingElse: TheKey")
}

Get().withHeaders(ApiTokenHeader("TheKey")) ~> routes ~> check {
  status should ===(StatusCodes.OK)
  responseAs[String] should ===("extracted> apiKey: TheKey")
}

Java

copysourceimport org.apache.pekko.http.javadsl.server.Directives;

import static org.apache.pekko.http.javadsl.server.Directives.headerValuePF;

final ApiTokenHeaderFactory apiTokenHeaderFactory = new ApiTokenHeaderFactory();
final PartialFunction<HttpHeader, String> extractFromCustomHeader =
    new JavaPartialFunction<HttpHeader, String>() {

      @Override
      public String apply(HttpHeader header, boolean isCheck) throws Exception {
        if (isCheck) return null;
        return apiTokenHeaderFactory
            .from(header)
            .map(apiTokenHeader -> "extracted> " + apiTokenHeader)
            .orElseGet(() -> "raw> " + header);
      }
    };

final Route route = headerValuePF(extractFromCustomHeader, Directives::complete);

testRoute(route)
    .run(HttpRequest.GET("/").addHeader(RawHeader.create("apiKey", "TheKey")))
    .assertStatusCode(StatusCodes.OK)
    .assertEntity("extracted> apiKey: TheKey");

testRoute(route)
    .run(HttpRequest.GET("/").addHeader(RawHeader.create("somethingElse", "TheKey")))
    .assertStatusCode(StatusCodes.OK)
    .assertEntity("raw> somethingElse: TheKey");

testRoute(route)
    .run(HttpRequest.GET("/").addHeader(apiTokenHeaderFactory.create("TheKey")))
    .assertStatusCode(StatusCodes.OK)
    .assertEntity("extracted> apiKey: TheKey");

Attributes

Sometimes it can be useful to keep track of some information associated with a request without explicitly closing over it. Such information can be attached to a request or response though message attributes:

Scala

copysourcecase class User(name: String)
object User {
  val attributeKey = AttributeKey[User]("user")
}

def determineUser(req: HttpRequest): HttpRequest = {
  val user = // ... somehow determine the user for this request

  // Add the attribute
  req.addAttribute(User.attributeKey, user)
}

// Retrieve the attribute
val user: Option[User] = requestWithAttribute.attribute(User.attributeKey)

Java

copysourceclass User {
  final String name;

  public User(String name) {
    this.name = name;
  }

  public static final AttributeKey<User> attributeKey = AttributeKey.create("user", User.class);
}

public HttpRequest determineUser(HttpRequest request) {
  User user = // ... somehow determine the user for this request

  // Add the attribute
  return request.addAttribute(User.attributeKey, user);
}

  // Retrieve the attribute
  Optional<User> user = requestWithAttribute.getAttribute(User.attributeKey);

Message attributes are only to be used within in your application, they are not present on the wire.

Parsing / Rendering

Parsing and rendering of HTTP data structures is heavily optimized and for most types there’s currently no public API provided to parse (or render to) Strings or byte arrays.

Registering Custom Media Types

Apache Pekko HTTP predefinespredefines most commonly encountered media types and emits them in their well-typed form while parsing http messages. Sometimes you may want to define a custom media type and inform the parser infrastructure about how to handle these custom media types, e.g. that application/custom is to be treated as NonBinary with WithFixedCharset. To achieve this you need to register the custom media type in the server’s settings by configuring ParserSettingsParserSettings like this:

Scala

copysource
// similarly in Java: `org.apache.pekko.http.javadsl.settings.[...]`
import org.apache.pekko
import pekko.http.scaladsl.settings.ParserSettings
import pekko.http.scaladsl.settings.ServerSettings

// define custom media type:
val utf8 = HttpCharsets.`UTF-8`
val `application/custom`: WithFixedCharset =
  MediaType.customWithFixedCharset("application", "custom", utf8)

// add custom media type to parser settings:
val parserSettings = ParserSettings.forServer(system).withCustomMediaTypes(`application/custom`)
val serverSettings = ServerSettings(system).withParserSettings(parserSettings)

val routes = extractRequest { r =>
  complete(r.entity.contentType.toString + " = " + r.entity.contentType.getClass)
}
val binding = Http().newServerAt("localhost", 0).withSettings(serverSettings).bind(routes)

Java

copysourceimport static org.apache.pekko.http.javadsl.server.Directives.complete;
import static org.apache.pekko.http.javadsl.server.Directives.extractRequest;

// Define custom media type:
final MediaType.WithFixedCharset applicationCustom =
    MediaTypes.customWithFixedCharset(
        "application",
        "custom", // The new Media Type name
        HttpCharsets.UTF_8, // The charset used
        new HashMap<>(), // Empty parameters
        false); // No arbitrary subtypes are allowed

// Add custom media type to parser settings:
final ParserSettings parserSettings =
    ParserSettings.forServer(system).withCustomMediaTypes(applicationCustom);
final ServerSettings serverSettings =
    ServerSettings.create(system).withParserSettings(parserSettings);

final Route route =
    extractRequest(
        req ->
            complete(
                req.entity().getContentType().toString()
                    + " = "
                    + req.entity().getContentType().getClass()));

final CompletionStage<ServerBinding> binding =
    Http.get(system).newServerAt(host, 0).withSettings(serverSettings).bind(route);

You may also want to read about MediaType Registration trees, in order to register your vendor specific media types in the right style / place.

Registering Custom Status Codes

Similarly to media types, Apache Pekko HTTP predefinespredefines well-known status codes, however sometimes you may need to use a custom one (or are forced to use an API which returns custom status codes). Similarly to the media types registration, you can register custom status codes by configuring ParserSettingsParserSettings like this:

Scala

copysource// similarly in Java: `org.apache.pekko.http.javadsl.settings.[...]`
import org.apache.pekko.http.scaladsl.settings.{ ParserSettings, ServerSettings }

// define custom status code:
val LeetCode = StatusCodes.custom(777, "LeetCode", "Some reason", isSuccess = true, allowsEntity = false)

// add custom method to parser settings:
val parserSettings = ParserSettings.forServer(system).withCustomStatusCodes(LeetCode)
val serverSettings = ServerSettings(system).withParserSettings(parserSettings)

val clientConSettings = ClientConnectionSettings(system).withParserSettings(parserSettings)
val clientSettings = ConnectionPoolSettings(system).withConnectionSettings(clientConSettings)

val routes =
  complete(HttpResponse(status = LeetCode))

// use serverSettings in server:
val binding = Http().newServerAt("127.0.0.1", 0).withSettings(serverSettings).bind(routes).futureValue

// use clientSettings in client:
val request = HttpRequest(uri = s"http://127.0.0.1:${binding.localAddress.getPort}/")
val response = Http().singleRequest(request, settings = clientSettings)

// futureValue is a ScalaTest helper:
response.futureValue.status should ===(LeetCode)

Java

copysourceimport static org.apache.pekko.http.javadsl.server.Directives.complete;
import static org.apache.pekko.http.javadsl.server.Directives.extractRequest;

// Define custom status code:
final StatusCode leetCode =
    StatusCodes.custom(
        777, // Our custom status code
        "LeetCode", // Our custom reason
        "Some reason", // Our custom default message
        true, // It should be considered a success response
        false); // Does not allow entities

// Add custom method to parser settings:
final ParserSettings parserSettings =
    ParserSettings.forServer(system).withCustomStatusCodes(leetCode);
final ServerSettings serverSettings =
    ServerSettings.create(system).withParserSettings(parserSettings);

final ClientConnectionSettings clientConSettings =
    ClientConnectionSettings.create(system).withParserSettings(parserSettings);
final ConnectionPoolSettings clientSettings =
    ConnectionPoolSettings.create(system).withConnectionSettings(clientConSettings);

final Route route = extractRequest(req -> complete(HttpResponse.create().withStatus(leetCode)));

// Use serverSettings in server:
final CompletionStage<ServerBinding> binding =
    Http.get(system).newServerAt(host, 0).withSettings(serverSettings).bind(route);

final ServerBinding serverBinding = binding.toCompletableFuture().get();

final int port = serverBinding.localAddress().getPort();

// Use clientSettings in client:
final HttpResponse response =
    Http.get(system)
        .singleRequest(
            HttpRequest.GET("http://" + host + ":" + port + "/"),
            ConnectionContext.https(SSLContext.getDefault()),
            clientSettings,
            system.log())
        .toCompletableFuture()
        .get();

// Check we get the right code back
assertEquals(leetCode, response.status());

Registering Custom HTTP Method

Apache Pekko HTTP also allows you to define custom HTTP methods, other than the well-known methods predefinedpredefined in Apache Pekko HTTP. To use a custom HTTP method, you need to define it, and then add it to parser settings like below:

Scala

copysourceimport org.apache.pekko.http.scaladsl.settings.{ ParserSettings, ServerSettings }

// define custom method type:
val BOLT = HttpMethod.custom("BOLT", safe = false,
  idempotent = true, requestEntityAcceptance = Expected)

// add custom method to parser settings:
val parserSettings = ParserSettings.forServer(system).withCustomMethods(BOLT)
val serverSettings = ServerSettings(system).withParserSettings(parserSettings)

val routes = extractMethod { method =>
  complete(s"This is a ${method.name} method request.")
}
val binding = Http().newServerAt(host, port).withSettings(serverSettings).bind(routes)

val request = HttpRequest(BOLT, s"http://$host:$port/", protocol = `HTTP/1.1`)

Java

copysourceimport static org.apache.pekko.http.javadsl.server.Directives.complete;
import static org.apache.pekko.http.javadsl.server.Directives.extractMethod;


// define custom method type:
HttpMethod BOLT = HttpMethods.custom("BOLT", false, true, Expected);

// add custom method to parser settings:
final ParserSettings parserSettings = ParserSettings.forServer(system).withCustomMethods(BOLT);
final ServerSettings serverSettings =
    ServerSettings.create(system).withParserSettings(parserSettings);

final Route routes =
    concat(extractMethod(method -> complete("This is a " + method.name() + " request.")));
final Http http = Http.get(system);
final CompletionStage<ServerBinding> binding =
    http.newServerAt(host, port)
        .withSettings(serverSettings)
        .logTo(loggingAdapter)
        .bind(routes);

HttpRequest request =
    HttpRequest.create()
        .withUri("http://" + host + ":" + Integer.toString(port))
        .withMethod(BOLT)
        .withProtocol(HTTP_1_1);

CompletionStage<HttpResponse> response = http.singleRequest(request);