Merge pull request #89 from kwongiho/main

Add support for deepseek-r1 model

.github/workflows/close-issue.yml

@@ -0,0 +1,29 @@
+name: Close inactive issues
+on:
+  workflow_dispatch: # for manual run
+  schedule:
+    - cron: "0 0 * * *" # Runs daily at midnight
+
+# Fine-grant permission
+# https://docs.github.com/en/actions/security-for-github-actions/security-guides/automatic-token-authentication#modifying-the-permissions-for-the-github_token
+permissions:
+  issues: write
+
+jobs:
+  close-issues:
+    runs-on: ubuntu-latest
+    permissions:
+      issues: write
+      pull-requests: write
+    steps:
+      - uses: actions/stale@v5
+        with:
+          exempt-issue-labels: "refactor,help wanted,good first issue,research,bug"
+          days-before-issue-stale: 7  # Issues become stale after 7 days
+          days-before-issue-close: 15 # Issues close 15 days after being marked stale
+          stale-issue-label: "stale"
+          close-issue-message: "This issue was closed because it has been inactive for 15 days since being marked as stale."
+          days-before-pr-stale: -1 # PRs are not handled
+          days-before-pr-close: -1 # PRs are not handled
+          operations-per-run: 10000
+          repo-token: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/label-issue-stale.yml

@@ -0,0 +1,28 @@
+name: Mark stale issues and pull requests
+
+on:
+  workflow_dispatch: # for manual run
+  schedule:
+    - cron: '0 0 * * *'  # Runs every day at midnight
+
+permissions:
+  contents: write # only for delete-branch option
+  issues: write
+  pull-requests: write
+
+jobs:
+  stale:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Mark stale issues and pull requests
+        uses: actions/stale@v8
+        with:
+          repo-token: ${{ github.token }}
+          days-before-stale: 15
+          stale-issue-message: 'This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.'
+          stale-pr-message: 'This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.'
+          days-before-close: 7
+          stale-issue-label: 'stale'
+          exempt-issue-labels: 'pinned,security'
+          stale-pr-label: 'stale'
+          exempt-pr-labels: 'work-in-progress'

README.md

@@ -283,6 +283,8 @@ If you like or are using this project to build your own, please give us a star.
 | 7 | Katie Backend     | An open-source AI-based question-answering platform for accessing private domain knowledge                    | [GitHub](https://github.com/wyona/katie-backend)                                                                                               |
 | 8 | TeleLlama3 Bot    | A question-answering Telegram bot                                                                             | [Repo](https://git.hiast.edu.sy/mohamadbashar.disoki/telellama3-bot)                                                                           |
 | 9 | moqui-wechat      | A moqui-wechat component                                                                                      | [GitHub](https://github.com/heguangyong/moqui-wechat)                                                                                          |
+| 10 | B4X      | A set of simple and powerful RAD tool for Desktop and Server development                                              | [Website](https://www.b4x.com/android/forum/threads/ollama4j-library-pnd_ollama4j-your-local-offline-llm-like-chatgpt.165003/)     |                                                              
+
 
 ## Traction
 

docs/docs/apis-generate/generate-with-tools.md

@@ -521,6 +521,23 @@ public class MyOllamaService{
 }
 ```
 
+Or, if one needs to provide an object instance directly:
+```java
+public class MyOllamaService{
+    
+    public void chatWithAnnotatedTool(){
+        ollamaAPI.registerAnnotatedTools(new BackendService());
+        OllamaChatRequest requestModel = builder
+                .withMessage(OllamaChatMessageRole.USER,
+                        "Compute the most important constant in the world using 5 digits")
+                .build();
+
+        OllamaChatResult chatResult = ollamaAPI.chat(requestModel);
+    }
+    
+}
+```
+
 The request should be the following:
 
 ```json

src/main/java/io/github/ollama4j/OllamaAPI.java

@@ -11,6 +11,7 @@ import io.github.ollama4j.models.embeddings.OllamaEmbeddingsRequestModel;
 import io.github.ollama4j.models.embeddings.OllamaEmbedResponseModel;
 import io.github.ollama4j.models.generate.OllamaGenerateRequest;
 import io.github.ollama4j.models.generate.OllamaStreamHandler;
+import io.github.ollama4j.models.generate.OllamaTokenHandler;
 import io.github.ollama4j.models.ps.ModelsProcessResponse;
 import io.github.ollama4j.models.request.*;
 import io.github.ollama4j.models.response.*;
@@ -785,15 +786,34 @@ public class OllamaAPI {
      * @throws InterruptedException if the operation is interrupted
      */
     public OllamaChatResult chat(OllamaChatRequest request, OllamaStreamHandler streamHandler) throws OllamaBaseException, IOException, InterruptedException {
+        return chatStreaming(request, new OllamaChatStreamObserver(streamHandler));
+    }
+
+    /**
+     * Ask a question to a model using an {@link OllamaChatRequest}. This can be constructed using an {@link OllamaChatRequestBuilder}.
+     * <p>
+     * Hint: the OllamaChatRequestModel#getStream() property is not implemented.
+     *
+     * @param request       request object to be sent to the server
+     * @param tokenHandler  callback handler to handle the last token from stream (caution: all previous messages from stream will be concatenated)
+     * @return {@link OllamaChatResult}
+     * @throws OllamaBaseException  any response code than 200 has been returned
+     * @throws IOException          in case the responseStream can not be read
+     * @throws InterruptedException in case the server is not reachable or network issues happen
+     * @throws OllamaBaseException  if the response indicates an error status
+     * @throws IOException          if an I/O error occurs during the HTTP request
+     * @throws InterruptedException if the operation is interrupted
+     */
+    public OllamaChatResult chatStreaming(OllamaChatRequest request, OllamaTokenHandler tokenHandler) throws OllamaBaseException, IOException, InterruptedException {
         OllamaChatEndpointCaller requestCaller = new OllamaChatEndpointCaller(host, basicAuth, requestTimeoutSeconds, verbose);
         OllamaChatResult result;
 
         // add all registered tools to Request
         request.setTools(toolRegistry.getRegisteredSpecs().stream().map(Tools.ToolSpecification::getToolPrompt).collect(Collectors.toList()));
 
-        if (streamHandler != null) {
+        if (tokenHandler != null) {
             request.setStream(true);
-            result = requestCaller.call(request, streamHandler);
+            result = requestCaller.call(request, tokenHandler);
         } else {
             result = requestCaller.callSync(request);
         }
@@ -810,8 +830,8 @@ public class OllamaAPI {
                 request.getMessages().add(new OllamaChatMessage(OllamaChatMessageRole.TOOL,"[TOOL_RESULTS]" + toolName + "(" + arguments.keySet() +") : " + res + "[/TOOL_RESULTS]"));
             }
 
-            if (streamHandler != null) {
+            if (tokenHandler != null) {
-                result = requestCaller.call(request, streamHandler);
+                result = requestCaller.call(request, tokenHandler);
             } else {
                 result = requestCaller.callSync(request);
             }
@@ -826,7 +846,9 @@ public class OllamaAPI {
         toolRegistry.addTool(toolSpecification.getFunctionName(), toolSpecification);
     }
 
+
     public void registerAnnotatedTools()  {
+        try {
             Class<?> callerClass = null;
             try {
                 callerClass = Class.forName(Thread.currentThread().getStackTrace()[2].getClassName());
@@ -835,14 +857,22 @@ public class OllamaAPI {
             }
 
             OllamaToolService ollamaToolServiceAnnotation = callerClass.getDeclaredAnnotation(OllamaToolService.class);
-        if(ollamaToolServiceAnnotation == null) {
+            if (ollamaToolServiceAnnotation == null) {
                 throw new IllegalStateException(callerClass + " is not annotated as " + OllamaToolService.class);
             }
 
             Class<?>[] providers = ollamaToolServiceAnnotation.providers();
+            for (Class<?> provider : providers) {
+                registerAnnotatedTools(provider.getDeclaredConstructor().newInstance());
+            }
+        } catch (InstantiationException | NoSuchMethodException | IllegalAccessException | InvocationTargetException e) {
+            throw new RuntimeException(e);
+        }
+    }
 
-        for(Class<?> provider : providers){
+    public void registerAnnotatedTools(Object object)  {
-            Method[] methods = provider.getMethods();
+        Class<?> objectClass = object.getClass();
+        Method[] methods = objectClass.getMethods();
         for(Method m : methods) {
             ToolSpec toolSpec = m.getDeclaredAnnotation(ToolSpec.class);
             if(toolSpec == null){
@@ -895,19 +925,10 @@ public class OllamaAPI {
                     )
                     .build();
 
-                try {
             ReflectionalToolFunction reflectionalToolFunction =
-                            new ReflectionalToolFunction(provider.getDeclaredConstructor().newInstance()
+                    new ReflectionalToolFunction(object, m, methodParams);
-                                    ,m
-                                    ,methodParams);
-
             toolSpecification.setToolFunction(reflectionalToolFunction);
             toolRegistry.addTool(toolSpecification.getFunctionName(),toolSpecification);
-                } catch (InstantiationException | IllegalAccessException | InvocationTargetException |
-                         NoSuchMethodException e) {
-                    throw new RuntimeException(e);
-                }
-            }
         }
 
     }

src/main/java/io/github/ollama4j/models/chat/OllamaChatStreamObserver.java

@@ -1,31 +1,19 @@
 package io.github.ollama4j.models.chat;
 
 import io.github.ollama4j.models.generate.OllamaStreamHandler;
+import io.github.ollama4j.models.generate.OllamaTokenHandler;
+import lombok.RequiredArgsConstructor;
 
-import java.util.ArrayList;
+@RequiredArgsConstructor
-import java.util.List;
+public class OllamaChatStreamObserver implements OllamaTokenHandler {
-
+    private final OllamaStreamHandler streamHandler;
-public class OllamaChatStreamObserver {
-
-    private OllamaStreamHandler streamHandler;
-
-    private List<OllamaChatResponseModel> responseParts = new ArrayList<>();
-
     private String message = "";
 
-    public OllamaChatStreamObserver(OllamaStreamHandler streamHandler) {
+    @Override
-        this.streamHandler = streamHandler;
+    public void accept(OllamaChatResponseModel token) {
-    }
+        if (streamHandler != null) {
-
+            message += token.getMessage().getContent();
-    public void notify(OllamaChatResponseModel currentResponsePart) {
-        responseParts.add(currentResponsePart);
-        handleCurrentResponsePart(currentResponsePart);
-    }
-
-    protected void handleCurrentResponsePart(OllamaChatResponseModel currentResponsePart) {
-        message = message + currentResponsePart.getMessage().getContent();
             streamHandler.accept(message);
         }
-
+    }
-
 }

src/main/java/io/github/ollama4j/models/generate/OllamaTokenHandler.java

@@ -0,0 +1,8 @@
+package io.github.ollama4j.models.generate;
+
+import io.github.ollama4j.models.chat.OllamaChatResponseModel;
+
+import java.util.function.Consumer;
+
+public interface OllamaTokenHandler extends Consumer<OllamaChatResponseModel> {
+}

src/main/java/io/github/ollama4j/models/request/OllamaChatEndpointCaller.java

@@ -4,9 +4,8 @@ import com.fasterxml.jackson.core.JsonProcessingException;
 import com.fasterxml.jackson.core.type.TypeReference;
 import io.github.ollama4j.exceptions.OllamaBaseException;
 import io.github.ollama4j.models.chat.*;
+import io.github.ollama4j.models.generate.OllamaTokenHandler;
 import io.github.ollama4j.models.response.OllamaErrorResponse;
-import io.github.ollama4j.models.generate.OllamaStreamHandler;
-import io.github.ollama4j.tools.Tools;
 import io.github.ollama4j.utils.Utils;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
@@ -29,7 +28,7 @@ public class OllamaChatEndpointCaller extends OllamaEndpointCaller {
 
     private static final Logger LOG = LoggerFactory.getLogger(OllamaChatEndpointCaller.class);
 
-    private OllamaChatStreamObserver streamObserver;
+    private OllamaTokenHandler tokenHandler;
 
     public OllamaChatEndpointCaller(String host, BasicAuth basicAuth, long requestTimeoutSeconds, boolean verbose) {
         super(host, basicAuth, requestTimeoutSeconds, verbose);
@@ -60,8 +59,8 @@ public class OllamaChatEndpointCaller extends OllamaEndpointCaller {
             OllamaChatMessage message = ollamaResponseModel.getMessage();
             if(message != null) {
                 responseBuffer.append(message.getContent());
-                if (streamObserver != null) {
+                if (tokenHandler != null) {
-                    streamObserver.notify(ollamaResponseModel);
+                    tokenHandler.accept(ollamaResponseModel);
                 }
             }
             return ollamaResponseModel.isDone();
@@ -71,9 +70,9 @@ public class OllamaChatEndpointCaller extends OllamaEndpointCaller {
         }
     }
 
-    public OllamaChatResult call(OllamaChatRequest body, OllamaStreamHandler streamHandler)
+    public OllamaChatResult call(OllamaChatRequest body, OllamaTokenHandler tokenHandler)
             throws OllamaBaseException, IOException, InterruptedException {
-        streamObserver = new OllamaChatStreamObserver(streamHandler);
+        this.tokenHandler = tokenHandler;
         return callSync(body);
     }
 
@@ -86,7 +85,7 @@ public class OllamaChatEndpointCaller extends OllamaEndpointCaller {
                         .POST(
                                 body.getBodyPublisher());
         HttpRequest request = requestBuilder.build();
-        if (isVerbose()) LOG.info("Asking model: " + body.toString());
+        if (isVerbose()) LOG.info("Asking model: " + body);
         HttpResponse<InputStream> response =
                 httpClient.send(request, HttpResponse.BodyHandlers.ofInputStream());
 

src/main/java/io/github/ollama4j/types/OllamaModelType.java

@@ -15,6 +15,7 @@ public class OllamaModelType {
     public static final String LLAMA3_1 = "llama3.1";
     public static final String MISTRAL = "mistral";
     public static final String MIXTRAL = "mixtral";
+    public static final String DEEPSEEK_R1 = "deepseek-r1";
     public static final String LLAVA = "llava";
     public static final String LLAVA_PHI3 = "llava-phi3";
     public static final String NEURAL_CHAT = "neural-chat";

src/test/java/io/github/ollama4j/integrationtests/TestRealAPIs.java

@@ -321,7 +321,7 @@ class TestRealAPIs {
             assertEquals(1, function.getArguments().size());
             Object noOfDigits = function.getArguments().get("noOfDigits");
             assertNotNull(noOfDigits);
-            assertEquals("5",noOfDigits);
+            assertEquals("5", noOfDigits.toString());
             assertTrue(chatResult.getChatHistory().size()>2);
             List<OllamaChatToolCalls> finalToolCalls = chatResult.getResponseModel().getMessage().getToolCalls();
             assertNull(finalToolCalls);
@@ -338,7 +338,7 @@ class TestRealAPIs {
             ollamaAPI.setVerbose(true);
             OllamaChatRequestBuilder builder = OllamaChatRequestBuilder.getInstance(config.getModel());
 
-            ollamaAPI.registerAnnotatedTools();
+            ollamaAPI.registerAnnotatedTools(new AnnotatedTool());
 
             OllamaChatRequest requestModel = builder
                     .withMessage(OllamaChatMessageRole.USER,