feat: Support websocket

.github/workflows/ci.yml

@@ -16,22 +16,47 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        java-version: [ "8", "11", "17" ]
-        os: [ "ubuntu-latest" ]
-        os-label: [ "Ubuntu" ]
+        java-version: [ "11", "17" ]
+        os: [ "ubuntu-latest", "windows-latest", "macos-latest" ]
+        os-label: [ "Ubuntu", "Windows", "macOS" ]
         include:
-          - { java-version: "11", os: "windows-latest", os-label: "Windows" }
-          - { java-version: "11", os: "macos-latest", os-label: "macOS" }
+          - { java-version: "8", os: "macos-latest", os-label: "macOS", distribution: "zulu" }
+        exclude:
+          - os: "windows-latest"
+            os-label: "Ubuntu"
+          - os: "windows-latest"
+            os-label: "macOS"
+          - os: "macos-latest"
+            os-label: "Ubuntu"
+          - os: "macos-latest"
+            os-label: "Windows"
+          - os: "ubuntu-latest"
+            os-label: "Windows"
+          - os: "ubuntu-latest"
+            os-label: "macOS"
     steps:
       - uses: actions/checkout@v4
       - name: Set up JDK ${{ matrix.java-version }}
         uses: actions/setup-java@v4
         with:
-          distribution: 'adopt'
+          distribution: ${{ matrix.java-version == '8' && matrix.os == 'macos-latest' && 'zulu' || 'temurin' }}
           java-version: ${{ matrix.java-version }}
           cache: 'maven'
+          java-package: 'jdk'
+      - name: Set JAVA_HOME (Windows)
+        if: runner.os == 'Windows'
+        run: |
+          echo "JAVA_HOME=${{ env.JAVA_HOME }}" >> $GITHUB_ENV
+          echo "${{ env.JAVA_HOME }}/bin" >> $GITHUB_PATH
+        shell: bash
       - name: Code style check
-        run: mvn spotless:check
+        run: |
+          echo "Java version:"
+          java -version
+          echo "Javac version:"
+          javac -version
+          mvn -v
+          mvn spotless:check
       - name: Build and Test with Coverage
         run: |
           mvn -pl api clean test-compile

.gitignore

@@ -7,6 +7,7 @@
 **/bin/*
 **/obj/*
 .mvn
+*.wav
 
 # Compiled class file
 *.class

README.md

@@ -71,6 +71,9 @@ dependencies {
 | how to handle exception       | [HandlerExceptionExample.java](example/src/main/java/example/service/HandlerExceptionExample.java)          |
 | get request log id            | [GetLogExample.java](example/src/main/java/example/service/GetLogExample.java)                              |
 | set timeout                   | [SetRequestTimeoutExample.java](example/src/main/java/example/service/SetRequestTimeoutExample.java)        |
+| websocket chat              | [ChatExample.java](example/src/main/java/example/websocket/chat/ChatExample.java)                           |
+| websocket speech synthesis  | [WebsocketAudioSpeechExample.java](example/src/main/java/example/websocket/audio/speech/WebsocketAudioSpeechExample.java) |
+| websocket transcription     | [WebsocketTranscriptionsExample.java](example/src/main/java/example/websocket/audio/transcriptions/WebsocketTranscriptionsExample.java) |
 
 ### Initialize the Coze Client 
 
@@ -984,4 +987,123 @@ while (iterator.hasNext()) {
 
 ```
 
+### WebSocket
+
+The SDK provides WebSocket interfaces for real-time chat, speech synthesis and speech transcription.
+
+You can check the official documentation for more information:
+https://www.coze.cn/open/docs/guides/websocket_openapi
+
+#### WebSocket Chat
+
+WebSocket chat allows real-time communication with bots, including text and audio interactions:
+
+```java
+WebsocketChatClient client = coze.websocket()
+    .chat()
+    .create(new WebsocketChatCreateReq(botID, new CallbackHandler()));
+
+// Send audio data
+String audioData = "..."; // Base64 encoded audio data
+client.inputAudioBufferAppend(audioData);
+client.inputAudioBufferComplete();
+
+// Handle responses in callback
+class CallbackHandler extends WebsocketChatCallbackHandler {
+    // Handle text responses
+    @Override
+    public void onConversationMessageDelta(WebsocketChatClient client, ConversationMessageDeltaEvent event) {
+        System.out.printf("Received: %s\n", event.getData().getContent());
+    }
+
+    // Handle audio responses  
+    @Override
+    public void onConversationAudioDelta(WebsocketChatClient client, ConversationAudioDeltaEvent event) {
+        byte[] audioData = event.getData().getAudio();
+        // Process audio data...
+    }
+}
+```
+
+#### Speech Synthesis
+
+WebSocket speech synthesis allows real-time text-to-speech conversion:
+
+```java
+WebsocketAudioSpeechClient client = coze.websocket()
+    .audio()
+    .speech()
+    .create(new WebsocketAudioSpeechCreateReq(new CallbackHandler()));
+
+// Configure audio output
+OutputAudio outputAudio = OutputAudio.builder()
+    .voiceId(voiceID)
+    .codec("pcm")
+    .speechRate(50)
+    .pcmConfig(PCMConfig.builder().sampleRate(24000).build())
+    .build();
+client.speechUpdate(new SpeechUpdateEventData(outputAudio));
+
+// Send text for synthesis
+client.inputTextBufferAppend("Hello world!");
+client.inputTextBufferComplete();
+
+// Handle synthesized audio in callback
+class CallbackHandler extends WebsocketAudioSpeechCallbackHandler {
+    @Override
+    public void onSpeechAudioUpdate(WebsocketAudioSpeechClient client, SpeechAudioUpdateEvent event) {
+        byte[] audioData = event.getDelta();
+        // Process audio data...
+    }
+}
+```
+
+#### Speech Transcription 
+
+WebSocket speech transcription provides real-time speech-to-text conversion:
+
+```java
+WebsocketAudioTranscriptionsClient client = coze.websocket()
+    .audio()
+    .transcriptions()
+    .create(new WebsocketAudioTranscriptionsCreateReq(new CallbackHandler()));
+
+// Configure audio input
+InputAudio inputAudio = InputAudio.builder()
+    .sampleRate(24000)
+    .codec("pcm")
+    .format("wav")
+    .channel(2)
+    .build();
+client.transcriptionsUpdate(new TranscriptionsUpdateEventData(inputAudio));
+
+// Send audio for transcription
+String audioData = "..."; // Base64 encoded audio data
+client.inputAudioBufferAppend(audioData);
+client.inputAudioBufferComplete();
+
+// Handle transcription results in callback
+class CallbackHandler extends WebsocketAudioTranscriptionsCallbackHandler {
+    @Override
+    public void onTranscriptionsMessageUpdate(
+        WebsocketAudioTranscriptionsClient client, 
+        TranscriptionsMessageUpdateEvent event) {
+        System.out.println(event.getData().getContent());
+    }
+}
+```
+
+All WebSocket clients support proper resource cleanup:
+
+```java
+try {
+    // Use the client...
+} finally {
+    if (client != null) {
+        client.close();
+    }
+    coze.shutdownExecutor();
+}
+```
+
 

api/pom.xml

@@ -43,7 +43,7 @@
     </parent>
 
     <artifactId>coze-api</artifactId>
-    <version>0.2.8</version>
+    <version>0.2.9</version>
 
     <scm>
         <connection>scm:git:git://github.com/coze-dev/coze-java.git</connection>
@@ -68,7 +68,7 @@
         <jackson.version>2.14.2</jackson.version>
         <jwt.version>0.11.5</jwt.version>
         <junit.version>5.10.2</junit.version>
-        <okhttp.version>4.8.1</okhttp.version>
+        <okhttp.version>3.14.9</okhttp.version>
     </properties>
 
     <dependencies>
@@ -280,17 +280,20 @@
             <plugin>
                 <groupId>org.apache.maven.plugins</groupId>
                 <artifactId>maven-surefire-plugin</artifactId>
-                <version>2.22.2</version>
+                <version>3.2.5</version>
                 <configuration>
                     <includes>
                         <include>**/*Test.java</include>
                     </includes>
                     <useSystemClassLoader>false</useSystemClassLoader>
                     <forkCount>1</forkCount>
                     <reuseForks>true</reuseForks>
-                    <argLine>@{argLine} -Djava.security.egd=file:/dev/./urandom</argLine>
+                    <argLine>@{argLine}</argLine>
                     <systemPropertyVariables>
-                        <okhttp.platform>jdk</okhttp.platform>
+                        <okhttp.platform>jdk-8</okhttp.platform>
+                        <java.security.egd>file:/dev/urandom</java.security.egd>
+                        <javax.net.ssl.trustStore>${java.home}/lib/security/cacerts</javax.net.ssl.trustStore>
+                        <javax.net.ssl.trustStorePassword>changeit</javax.net.ssl.trustStorePassword>
                     </systemPropertyVariables>
                 </configuration>
             </plugin>

api/src/main/java/com/coze/openapi/api/AudioTranscriptionAPI.java

@@ -0,0 +1,19 @@
+package com.coze.openapi.api;
+
+import com.coze.openapi.client.audio.transcriptions.CreateTranscriptionsResp;
+import com.coze.openapi.client.common.BaseReq;
+import com.coze.openapi.client.common.BaseResponse;
+
+import okhttp3.MultipartBody;
+import retrofit2.Call;
+import retrofit2.http.Multipart;
+import retrofit2.http.POST;
+import retrofit2.http.Part;
+import retrofit2.http.Tag;
+
+public interface AudioTranscriptionAPI {
+  @Multipart
+  @POST("/v1/audio/transcriptions")
+  Call<BaseResponse<CreateTranscriptionsResp>> create(
+      @Part MultipartBody.Part file, @Tag BaseReq baseReq);
+}

api/src/main/java/com/coze/openapi/client/audio/speech/CreateSpeechReq.java

@@ -28,4 +28,7 @@ public class CreateSpeechReq extends BaseReq {
   @JsonProperty("speed")
   @Builder.Default
   private float speed = 1.0f;
+
+  @JsonProperty("sample_rate")
+  private int sampleRate;
 }

api/src/main/java/com/coze/openapi/client/audio/transcriptions/CreateTranscriptionsReq.java

@@ -0,0 +1,42 @@
+package com.coze.openapi.client.audio.transcriptions;
+
+import java.io.File;
+
+import com.coze.openapi.client.common.BaseReq;
+
+import lombok.EqualsAndHashCode;
+import lombok.Getter;
+import lombok.NoArgsConstructor;
+import lombok.experimental.SuperBuilder;
+
+@Getter
+@SuperBuilder
+@NoArgsConstructor
+@EqualsAndHashCode(callSuper = true)
+public class CreateTranscriptionsReq extends BaseReq {
+  /** local file path */
+  private String filePath;
+
+  /*
+   * file byte array
+   */
+  private byte[] fileBytes;
+
+  /** file name */
+  private String fileName;
+
+  /** file object */
+  private File file;
+
+  public static CreateTranscriptionsReq of(String fileName, byte[] fileBytes) {
+    return CreateTranscriptionsReq.builder().fileName(fileName).fileBytes(fileBytes).build();
+  }
+
+  public static CreateTranscriptionsReq of(File file) {
+    return CreateTranscriptionsReq.builder().file(file).build();
+  }
+
+  public static CreateTranscriptionsReq of(String filePath) {
+    return CreateTranscriptionsReq.builder().filePath(filePath).build();
+  }
+}

api/src/main/java/com/coze/openapi/client/audio/transcriptions/CreateTranscriptionsResp.java

@@ -0,0 +1,17 @@
+package com.coze.openapi.client.audio.transcriptions;
+
+import com.coze.openapi.client.common.BaseResp;
+import com.fasterxml.jackson.annotation.JsonProperty;
+
+import lombok.*;
+
+@Data
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+@EqualsAndHashCode(callSuper = true)
+@ToString(callSuper = true)
+public class CreateTranscriptionsResp extends BaseResp {
+  @JsonProperty("text")
+  private String text;
+}

api/src/main/java/com/coze/openapi/client/bots/CreateBotReq.java

@@ -1,7 +1,5 @@
 package com.coze.openapi.client.bots;
 
-import org.jetbrains.annotations.NotNull;
-
 import com.coze.openapi.client.bots.model.BotOnboardingInfo;
 import com.coze.openapi.client.bots.model.BotPromptInfo;
 import com.coze.openapi.client.common.BaseReq;
@@ -11,6 +9,7 @@
 import lombok.Data;
 import lombok.EqualsAndHashCode;
 import lombok.NoArgsConstructor;
+import lombok.NonNull;
 import lombok.experimental.SuperBuilder;
 
 @Data
@@ -19,11 +18,11 @@
 @AllArgsConstructor
 @EqualsAndHashCode(callSuper = true)
 public class CreateBotReq extends BaseReq {
-  @NotNull
+  @NonNull
   @JsonProperty("space_id")
   String spaceID;
 
-  @NotNull
+  @NonNull
   @JsonProperty("name")
   String name;
 

api/src/main/java/com/coze/openapi/client/bots/UpdateBotReq.java

@@ -1,17 +1,12 @@
 package com.coze.openapi.client.bots;
 
-import org.jetbrains.annotations.NotNull;
-
 import com.coze.openapi.client.bots.model.BotKnowledge;
 import com.coze.openapi.client.bots.model.BotOnboardingInfo;
 import com.coze.openapi.client.bots.model.BotPromptInfo;
 import com.coze.openapi.client.common.BaseReq;
 import com.fasterxml.jackson.annotation.JsonProperty;
 
-import lombok.AllArgsConstructor;
-import lombok.Data;
-import lombok.EqualsAndHashCode;
-import lombok.NoArgsConstructor;
+import lombok.*;
 import lombok.experimental.SuperBuilder;
 
 @Data
@@ -21,7 +16,7 @@
 @EqualsAndHashCode(callSuper = true)
 public class UpdateBotReq extends BaseReq {
 
-  @NotNull
+  @NonNull
   @JsonProperty("bot_id")
   private String botID;
 

api/src/main/java/com/coze/openapi/client/chat/model/ChatUsage.java

@@ -23,6 +23,12 @@ public class ChatUsage {
   @JsonProperty("output_count")
   private int outputCount;
 
+  @JsonProperty("output_tokens")
+  private int outputTokens;
+
+  @JsonProperty("input_tokens")
+  private int inputTokens;
+
   /** The total number of Tokens consumed for the input part. */
   @JsonProperty("input_count")
   private int inputCount;