AttributeError: 'super' object has no attribute 'parse_obj' when using OpenAPISpec.from_file

elliexcoding commented 10 months ago

System Info

Langchain version 0.0.268

Who can help?

No response

Information

[ ] The official example notebooks/scripts
[ ] My own modified scripts

Related Components

[ ] LLMs/Chat Models
[ ] Embedding Models
[ ] Prompts / Prompt Templates / Prompt Selectors
[ ] Output Parsers
[ ] Document Loaders
[ ] Vector Stores / Retrievers
[ ] Memory
[ ] Agents / Agent Executors
[X] Tools / Toolkits
[ ] Chains
[ ] Callbacks/Tracing
[ ] Async

Reproduction

Load in .yaml file containing the OpenAPI specification.
Call the OpenAPISpec with open_api_spec = OpenAPISpec.from_file(file_path)

Error shows that in openapi.py line 202 ,return super().parse_obj(obj) no longer inherits the super method from Pydantic's BaseModel parse_obj.

Expected behavior

The OpenAPISpec should be delivered so that it can be passed on to NLAToolkit.from_llm_and_spec.

dosubot[bot] commented 10 months ago

🤖

Hello,

The error you're encountering, 'super' object has no attribute 'parse_obj', is likely due to the fact that the 'parse_obj' method is not available in the superclass of 'OpenAPISpec'.

In LangChain, 'OpenAPISpec' is a subclass of 'OpenAPI', which is imported from the 'openapi_schema_pydantic' package. The 'parse_obj' method is called on the superclass in the 'parse_obj' method of 'OpenAPISpec':

@classmethod
def parse_obj(cls, obj: dict) -> OpenAPISpec:
    try:
        cls._alert_unsupported_spec(obj)
        return super().parse_obj(obj)
    except ValidationError as e:
        # We are handling possibly misconfigured specs and
        # want to do a best-effort job to get a reasonable interface out of it.
        new_obj = copy.deepcopy(obj)
        for error in e.errors():
            keys = error["loc"]
            item = new_obj
            for key in keys[:-1]:
                item = item[key]
            item.pop(keys[-1], None)
        return cls.parse_obj(new_obj)

The error suggests that the 'OpenAPI' class does not have a 'parse_obj' method. This could be due to a version mismatch between the 'openapi_schema_pydantic' package used in your environment and the one expected by LangChain.

LangChain v0.0.268 is compatible with the 'openapi_schema_pydantic' package version "^1.2". This means any version that is compatible with version 1.2. The caret (^) symbol in "^1.2" indicates compatibility with versions up to but not including 2.0.0. This information is found in the 'pyproject.toml' file under the '[tool.poetry.dependencies]' section:

openapi-schema-pydantic = {version = "^1.2", optional = true}

Please ensure that you have the correct version of 'openapi_schema_pydantic' installed. If the issue persists, it might be a bug in LangChain or 'openapi_schema_pydantic', and you should consider reporting it to the maintainers.

I hope this helps! Let me know if you have any other questions.

Sources

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

kreneskyp commented 10 months ago

Also affecting

OpenAPISpec.from_url
get_openapi_chain

The get_openapi_chain error is hidden, but I confirmed it's the same error as above.

File "/usr/local/lib/python3.11/site-packages/langchain/chains/openai_functions/openapi.py", line 272, in get_openapi_chain
    raise ValueError(f"Unable to parse spec from source {spec}")

kreneskyp commented 10 months ago

cc @baskaryan @eyurtsev

Tracked this regression to openapi-schema-pydantic being switched to an optional dependency (https://github.com/langchain-ai/langchain/pull/9408).

Work around:

install openapi-schema-pydantic into your environment.

Root cause:

OpenAPISpec's parent class defaults to object when openapi-schema-pydantic is unavailable.

https://github.com/langchain-ai/langchain/blob/master/libs/langchain/langchain/utilities/openapi.py#L54-L57
This leads to an exception in OpenAPISpec.parse_obj when it calls super().parse_obj().

https://github.com/langchain-ai/langchain/blob/master/libs/langchain/langchain/utilities/openapi.py#L201-L202

batmanscode commented 10 months ago

Thanks for looking into it and figuring out a workaround @kreneskyp

Unfortunately for me installing openapi-schema-pydantic didn't work

EnkrateiaLucca commented 9 months ago

Having the same issue, installing openaiapi-schema-pydantic did not work. :(

kreneskyp commented 8 months ago

Having the same issue, installing openaiapi-schema-pydantic did not work. :(

I ran into this issue again when using Pydantic 2.0 with LangChain.

Try langchain==0.0.319 with openapi-pydantic==0.3.2. Hopefully fixes all the issues now!

PizBernina commented 8 months ago

Having the same issue, installing openaiapi-schema-pydantic did not work. :(

I ran into this issue again when using Pydantic 2.0 with LangChain.

Try langchain==0.0.319 with openapi-pydantic==0.3.2. Hopefully fixes all the issues now!

This resolved the issue for me, cheers!

borisko123 commented 8 months ago

openapi_schema_pydantic is changed to: openapi-pydantic 0.3.2 (last version) install it when using langchain

jepsenwan commented 7 months ago

openapi_schema_pydantic is changed to: openapi-pydantic 0.3.2 (last version) install it when using langchain

This works perfectly fine, thanks for the advise.

marufaytekin commented 6 months ago

openapi_schema_pydantic is changed to: openapi-pydantic 0.3.2 (last version) install it when using langchain

This worked fine, thanks for the advise.

payamsaremi commented 5 months ago

"pip install openapi-pydantic" fixed it for me

xinyangfan commented 3 months ago

langchain 0.1.9 openapi-pydantic 0.4.0 Still having this issue

prova commented 3 months ago

My colleague is facing the same issue on Python 3.10.12 and langchain==0.1.12 plus openapi-pydantic==0.4.0 (and also tried 0.3.2). On my PC, all is good with the same versions and Python. This is some random stuff.

ChanduTadanki commented 3 months ago

openapi_schema_pydantic is changed to: openapi-pydantic 0.3.2 (last version) install it when using langchain

That solved my issue. Thanks. Note: (1) I also had to restart the kernel, (2) remove the imports from openapi_schema_pydantic, and (3) include imports from openapi-pydantic only. More details here: https://community.deeplearning.ai/t/attributeerror-super-object-has-no-attribute-parse-obj/481376/11

bigrobinson commented 6 days ago

Thanks, @ChanduTadanki for reminding me to restart kernel. That was the last step to get things working for me.

langchain-ai / langchain