feat: Add query factory functions (text_query, vector_query)

[jhamon]

Summary

Adds text_query() and vector_query() factory functions that provide a simpler, more ergonomic API for constructing query objects to use with search_documents().

Problem

Users need to import and instantiate TextQuery and VectorQuery classes directly, which is verbose for simple use cases.

Solution

Added two factory functions that wrap the dataclass constructors:

• text_query(field, query, boost=None, slop=None) - Creates a TextQuery for full-text search
• vector_query(field, values=None, sparse_values=None) - Creates a VectorQuery for similarity search

Both functions are exported from the pinecone package for convenient imports.

Usage Examples

from pinecone import text_query, vector_query, SparseValues

# Simple text search
results = index.search_documents(
    namespace="movies",
    score_by=text_query("title", "pink panther"),
    top_k=10,
)

# Phrase match with quotes
results = index.search_documents(
    namespace="movies",
    score_by=text_query("title", '"pink panther"'),
    top_k=10,
)

# With boost and slop
results = index.search_documents(
    namespace="movies",
    score_by=text_query("title", '"pink panther"', boost=3.0, slop=2),
    top_k=10,
)

# Dense vector search
results = index.search_documents(
    namespace="movies",
    score_by=vector_query("embedding", values=[0.1, 0.2, 0.3]),
    top_k=10,
)

# Sparse vector search
results = index.search_documents(
    namespace="movies",
    score_by=vector_query(
        "sparse_embedding",
        sparse_values=SparseValues(indices=[1, 5, 10], values=[0.5, 0.3, 0.2]),
    ),
    top_k=10,
)

Related Issues

• Linear: SDK-109

---

Note

Low Risk
Low risk: adds thin wrapper factory functions and exposes them via top-level lazy imports, with unit coverage; main impact is expanded public API surface and import/export behavior.

Overview
Adds new factory helpers text_query() and vector_query() (in pinecone/db_data/query_helpers.py) to simplify constructing TextQuery/VectorQuery objects for search_documents().

Exports these helpers from the top-level pinecone package via the existing lazy-import mechanism, and adds unit tests validating parameter handling and .as_dict() serialization plus top-level importability.

^{Written by Cursor Bugbot for commit 0dd0261. This will update automatically on new commits. Configure here.}