:hedgehog: PostHog is the leading platform for building self-driving products. Our developer tools – AI observability, analytics, session replay, flags, experiments, error tracking, logs, and more – capture all the context agents need to diagnose problems, uncover opportunities, and ship fixes. Steer it all from Slack, web, desktop, or the MCP.
Python ab-testingai-analyticsanalyticscdpdata-warehouseexperimentsfeature-flagsjavascriptproduct-analyticspythonreactsession-replaysurveystypescriptweb-analytics
from __future__ import annotations
import json
from dataclasses import dataclass
from pathlib import Path
from typing import Any
import pytest
import yaml
from owners_yaml.resolver import OwnersResolver, Purpose
CONFORMANCE_DIR = Path(__file__).parent.parent / "conformance"
CASE_FILES = sorted((CONFORMANCE_DIR / "cases").glob("*.yaml"))
@dataclass(frozen=True)
class ConformanceCase:
name: str
files: dict[str, str]
purpose: Purpose
producer: str | None
expect: dict[str, dict[str, Any]]
def load_cases() -> list[pytest.ParameterSet]:
params = []
for case_file in CASE_FILES:
for raw in yaml.safe_load(case_file.read_text())["cases"]:
case = ConformanceCase(
name=raw["name"],
files=raw["files"],
purpose=raw.get("purpose", "slack"),
producer=raw.get("producer"),
expect=raw["expect"],
)
params.append(pytest.param(case, id=f"{case_file.name}::{case.name}"))
return params
@pytest.mark.parametrize("case", load_cases())
def test_conformance_case(tmp_path: Path, case: ConformanceCase) -> None:
for rel, text in case.files.items():
target = tmp_path / rel
target.parent.mkdir(parents=True, exist_ok=True)
target.write_text(text)
resolver = OwnersResolver(repo_root=tmp_path, purpose=case.purpose, producer=case.producer)
actual = {}
for path in case.expect:
resolution = resolver.resolve(path)
actual[path] = {
"owners": resolution.owners or [],
"unowned_by_design": resolution.unowned_by_design,
"status": resolution.status,
"source": resolution.source,
"slack": resolution.slack,
"additions": resolution.additions,
"sensitive": resolution.sensitive,
}
# A case that leaves `additions` or `sensitive` out expects the default, so the cases written
# before a field existed also check that nothing leaks into it.
expected = {path: {"additions": [], "sensitive": False, **resolution} for path, resolution in case.expect.items()}
assert actual == expected
@pytest.mark.parametrize("case_file", CASE_FILES, ids=[f.name for f in CASE_FILES])
def test_case_file_matches_schema(case_file: Path) -> None:
jsonschema = pytest.importorskip("jsonschema")
schema = json.loads((CONFORMANCE_DIR / "case.schema.json").read_text())
document = yaml.safe_load(case_file.read_text())
jsonschema.validate(document, schema, cls=jsonschema.Draft202012Validator)
names = [case["name"] for case in document["cases"]]
assert len(names) == len(set(names)), "case names must be unique within a file"