scholarr/tests/unit/test_scholar_search_safety.py
Justin Visser 87aa89b58c fix: resolve mypy errors, fix arxiv test session isolation, and add premerge script
- Remove unused type:ignore comment in scholars.py
- Fix 51 mypy errors across test files (Any return types, cast(), protocol stubs)
- Fix arxiv unit test failures caused by session isolation (patch_session_factory fixture)
- Fix integration test rate-limit bleed between search and create
- Add scripts/premerge.sh for local CI verification
- Add docs/website/.docusaurus/ to .gitignore

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-02 12:46:10 +01:00

278 lines
8.8 KiB
Python

from __future__ import annotations
import pytest
from sqlalchemy.ext.asyncio import AsyncSession
from app.services.scholar.parser import ParseState
from app.services.scholar.source import FetchResult
from app.services.scholars import application as scholar_service
pytestmark = [pytest.mark.integration, pytest.mark.db]
class StubScholarSource:
def __init__(self, fetch_results: list[FetchResult]) -> None:
self._fetch_results = list(fetch_results)
self.calls = 0
async def fetch_author_search_html(self, query: str, *, start: int = 0) -> FetchResult:
assert start == 0
self.calls += 1
if not self._fetch_results:
raise RuntimeError("No stub fetch results configured.")
index = min(self.calls - 1, len(self._fetch_results) - 1)
return self._fetch_results[index]
async def fetch_profile_html(self, scholar_id: str) -> FetchResult:
raise NotImplementedError
async def fetch_profile_page_html(self, scholar_id: str, *, cstart: int, pagesize: int) -> FetchResult:
raise NotImplementedError
def _ok_author_search_fetch() -> FetchResult:
body = (
"<html><body>"
'<div class="gsc_1usr">'
'<img src="/citations/images/avatar_scholar_256.png" />'
'<a class="gs_ai_name" href="/citations?hl=en&user=abcDEF123456">Ada Lovelace</a>'
'<div class="gs_ai_aff">Analytical Engine</div>'
'<div class="gs_ai_eml">Verified email at computing.example</div>'
'<div class="gs_ai_cby">Cited by 42</div>'
'<a class="gs_ai_one_int">Mathematics</a>'
"</div>"
"</body></html>"
)
return FetchResult(
requested_url="https://scholar.google.com/citations?hl=en&view_op=search_authors&mauthors=ada",
status_code=200,
final_url="https://scholar.google.com/citations?hl=en&view_op=search_authors&mauthors=ada",
body=body,
error=None,
)
def _blocked_author_search_fetch() -> FetchResult:
return FetchResult(
requested_url="https://scholar.google.com/citations?hl=en&view_op=search_authors&mauthors=ada",
status_code=200,
final_url=(
"https://accounts.google.com/v3/signin/identifier?continue=https%3A%2F%2Fscholar.google.com%2Fcitations"
),
body="<html><body>Sign in</body></html>",
error=None,
)
def _network_timeout_fetch() -> FetchResult:
return FetchResult(
requested_url="https://scholar.google.com/citations?hl=en&view_op=search_authors&mauthors=ada",
status_code=None,
final_url=None,
body="",
error="timed out",
)
@pytest.mark.asyncio
async def test_search_author_candidates_serves_cached_response_for_same_query(
db_session: AsyncSession,
) -> None:
source = StubScholarSource([_ok_author_search_fetch()])
first = await scholar_service.search_author_candidates(
source=source,
db_session=db_session,
query="Ada Lovelace",
limit=10,
network_error_retries=0,
retry_backoff_seconds=0.0,
search_enabled=True,
cache_ttl_seconds=600,
blocked_cache_ttl_seconds=60,
cache_max_entries=64,
min_interval_seconds=0.0,
interval_jitter_seconds=0.0,
cooldown_block_threshold=3,
cooldown_seconds=300,
)
second = await scholar_service.search_author_candidates(
source=source,
db_session=db_session,
query="Ada Lovelace",
limit=10,
network_error_retries=0,
retry_backoff_seconds=0.0,
search_enabled=True,
cache_ttl_seconds=600,
blocked_cache_ttl_seconds=60,
cache_max_entries=64,
min_interval_seconds=0.0,
interval_jitter_seconds=0.0,
cooldown_block_threshold=3,
cooldown_seconds=300,
)
assert first.state == ParseState.OK
assert second.state == ParseState.OK
assert source.calls == 1
assert len(second.candidates) == 1
assert second.candidates[0].scholar_id == "abcDEF123456"
assert "author_search_served_from_cache" in second.warnings
@pytest.mark.asyncio
async def test_search_author_candidates_trips_cooldown_after_blocked_responses(
db_session: AsyncSession,
) -> None:
source = StubScholarSource([_blocked_author_search_fetch()])
first = await scholar_service.search_author_candidates(
source=source,
db_session=db_session,
query="Blocked Query",
limit=10,
network_error_retries=0,
retry_backoff_seconds=0.0,
search_enabled=True,
cache_ttl_seconds=0,
blocked_cache_ttl_seconds=0,
cache_max_entries=64,
min_interval_seconds=0.0,
interval_jitter_seconds=0.0,
cooldown_block_threshold=1,
cooldown_seconds=300,
)
second = await scholar_service.search_author_candidates(
source=source,
db_session=db_session,
query="Blocked Query",
limit=10,
network_error_retries=0,
retry_backoff_seconds=0.0,
search_enabled=True,
cache_ttl_seconds=0,
blocked_cache_ttl_seconds=0,
cache_max_entries=64,
min_interval_seconds=0.0,
interval_jitter_seconds=0.0,
cooldown_block_threshold=1,
cooldown_seconds=300,
)
assert first.state == ParseState.BLOCKED_OR_CAPTCHA
assert "author_search_circuit_breaker_armed" in first.warnings
assert source.calls == 1
assert second.state == ParseState.BLOCKED_OR_CAPTCHA
assert second.state_reason == scholar_service.SEARCH_COOLDOWN_REASON
assert "author_search_cooldown_active" in second.warnings
@pytest.mark.asyncio
async def test_search_author_candidates_emits_cooldown_alert_warning_after_threshold(
db_session: AsyncSession,
) -> None:
source = StubScholarSource([_blocked_author_search_fetch()])
_ = await scholar_service.search_author_candidates(
source=source,
db_session=db_session,
query="Blocked Query",
limit=10,
network_error_retries=0,
retry_backoff_seconds=0.0,
search_enabled=True,
cache_ttl_seconds=0,
blocked_cache_ttl_seconds=0,
cache_max_entries=64,
min_interval_seconds=0.0,
interval_jitter_seconds=0.0,
cooldown_block_threshold=1,
cooldown_seconds=300,
cooldown_rejection_alert_threshold=1,
)
second = await scholar_service.search_author_candidates(
source=source,
db_session=db_session,
query="Blocked Query",
limit=10,
network_error_retries=0,
retry_backoff_seconds=0.0,
search_enabled=True,
cache_ttl_seconds=0,
blocked_cache_ttl_seconds=0,
cache_max_entries=64,
min_interval_seconds=0.0,
interval_jitter_seconds=0.0,
cooldown_block_threshold=1,
cooldown_seconds=300,
cooldown_rejection_alert_threshold=1,
)
assert second.state == ParseState.BLOCKED_OR_CAPTCHA
assert "author_search_cooldown_alert_threshold_exceeded" in second.warnings
@pytest.mark.asyncio
async def test_search_author_candidates_adds_retry_threshold_warning(
db_session: AsyncSession,
) -> None:
source = StubScholarSource(
[
_network_timeout_fetch(),
_network_timeout_fetch(),
_ok_author_search_fetch(),
]
)
result = await scholar_service.search_author_candidates(
source=source,
db_session=db_session,
query="Ada Lovelace",
limit=10,
network_error_retries=2,
retry_backoff_seconds=0.0,
search_enabled=True,
cache_ttl_seconds=0,
blocked_cache_ttl_seconds=0,
cache_max_entries=64,
min_interval_seconds=0.0,
interval_jitter_seconds=0.0,
cooldown_block_threshold=3,
cooldown_seconds=300,
retry_alert_threshold=2,
)
assert result.state == ParseState.OK
assert "author_search_retry_threshold_exceeded_2" in result.warnings
assert source.calls == 3
@pytest.mark.asyncio
async def test_search_author_candidates_can_be_disabled_by_configuration(
db_session: AsyncSession,
) -> None:
source = StubScholarSource([_ok_author_search_fetch()])
parsed = await scholar_service.search_author_candidates(
source=source,
db_session=db_session,
query="Ada Lovelace",
limit=5,
network_error_retries=0,
retry_backoff_seconds=0.0,
search_enabled=False,
cache_ttl_seconds=600,
blocked_cache_ttl_seconds=60,
cache_max_entries=64,
min_interval_seconds=0.0,
interval_jitter_seconds=0.0,
cooldown_block_threshold=1,
cooldown_seconds=300,
)
assert parsed.state == ParseState.BLOCKED_OR_CAPTCHA
assert parsed.state_reason == scholar_service.SEARCH_DISABLED_REASON
assert source.calls == 0
assert "author_search_disabled_by_configuration" in parsed.warnings