- Updatethe `get_function_custom` function to use the `validate.extract_class_name` function to extract class name from the `code` parameter.
- Modify `instantiate_tool` function to update the `class_object` with the return value of `get_function_custom` and call the `build` method on the instantiated object with the remaining params.
- Add a new function `extract_class_name` in the `validate` module to extract class name from the `code` parameter.
This commit removes the file `extract_info_from_class.py` which contained unnecessary code.
[v1/endpoints.py] Fix error handling in custom_component endpoint
This commit fixes the error handling in the `custom_component` endpoint in `endpoints.py`. If the class template extracted from the code is not valid, an error message is printed.
[importing/utils.py] Comment out unused code
This commit comments out unused code in `get_function_custom` function in `utils.py` file.
[initialize/loading.py] Comment out unused code
This commit comments out unused code in the `instantiate_tool` function in `loading.py` file.
[interface/tools/custom.py] Refactor code and add properties
This commit refactors the code in `CustomComponent` class in `custom.py` file. It adds properties for `data`, `is_valid`, and `args_and_return_type`.
[interface/types.py] Add base classes to custom component template
This commit adds base classes to the custom component template in the `build_langchain_template_custom_component` function in `types.py` file.
[utils/constants.py] Remove unnecessary import
This commit removes an unnecessary import in `DEFAULT_CUSTOM_COMPONENT_CODE` constant in `constants.py` file.
The condition to check for the presence of "handle_keys" has been fixed to correctly check for its existence in the format_kwargs dictionary instead of the params dictionary. This ensures that the "handle_keys" list is properly populated when necessary.
The code was missing a condition to check if the variable is an instance of BaseOutputParser and has the "get_format_instructions" attribute. This condition is necessary to properly handle the variable and avoid potential errors.
🐛 fix(interface/initialize/loading.py): handle BaseOutputParser instances in instantiate_prompt
The `optional_inputs` list in the `Vertex` class is now extended with the `input_types` from the `template_dicts` dictionary. This ensures that the `optional_inputs` list contains all the input types that are not required. In the `instantiate_prompt` function, if the `variable` is an instance of `BaseOutputParser` and has a `get_format_instructions` method, the `format_kwargs` dictionary is updated with the result of calling `get_format_instructions()`. This allows handling `BaseOutputParser` instances correctly in the `instantiate_prompt` function.
This commit adds support for Custom Component in the Langflow interface. It introduces a new class `CustomComponent`, which takes in a `code` as a parameter and validates it. The `CustomComponent` class also provides a method to get the function specified in the code.
The commit also makes some modifications in `initialize/loading.py` file to handle the new `CustomComponent` class. It adds a new helper function `get_function_custom` which creates a function using `validate.create_function` and the `build` function name.
The code now checks if the "handle_keys" key is present in the params dictionary. If it is not present, it creates an empty list and adds the input_variable to it. This ensures that the "handle_keys" list is always present and the input_variable is correctly added to it.
🔧 fix(loading.py): add import statement for output_parser_creator
🔧 fix(listing.py): add import statement for output_parser_creator
🔧 fix(types.py): add import statement for output_parser_creator
The import statements for the `output_parsers` module were missing in the respective files. This fix adds the necessary import statements to ensure that the `output_parser_creator` is accessible and can be used in the code.
The instantiate_prompt function now properly formats the input variables before passing them to the prompt object. If the input variable is a string, it is passed as is. If the input variable is a list of Document objects, it is formatted to contain the page_content and metadata as one string separated by a newline. This ensures that the prompt object receives the correct input format.
- Added support for the CustomComponent tool in the Langflow API.
- The tool has been added to the config.yaml file.
- The CustomComponentNode class has been implemented in the frontend nodes.
- The code changes include modifications in various files for the implementation of the CustomComponent tool.
- The code changes include the addition of a new field "code" in the TemplateField class.
- The build_langchain_template_custom_component function has been implemented to build the template for the CustomComponent tool.
- New custom fields "my_id", "year", and "other_field" have been added to the template for the CustomComponent tool.
💡 chore(loading.py): improve code readability and maintainability by simplifying the instantiation of the prompt object and assignment of format_kwargs dictionary.
The code has been refactored to use a dictionary comprehension to create the format_kwargs dictionary. This simplifies the logic and improves code readability. The commented out code for the prompt.partial() method has been removed as it is no longer necessary.
🐛 fix(vertex/base.py): update instantiation logic to handle tuple result from loading.instantiate_class()
🐛 fix(loading.py): update return value of instantiate_prompt() to return a tuple of prompt and format_kwargs
The Vertex class now has a new attribute called artifacts, which is a dictionary used to store additional data related to the vertex. The instantiation logic in the Vertex class has been updated to handle the case where loading.instantiate_class() returns a tuple containing the built object and additional artifacts. The loading.instantiate_prompt() function now returns a tuple containing the prompt and format_kwargs. These changes fix issues related to storing and handling additional data in the vertex and loading modules.
🐛 fix(loading.py): set input variable values if they are present in params
The "input_variables" field in the prompt's frontend_node template is now updated with the new list of variables, if it exists. This ensures that the input variables are correctly set when validating the prompt. In the loading module, the input variable values are now set if they are present in the params dictionary. This ensures that the prompt is correctly instantiated with the provided input variable values.
The config.yaml file has been updated to include new prompt templates. In types.py, the input_variables list was not being cleared when prompt or messages were present, which has been fixed. The instantiate_prompt function in loading.py has been updated to support the new prompt templates.
🔀 chore(config): add new prompt templates to config.yaml
🐛 fix(types.py): fix input_variables not being cleared when prompt or messages are present
The `instantiate_documentloader` function now supports filtering files by extension using a `file_filter` parameter. The parameter is a string of comma-separated extensions, and the function now converts it into a lambda function that filters files based on whether their name contains any of the specified extensions. This improves the flexibility of the document loader by allowing it to load only specific types of files.
The `instantiate_vectorstore` function now uses a dictionary to initialize vector stores instead of a series of if-else statements. This improves the readability and maintainability of the code. A new dictionary `vecstore_initializer` is added to `vector_store.py` to map the class names of vector stores to their respective initialization functions.
The `instantiate_vectorstore` function now supports the `MongoDBAtlasVectorSearch` vector store. This allows for the use of MongoDB Atlas as a vector store for Langflow. The `search_kwargs` parameter is now supported for all vector stores that have a `as_retriever` method. This allows for the configuration of the vector store's search parameters.
The type hinting for allowed_tools variable is unnecessary as it is already defined in the previous line. Removing the type hinting improves the readability of the code.
🚀 feat(vector_store.py): add support for initializing SupabaseVectorStore
This commit adds support for initializing new vector store types in the loading.py file. Specifically, the initialize_weaviate, initialize_faiss, and initialize_supabase functions were added to support the Weaviate, FAISS, and SupabaseVectorStore vector stores, respectively. The vector_store.py file was also updated to include the SupabaseVectorStore class and the initialize_supabase function. This allows for more flexibility in choosing vector stores for the application.
📝 docs(initialize): add empty __init__.py file to initialize directory
The initialize directory was added to the project, but it was missing an __init__.py file. This file is necessary to make the directory a package and allow importing modules from it. An empty __init__.py file was added to the directory to fix this issue.
2023-06-22 13:45:04 -03:00
Renamed from src/backend/langflow/interface/loading.py (Browse further)
