def test_pypi_visit_1_release_with_2_artifacts(swh_storage, requests_mock_datadir):
"""With no prior visit, load a pypi project ends up with 1 snapshot"""
url = "https://pypi.org/project/nexter"
loader = PyPILoader(swh_storage, url)
actual_load_status = loader.load()
expected_snapshot_id = hash_to_bytes("1394b2e59351a944cc763bd9d26d90ce8e8121a8")
assert actual_load_status == {
"status": "eventful",
"snapshot_id": expected_snapshot_id.hex(),
}
assert_last_visit_matches(
swh_storage, url, status="full", type="pypi", snapshot=expected_snapshot_id
)
expected_snapshot = Snapshot(
id=expected_snapshot_id,
branches={
b"releases/1.1.0/nexter-1.1.0.zip": SnapshotBranch(
target=hash_to_bytes("f7d43faeb65b64d3faa67e4f46559db57d26b9a4"),
target_type=TargetType.RELEASE,
),
b"releases/1.1.0/nexter-1.1.0.tar.gz": SnapshotBranch(
target=hash_to_bytes("732bb9dc087e6015884daaebb8b82559be729b5a"),
target_type=TargetType.RELEASE,
),
},
)
check_snapshot(expected_snapshot, swh_storage)
def test_pypi_release_with_traceback(swh_storage, requests_mock_datadir):
url = "https://pypi.org/project/0805nexter"
with patch(
"swh.loader.package.pypi.loader.PyPILoader.last_snapshot",
side_effect=ValueError("Fake problem to fail the visit"),
):
loader = PyPILoader(swh_storage, url)
actual_load_status = loader.load()
assert actual_load_status == {"status": "failed"}
assert_last_visit_matches(swh_storage, url, status="failed", type="pypi")
stats = get_stats(swh_storage)
assert {
"content": 0,
"directory": 0,
"origin": 1,
"origin_visit": 1,
"release": 0,
"revision": 0,
"skipped_content": 0,
"snapshot": 0,
} == stats
def test_pypi_no_release_artifact(swh_storage, requests_mock_datadir_missing_all):
"""Load a pypi project with all artifacts missing ends up with no snapshot"""
url = "https://pypi.org/project/0805nexter"
loader = PyPILoader(swh_storage, url)
actual_load_status = loader.load()
assert actual_load_status["status"] == "uneventful"
assert actual_load_status["snapshot_id"] is not None
empty_snapshot = Snapshot(branches={})
assert_last_visit_matches(
swh_storage, url, status="partial", type="pypi", snapshot=empty_snapshot.id
)
stats = get_stats(swh_storage)
assert {
"content": 0,
"directory": 0,
"origin": 1,
"origin_visit": 1,
"release": 0,
"revision": 0,
"skipped_content": 0,
"snapshot": 1,
} == stats
def test_pypi_origin_not_found(swh_storage, requests_mock_datadir):
url = "https://pypi.org/project/unknown"
loader = PyPILoader(swh_storage, url)
assert loader.load() == {"status": "failed"}
assert_last_visit_matches(
swh_storage, url, status="not_found", type="pypi", snapshot=None
)
def test_pypi_visit_with_missing_artifact(
swh_storage, requests_mock_datadir_missing_one
):
"""Load a pypi project with some missing artifacts ends up with 1 snapshot"""
url = "https://pypi.org/project/0805nexter"
loader = PyPILoader(swh_storage, url)
actual_load_status = loader.load()
expected_snapshot_id = hash_to_bytes("00785a38479abe5fbfa402df96be26d2ddf89c97")
assert actual_load_status == {
"status": "eventful",
"snapshot_id": expected_snapshot_id.hex(),
}
assert_last_visit_matches(
swh_storage,
url,
status="partial",
type="pypi",
snapshot=expected_snapshot_id,
)
expected_snapshot = Snapshot(
id=hash_to_bytes(expected_snapshot_id),
branches={
b"releases/1.2.0": SnapshotBranch(
target=hash_to_bytes("fbbcb817f01111b06442cdcc93140ab3cc777d68"),
target_type=TargetType.RELEASE,
),
b"HEAD": SnapshotBranch(
target=b"releases/1.2.0",
target_type=TargetType.ALIAS,
),
},
)
check_snapshot(expected_snapshot, storage=swh_storage)
stats = get_stats(swh_storage)
assert {
"content": 3,
"directory": 2,
"origin": 1,
"origin_visit": 1,
"release": 1,
"revision": 0,
"skipped_content": 0,
"snapshot": 1,
} == stats
def test_pypi_visit_with_1_release_artifact(swh_storage, requests_mock_datadir):
"""With no prior visit, load a pypi project ends up with 1 snapshot"""
url = "https://pypi.org/project/0805nexter"
loader = PyPILoader(swh_storage, url)
actual_load_status = loader.load()
expected_snapshot_id = hash_to_bytes("3dd50c1a0e48a7625cf1427e3190a65b787c774e")
assert actual_load_status == {
"status": "eventful",
"snapshot_id": expected_snapshot_id.hex(),
}
assert_last_visit_matches(
swh_storage, url, status="full", type="pypi", snapshot=expected_snapshot_id
)
expected_snapshot = Snapshot(
id=expected_snapshot_id,
branches={
b"releases/1.1.0": SnapshotBranch(
target=hash_to_bytes("f8789ff3ed70a5f570c35d885c7bcfda7b23b091"),
target_type=TargetType.RELEASE,
),
b"releases/1.2.0": SnapshotBranch(
target=hash_to_bytes("fbbcb817f01111b06442cdcc93140ab3cc777d68"),
target_type=TargetType.RELEASE,
),
b"HEAD": SnapshotBranch(
target=b"releases/1.2.0",
target_type=TargetType.ALIAS,
),
},
)
check_snapshot(expected_snapshot, swh_storage)
stats = get_stats(swh_storage)
assert {
"content": 6,
"directory": 4,
"origin": 1,
"origin_visit": 1,
"release": 2,
"revision": 0,
"skipped_content": 0,
"snapshot": 1,
} == stats
def test_pypi_artifact_with_no_intrinsic_metadata(swh_storage, requests_mock_datadir):
"""Skip artifact with no intrinsic metadata during ingestion"""
url = "https://pypi.org/project/upymenu"
loader = PyPILoader(swh_storage, url)
actual_load_status = loader.load()
expected_snapshot_id = hash_to_bytes("1a8893e6a86f444e8be8e7bda6cb34fb1735a00e")
assert actual_load_status == {
"status": "eventful",
"snapshot_id": expected_snapshot_id.hex(),
}
# no branch as one artifact without any intrinsic metadata
expected_snapshot = Snapshot(id=expected_snapshot_id, branches={})
assert_last_visit_matches(
swh_storage, url, status="full", type="pypi", snapshot=expected_snapshot.id
)
check_snapshot(expected_snapshot, swh_storage)
def test_pypi_incremental_visit(swh_storage, requests_mock_datadir_visits):
"""With prior visit, 2nd load will result with a different snapshot"""
url = "https://pypi.org/project/0805nexter"
loader = PyPILoader(swh_storage, url)
visit1_actual_load_status = loader.load()
visit1_stats = get_stats(swh_storage)
expected_snapshot_id = hash_to_bytes("3dd50c1a0e48a7625cf1427e3190a65b787c774e")
assert visit1_actual_load_status == {
"status": "eventful",
"snapshot_id": expected_snapshot_id.hex(),
}
assert_last_visit_matches(
swh_storage, url, status="full", type="pypi", snapshot=expected_snapshot_id
)
assert {
"content": 6,
"directory": 4,
"origin": 1,
"origin_visit": 1,
"release": 2,
"revision": 0,
"skipped_content": 0,
"snapshot": 1,
} == visit1_stats
# Reset internal state
del loader._cached__raw_info
del loader._cached_info
visit2_actual_load_status = loader.load()
visit2_stats = get_stats(swh_storage)
assert visit2_actual_load_status["status"] == "eventful", visit2_actual_load_status
expected_snapshot_id2 = hash_to_bytes("77febe6ff0faf6cc00dd015a6c9763579a9fb6c7")
assert visit2_actual_load_status == {
"status": "eventful",
"snapshot_id": expected_snapshot_id2.hex(),
}
assert_last_visit_matches(
swh_storage, url, status="full", type="pypi", snapshot=expected_snapshot_id2
)
expected_snapshot = Snapshot(
id=expected_snapshot_id2,
branches={
b"releases/1.1.0": SnapshotBranch(
target=hash_to_bytes("f8789ff3ed70a5f570c35d885c7bcfda7b23b091"),
target_type=TargetType.RELEASE,
),
b"releases/1.2.0": SnapshotBranch(
target=hash_to_bytes("fbbcb817f01111b06442cdcc93140ab3cc777d68"),
target_type=TargetType.RELEASE,
),
b"releases/1.3.0": SnapshotBranch(
target=hash_to_bytes("a21b09cbec8e31f47307f196bb1f939effc26e11"),
target_type=TargetType.RELEASE,
),
b"HEAD": SnapshotBranch(
target=b"releases/1.3.0",
target_type=TargetType.ALIAS,
),
},
)
assert_last_visit_matches(
swh_storage, url, status="full", type="pypi", snapshot=expected_snapshot.id
)
check_snapshot(expected_snapshot, swh_storage)
assert {
"content": 6 + 1, # 1 more content
"directory": 4 + 2, # 2 more directories
"origin": 1,
"origin_visit": 1 + 1,
"release": 2 + 1, # 1 more release
"revision": 0,
"skipped_content": 0,
"snapshot": 1 + 1, # 1 more snapshot
} == visit2_stats
urls = [
m.url
for m in requests_mock_datadir_visits.request_history
if m.url.startswith("https://files.pythonhosted.org")
]
# visited each artifact once across 2 visits
assert len(urls) == len(set(urls))
def test_pypi_multiple_visits_with_no_change(swh_storage, requests_mock_datadir):
"""Multiple visits with no changes results in 1 same snapshot"""
url = "https://pypi.org/project/0805nexter"
loader = PyPILoader(swh_storage, url)
actual_load_status = loader.load()
snapshot_id = hash_to_bytes("3dd50c1a0e48a7625cf1427e3190a65b787c774e")
assert actual_load_status == {
"status": "eventful",
"snapshot_id": snapshot_id.hex(),
}
assert_last_visit_matches(
swh_storage, url, status="full", type="pypi", snapshot=snapshot_id
)
expected_snapshot = Snapshot(
id=snapshot_id,
branches={
b"releases/1.1.0": SnapshotBranch(
target=hash_to_bytes("f8789ff3ed70a5f570c35d885c7bcfda7b23b091"),
target_type=TargetType.RELEASE,
),
b"releases/1.2.0": SnapshotBranch(
target=hash_to_bytes("fbbcb817f01111b06442cdcc93140ab3cc777d68"),
target_type=TargetType.RELEASE,
),
b"HEAD": SnapshotBranch(
target=b"releases/1.2.0",
target_type=TargetType.ALIAS,
),
},
)
check_snapshot(expected_snapshot, swh_storage)
stats = get_stats(swh_storage)
assert {
"content": 6,
"directory": 4,
"origin": 1,
"origin_visit": 1,
"release": 2,
"revision": 0,
"skipped_content": 0,
"snapshot": 1,
} == stats
actual_load_status2 = loader.load()
assert actual_load_status2 == {
"status": "uneventful",
"snapshot_id": actual_load_status2["snapshot_id"],
}
visit_status2 = assert_last_visit_matches(
swh_storage, url, status="full", type="pypi"
)
stats2 = get_stats(swh_storage)
expected_stats2 = stats.copy()
expected_stats2["origin_visit"] = 1 + 1
assert expected_stats2 == stats2
# same snapshot
assert visit_status2.snapshot == snapshot_id
def test_pypi_release_metadata_structure(
swh_storage, requests_mock_datadir, _0805nexter_api_info
):
url = "https://pypi.org/project/0805nexter"
loader = PyPILoader(swh_storage, url)
actual_load_status = loader.load()
assert actual_load_status["status"] == "eventful"
assert actual_load_status["snapshot_id"] is not None
expected_release_id = hash_to_bytes("fbbcb817f01111b06442cdcc93140ab3cc777d68")
expected_snapshot = Snapshot(
id=hash_to_bytes(actual_load_status["snapshot_id"]),
branches={
b"HEAD": SnapshotBranch(
target=b"releases/1.2.0",
target_type=TargetType.ALIAS,
),
b"releases/1.1.0": SnapshotBranch(
target=hash_to_bytes("f8789ff3ed70a5f570c35d885c7bcfda7b23b091"),
target_type=TargetType.RELEASE,
),
b"releases/1.2.0": SnapshotBranch(
target=expected_release_id,
target_type=TargetType.RELEASE,
),
},
)
assert_last_visit_matches(
swh_storage, url, status="full", type="pypi", snapshot=expected_snapshot.id
)
check_snapshot(expected_snapshot, swh_storage)
release = swh_storage.release_get([expected_release_id])[0]
assert release is not None
release_swhid = CoreSWHID(
object_type=ObjectType.RELEASE, object_id=expected_release_id
)
directory_swhid = ExtendedSWHID(
object_type=ExtendedObjectType.DIRECTORY, object_id=release.target
)
metadata_authority = MetadataAuthority(
type=MetadataAuthorityType.FORGE,
url="https://pypi.org/",
)
expected_metadata = [
RawExtrinsicMetadata(
target=directory_swhid,
authority=metadata_authority,
fetcher=MetadataFetcher(
name="swh.loader.package.pypi.loader.PyPILoader",
version=__version__,
),
discovery_date=loader.visit_date,
format="pypi-project-json",
metadata=json.dumps(
json.loads(_0805nexter_api_info)["releases"]["1.2.0"][0]
).encode(),
origin=url,
release=release_swhid,
)
]
assert swh_storage.raw_extrinsic_metadata_get(
directory_swhid,
metadata_authority,
) == PagedResult(
next_page_token=None,
results=expected_metadata,
)