This is an automated email from the ASF dual-hosted git repository. xtsong pushed a commit to branch main in repository https://gitbox.apache.org/repos/asf/flink-agents.git
commit fa5bb7eb7d8041e25c783525570f22b65c916816 Author: youjin <[email protected]> AuthorDate: Mon Jan 26 12:02:40 2026 +0800 [doc] Introduce Hierarchical ResourceClassName for Cross-Language Resource Creation --- docs/content/docs/development/chat_models.md | 73 +++++++++++----------- docs/content/docs/development/embedding_models.md | 46 ++++++-------- docs/content/docs/development/mcp.md | 18 +++--- docs/content/docs/development/react_agent.md | 4 +- docs/content/docs/development/vector_stores.md | 51 +++++++-------- docs/content/docs/development/workflow_agent.md | 4 +- .../docs/get-started/quickstart/react_agent.md | 6 +- .../docs/get-started/quickstart/workflow_agent.md | 4 +- 8 files changed, 97 insertions(+), 109 deletions(-) diff --git a/docs/content/docs/development/chat_models.md b/docs/content/docs/development/chat_models.md index 776beaa9..1bcbcd7f 100644 --- a/docs/content/docs/development/chat_models.md +++ b/docs/content/docs/development/chat_models.md @@ -65,7 +65,7 @@ class MyAgent(Agent): @staticmethod def ollama_connection() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OLLAMA_CHAT_MODEL_CONNECTION, + clazz=ResourceName.ChatModel.OLLAMA_CONNECTION, base_url="http://localhost:11434";, request_timeout=30.0 ) @@ -74,7 +74,7 @@ class MyAgent(Agent): @staticmethod def ollama_chat_model() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OLLAMA_CHAT_MODEL_SETUP, + clazz=ResourceName.ChatModel.OLLAMA_SETUP, connection="ollama_connection", model="qwen3:8b", temperature=0.7 @@ -106,14 +106,14 @@ class MyAgent(Agent): public class MyAgent extends Agent { @ChatModelConnection public static ResourceDescriptor ollamaConnection() { - return ResourceDescriptor.Builder.newBuilder(Constant.OLLAMA_CHAT_MODEL_CONNECTION) + return ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.OLLAMA_CONNECTION) .addInitialArgument("endpoint", "http://localhost:11434";) .build(); } @ChatModelSetup public static ResourceDescriptor ollamaChatModel() { - return ResourceDescriptor.Builder.newBuilder(Constant.OLLAMA_CHAT_MODEL_SETUP) + return ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.OLLAMA_SETUP) .addInitialArgument("connection", "ollamaConnection") .addInitialArgument("model", "qwen3:8b") .build(); @@ -196,7 +196,7 @@ Azure AI is only supported in Java currently. To use Azure AI from Python agents public class MyAgent extends Agent { @ChatModelConnection public static ResourceDescriptor azureAIConnection() { - return ResourceDescriptor.Builder.newBuilder(Constant.AZURE_CHAT_MODEL_CONNECTION) + return ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.AZURE_CONNECTION) .addInitialArgument("endpoint", "https://your-resource.inference.ai.azure.com";) .addInitialArgument("apiKey", "your-api-key-here") .build(); @@ -204,7 +204,7 @@ public class MyAgent extends Agent { @ChatModelSetup public static ResourceDescriptor azureAIChatModel() { - return ResourceDescriptor.Builder.newBuilder(Constant.AZURE_CHAT_MODEL_SETUP) + return ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.AZURE_SETUP) .addInitialArgument("connection", "azureAIConnection") .addInitialArgument("model", "gpt-4o") .build(); @@ -314,7 +314,7 @@ class MyAgent(Agent): @staticmethod def anthropic_connection() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.ANTHROPIC_CHAT_MODEL_CONNECTION, + clazz=ResourceName.ChatModel.ANTHROPIC_CONNECTION, api_key="<your-api-key>", max_retries=3, timeout=60.0 @@ -324,7 +324,7 @@ class MyAgent(Agent): @staticmethod def anthropic_chat_model() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.ANTHROPIC_CHAT_MODEL_SETUP, + clazz=ResourceName.ChatModel.ANTHROPIC_SETUP, connection="anthropic_connection", model="claude-sonnet-4-20250514", max_tokens=2048, @@ -340,7 +340,7 @@ class MyAgent(Agent): public class MyAgent extends Agent { @ChatModelConnection public static ResourceDescriptor anthropicConnection() { - return ResourceDescriptor.Builder.newBuilder(AnthropicChatModelConnection.class.getName()) + return ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.ANTHROPIC_CONNECTION) .addInitialArgument("api_key", "<your-api-key>") .addInitialArgument("timeout", 120) .addInitialArgument("max_retries", 3) @@ -349,7 +349,7 @@ public class MyAgent extends Agent { @ChatModelSetup public static ResourceDescriptor anthropicChatModel() { - return ResourceDescriptor.Builder.newBuilder(AnthropicChatModelSetup.class.getName()) + return ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.ANTHROPIC_SETUP) .addInitialArgument("connection", "anthropicConnection") .addInitialArgument("model", "claude-sonnet-4-20250514") .addInitialArgument("temperature", 0.7d) @@ -456,7 +456,7 @@ class MyAgent(Agent): @staticmethod def ollama_connection() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OLLAMA_CHAT_MODEL_CONNECTION, + clazz=ResourceName.ChatModel.OLLAMA_CONNECTION, base_url="http://localhost:11434";, request_timeout=120.0 ) @@ -465,7 +465,7 @@ class MyAgent(Agent): @staticmethod def my_chat_model() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OLLAMA_CHAT_MODEL_CONNECTION, + clazz=ResourceName.ChatModel.OLLAMA_CONNECTION, connection="ollama_connection", model="qwen3:8b", temperature=0.7, @@ -483,7 +483,7 @@ class MyAgent(Agent): public class MyAgent extends Agent { @ChatModelConnection public static ResourceDescriptor ollamaConnection() { - return ResourceDescriptor.Builder.newBuilder(Constant.OLLAMA_CHAT_MODEL_CONNECTION) + return ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.OLLAMA_CONNECTION) .addInitialArgument("endpoint", "http://localhost:11434";) .addInitialArgument("requestTimeout", 120) .build(); @@ -491,7 +491,7 @@ public class MyAgent extends Agent { @ChatModelSetup public static ResourceDescriptor ollamaChatModel() { - return ResourceDescriptor.Builder.newBuilder(Constant.OLLAMA_CHAT_MODEL_SETUP) + return ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.OLLAMA_SETUP) .addInitialArgument("connection", "ollamaConnection") .addInitialArgument("model", "qwen3:8b") .build(); @@ -614,7 +614,7 @@ class MyAgent(Agent): @staticmethod def openai_connection() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OPENAI_CHAT_MODEL_CONNECTION, + clazz=ResourceName.ChatModel.OPENAI_CONNECTION, api_key="<your-api-key>", api_base_url="https://api.openai.com/v1";, max_retries=3, @@ -625,7 +625,7 @@ class MyAgent(Agent): @staticmethod def openai_chat_model() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OPENAI_CHAT_MODEL_SETUP, + clazz=ResourceName.ChatModel.OPENAI_SETUP, connection="openai_connection", model="gpt-4", temperature=0.7, @@ -641,7 +641,7 @@ class MyAgent(Agent): public class MyAgent extends Agent { @ChatModelConnection public static ResourceDescriptor openaiConnection() { - return ResourceDescriptor.Builder.newBuilder(OpenAIChatModelConnection.class.getName()) + return ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.OPENAI_CONNECTION) .addInitialArgument("api_key", "<your-api-key>") .addInitialArgument("api_base_url", "https://api.openai.com/v1";) .addInitialArgument("timeout", 60) @@ -651,7 +651,7 @@ public class MyAgent extends Agent { @ChatModelSetup public static ResourceDescriptor openaiChatModel() { - return ResourceDescriptor.Builder.newBuilder(OpenAIChatModelSetup.class.getName()) + return ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.OPENAI_SETUP) .addInitialArgument("connection", "openaiConnection") .addInitialArgument("model", "gpt-4") .addInitialArgument("temperature", 0.7d) @@ -719,7 +719,7 @@ class MyAgent(Agent): @staticmethod def tongyi_connection() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.TONGYI_CHAT_MODEL_CONNECTION, + clazz=ResourceName.ChatModel.TONGYI_CONNECTION, api_key="your-api-key-here", # Or set DASHSCOPE_API_KEY env var request_timeout=60.0 ) @@ -728,7 +728,7 @@ class MyAgent(Agent): @staticmethod def tongyi_chat_model() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.TONGYI_CHAT_MODEL_SETUP, + clazz=ResourceName.ChatModel.Python.TONGYI_SETUP, connection="tongyi_connection", model="qwen-plus", temperature=0.7, @@ -758,6 +758,7 @@ Flink Agents supports cross-language chat model integration, allowing you to use {{< hint warning >}} **Limitations:** + - Cross-language resources are currently supported only when [running in Flink]({{< ref "docs/operations/deployment#run-in-flink" >}}), not in local development mode - Complex object serialization between languages may have limitations {{< /hint >}} @@ -766,8 +767,8 @@ Flink Agents supports cross-language chat model integration, allowing you to use To leverage chat model supports provided in a different language, you need to declare the resource within a built-in cross-language wrapper, and specify the target provider as an argument: -- **Using Java chat models in Python**: Use `Constant.JAVA_CHAT_MODEL_CONNECTION` and `Constant.JAVA_CHAT_MODEL_SETUP`, specifying the Java provider class via the `java_clazz` parameter -- **Using Python chat models in Java**: Use `Constant.PYTHON_CHAT_MODEL_CONNECTION` and `Constant.PYTHON_CHAT_MODEL_SETUP`, specifying the Python provider via `module` and `clazz` parameters +- **Using Java chat models in Python**: Use `ResourceName.ChatModel.JAVA_WRAPPER_CONNECTION` and `ResourceName.ChatModel.JAVA_WRAPPER_SETUP`, specifying the Java provider class via the `java_clazz` parameter +- **Using Python chat models in Java**: Use `ResourceName.ChatModel.PYTHON_WRAPPER_CONNECTION` and `ResourceName.ChatModel.PYTHON_WRAPPER_SETUP`, specifying the Python provider via the `pythonClazz` parameter @@ -784,13 +785,13 @@ class MyAgent(Agent): def java_chat_model_connection() -> ResourceDescriptor: # In pure Java, the equivalent ResourceDescriptor would be: # ResourceDescriptor.Builder - # .newBuilder(Constant.OllamaChatModelConnection) + # .newBuilder(ResourceName.ChatModel.OLLAMA_CONNECTION) # .addInitialArgument("endpoint", "http://localhost:11434";) # .addInitialArgument("requestTimeout", 120) # .build(); return ResourceDescriptor( - clazz=Constant.JAVA_CHAT_MODEL_CONNECTION, - java_clazz="org.apache.flink.agents.integrations.chatmodels.ollama.OllamaChatModelConnection", + clazz=ResourceName.ChatModel.JAVA_WRAPPER_CONNECTION, + java_clazz=ResourceName.ChatModel.Java.OLLAMA_CONNECTION, endpoint="http://localhost:11434";, requestTimeout=120, ) @@ -801,7 +802,7 @@ class MyAgent(Agent): def java_chat_model() -> ResourceDescriptor: # In pure Java, the equivalent ResourceDescriptor would be: # ResourceDescriptor.Builder - # .newBuilder(Constant.OllamaChatModelSetup) + # .newBuilder(ResourceName.ChatModel.OLLAMA_SETUP) # .addInitialArgument("connection", "java_chat_model_connection") # .addInitialArgument("model", "qwen3:8b") # .addInitialArgument("prompt", "my_prompt") @@ -809,8 +810,8 @@ class MyAgent(Agent): # .addInitialArgument("extractReasoning", true) # .build(); return ResourceDescriptor( - clazz=Constant.JAVA_CHAT_MODEL_SETUP, - java_clazz="org.apache.flink.agents.integrations.chatmodels.ollama.OllamaChatModelSetup", + clazz=ResourceName.ChatModel.JAVA_WRAPPER_SETUP, + java_clazz=ResourceName.ChatModel.Java.OLLAMA_SETUP, connection="java_chat_model_connection", model="qwen3:8b", prompt="my_prompt", @@ -847,13 +848,11 @@ public class MyAgent extends Agent { public static ResourceDescriptor pythonChatModelConnection() { // In pure Python, the equivalent ResourceDescriptor would be: // ResourceDescriptor( - // clazz=Constant.OLLAMA_CHAT_MODEL_CONNECTION, + // clazz=ResourceName.ChatModel.OLLAMA_CONNECTION, // request_timeout=120.0 // ) - return ResourceDescriptor.Builder.newBuilder(Constant.PYTHON_CHAT_MODEL_CONNECTION) - .addInitialArgument( - "module", "flink_agents.integrations.chat_models.ollama_chat_model") - .addInitialArgument("clazz", "OllamaChatModelConnection") + return ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.PYTHON_WRAPPER_CONNECTION) + .addInitialArgument("pythonClazz", ResourceName.ChatModel.Python.OLLAMA_CONNECTION) .addInitialArgument("request_timeout", 120.0) .build(); } @@ -862,16 +861,14 @@ public class MyAgent extends Agent { public static ResourceDescriptor pythonChatModel() { // In pure Python, the equivalent ResourceDescriptor would be: // ResourceDescriptor( - // clazz=Constant.OLLAMA_CHAT_MODEL_SETUP, + // clazz=ResourceName.ChatModel.OLLAMA_SETUP, // connection="pythonChatModelConnection", // model="qwen3:8b", // tools=["tool1", "tool2"], // extract_reasoning=True // ) - return ResourceDescriptor.Builder.newBuilder(Constant.PYTHON_CHAT_MODEL_SETUP) - .addInitialArgument( - "module", "flink_agents.integrations.chat_models.ollama_chat_model") - .addInitialArgument("clazz", "OllamaChatModelSetup") + return ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.PYTHON_WRAPPER_SETUP) + .addInitialArgument("pythonClazz", ResourceName.ChatModel.Python.OLLAMA_SETUP) .addInitialArgument("connection", "pythonChatModelConnection") .addInitialArgument("model", "qwen3:8b") .addInitialArgument("tools", List.of("tool1", "tool2")) diff --git a/docs/content/docs/development/embedding_models.md b/docs/content/docs/development/embedding_models.md index 400c477d..dc0af66b 100644 --- a/docs/content/docs/development/embedding_models.md +++ b/docs/content/docs/development/embedding_models.md @@ -65,7 +65,7 @@ class MyAgent(Agent): @staticmethod def openai_connection() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OPENAI_EMBEDDING_MODEL_CONNECTION, + clazz=ResourceName.EmbeddingModel.OPENAI_CONNECTION, api_key="your-api-key-here", base_url="https://api.openai.com/v1";, request_timeout=30.0 @@ -75,7 +75,7 @@ class MyAgent(Agent): @staticmethod def openai_embedding() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OPENAI_EMBEDDING_MODEL_SETUP, + clazz=ResourceName.EmbeddingModel.OPENAI_SETUP, connection="openai_connection", model="your-embedding-model-here" ) @@ -132,7 +132,7 @@ class MyAgent(Agent): @staticmethod def ollama_connection() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OLLAMA_EMBEDDING_MODEL_CONNECTION, + clazz=ResourceName.EmbeddingModel.OLLAMA_CONNECTION, base_url="http://localhost:11434";, request_timeout=30.0 ) @@ -141,7 +141,7 @@ class MyAgent(Agent): @staticmethod def ollama_embedding() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OLLAMA_EMBEDDING_MODEL_SETUP, + clazz=ResourceName.EmbeddingModel.OLLAMA_SETUP, connection="ollama_connection", model="nomic-embed-text", truncate=True, @@ -185,7 +185,7 @@ class MyAgent(Agent): @staticmethod def openai_connection() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OPENAI_EMBEDDING_MODEL_CONNECTION, + clazz=ResourceName.EmbeddingModel.OPENAI_CONNECTION, api_key="your-api-key-here", base_url="https://api.openai.com/v1";, request_timeout=30.0, @@ -196,7 +196,7 @@ class MyAgent(Agent): @staticmethod def openai_embedding() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OPENAI_CHAT_MODEL_SETUP, + clazz=ResourceName.EmbeddingModel.OPENAI_SETUP, connection="openai_connection", model="your-embedding-model-here", encoding_format="float" @@ -252,8 +252,8 @@ Flink Agents supports cross-language embedding model integration, allowing you t To leverage embedding model supports provided in a different language, you need to declare the resource within a built-in cross-language wrapper, and specify the target provider as an argument: -- **Using Java embedding models in Python**: Use `Constant.JAVA_EMBEDDING_MODEL_CONNECTION` and `Constant.JAVA_EMBEDDING_MODEL_SETUP`, specifying the Java provider class via the `java_clazz` parameter -- **Using Python embedding models in Java**: Use `Constant.PYTHON_EMBEDDING_MODEL_CONNECTION` and `Constant.PYTHON_EMBEDDING_MODEL_SETUP`, specifying the Python provider via `module` and `clazz` parameters +- **Using Java embedding models in Python**: Use `ResourceName.EmbeddingModel.JAVA_WRAPPER_CONNECTION` and `ResourceName.EmbeddingModel.JAVA_WRAPPER_SETUP`, specifying the Java provider class via the `java_clazz` parameter +- **Using Python embedding models in Java**: Use `ResourceName.EmbeddingModel.PYTHON_WRAPPER_CONNECTION` and `ResourceName.EmbeddingModel.PYTHON_WRAPPER_SETUP`, specifying the Python provider via the `pythonClazz` parameter ### Usage Example @@ -269,12 +269,12 @@ class MyAgent(Agent): def java_embedding_connection() -> ResourceDescriptor: # In pure Java, the equivalent ResourceDescriptor would be: # ResourceDescriptor.Builder - # .newBuilder(Constant.OllamaEmbeddingModelConnection) + # .newBuilder(ResourceName.EmbeddingModel.OLLAMA_CONNECTION) # .addInitialArgument("host", "http://localhost:11434";) # .build(); return ResourceDescriptor( - clazz=Constant.JAVA_EMBEDDING_MODEL_CONNECTION, - java_clazz="org.apache.flink.agents.integrations.embeddingmodels.ollama.OllamaEmbeddingModelConnection", + clazz=ResourceName.EmbeddingModel.JAVA_WRAPPER_CONNECTION, + java_clazz=ResourceName.EmbeddingModel.Java.OLLAMA_CONNECTION, host="http://localhost:11434"; ) @@ -283,13 +283,13 @@ class MyAgent(Agent): def java_embedding_model() -> ResourceDescriptor: # In pure Java, the equivalent ResourceDescriptor would be: # ResourceDescriptor.Builder - # .newBuilder(Constant.OllamaEmbeddingModelSetup) + # .newBuilder(ResourceName.EmbeddingModel.OLLAMA_SETUP) # .addInitialArgument("connection", "java_embedding_connection") # .addInitialArgument("model", "nomic-embed-text") # .build(); return ResourceDescriptor( - clazz=Constant.JAVA_EMBEDDING_MODEL_SETUP, - java_clazz="org.apache.flink.agents.integrations.embeddingmodels.ollama.OllamaEmbeddingModelSetup", + clazz=ResourceName.EmbeddingModel.JAVA_WRAPPER_SETUP, + java_clazz=ResourceName.EmbeddingModel.Java.OLLAMA_Setup, connection="java_embedding_connection", model="nomic-embed-text" ) @@ -314,14 +314,11 @@ public class MyAgent extends Agent { public static ResourceDescriptor pythonEmbeddingConnection() { // In pure Python, the equivalent ResourceDescriptor would be: // ResourceDescriptor( - // clazz=Constant.OLLAMA_EMBEDDING_MODEL_CONNECTION, + // clazz=ResourceName.EmbeddingModel.OLLAMA_CONNECTION, // base_url="http://localhost:11434"; // ) - return ResourceDescriptor.Builder.newBuilder(Constant.PYTHON_EMBEDDING_MODEL_CONNECTION) - .addInitialArgument( - "module", - "flink_agents.integrations.embedding_models.local.ollama_embedding_model") - .addInitialArgument("clazz", "OllamaEmbeddingModelConnection") + return ResourceDescriptor.Builder.newBuilder(ResourceName.EmbeddingModel.PYTHON_WRAPPER_CONNECTION) + .addInitialArgument("pythonClazz", ResourceName.EmbeddingModel.Python.OLLAMA_CONNECTION) .addInitialArgument("base_url", "http://localhost:11434";) .build(); } @@ -330,15 +327,12 @@ public class MyAgent extends Agent { public static ResourceDescriptor pythonEmbeddingModel() { // In pure Python, the equivalent ResourceDescriptor would be: // ResourceDescriptor( - // clazz=Constant.OLLAMA_EMBEDDING_MODEL_SETUP, + // clazz=ResourceName.EmbeddingModel.OLLAMA_SETUP, // connection="ollama_connection", // model="nomic-embed-text" // ) - return ResourceDescriptor.Builder.newBuilder(Constant.PYTHON_EMBEDDING_MODEL_SETUP) - .addInitialArgument( - "module", - "flink_agents.integrations.embedding_models.local.ollama_embedding_model") - .addInitialArgument("clazz", "OllamaEmbeddingModelSetup") + return ResourceDescriptor.Builder.newBuilder(ResourceName.EmbeddingModel.PYTHON_WRAPPER_SETUP) + .addInitialArgument("pythonClazz", ResourceName.EmbeddingModel.Python.OLLAMA_SETUP) .addInitialArgument("connection", "pythonEmbeddingConnection") .addInitialArgument("model", "nomic-embed-text") .build(); diff --git a/docs/content/docs/development/mcp.md b/docs/content/docs/development/mcp.md index d27b3433..d7aae5ea 100644 --- a/docs/content/docs/development/mcp.md +++ b/docs/content/docs/development/mcp.md @@ -1,4 +1,4 @@ ---- +from python.flink_agents.api.resource import ResourceName--- title: MCP weight: 9 type: docs @@ -40,7 +40,7 @@ class ReviewAnalysisAgent(Agent): @staticmethod def my_mcp_server() -> ResourceDescriptor: """Define MCP server connection.""" - return ResourceDescriptor(clazz=Constant.MCP_SERVER, + return ResourceDescriptor(clazz=ResourceName.MCP_SERVER, endpoint="http://127.0.0.1:8000/mcp";) ``` {{< /tab >}} @@ -51,7 +51,7 @@ public class ReviewAnalysisAgent extends Agent { @MCPServer public static ResourceDescriptor myMcp() { - return ResourceDescriptor.Builder.newBuilder(Constant.MCP_SERVER) + return ResourceDescriptor.Builder.newBuilder(ResourceName.MCP_SERVER) .addInitialArgument("endpoint", MCP_ENDPOINT) .addInitialArgument("timeout", 30) .build(); @@ -80,7 +80,7 @@ MCP servers can be configured with authentication: @staticmethod def authenticated_mcp_server() -> MCPServer: """Connect to MCP server with authentication.""" - return ResourceDescriptor(clazz=Constant.MCP_SERVER, + return ResourceDescriptor(clazz=ResourceName.MCP_SERVER, endpoint="http://api.example.com/mcp";, headers={"Authorization": "Bearer your-token"}) # Or using Basic Authentication @@ -97,7 +97,7 @@ def authenticated_mcp_server() -> MCPServer: @MCPServer public static org.apache.flink.agents.integrations.mcp.MCPServer authenticatedMcpServer() { // Using Bearer Token Authentication - return ResourceDescriptor.Builder.newBuilder(Constant.MCP_SERVER) + return ResourceDescriptor.Builder.newBuilder(ResourceName.MCP_SERVER) .addInitialArgument("endpoint", "http://api.example.com/mcp";) .addInitialArgument("timeout", 30) .addInitialArgument("auth", new BearerTokenAuth("your-oauth-token")) @@ -134,14 +134,14 @@ class ReviewAnalysisAgent(Agent): @staticmethod def review_mcp_server() -> ResourceDescriptor: """Connect to MCP server.""" - return ResourceDescriptor(clazz=Constant.MCP_SERVER, + return ResourceDescriptor(clazz=ResourceName.MCP_SERVER, endpoint="http://127.0.0.1:8000/mcp";) @chat_model_setup @staticmethod def review_model() -> ResourceDescriptor: return ResourceDescriptor( - clazz=OllamaChatModelSetup, + clazz=ResourceName.ChatModel.OLLAMA_SETUP, connection="ollama_server", model="qwen3:8b", # Reference MCP prompt by name like local prompt @@ -158,7 +158,7 @@ public class ReviewAnalysisAgent extends Agent { @MCPServer public static ResourceDescriptor myMcp() { - return ResourceDescriptor.Builder.newBuilder(Constant.MCP_SERVER) + return ResourceDescriptor.Builder.newBuilder(ResourceName.MCP_SERVER) .addInitialArgument("endpoint", "http://127.0.0.1:8000/mcp";) .addInitialArgument("timeout", 30) .build(); @@ -166,7 +166,7 @@ public class ReviewAnalysisAgent extends Agent { @ChatModelSetup public static ResourceDescriptor reviewModel() { - return ResourceDescriptor.Builder.newBuilder(OllamaChatModelSetup.class.getName()) + return ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.OLLAMA_SETUP) .addInitialArgument("connection", "ollamaChatModelConnection") .addInitialArgument("model", "qwen3:8b") // Reference MCP prompt by name like local prompt diff --git a/docs/content/docs/development/react_agent.md b/docs/content/docs/development/react_agent.md index 3bcd87c8..e559effc 100644 --- a/docs/content/docs/development/react_agent.md +++ b/docs/content/docs/development/react_agent.md @@ -69,7 +69,7 @@ We use `ResourceDescriptor` to describe the chat model, includes chat model type {{< tab "Python" >}} ```python chat_model_descriptor = ResourceDescriptor( - clazz=Constant.OLLAMA_CHAT_MODEL_SETUP, + clazz=ResourceName.ChatModel.OLLAMA_SETUP, connection="my_ollama_connection", model="qwen3:8b", tools=["my_tool1, my_tool2"], @@ -80,7 +80,7 @@ chat_model_descriptor = ResourceDescriptor( {{< tab "Java" >}} ```java ResourceDescriptor chatModelDescriptor = - ResourceDescriptor.Builder.newBuilder(Constant.OLLAMA_CHAT_MODEL_SETUP) + ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.OLLAMA_SETUP) .addInitialArgument("connection", "myOllamaConnection") .addInitialArgument("model", "qwen3:8b") .addInitialArgument( diff --git a/docs/content/docs/development/vector_stores.md b/docs/content/docs/development/vector_stores.md index a51a925c..2940da18 100644 --- a/docs/content/docs/development/vector_stores.md +++ b/docs/content/docs/development/vector_stores.md @@ -59,7 +59,7 @@ Flink Agents provides decorators/annotations to simplify vector store setup with @staticmethod def my_vector_store() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.CHROMA_VECTOR_STORE, + clazz=ResourceName.VectorStore.CHROMA_VECTOR_STORE, embedding_model="embedding_model", collection="my_chroma_store" ) @@ -70,7 +70,7 @@ def my_vector_store() -> ResourceDescriptor: ```java @VectorStore public static ResourceDescriptor vectorStore() { - return ResourceDescriptor.Builder.newBuilder(Constant.ELASTICSEARCH_VECTOR_STORE) + return ResourceDescriptor.Builder.newBuilder(ResourceName.VectorStore.ELASTICSEARCH_VECTOR_STORE) .addInitialArgument("embedding_model", "embeddingModel") .addInitialArgument("host", "http://localhost:9200";) .addInitialArgument("index", "my_documents") @@ -300,7 +300,7 @@ class MyAgent(Agent): @staticmethod def openai_connection() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OPENAI_EMBEDDING_MODEL_CONNECTION, + clazz=ResourceName.EmbeddingModel.OPENAI_CONNECTION, api_key="your-api-key-here" ) @@ -308,7 +308,7 @@ class MyAgent(Agent): @staticmethod def openai_embedding() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OPENAI_EMBEDDING_MODEL_SETUP, + clazz=ResourceName.EmbeddingModel.OPENAI_SETUP, connection="openai_connection", model="your-embedding-model-here" ) @@ -318,7 +318,7 @@ class MyAgent(Agent): @staticmethod def chroma_store() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.CHROMA_VECTOR_STORE, + clazz=ResourceName.VectorStore.CHROMA_VECTOR_STORE, embedding_model="openai_embedding", collection="my_chroma_store" ) @@ -352,14 +352,14 @@ public class MyAgent extends Agent { @EmbeddingModelConnection public static ResourceDescriptor embeddingConnection() { - return ResourceDescriptor.Builder.newBuilder(Constant.OLLAMA_EMBEDDING_MODEL_CONNECTION) + return ResourceDescriptor.Builder.newBuilder(ResourceName.EmbeddingModel.OLLAMA_CONNECTION) .addInitialArgument("host", "http://localhost:11434";) .build(); } @EmbeddingModelSetup public static ResourceDescriptor embeddingModel() { - return ResourceDescriptor.Builder.newBuilder(Constant.OLLAMA_EMBEDDING_MODEL_SETUP) + return ResourceDescriptor.Builder.newBuilder(ResourceName.EmbeddingModel.OLLAMA_SETUP) .addInitialArgument("connection", "embeddingConnection") .addInitialArgument("model", "nomic-embed-text") .build(); @@ -367,7 +367,7 @@ public class MyAgent extends Agent { @VectorStore public static ResourceDescriptor vectorStore() { - return ResourceDescriptor.Builder.newBuilder(Constant.ELASTICSEARCH_VECTOR_STORE) + return ResourceDescriptor.Builder.newBuilder(ResourceName.VectorStore.ELASTICSEARCH_VECTOR_STORE) .addInitialArgument("embedding_model", "embeddingModel") .addInitialArgument("host", "http://localhost:9200";) .addInitialArgument("index", "my_documents") @@ -446,7 +446,7 @@ class MyAgent(Agent): @staticmethod def openai_connection() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OPENAI_EMBEDDING_MODEL_CONNECTION, + clazz=ResourceName.EmbeddingModel.OPENAI_CONNECTION, api_key="your-api-key-here" ) @@ -454,7 +454,7 @@ class MyAgent(Agent): @staticmethod def openai_embedding() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.OPENAI_EMBEDDING_MODEL_SETUP, + clazz=ResourceName.EmbeddingModel.OPENAI_SETUP, connection="openai_connection", model="your-embedding-model-here" ) @@ -464,7 +464,7 @@ class MyAgent(Agent): @staticmethod def chroma_store() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.CHROMA_VECTOR_STORE, + clazz=ResourceName.VectorStore.CHROMA_VECTOR_STORE, embedding_model="openai_embedding", persist_directory="/path/to/chroma/data", # For persistent storage collection="my_documents", @@ -491,7 +491,7 @@ ChromaDB supports multiple deployment modes: @staticmethod def chroma_store() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.CHROMA_VECTOR_STORE, + clazz=ResourceName.VectorStore.CHROMA_VECTOR_STORE, embedding_model="your_embedding_model", collection="my_documents" # No connection configuration needed for in-memory mode @@ -504,7 +504,7 @@ def chroma_store() -> ResourceDescriptor: @staticmethod def chroma_store() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.CHROMA_VECTOR_STORE, + clazz=ResourceName.VectorStore.CHROMA_VECTOR_STORE, embedding_model="your_embedding_model", persist_directory="/path/to/chroma/data", collection="my_documents" @@ -517,7 +517,7 @@ def chroma_store() -> ResourceDescriptor: @staticmethod def chroma_store() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.CHROMA_VECTOR_STORE, + clazz=ResourceName.VectorStore.CHROMA_VECTOR_STORE, embedding_model="your_embedding_model", host="your-chroma-server.com", port=8000, @@ -531,7 +531,7 @@ def chroma_store() -> ResourceDescriptor: @staticmethod def chroma_store() -> ResourceDescriptor: return ResourceDescriptor( - clazz=Constant.CHROMA_VECTOR_STORE, + clazz=ResourceName.VectorStore.CHROMA_VECTOR_STORE, embedding_model="your_embedding_model", api_key="your-chroma-cloud-api-key", collection="my_documents" @@ -585,7 +585,7 @@ Here's how to define an Elasticsearch vector store in your Java agent: ```java @VectorStore public static ResourceDescriptor vectorStore() { - return ResourceDescriptor.Builder.newBuilder(Constant.ELASTICSEARCH_VECTOR_STORE) + return ResourceDescriptor.Builder.newBuilder(ResourceName.VectorStore.ELASTICSEARCH_VECTOR_STORE) .addInitialArgument("embedding_model", "embeddingModel") .addInitialArgument("host", "http://localhost:9200";) .addInitialArgument("index", "my_documents") @@ -616,8 +616,8 @@ Flink Agents supports cross-language vector store integration, allowing you to u To leverage vector store supports provided in a different language, you need to declare the resource within a built-in cross-language wrapper, and specify the target provider as an argument: -- **Using Java vector stores in Python**: Use `Constant.JAVA_COLLECTION_MANAGEABLE_VECTOR_STORE`, specifying the Java provider class via the `java_clazz` parameter -- **Using Python vector stores in Java**: Use `Constant.PYTHON_COLLECTION_MANAGEABLE_VECTOR_STORE`, specifying the Python provider via `module` and `clazz` parameters +- **Using Java vector stores in Python**: Use `ResourceName.VectorStore.JAVA_WRAPPER_COLLECTION_MANAGEABLE_VECTOR_STORE`, specifying the Java provider class via the `java_clazz` parameter +- **Using Python vector stores in Java**: Use `ResourceName.VectorStore.PYTHON_WRAPPER_COLLECTION_MANAGEABLE_VECTOR_STORE`, specifying the Python provider via the `pythonClazz` parameter ### Usage Example @@ -647,15 +647,15 @@ class MyAgent(Agent): def java_vector_store() -> ResourceDescriptor: # In pure Java, the equivalent ResourceDescriptor would be: # ResourceDescriptor.Builder - # .newBuilder(Constant.ElasticsearchVectorStore) + # .newBuilder(ResourceName.VectorStore.ELASTICSEARCH_VECTOR_STORE) # .addInitialArgument("embedding_model", "my_embedding_model") # .addInitialArgument("host", "http://localhost:9200";) # .addInitialArgument("index", "my_documents") # .addInitialArgument("dims", 768) # .build(); return ResourceDescriptor( - clazz=Constant.JAVA_COLLECTION_MANAGEABLE_VECTOR_STORE, - java_clazz="org.apache.flink.agents.integrations.vectorstores.elasticsearch.ElasticsearchVectorStore", + clazz=ResourceName.VectorStore.Java_WRAPPER_COLLECTION_MANAGEABLE_VECTOR_STORE, + java_clazz=ResourceName.VectorStore.Java.ELASTICSEARCH_VECTOR_STORE, embedding_model="my_embedding_model", host="http://localhost:9200";, index="my_documents", @@ -699,14 +699,11 @@ public class MyAgent extends Agent { public static ResourceDescriptor pythonVectorStore() { // In pure Python, the equivalent ResourceDescriptor would be: // ResourceDescriptor( - // clazz=Constant.CHROMA_VECTOR_STORE, + // clazz=ResourceName.VectorStore.CHROMA_VECTOR_STORE, // embedding_model="my_embedding_model", // ) - return ResourceDescriptor.Builder.newBuilder(PYTHON_COLLECTION_MANAGEABLE_VECTOR_STORE) - .addInitialArgument( - "module", - "flink_agents.integrations.vector_stores.chroma.chroma_vector_store") - .addInitialArgument("clazz", "ChromaVectorStore") + return ResourceDescriptor.Builder.newBuilder(ResourceName.VectorStore.PYTHON_WRAPPER_COLLECTION_MANAGEABLE_VECTOR_STORE) + .addInitialArgument("pythonClazz", ResourceName.VectorStore.Python.CHROMA_VECTOR_STORE) .addInitialArgument("embedding_model", "myEmbeddingModel") .build(); } diff --git a/docs/content/docs/development/workflow_agent.md b/docs/content/docs/development/workflow_agent.md index 230b655e..cea7f479 100644 --- a/docs/content/docs/development/workflow_agent.md +++ b/docs/content/docs/development/workflow_agent.md @@ -77,7 +77,7 @@ class ReviewAnalysisAgent(Agent): def review_analysis_model() -> ResourceDescriptor: """ChatModel which focus on review analysis.""" return ResourceDescriptor( - clazz=Constant.OLLAMA_CHAT_MODEL_SETUP, + clazz=ResourceName.ChatModel.OLLAMA_SETUP, connection="ollama_server", model="qwen3:8b", prompt="review_analysis_prompt", @@ -143,7 +143,7 @@ public class ReviewAnalysisAgent extends Agent { @ChatModelSetup public static ResourceDescriptor reviewAnalysisModel() { - return ResourceDescriptor.Builder.newBuilder(Constant.OLLAMA_CHAT_MODEL_SETUP) + return ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.OLLAMA_SETUP) .addInitialArgument("connection", "ollamaChatModelConnection") .addInitialArgument("model", "qwen3:8b") .addInitialArgument("prompt", "reviewAnalysisPrompt") diff --git a/docs/content/docs/get-started/quickstart/react_agent.md b/docs/content/docs/get-started/quickstart/react_agent.md index e3fa22e1..48cfaff2 100644 --- a/docs/content/docs/get-started/quickstart/react_agent.md +++ b/docs/content/docs/get-started/quickstart/react_agent.md @@ -49,7 +49,7 @@ agents_env = AgentsExecutionEnvironment.get_execution_environment(env) agents_env.add_resource( "ollama_server", ResourceType.CHAT_MODEL_CONNECTION, - ResourceDescriptor(clazz=Constant.OLLAMA_CHAT_MODEL_CONNECTION, request_timeout=120), + ResourceDescriptor(clazz=ResourceName.ChatModel.OLLAMA_CONNECTION, request_timeout=120), ).add_resource( "notify_shipping_manager", ResourceType.TOOL, Tool.from_callable(notify_shipping_manager) ) @@ -90,7 +90,7 @@ Create the ReAct Agent instance, configure the chat model, prompt and the output ```python review_analysis_react_agent = ReActAgent( chat_model=ResourceDescriptor( - clazz=Constant.OLLAMA_CHAT_MODEL_SETUP, + clazz=ResourceName.ChatModel.OLLAMA_SETUP, connection="ollama_server", model="qwen3:8b", tools=["notify_shipping_manager"], @@ -108,7 +108,7 @@ ReActAgent reviewAnalysisReactAgent = getReActAgent(); private static ReActAgent getReActAgent() { return new ReActAgent( - ResourceDescriptor.Builder.newBuilder(Constant.OLLAMA_CHAT_MODEL_SETUP) + ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.OLLAMA_SETUP) .addInitialArgument("connection", "ollamaChatModelConnection") .addInitialArgument("model", "qwen3:8b") .addInitialArgument( diff --git a/docs/content/docs/get-started/quickstart/workflow_agent.md b/docs/content/docs/get-started/quickstart/workflow_agent.md index 43c3e3fc..6dc3282c 100644 --- a/docs/content/docs/get-started/quickstart/workflow_agent.md +++ b/docs/content/docs/get-started/quickstart/workflow_agent.md @@ -119,7 +119,7 @@ class ReviewAnalysisAgent(Agent): def review_analysis_model() -> ResourceDescriptor: """ChatModel which focus on review analysis.""" return ResourceDescriptor( - clazz=Constant.OLLAMA_CHAT_MODEL_SETUP, + clazz=ResourceName.ChatModel.OLLAMA_SETUP, connection="ollama_server", model="qwen3:8b", prompt="review_analysis_prompt", @@ -185,7 +185,7 @@ public class ReviewAnalysisAgent extends Agent { @ChatModelSetup public static ResourceDescriptor reviewAnalysisModel() { - return ResourceDescriptor.Builder.newBuilder(Constant.OLLAMA_CHAT_MODEL_SETUP) + return ResourceDescriptor.Builder.newBuilder(ResourceName.ChatModel.OLLAMA_SETUP) .addInitialArgument("connection", "ollamaChatModelConnection") .addInitialArgument("model", "qwen3:8b") .addInitialArgument("prompt", "reviewAnalysisPrompt")
