dosubot[bot]
commented
4 months ago Hey @ritzshah, I'm here to assist you with the issue you've encountered. I'm a bot designed to help with bug fixes, answer questions, and guide you through becoming a contributor. Let's tackle this together!
To resolve the error "'CacheMiss' object has no attribute 'get'" when trying to build a vertex in the Langflow application using PGVector with PostgreSQL, ensure that the inputs parameter is correctly passed and handled. The error suggests an issue with how the inputs parameter is being accessed or used.
Here is the relevant code snippet from the PGVectorComponent class:
def build(
self,
embedding: Embeddings,
pg_server_url: str,
collection_name: str,
inputs: Optional[Record] = None,
) -> Union[VectorStore, BaseRetriever]:
documents = []
for _input in inputs or []:
if isinstance(_input, Record):
documents.append(_input.to_lc_document())
else:
documents.append(_input)
try:
if documents is None:
vector_store = PGVector.from_existing_index(
embedding=embedding,
collection_name=collection_name,
connection_string=pg_server_url,
)
else:
vector_store = PGVector.from_documents(
embedding=embedding,
documents=documents, # type: ignore
collection_name=collection_name,
connection_string=pg_server_url,
)
except Exception as e:
raise RuntimeError(f"Failed to build PGVector: {e}")
return vector_storeEnsure that the inputs parameter is correctly passed as a list of Record objects. The inputs parameter should not be None or an incorrect type that does not have the required attributes.
If the inputs parameter is being fetched from a cache or another source, make sure that the fetched object is indeed a list of Record objects and not a CacheMiss object or any other type that does not have the required attributes.
Here is an example of how you might call the build method correctly:
embedding = ... # Your Embeddings object
pg_server_url = "your_postgresql_server_url"
collection_name = "your_collection_name"
inputs = [...] # A list of Record objects
pg_vector_component = PGVectorComponent()
vector_store = pg_vector_component.build(
embedding=embedding,
pg_server_url=pg_server_url,
collection_name=collection_name,
inputs=inputs
)Make sure that inputs is a list of Record objects and not None or any other incorrect type. Additionally, refer to the Langflow documentation for more details on the correct usage and configuration of the PGVectorComponent [1][2][3].
To continue talking to Dosu, mention @dosu.
ritzshah
anovazzi1
carlosrcoelho
Describe the bug A clear and concise description of what the bug is. While implementing a Vector Store using PostgreSQL (PGVector) connecting to localhost:5432, get error "'CacheMiss' object has no attribute 'get'"
Browser and Version
Browser [e.g. chrome, safari] - Google Chrome
Version [e.g. 22] - 124.0.6367.210
Langflow version - v1.0.0a43...
To Reproduce Steps to reproduce the behavior:
& this is the error message
Screenshots If applicable, add screenshots to help explain your problem.
Error message on langflow terminal venv/venv310/lib/python3.10/site-packages/langchain_community/vectorstores/pgvector.py:322: LangChainPendingDeprecationWarning: Please use JSONB instead of JSON for metadata. This change will allow for more efficient querying that involves filtering based on metadata.Please note that filtering operators have been changed when using JSOB metadata to be prefixed with a $ sign to avoid name collisions with columns. If you're using an existing database, you will need to create adb migration for your metadata column to be JSONB and update your queries to use the new operators. warn_deprecated( objc[34747]: +[NSCFConstantString initialize] may have been in progress in another thread when fork() was called. objc[34747]: +[NSCFConstantString initialize] may have been in progress in another thread when fork() was called. We cannot safely call it or ignore it in the fork() child process. Crashing instead. Set a breakpoint on objc_initializeAfterForkError to debug. [06/03/24 11:09:10] ERROR 2024-06-03 11:09:10 - ERROR - glogging - Worker (pid:34747) was sent SIGABRT! glogging.py:272 Starting Langflow v1.0.0a41... [06/03/24 11:09:14] ERROR 2024-06-03 11:09:14 - ERROR - chat - Error building vertex: 'CacheMiss' object has no attribute 'get' chat.py:232 ERROR 2024-06-03 11:09:14 - ERROR - chat - 'CacheMiss' object has no attribute 'get' chat.py:233 Traceback (most recent call last):
Additional context Add any other context about the problem here.