AttributeError: 'WeaviateClient' object has no attribute 'schema'

[zohaibdr]

Checked other resources

• [X] I added a very descriptive title to this issue.
• [X] I searched the LangChain documentation with the integrated search.
• [X] I used the GitHub search to find a similar question and didn't find it.
• [X] I am sure that this is a bug in LangChain rather than my code.
• [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).

Example Code

from langchain_community.vectorstores import Weaviate
from langchain_openai import OpenAIEmbeddings
embeddings = OpenAIEmbeddings()
vectorstore = Weaviate.from_documents(
    text_chunks, 
    embeddings, 
    client=client, 
    by_text=False
)

document_content_description = "description"

### Error Message and Stack Trace (if applicable)

```python
AttributeError: 'WeaviateClient' object has no attribute 'schema'
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
File <command-3980657862137734>, line 6
      2 from langchain_openai import OpenAIEmbeddings
      4 embeddings = OpenAIEmbeddings()
----> 6 vectorstore = Weaviate.from_documents(
      7     text_chunks, 
      8     embeddings, 
      9     client=client, 
     10     by_text=False
     11 )
     13 document_content_description = "telecom documentation"

File /local_disk0/.ephemeral_nfs/envs/pythonEnv-bb42155d-01ee-4da3-9c8c-ce4489c59c93/lib/python3.10/site-packages/langchain_core/vectorstores.py:528, in VectorStore.from_documents(cls, documents, embedding, **kwargs)
    526 texts = [d.page_content for d in documents]
    527 metadatas = [d.metadata for d in documents]
--> 528 return cls.from_texts(texts, embedding, metadatas=metadatas, **kwargs)

File /local_disk0/.ephemeral_nfs/envs/pythonEnv-bb42155d-01ee-4da3-9c8c-ce4489c59c93/lib/python3.10/site-packages/langchain_community/vectorstores/weaviate.py:465, in Weaviate.from_texts(cls, texts, embedding, metadatas, client, weaviate_url, weaviate_api_key, batch_size, index_name, text_key, by_text, relevance_score_fn, **kwargs)
    463 schema = _default_schema(index_name, text_key)
    464 # check whether the index already exists
--> 465 if not client.schema.exists(index_name):
    466     client.schema.create_class(schema)
    468 embeddings = embedding.embed_documents(texts) if embedding else None

AttributeError: 'WeaviateClient' object has no attribute 'schema'

### Description

Trying to ingest langchain Document classes into Weaviate Cloud. 
Using the up-to-date version of all libs. 
Followed the tutorial here: https://python.langchain.com/docs/integrations/vectorstores/weaviate
### System Info

langchain==0.1.7
weaviate-client==4.5.1 
weaviate cloud ver. 1.24.1 and 1.23.10 (2 different clusters)

Running in Windows as well on databricks. 

[gingters]

Langchain currently still uses the Weaviate client version 3. You have installed the Weaviate client in version 4, which had many breaking changes in its API.

So when downgrading the client to the latest 3.x, that should work.

[xindexer]

This is true, but V4 is the future of python Weaviate and they have officially moved it out of Beta and are have deprecated V3.

This is the connection message using v3

DeprecationWarning: Dep016: You are using the Weaviate v3 client, which is deprecated. Consider upgrading to the new and improved v4 client instead! See here for usage: https://weaviate.io/developers/weaviate/client-libraries/python

[zohaibdr]

I ended up creating my own retriever wrapping weaviate methods inside langchain's base class.

[jansaidashaik1995]

@zohaibdr can you please share more details how exactly solved

[gingters]

The new weaviate client is on v4: https://python.langchain.com/docs/integrations/vectorstores/weaviate/