test_trim_messages_exact_token_boundary() — langchain Function Reference

Architecture documentation for the test_trim_messages_exact_token_boundary() function in test_utils.py from the langchain codebase.

Function python CoreAbstractions MessageSchema

Entity Profile

CoreAbstractions→ MessageSchema→ test_trim_messages_exact_token_boundary() — langchain Function Reference

Dependency Diagram

graph TD
  07f0697c_07e6_fea7_eb8d_f1d8244a70ad["test_trim_messages_exact_token_boundary()"]
  ea3f8d89_f44b_6738_3cb9_a740a73cfca4["test_utils.py"]
  07f0697c_07e6_fea7_eb8d_f1d8244a70ad -->|defined in| ea3f8d89_f44b_6738_3cb9_a740a73cfca4
  style 07f0697c_07e6_fea7_eb8d_f1d8244a70ad fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

libs/core/tests/unit_tests/messages/test_utils.py lines 632–656

def test_trim_messages_exact_token_boundary() -> None:
    messages = [
        SystemMessage(content="10 tokens exactly."),
        HumanMessage(content="Another 10 tokens."),
    ]

    # First message only
    result1 = trim_messages(
        messages,
        max_tokens=10,  # Exactly the size of first message
        token_counter=dummy_token_counter,
        strategy="first",
    )
    assert len(result1) == 1
    assert result1[0].content == "10 tokens exactly."

    # Both messages exactly fit
    result2 = trim_messages(
        messages,
        max_tokens=20,  # Exactly the size of both messages
        token_counter=dummy_token_counter,
        strategy="first",
    )
    assert len(result2) == 2
    assert result2 == messages