microsoft/graphrag · error · ValueError

DRIFT Search query cannot be empty.

Error message

DRIFT Search query cannot be empty.

What it means

DriftSearch.search validates up front that the query string is non-empty (query == "") and raises this ValueError otherwise. It is a simple input guard so the search pipeline never runs on an empty prompt.

Source

Thrown at packages/graphrag/graphrag/query/structured_search/drift_search/search.py:213

        """
        Perform an asynchronous DRIFT search.

        Args:
            query (str): The query to search for.
            conversation_history (Any, optional): The conversation history, if any.
            reduce (bool, optional): Whether to reduce the response to a single comprehensive response.

        Returns
        -------
        SearchResult: The search result containing the response and context data.

        Raises
        ------
        ValueError: If the query is empty.
        """
        if query == "":
            error_msg = "DRIFT Search query cannot be empty."
            raise ValueError(error_msg)

        llm_calls, prompt_tokens, output_tokens = {}, {}, {}

        start_time = time.perf_counter()

        # Check if query state is empty
        if not self.query_state.graph:
            # Prime the search with the primer
            primer_context, token_ct = await self.context_builder.build_context(query)
            llm_calls["build_context"] = token_ct["llm_calls"]
            prompt_tokens["build_context"] = token_ct["prompt_tokens"]
            output_tokens["build_context"] = token_ct["output_tokens"]

            primer_response = await self.primer.search(
                query=query, top_k_reports=primer_context
            )
            llm_calls["primer"] = primer_response.llm_calls
            prompt_tokens["primer"] = primer_response.prompt_tokens

View on GitHub (pinned to f40e9a26ce)

Solutions

1. Validate the query is a non-empty, non-whitespace string before calling search
2. Default to a meaningful placeholder or skip empty inputs in batch processing
3. Strip and check the string: if not query.strip(): raise/skip in your own code first

Example fix

# before
result = await drift_search.aresolve(user_query)  # user_query may be ''

# after
if not user_query or not user_query.strip():
    raise ValueError('Query is required')
result = await drift_search.aresolve(user_query)

if not query or not query.strip():
    raise ValueError("Query must be a non-empty string")
result = await drift_search.aresolve(query)

def is_valid_query(q) -> bool:
    return isinstance(q, str) and len(q.strip()) > 0

When it happens

Trigger: Calling DriftSearch.search(''), DriftSearch.stream_search(''), or _search_step with an empty string, e.g. because a variable holding the user question was never set or was stripped to ''.

Common situations: Programmatic query loops where some rows of an input file have blank questions, frontends forwarding an empty input box, or whitespace-only strings that were stripped before the call.

Related errors

• No community reports available. Please provide a list of rep
• Query and document embeddings are not compatible. Please ens
• No intermediate answers found in primer response. Ensure tha
• No follow-up queries found in primer response. Ensure that t
• Response must be a list of dictionaries.

AI-assisted analysis of microsoft/graphrag@f40e9a26ce (2026-08-27). Data as JSON: /api/errors/3a1f75112a713180. Report an issue: GitHub.