c9d5c0caed
- 6 route integration tests (mock sidecar via vitest vi.mock) - 12 task CRUD route tests (mock repository) - 29 Python tests: 10 extractor, 12 models, 7 app endpoints - Fix extractor.py: correct lx.extract() API (text_or_documents positional, prompt_description) - Mock fallback when no GEMINI_API_KEY or USE_MOCK_EXTRACTOR=true - 46 TS tests + 29 Python tests = 75 total
125 lines
2.9 KiB
Python
125 lines
2.9 KiB
Python
"""
|
|
Unit tests for Pydantic models.
|
|
"""
|
|
|
|
import pytest
|
|
from pydantic import ValidationError
|
|
|
|
from src.models import (
|
|
Extraction,
|
|
ExtractionExample,
|
|
ExtractRequest,
|
|
BatchExtractRequest,
|
|
ExtractMetadata,
|
|
ExtractResponse,
|
|
HealthResponse,
|
|
)
|
|
|
|
|
|
def test_extraction_basic():
|
|
e = Extraction(extraction_class="topic", extraction_text="meeting")
|
|
assert e.extraction_class == "topic"
|
|
assert e.attributes is None
|
|
|
|
|
|
def test_extraction_with_attributes():
|
|
e = Extraction(
|
|
extraction_class="emotion",
|
|
extraction_text="stressed",
|
|
attributes={"valence": "negative"},
|
|
)
|
|
assert e.attributes["valence"] == "negative"
|
|
|
|
|
|
def test_extraction_example():
|
|
ex = ExtractionExample(
|
|
text="sample text",
|
|
extractions=[
|
|
Extraction(extraction_class="topic", extraction_text="sample"),
|
|
],
|
|
)
|
|
assert len(ex.extractions) == 1
|
|
|
|
|
|
def test_extract_request_minimal():
|
|
req = ExtractRequest(text="Hello world")
|
|
assert req.text == "Hello world"
|
|
assert req.task_id is None
|
|
assert req.model_id is None
|
|
|
|
|
|
def test_extract_request_full():
|
|
req = ExtractRequest(
|
|
text="Test text",
|
|
task_id="triage",
|
|
task_prompt="Extract entities",
|
|
model_id="gemini-2.5-flash",
|
|
extraction_passes=2,
|
|
max_workers=5,
|
|
max_char_buffer=500,
|
|
)
|
|
assert req.extraction_passes == 2
|
|
assert req.max_workers == 5
|
|
|
|
|
|
def test_extract_request_rejects_empty_text():
|
|
with pytest.raises(ValidationError):
|
|
ExtractRequest(text="")
|
|
|
|
|
|
def test_extract_request_rejects_oversized_text():
|
|
with pytest.raises(ValidationError):
|
|
ExtractRequest(text="a" * 50_001)
|
|
|
|
|
|
def test_extract_request_rejects_invalid_passes():
|
|
with pytest.raises(ValidationError):
|
|
ExtractRequest(text="test", extraction_passes=10)
|
|
|
|
|
|
def test_batch_extract_request():
|
|
batch = BatchExtractRequest(
|
|
requests=[
|
|
ExtractRequest(text="doc 1"),
|
|
ExtractRequest(text="doc 2"),
|
|
]
|
|
)
|
|
assert len(batch.requests) == 2
|
|
|
|
|
|
def test_batch_extract_request_rejects_empty():
|
|
with pytest.raises(ValidationError):
|
|
BatchExtractRequest(requests=[])
|
|
|
|
|
|
def test_extract_metadata():
|
|
meta = ExtractMetadata(
|
|
model_id="gemini-2.5-flash",
|
|
duration_ms=150.5,
|
|
char_count=42,
|
|
)
|
|
assert meta.token_count is None
|
|
assert meta.duration_ms == 150.5
|
|
|
|
|
|
def test_extract_response():
|
|
resp = ExtractResponse(
|
|
extractions=[
|
|
Extraction(extraction_class="topic", extraction_text="AI"),
|
|
],
|
|
metadata=ExtractMetadata(
|
|
model_id="gemini-2.5-flash",
|
|
duration_ms=100,
|
|
char_count=20,
|
|
),
|
|
)
|
|
assert len(resp.extractions) == 1
|
|
assert resp.metadata.model_id == "gemini-2.5-flash"
|
|
|
|
|
|
def test_health_response_defaults():
|
|
h = HealthResponse()
|
|
assert h.status == "ok"
|
|
assert h.version == "0.1.0"
|
|
assert h.sidecar == "langextract"
|