Merge main into timestamping_enricher

This commit is contained in:
Patrick Robertson
2025-03-24 15:09:29 +04:00
219 changed files with 11049 additions and 2933 deletions

View File

@@ -1,6 +1,7 @@
"""
pytest conftest file, for shared fixtures and configuration
"""
import os
import pickle
from datetime import datetime, timezone
@@ -16,32 +17,36 @@ from auto_archiver.core.module import ModuleFactory
# that you only want to run if everything else succeeds (e.g. API calls). The order here is important
# what comes first will be run first (at the end of all other tests not mentioned)
# format is the name of the module (python file) without the .py extension
TESTS_TO_RUN_LAST = ['test_twitter_api_archiver']
TESTS_TO_RUN_LAST = ["test_twitter_api_archiver"]
@pytest.fixture
def setup_module(request):
def _setup_module(module_name, config={}):
def _setup_module(module_name, config=None):
if config is None:
config = {}
module_factory = ModuleFactory()
if isinstance(module_name, type):
# get the module name:
# if the class does not have a .name, use the name of the parent folder
module_name = module_name.__module__.rsplit(".",2)[-2]
module_name = module_name.__module__.rsplit(".", 2)[-2]
m = module_factory.get_module(module_name, {module_name: config})
# add the tmp_dir to the module
tmp_dir = TemporaryDirectory()
m.tmp_dir = tmp_dir.name
def cleanup():
tmp_dir.cleanup()
request.addfinalizer(cleanup)
return m
return _setup_module
@pytest.fixture
def check_hash():
def _check_hash(filename: str, hash: str):
@@ -51,6 +56,7 @@ def check_hash():
return _check_hash
@pytest.fixture
def make_item():
def _make_item(url: str, **kwargs) -> Metadata:
@@ -62,7 +68,6 @@ def make_item():
return _make_item
def pytest_collection_modifyitems(items):
module_mapping = {item: item.module.__name__.split(".")[-1] for item in items}
@@ -78,13 +83,13 @@ def pytest_collection_modifyitems(items):
items[:] = sorted_items
# Incremental testing - fail tests in a class if any previous test fails
# taken from https://docs.pytest.org/en/latest/example/simple.html#incremental-testing-test-steps
# store history of failures per test class name and per index in parametrize (if parametrize used)
_test_failed_incremental: Dict[str, Dict[Tuple[int, ...], str]] = {}
def pytest_runtest_makereport(item, call):
if "incremental" in item.keywords:
# incremental marker is used
@@ -93,17 +98,11 @@ def pytest_runtest_makereport(item, call):
# retrieve the class name of the test
cls_name = str(item.cls)
# retrieve the index of the test (if parametrize is used in combination with incremental)
parametrize_index = (
tuple(item.callspec.indices.values())
if hasattr(item, "callspec")
else ()
)
parametrize_index = tuple(item.callspec.indices.values()) if hasattr(item, "callspec") else ()
# retrieve the name of the test function
test_name = item.originalname or item.name
# store in _test_failed_incremental the original name of the failed test
_test_failed_incremental.setdefault(cls_name, {}).setdefault(
parametrize_index, test_name
)
_test_failed_incremental.setdefault(cls_name, {}).setdefault(parametrize_index, test_name)
def pytest_runtest_setup(item):
@@ -119,16 +118,17 @@ def pytest_runtest_setup(item):
pytest.xfail(f"previous test failed ({test_name})")
@pytest.fixture()
@pytest.fixture
def unpickle():
"""
Returns a helper function that unpickles a file
** gets the file from the test_files directory: tests/data/ **
"""
def _unpickle(path):
with open(os.path.join("tests/data", path), "rb") as f:
return pickle.load(f)
return _unpickle
@@ -151,9 +151,9 @@ def sample_datetime():
return datetime(2023, 1, 1, 12, 0, tzinfo=timezone.utc)
@pytest.fixture(autouse=True)
@pytest.fixture
def mock_sleep(mocker):
"""Globally mock time.sleep to avoid delays."""
"""Mock time.sleep to avoid delays."""
return mocker.patch("time.sleep")
@@ -162,4 +162,4 @@ def metadata():
metadata = Metadata()
metadata.set("_processed_at", "2021-01-01T00:00:00")
metadata.set_url("https://example.com")
return metadata
return metadata