langchain Architecture Docs
Home / File/ test_prompt_cache_key.py — langchain Source File

test_prompt_cache_key.py — langchain Source File

Architecture documentation for test_prompt_cache_key.py, a python file in the langchain codebase. 2 imports, 0 dependents.

File python CoreAbstractions MessageSchema 2 imports 5 functions

Entity Profile

Dependency Diagram

graph LR
  88675377_c5a7_deb4_6393_67dec22d158b["test_prompt_cache_key.py"]
  d758344f_537f_649e_f467_b9d7442e86df["langchain_core.messages"]
  88675377_c5a7_deb4_6393_67dec22d158b --> d758344f_537f_649e_f467_b9d7442e86df
  0b28cff6_d823_1571_d2bb_ec61508cc89c["langchain_openai"]
  88675377_c5a7_deb4_6393_67dec22d158b --> 0b28cff6_d823_1571_d2bb_ec61508cc89c
  style 88675377_c5a7_deb4_6393_67dec22d158b fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

"""Unit tests for prompt_cache_key parameter."""

from langchain_core.messages import HumanMessage

from langchain_openai import ChatOpenAI


def test_prompt_cache_key_parameter_inclusion() -> None:
    """Test that prompt_cache_key parameter is properly included in request payload."""
    chat = ChatOpenAI(model="gpt-4o-mini", max_completion_tokens=10)
    messages = [HumanMessage("Hello")]

    payload = chat._get_request_payload(messages, prompt_cache_key="test-cache-key")
    assert "prompt_cache_key" in payload
    assert payload["prompt_cache_key"] == "test-cache-key"


def test_prompt_cache_key_parameter_exclusion() -> None:
    """Test that prompt_cache_key parameter behavior matches OpenAI API."""
    chat = ChatOpenAI(model="gpt-4o-mini", max_completion_tokens=10)
    messages = [HumanMessage("Hello")]

    # Test with explicit None (OpenAI should accept None values (marked Optional))
    payload = chat._get_request_payload(messages, prompt_cache_key=None)
    assert "prompt_cache_key" in payload
    assert payload["prompt_cache_key"] is None


def test_prompt_cache_key_per_call() -> None:
    """Test that prompt_cache_key can be passed per-call with different values."""
    chat = ChatOpenAI(model="gpt-4o-mini", max_completion_tokens=10)
    messages = [HumanMessage("Hello")]

    # Test different cache keys per call
    payload1 = chat._get_request_payload(messages, prompt_cache_key="cache-v1")
    payload2 = chat._get_request_payload(messages, prompt_cache_key="cache-v2")

    assert payload1["prompt_cache_key"] == "cache-v1"
    assert payload2["prompt_cache_key"] == "cache-v2"

    # Test dynamic cache key assignment
    cache_keys = ["customer-v1", "support-v1", "feedback-v1"]

    for cache_key in cache_keys:
        payload = chat._get_request_payload(messages, prompt_cache_key=cache_key)
        assert "prompt_cache_key" in payload
        assert payload["prompt_cache_key"] == cache_key


def test_prompt_cache_key_model_kwargs() -> None:
    """Test prompt_cache_key via model_kwargs and method precedence."""
    messages = [HumanMessage("Hello world")]

    # Test model-level via model_kwargs
    chat = ChatOpenAI(
        model="gpt-4o-mini",
        max_completion_tokens=10,
        model_kwargs={"prompt_cache_key": "model-level-cache"},
    )
    payload = chat._get_request_payload(messages)
    assert "prompt_cache_key" in payload
    assert payload["prompt_cache_key"] == "model-level-cache"

    # Test that per-call cache key overrides model-level
    payload_override = chat._get_request_payload(
        messages, prompt_cache_key="per-call-cache"
    )
    assert payload_override["prompt_cache_key"] == "per-call-cache"


def test_prompt_cache_key_responses_api() -> None:
    """Test that prompt_cache_key works with Responses API."""
    chat = ChatOpenAI(
        model="gpt-4o-mini",
        use_responses_api=True,
        output_version="responses/v1",
        max_completion_tokens=10,
    )

    messages = [HumanMessage("Hello")]
    payload = chat._get_request_payload(
        messages, prompt_cache_key="responses-api-cache-v1"
    )

    # prompt_cache_key should be present regardless of API type
    assert "prompt_cache_key" in payload
    assert payload["prompt_cache_key"] == "responses-api-cache-v1"

Domain

Subdomains

Functions

Dependencies

  • langchain_core.messages
  • langchain_openai

Source

Frequently Asked Questions

What does test_prompt_cache_key.py do?
test_prompt_cache_key.py is a source file in the langchain codebase, written in python. It belongs to the CoreAbstractions domain, MessageSchema subdomain.
What functions are defined in test_prompt_cache_key.py?
test_prompt_cache_key.py defines 5 function(s): test_prompt_cache_key_model_kwargs, test_prompt_cache_key_parameter_exclusion, test_prompt_cache_key_parameter_inclusion, test_prompt_cache_key_per_call, test_prompt_cache_key_responses_api.
What does test_prompt_cache_key.py depend on?
test_prompt_cache_key.py imports 2 module(s): langchain_core.messages, langchain_openai.
Where is test_prompt_cache_key.py in the architecture?
test_prompt_cache_key.py is located at libs/partners/openai/tests/unit_tests/chat_models/test_prompt_cache_key.py (domain: CoreAbstractions, subdomain: MessageSchema, directory: libs/partners/openai/tests/unit_tests/chat_models).

Analyze Your Own Codebase

Get architecture documentation, dependency graphs, and domain analysis for your codebase in minutes.

Try Supermodel Free