test_global_cache_async() — langchain Function Reference

Architecture documentation for the test_global_cache_async() function in test_cache.py from the langchain codebase.

Function python CoreAbstractions MessageSchema

Entity Profile

CoreAbstractions→ MessageSchema→ test_global_cache_async() — langchain Function Reference

Dependency Diagram

graph TD
  ef578796_c25a_2172_ece2_fc1f1b6709cd["test_global_cache_async()"]
  51f634bf_713d_3f19_d694_5c6ef3e59c57["test_cache.py"]
  ef578796_c25a_2172_ece2_fc1f1b6709cd -->|defined in| 51f634bf_713d_3f19_d694_5c6ef3e59c57
  style ef578796_c25a_2172_ece2_fc1f1b6709cd fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

libs/core/tests/unit_tests/language_models/chat_models/test_cache.py lines 130–153

async def test_global_cache_async() -> None:
    """Test that the global cache gets populated when cache = True."""
    global_cache = InMemoryCache()
    try:
        set_llm_cache(global_cache)
        chat_model = FakeListChatModel(
            cache=True, responses=["hello", "goodbye", "meow", "woof"]
        )
        assert (await chat_model.ainvoke("How are you?")).content == "hello"
        # If the cache works we should get the same response since
        # the prompt is the same
        assert (await chat_model.ainvoke("How are you?")).content == "hello"
        # The global cache should be populated
        assert len(global_cache._cache) == 1
        llm_result = list(global_cache._cache.values())
        chat_generation = llm_result[0][0]
        assert isinstance(chat_generation, ChatGeneration)
        assert chat_generation.message.content == "hello"
        # Verify that another prompt will trigger the call to the model
        assert chat_model.invoke("nice").content == "goodbye"
        # The local cache should be populated
        assert len(global_cache._cache) == 2
    finally:
        set_llm_cache(None)