nanobot/tests/providers/test_responses_circuit_breaker.py

"""Tests for Responses API circuit breaker in OpenAICompatProvider."""

import time

import pytest

from nanobot.providers.base import ProviderCallContext
from nanobot.providers.openai_compat_provider import (
    _RESPONSES_FAILURE_THRESHOLD,
    _RESPONSES_PROBE_INTERVAL_S,
    OpenAICompatProvider,
)
from nanobot.providers.registry import (
    ProviderSpec,
    ResponsesCapabilities,
    find_by_name,
)


@pytest.fixture()
def provider():
    """A direct-OpenAI provider with Responses API support."""
    p = OpenAICompatProvider.__new__(OpenAICompatProvider)
    p.default_model = "gpt-5"
    p._spec = find_by_name("openai")
    p._effective_base = "https://api.openai.com/v1"
    p._api_type = "auto"
    p._responses_failures = {}
    p._responses_tripped_at = {}
    return p


def test_responses_api_available_by_default(provider):
    assert provider._should_use_responses_api("gpt-5", None) is True


def test_deepseek_v4_flash_uses_responses_by_model(provider):
    provider._spec = find_by_name("deepseek")
    provider._effective_base = "https://api.deepseek.com"
    provider.default_model = "deepseek-v4-flash"

    assert provider._should_use_responses_api("deepseek-v4-flash", None) is True
    assert provider._should_use_responses_api("deepseek-v4-pro", None) is False


def test_deepseek_v4_flash_matches_provider_prefixed_model(provider):
    provider._spec = find_by_name("deepseek")
    provider._effective_base = "https://api.deepseek.com"

    assert provider._should_use_responses_api("deepseek/deepseek-v4-flash", None) is True


def test_responses_behavior_is_declared_by_capabilities(provider):
    provider._spec = ProviderSpec(
        name="example",
        keywords=("example",),
        env_key="EXAMPLE_API_KEY",
        responses=ResponsesCapabilities(
            models=("example-o3",),
            reasoning_replay="plaintext",
        ),
    )
    provider._effective_base = "https://example.test"

    assert provider._should_use_responses_api("example-o3", None) is True

    body = provider._build_responses_body(
        messages=[
            {"role": "user", "content": "question"},
            {
                "role": "assistant",
                "reasoning_content": "think first",
                "content": "answer",
            },
            {"role": "user", "content": "follow-up"},
        ],
        tools=None,
        model="example-o3",
        max_tokens=100,
        temperature=0.1,
        reasoning_effort="high",
        tool_choice=None,
    )

    assert {"type": "reasoning", "content": "think first"} in body["input"]
    assert "include" not in body


def test_direct_openai_enables_server_compaction(provider):
    provider._extra_body = {}

    body = provider._build_responses_body(
        messages=[{"role": "user", "content": "hello"}],
        tools=None,
        model="gpt-5.6",
        max_tokens=30_000,
        temperature=0.1,
        reasoning_effort="high",
        tool_choice=None,
        provider_context=ProviderCallContext(context_window_tokens=100_000),
    )

    assert body["context_management"] == [{
        "type": "compaction",
        "compact_threshold": 70_000,
    }]
    assert body["include"] == ["reasoning.encrypted_content"]


def test_api_type_chat_completions_disables_responses(provider):
    provider._api_type = "chat_completions"
    assert provider._should_use_responses_api("gpt-5", None) is False


def test_api_type_responses_forces_responses_for_openai(provider):
    provider.default_model = "gpt-4o"
    provider._api_type = "responses"
    assert provider._should_use_responses_api("gpt-4o", None) is True


def test_api_type_responses_ignores_circuit_breaker(provider):
    provider.default_model = "gpt-4o"
    provider._api_type = "responses"
    provider._responses_failures = {"gpt-4o|gpt-4o|": _RESPONSES_FAILURE_THRESHOLD}
    provider._responses_tripped_at = {"gpt-4o|gpt-4o|": 0.0}

    assert provider._should_use_responses_api("gpt-4o", None) is True


def test_api_type_responses_does_not_force_non_openai(provider):
    provider._spec = find_by_name("custom")
    provider._api_type = "responses"

    assert provider._should_use_responses_api("gpt-4o", None) is False


def test_circuit_opens_after_threshold(provider):
    for _ in range(_RESPONSES_FAILURE_THRESHOLD):
        provider._record_responses_failure("gpt-5", None)
    assert provider._should_use_responses_api("gpt-5", None) is False


def test_circuit_does_not_affect_other_models(provider):
    for _ in range(_RESPONSES_FAILURE_THRESHOLD):
        provider._record_responses_failure("gpt-5", None)
    assert provider._should_use_responses_api("o4-mini", None) is True


def test_success_resets_circuit(provider):
    for _ in range(_RESPONSES_FAILURE_THRESHOLD):
        provider._record_responses_failure("gpt-5", None)
    assert provider._should_use_responses_api("gpt-5", None) is False
    provider._record_responses_success("gpt-5", None)
    assert provider._should_use_responses_api("gpt-5", None) is True


def test_probe_after_interval(provider, monkeypatch):
    for _ in range(_RESPONSES_FAILURE_THRESHOLD):
        provider._record_responses_failure("gpt-5", None)
    assert provider._should_use_responses_api("gpt-5", None) is False

    # Fast-forward past the probe interval
    key = "gpt-5:"
    provider._responses_tripped_at[key] = time.monotonic() - _RESPONSES_PROBE_INTERVAL_S - 1
    assert provider._should_use_responses_api("gpt-5", None) is True


def test_below_threshold_still_allows(provider):
    provider._record_responses_failure("gpt-5", None)
    provider._record_responses_failure("gpt-5", None)
    assert provider._should_use_responses_api("gpt-5", None) is True


def test_reasoning_effort_keyed_separately(provider):
    for _ in range(_RESPONSES_FAILURE_THRESHOLD):
        provider._record_responses_failure("o3", "high")
    assert provider._should_use_responses_api("o3", "high") is False
    assert provider._should_use_responses_api("o3", "low") is True


def test_reasoning_effort_key_is_case_insensitive(provider):
    for _ in range(_RESPONSES_FAILURE_THRESHOLD):
        provider._record_responses_failure("o3", "High")
    assert provider._should_use_responses_api("o3", "high") is False