hermes-agent/tests/agent/test_gemini_free_tier_gate.py
Teknium 6b81590c55
test: prune low-value tests suite-wide (wave 1) — 46,820 → 28,106 test functions
Systematic prune per AGENTS.md test policy, one pass over every major
test tree (gateway, hermes_cli, tools, agent, run_agent, plugins, cli,
cron, tui_gateway, honcho/openviking, root-level):

- DELETE: source-reading tests (read_text/getsource on prod files),
  change-detector tests (exact catalog counts, model-name snapshots,
  config version literals), mock-echo tests (assert a mock returns what
  it was told), assertion-free/trivial tests, near-duplicate
  parametrizations (boundaries + one representative kept), async/sync
  twin duplicates, cosmetic within-file variations.
- KEEP (mandatory): security/redaction/approval guards, message-role
  alternation invariants, prompt-caching/deterministic-call-id
  invariants, issue-number regression tests (deduped), E2E tests.
- 6 test files deleted outright (script-style/no-assert or fully
  redundant); conftest.py, fakes/, fixtures/ untouched.
- tests/acp/conftest.py added: autouse fixture stubs the live
  models.dev/GitHub/Copilot/Anthropic inventory fetches that ACP server
  tests performed on every session create — test_server.py 147s → 3.4s,
  and the tests are now genuinely hermetic.
- Sleep-based slowness shrunk where safe (codex_ttfb_watchdog,
  compression_concurrent_fork, etc.); no wall-clock assertion tightened.

Verification: full hermetic suite via scripts/run_tests.sh —
2439 files, 31,130 tests passed, 0 failed, 0 flaky retries, 315s wall
(baseline: 583s wall, 13,564s subprocess CPU).
2026-07-29 13:10:23 -07:00

98 lines
2.7 KiB
Python

"""Tests for Gemini free-tier detection and blocking."""
from __future__ import annotations
from unittest.mock import MagicMock, patch
from agent.gemini_native_adapter import (
gemini_http_error,
is_free_tier_quota_error,
probe_gemini_tier,
)
def _mock_response(status: int, headers: dict | None = None, text: str = "") -> MagicMock:
resp = MagicMock()
resp.status_code = status
resp.headers = headers or {}
resp.text = text
return resp
def _run_probe(resp: MagicMock) -> str:
with patch("agent.gemini_native_adapter.httpx.Client") as MC:
inst = MagicMock()
inst.post.return_value = resp
MC.return_value.__enter__.return_value = inst
return probe_gemini_tier("fake-key")
class TestProbeGeminiTier:
"""Verify the tier probe classifies keys correctly."""
def test_free_tier_via_429_body(self):
body = (
'{"error":{"code":429,"message":"Quota exceeded for metric: '
'generativelanguage.googleapis.com/generate_content_free_tier_requests, '
'limit: 20"}}'
)
resp = _mock_response(429, {}, body)
assert _run_probe(resp) == "free"
def test_successful_200_without_rpd_header_is_paid(self):
resp = _mock_response(200, {}, '{"candidates":[]}')
assert _run_probe(resp) == "paid"
class TestIsFreeTierQuotaError:
def test_detects_free_tier_marker(self):
assert is_free_tier_quota_error(
"Quota exceeded for metric: generate_content_free_tier_requests"
)
def test_no_free_tier_marker(self):
assert not is_free_tier_quota_error("rate limited")
def test_none(self):
assert not is_free_tier_quota_error(None) # type: ignore[arg-type]
class TestGeminiHttpErrorFreeTierGuidance:
"""gemini_http_error should append free-tier guidance for free-tier 429s."""
class _FakeResp:
def __init__(self, status: int, text: str):
self.status_code = status
self.headers: dict = {}
self.text = text
def test_free_tier_429_appends_guidance(self):
body = (
'{"error":{"code":429,"message":"Quota exceeded for metric: '
"generativelanguage.googleapis.com/generate_content_free_tier_requests, "
'limit: 20","status":"RESOURCE_EXHAUSTED"}}'
)
err = gemini_http_error(self._FakeResp(429, body))
msg = str(err)
assert "free tier" in msg.lower()
assert "aistudio.google.com/apikey" in msg
def test_paid_429_has_no_billing_url(self):
body = '{"error":{"code":429,"message":"Rate limited","status":"RESOURCE_EXHAUSTED"}}'
err = gemini_http_error(self._FakeResp(429, body))
assert "aistudio.google.com/apikey" not in str(err)