Added new OpenAI chat models to the CHAT_OPENAI_MODELS list to improve the quality of the chatbot responses. The new models are "gpt-3.5-turbo-0613", "gpt-3.5-turbo-16k-0613", "gpt-4-0613", and "gpt-4-32k-0613". The existing models "gpt-3.5-turbo" and "gpt-4-32k" were also updated to their latest versions.
The status codes have been added to the API endpoints to improve the readability of the code and to make it more explicit what the expected response codes are. The status codes have been added to the following endpoints: create_flow, read_flows, read_flow, update_flow, delete_flow, create_flows, upload_file, and download_file.
✨ feat(endpoints.py): add flow_id parameter to predict_flow function to allow for running a flow by ID
The predict_flow function now includes a flow_id parameter and a session dependency to allow for running a flow by ID. The flow object is retrieved from the session using the flow_id parameter. If the flow is not found, a ValueError is raised.
🔨 refactor(constants.tsx): change API_URL constant to BASE_API_URL and add flow_id parameter to run_flow function
The API_URL constant has been renamed to BASE_API_URL to better reflect its purpose. The run_flow function now includes a flow_id parameter to allow for running a flow by ID. The flow_id parameter is used to construct the API URL.
The process_tweaks function has been refactored to improve readability and maintainability. The function now takes in two parameters, graph_data and tweaks, and returns the modified graph_data. The tweaks parameter is a dictionary of dictionaries, where the key is the node id and the value is a dictionary of the tweaks. The function processes the graph data to add the tweaks by iterating over the nodes and checking if the node id is in the tweaks dictionary. If it is, the function applies the tweaks to the node by updating the template data with the new values. The function also prints a message to the console to indicate that a tweak has been applied.
The PredictRequest schema now includes an optional tweaks field, which is a dictionary of dictionaries. This field allows for additional customization of the prediction request, such as specifying tool names or descriptions. The tweaks field is optional, and if not provided, the default value is an empty dictionary. The schema_extra attribute has also been updated to include an example of the new tweaks field.
🐛 fix(endpoints.py): change predict endpoint to use Flow object instead of flow_id
🐛 fix(endpoints.py): add support for processing tweaks in predict endpoint
The predict endpoint now requires authentication using HTTPBearer. The flow_id is now extracted from the bearer token instead of being passed as a parameter. This improves security as the flow_id is not exposed in the URL. The predict endpoint now uses the Flow object instead of the flow_id to retrieve the graph data. This improves code readability and reduces the number of database queries. The predict endpoint now supports processing tweaks, which allows for more flexibility in the processing of messages.
The graph_data variable was previously assigned to flow_obj.flow, which is incorrect as the flow_obj object does not have a flow attribute. Instead, the correct attribute to use is flow_obj.data.
The create_db_and_tables function is now called on app startup using the FastAPI on_event decorator. This ensures that the database and tables are created before the app starts listening for requests.
This commit extracts the formatting methods for Azure and Llama fields from the `format_field` method to improve readability and maintainability of the code. The `format_azure_field` method formats the fields for Azure, while the `format_llama_field` method formats the fields for Llama. These methods are called conditionally based on the name of the field.
🐛 fix(__main__.py): rename path variable to frontend_path for clarity
The first change fixes a type error for the multiprocess import. The second change renames the path variable to frontend_path to improve clarity and readability of the code.
🚀 feat(__main__.py): add optional open_browser argument to serve function to specify whether to open browser after starting server
The serve function now accepts an optional path argument to specify the path to the frontend directory containing build files. This is useful for development purposes only. The function also accepts an optional open_browser argument to specify whether to open the browser after starting the server. This is useful when running the server locally.
The code now checks if the streaming and stream attributes are boolean before setting them to True. This ensures that the attributes are not set to True if they are not boolean, which could cause errors in the code.
✨ feat(__main__.py): add banner with title and info text to be displayed on server start
🐛 fix(App.tsx): fix API endpoint URL
The sendAll function URL had an extra forward slash. The API endpoint URL in App.tsx was incorrect and has been fixed. A banner with a title and info text has been added to be displayed on server start to provide users with more information about the application.
The PythonFunction tool has been added to the list of available tools in the config.yaml file. This allows the backend to use Python functions as part of the language processing pipeline.
🧪 test(test_database.py): add test case for creating flows without a flow
The flow field is now optional to allow creation of flows without a flow. This is useful when creating a flow that will be populated later. A test case was added to ensure that flows can be created without a flow.
🐛 fix(llms.py): add name check before checking if "azure" is in name.lower()
🔨 refactor(test_database.py): rename updated_flow_style variable to to_update_flow_style for clarity
The update_settings function now has an optional database_url parameter to allow for more flexibility in updating settings. The llms.py file now checks if the name variable is not None before checking if "azure" is in name.lower(). In test_database.py, the updated_flow_style variable is renamed to to_update_flow_style for clarity.
The FlowListRead schema is added to support reading a list of flows with their styles. The SQLModelSerializable base model is added to support serialization of SQLModel objects to JSON using orjson. This improves performance and reduces memory usage.
🐛 fix(flow.py): add optional style relationship to Flow model
The style relationship is now optional to allow for flows without styles. This is achieved by setting the uselist parameter of the sa_relationship_kwargs to False.
✨ feat(flow.py): add FlowReadWithStyle and FlowUpdate models
The FlowReadWithStyle model is added to support reading a flow with its style. The FlowUpdate model is added to support updating a flow.
The FlowStyle model is added to the project, which represents the style of a flow. It has a color and an emoji field, and a foreign key to the Flow model. The CRUD classes are also added to the file, which are FlowStyleCreate, FlowStyleRead, and FlowStyleUpdate. These classes are used to create, read, and update FlowStyle instances respectively.
The imports for the deleted FlowStyle model are removed from flow_styles.py. The comments for the FlowStyleCreate class are updated to reflect the fields it contains.
✨ feat(router.py): add new routers for flows and flow styles
🔧 refactor(__init__.py): add new routers to __all__ list
🔧 refactor(conftest.py): update import statement for get_session function
The unused code and endpoints related to flows have been removed from the database.py file. New routers for flows and flow styles have been added to the router.py file. The __all__ list in the __init__.py file has been updated to include the new routers. The import statement for the get_session function in the conftest.py file has been updated to reflect the new location of the function.
The code was updated to add a null check for the name variable before checking if it contains the string "azure". This prevents a potential runtime error if the name variable is null.
🚀 feat(loading.py): add support for PythonFunction node type
🚀 feat(constants.py): add PythonFunction to CUSTOM_TOOLS
🚀 feat(custom.py): add PythonFunction class
🚀 feat(frontend_node/tools.py): add PythonFunctionNode class
🧪 test(test_custom_types.py): add test for PythonFunction class
🧪 test(test_llms_template.py): comment out tests for AzureOpenAI and AzureChatOpenAI
The changes add support for a new node type, PythonFunction, which allows users to define a Python function to be executed. The node type is added to CUSTOM_NODES in customs.py, and support for the node type is added to loading.py. The node type is also added to CUSTOM_TOOLS in constants.py, and the PythonFunction class is added to custom.py. The PythonFunctionNode class is added to frontend_node/tools.py. Tests for the new PythonFunction class are added to test_custom_types.py. Tests for AzureOpenAI and AzureChatOpenAI are commented out in test_llms_template.py.
🔨 refactor(custom.py, loading.py, prompts/custom.py, run.py): update import statements to use extract_input_variables_from_prompt from interface.utils module
🔨 refactor(run.py): remove unused imports and functions
🔨 refactor(utils.py): add type hinting to extract_input_variables_from_prompt function and remove unused imports
The extract_input_variables_from_prompt function has been moved to the interface.utils module to improve code organization. The import statements in the affected modules have been updated to reflect this change. Unused imports and functions have been removed from the run.py module. Type hinting has been added to the extract_input_variables_from_prompt function in the interface.utils module.
🚀 feat(processing): add processing module with get_result_and_steps and fix_memory_inputs functions
The processing module was added to the project with two functions: get_result_and_steps and fix_memory_inputs. The get_result_and_steps function extracts the result and thought from a LangChain object and returns them. The fix_memory_inputs function checks if a LangChain object has a memory attribute and if that memory key exists in the object's input variables. If not, it gets a possible new memory key using the get_memory_key function and updates the memory keys using the update_memory_keys function.
🚀 feat(utils.py): import extract_input_variables_from_prompt from langflow.interface.utils
The `from_payload` class method is added to the `Graph` class to create a graph from a payload. This method takes a dictionary as input and returns a `Graph` object. The `extract_input_variables_from_prompt` function is imported from `langflow.interface.utils` to extract input variables from a prompt. This function is used in other parts of the codebase to extract input variables from prompts.
✨ feat(utils.py): add process_graph function to process graph data and generate result and thought
The ChatManager class manages active connections and chat history. The ChatHistory class manages the chat history for a client. The process_graph function processes graph data and generates a result and thought. This function is used in the ChatManager class to generate a response back to the frontend.
This commit adds new API endpoints for chat, validation, and version. The chat endpoint is a websocket endpoint for chat. The validation endpoint has three sub-endpoints for validating code, prompt, and node. The version endpoint returns the version of LangFlow.
The base.py file contains the following classes and functions:
- CacheResponse: a pydantic BaseModel that represents a response containing a dictionary of data
- Code: a pydantic BaseModel that represents a code string
- Prompt: a pydantic BaseModel that represents a prompt template string
- CodeValidationResponse: a pydantic BaseModel that represents a response containing the validation results of code
- PromptValidationResponse: a pydantic BaseModel that represents a response containing the validation results of a prompt
- validate_prompt: a function that validates a prompt template string and returns a PromptValidationResponse object
- check_input_variables: a function that checks if input variables contain invalid characters and returns a list of fixed input variables
The callback.py file contains the following classes:
- AsyncStreamingLLMCallbackHandler: an AsyncCallbackHandler that handles streaming LLM responses asynchronously
- StreamingLLMCallbackHandler: a BaseCallbackHandler that handles streaming LLM responses
These files were added to provide support for Langflow's backend API.
The API now has versioning, with the prefix "/api/v1". The router has been restructured to include the chat, endpoints, and validate routers. This improves the organization of the code and makes it easier to add new routers in the future.
The routers for the langflow API have been moved to a single file for better organization and maintainability. The routers have been imported and included in the main.py file using the new file. A new health check endpoint has been added to the API to check the status of the application.
This commit refactors the FrontendNode class by extracting two methods to handle specific field values related to models and API keys. The _handle_model_specific_field_values method handles the options and is_list properties for fields related to models, while the _handle_api_key_specific_field_values method handles the display_name and required properties for fields related to API keys. This improves the readability and maintainability of the code.
✨ feat(flow.py): add validator to ensure flow field is a valid JSON object with required fields
The flow field in the FlowBase model has been changed from a string to a dictionary to allow for JSON data. A validator has been added to ensure that the flow field is a valid JSON object with the required fields. The tests have been updated to reflect these changes.
There are still some rough edges due to underlying langchain and
openai API limitations, e.g. hwchase17/langchain#3769 and
openai/openai-python#411. Notably, you can't use the Azure and
non-Azure node types in the same server, since there's global openai
configuration needed to choose between the two. So it's probably best
to still leave the Azure node types commented out in the default
config. But with this PR, if you uncomment those nodes and start the
server with OPENAI_API_TYPE=azure, you will have working Azure nodes.
✨ feat(database.py): add default argument to json.dumps to handle datetime objects
🚨 test(database.py): add tests for batch flow creation, file upload, and file download
The fix in database.py handles the case where the data dictionary does not contain the "flows" key. This is important because the code assumes that the "flows" key is present and will raise an exception if it is not. The fix adds a check to see if the "flows" key is present and if not, it creates a new FlowListCreate object with the data as a list of FlowCreate objects.
The feature in database.py adds a default argument to the json.dumps function to handle datetime objects. This is important because the default json encoder does not handle datetime objects and will raise an exception if it encounters one.
The tests in test_database.py cover the batch creation of flows, uploading a file containing flows, and downloading a file containing flows. These tests ensure that the endpoints are working as expected and that the data is being handled correctly.
🚀 feat(flowstyle.py): add FlowStyle model
🚀 feat(flowstyle.py): add FlowStyleCreate and FlowStyleRead models
🐛 fix(settings.py): correct typo in database_url variable name
The Flow model now has a relationship to the FlowStyle model, which allows for the creation of a FlowStyle object that is associated with a Flow object. The FlowStyle model is a new model that contains the color and emoji fields, which are used to style the Flow object. The FlowStyleCreate and FlowStyleRead models are used to create and read FlowStyle objects respectively. The typo in the database_url variable name has been corrected to ensure that the application connects to the correct database.
