Azure / Azure/azure-search-vector-samples
Issue while trying to create index with error The request is invalid. Details: An unexpected 'StartArray' node was found when reading from the JSON reader. A 'PrimitiveValue' node was expected.
- Dominant language
- Jupyter Notebook
- Stars
- 910
- Forks
- 377
- PR merge metrics
- No merged PRs in 30d
Description
index_client = SearchIndexClient(
endpoint=service_endpoint,
credential=cognitive_search_credential,
)
# Index name to use
index_name = f"vs_spn_chatrules"
search_client = SearchClient(
endpoint=service_endpoint,
index_name=index_name,
credential=cognitive_search_credential,
)
# Creating an Azure AI Search Vector Store
vector_store = AzureAISearchVectorStore(
search_or_index_client=index_client,
filterable_metadata_field_keys=metadata_fields,
index_name=index_name,
index_management=IndexManagement.CREATE_IF_NOT_EXISTS,
id_field_key="id",
chunk_field_key="chunk", #content
embedding_field_key="embedding", #content_vector
embedding_dimensionality=1536,
metadata_string_field_key="metadata",
doc_id_field_key="doc_id",
language_analyzer="en.lucene",
vector_algorithm_type="exhaustiveKnn", #HNSW focuses on approximate methods for efficiency, KNN ensures exactness through exhaustive searches.
)
Settings.llm = llm
Settings.embed_model = embed_model
storage_context = StorageContext.from_defaults(vector_store=vector_store)
# VectorStoreIndex.from_documents
index = VectorStoreIndex.from_documents(
all_docs, storage_context=storage_context
)
all_docs is a list of documents with document metadata and documents
while im trying to create index its gives error
HttpResponseError: () The request is invalid. Details: An unexpected 'StartArray' node was found when reading from the JSON reader. A 'PrimitiveValue' node was expected.
Code:
Message: The request is invalid. Details: An unexpected 'StartArray' node was found when reading from the JSON reader. A 'PrimitiveValue' node was expected.
any guidance on this
Contributor guide
No contributing guide indexed for this repository
Research direction
Start with the AzureAISearchVectorStore construction and the VectorStoreIndex.from_documents call, then compare the supplied metadata_fields and document metadata with the index schema. Reproduce the request using the shown SearchIndexClient setup and identify which field is serialized as an array; done means index creation succeeds without the StartArray/PrimitiveValue error.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- azure, python
- Domain
- backend, search
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100