Modify FuncchainSettings.llm to `str` type #33

danielpodrazka · 2025-01-29T18:49:09Z

The current union of BaseChatModel | str in FuncchainSettings is causing a failure when we try to set LLM in .env because it tries to call a BaseChatModel function on validation, which doesn't exist if the passed value is of str type. Changing it to str fixes the issue although I've noticed that llm is assigned BaseChatModel in some places so changing it to str doesn't fully describe what the variable can be. However, for the sake of the fix and for a working .env reading, I think it's a good trade-off.

The error that this change is fixing:

Traceback (most recent call last):
  File "/home/daniel/pp/venvs/funcchain/lib/python3.11/site-packages/IPython/core/interactiveshell.py", line 3577, in run_code
    exec(code_obj, self.user_global_ns, self.user_ns)
  File "<ipython-input-2-5eb8bcd049fd>", line 1, in <module>
    from funcchain import settings
  File "/home/daniel/.local/share/JetBrains/Toolbox/apps/pycharm-professional/plugins/python-ce/helpers/pydev/_pydev_bundle/pydev_import_hook.py", line 21, in do_import
    module = self._system_import(name, *args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/daniel/pp/funcchain/src/funcchain/__init__.py", line 3, in <module>
    from .backend.settings import settings
  File "/home/daniel/.local/share/JetBrains/Toolbox/apps/pycharm-professional/plugins/python-ce/helpers/pydev/_pydev_bundle/pydev_import_hook.py", line 21, in do_import
    module = self._system_import(name, *args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/daniel/pp/funcchain/src/funcchain/backend/settings.py", line 100, in <module>
    settings = FuncchainSettings()
               ^^^^^^^^^^^^^^^^^^^
  File "/home/daniel/pp/venvs/funcchain/lib/python3.11/site-packages/pydantic_settings/main.py", line 167, in __init__
    super().__init__(
  File "/home/daniel/pp/venvs/funcchain/lib/python3.11/site-packages/pydantic/main.py", line 212, in __init__
    validated_self = self.__pydantic_validator__.validate_python(data, self_instance=self)
                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/daniel/pp/venvs/funcchain/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py", line 236, in raise_deprecation
    if values.get("callback_manager") is not None:
       ^^^^^^^^^^
AttributeError: 'str' object has no attribute 'get'

To replicate the error:

Add .env file with content: LLM=something
Call from funcchain import settings

The current union of `BaseChatModel | str` is causing a failure when we try to set LLM in `.env` because it tries to call a `BaseChatModel` function on validation, which doesn't exist if the passed value is of `str` type

danielpodrazka · 2025-01-30T08:10:48Z

@shroominic

Modify FuncchainSettings.llm to str type.

f609462

The current union of `BaseChatModel | str` is causing a failure when we try to set LLM in `.env` because it tries to call a `BaseChatModel` function on validation, which doesn't exist if the passed value is of `str` type

danielpodrazka changed the title ~~Modify FuncchainSettings.llm to str type.~~ Modify FuncchainSettings.llm to str type Jan 29, 2025

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Modify FuncchainSettings.llm to `str` type #33

Modify FuncchainSettings.llm to `str` type #33

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Modify FuncchainSettings.llm to str type #33

Are you sure you want to change the base?

Modify FuncchainSettings.llm to str type #33

Uh oh!

Conversation

Uh oh!

Uh oh!

Uh oh!

Modify FuncchainSettings.llm to `str` type #33

Modify FuncchainSettings.llm to `str` type #33