diff --git a/swh/loader/package/archive/tests/test_archive.py b/swh/loader/package/archive/tests/test_archive.py index f5682bc..7cb2cfa 100644 --- a/swh/loader/package/archive/tests/test_archive.py +++ b/swh/loader/package/archive/tests/test_archive.py @@ -1,381 +1,380 @@ # Copyright (C) 2019-2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information import attr from swh.model.hashutil import hash_to_bytes from swh.model.model import Snapshot, SnapshotBranch, TargetType from swh.loader.package.archive.loader import ArchiveLoader, ArchivePackageInfo from swh.loader.package.tests.common import check_metadata_paths from swh.loader.tests import ( assert_last_visit_matches, check_snapshot, get_stats, ) URL = "https://ftp.gnu.org/gnu/8sync/" GNU_ARTIFACTS = [ { "time": 944729610, "url": "https://ftp.gnu.org/gnu/8sync/8sync-0.1.0.tar.gz", "length": 221837, "filename": "8sync-0.1.0.tar.gz", "version": "0.1.0", } ] _expected_new_contents_first_visit = [ "e9258d81faf5881a2f96a77ba609396f82cb97ad", "1170cf105b04b7e2822a0e09d2acf71da7b9a130", "fbd27c3f41f2668624ffc80b7ba5db9b92ff27ac", "0057bec9b5422aff9256af240b177ac0e3ac2608", "2b8d0d0b43a1078fc708930c8ddc2956a86c566e", "27de3b3bc6545d2a797aeeb4657c0e215a0c2e55", "2e6db43f5cd764e677f416ff0d0c78c7a82ef19b", "ae9be03bd2a06ed8f4f118d3fe76330bb1d77f62", "edeb33282b2bffa0e608e9d2fd960fd08093c0ea", "d64e64d4c73679323f8d4cde2643331ba6c20af9", "7a756602914be889c0a2d3952c710144b3e64cb0", "84fb589b554fcb7f32b806951dcf19518d67b08f", "8624bcdae55baeef00cd11d5dfcfa60f68710a02", "e08441aeab02704cfbd435d6445f7c072f8f524e", "f67935bc3a83a67259cda4b2d43373bd56703844", "809788434b433eb2e3cfabd5d591c9a659d5e3d8", "7d7c6c8c5ebaeff879f61f37083a3854184f6c41", "b99fec102eb24bffd53ab61fc30d59e810f116a2", "7d149b28eaa228b3871c91f0d5a95a2fa7cb0c68", "f0c97052e567948adf03e641301e9983c478ccff", "7fb724242e2b62b85ca64190c31dcae5303e19b3", "4f9709e64a9134fe8aefb36fd827b84d8b617ab5", "7350628ccf194c2c3afba4ac588c33e3f3ac778d", "0bb892d9391aa706dc2c3b1906567df43cbe06a2", "49d4c0ce1a16601f1e265d446b6c5ea6b512f27c", "6b5cc594ac466351450f7f64a0b79fdaf4435ad3", "3046e5d1f70297e2a507b98224b6222c9688d610", "1572607d456d7f633bc6065a2b3048496d679a31", ] _expected_new_directories_first_visit = [ "daabc65ec75d487b1335ffc101c0ac11c803f8fc", "263be23b4a8101d3ad0d9831319a3e0f2b065f36", "7f6e63ba6eb3e2236f65892cd822041f1a01dd5c", "4db0a3ecbc976083e2dac01a62f93729698429a3", "dfef1c80e1098dd5deda664bb44a9ab1f738af13", "eca971d346ea54d95a6e19d5051f900237fafdaa", "3aebc29ed1fccc4a6f2f2010fb8e57882406b528", ] _expected_new_revisions_first_visit = { "44183488c0774ce3c957fa19ba695cf18a4a42b3": ( "3aebc29ed1fccc4a6f2f2010fb8e57882406b528" ) } def visit_with_no_artifact_found(swh_config, requests_mock_datadir): url = URL unknown_artifact_url = "https://ftp.g.o/unknown/8sync-0.1.0.tar.gz" loader = ArchiveLoader( url, artifacts=[ { "time": 944729610, "url": unknown_artifact_url, # unknown artifact "length": 221837, "filename": "8sync-0.1.0.tar.gz", "version": "0.1.0", } ], ) actual_load_status = loader.load() assert actual_load_status["status"] == "uneventful" assert actual_load_status["snapshot_id"] is not None stats = get_stats(loader.storage) assert { "content": 0, "directory": 0, "origin": 1, "origin_visit": 1, "person": 0, "release": 0, "revision": 0, "skipped_content": 0, "snapshot": 1, } == stats assert_last_visit_matches(loader.storage, url, status="partial", type="tar") def test_check_revision_metadata_structure(swh_config, requests_mock_datadir): loader = ArchiveLoader(url=URL, artifacts=GNU_ARTIFACTS) actual_load_status = loader.load() assert actual_load_status["status"] == "eventful" assert actual_load_status["snapshot_id"] is not None assert_last_visit_matches(loader.storage, URL, status="full", type="tar") expected_revision_id = hash_to_bytes("44183488c0774ce3c957fa19ba695cf18a4a42b3") - revision = list(loader.storage.revision_get([expected_revision_id]))[0] - + revision = loader.storage.revision_get([expected_revision_id])[0] assert revision is not None check_metadata_paths( - revision["metadata"], + revision.metadata, paths=[ ("intrinsic", dict), ("extrinsic.provider", str), ("extrinsic.when", str), ("extrinsic.raw", dict), ("original_artifact", list), ], ) - for original_artifact in revision["metadata"]["original_artifact"]: + for original_artifact in revision.metadata["original_artifact"]: check_metadata_paths( original_artifact, paths=[("filename", str), ("length", int), ("checksums", dict),], ) def test_visit_with_release_artifact_no_prior_visit(swh_config, requests_mock_datadir): """With no prior visit, load a gnu project ends up with 1 snapshot """ loader = ArchiveLoader(url=URL, artifacts=GNU_ARTIFACTS) actual_load_status = loader.load() assert actual_load_status["status"] == "eventful" expected_snapshot_first_visit_id = hash_to_bytes( "c419397fd912039825ebdbea378bc6283f006bf5" ) assert ( hash_to_bytes(actual_load_status["snapshot_id"]) == expected_snapshot_first_visit_id ) assert_last_visit_matches(loader.storage, URL, status="full", type="tar") stats = get_stats(loader.storage) assert { "content": len(_expected_new_contents_first_visit), "directory": len(_expected_new_directories_first_visit), "origin": 1, "origin_visit": 1, "person": 1, "release": 0, "revision": len(_expected_new_revisions_first_visit), "skipped_content": 0, "snapshot": 1, } == stats expected_contents = map(hash_to_bytes, _expected_new_contents_first_visit) assert list(loader.storage.content_missing_per_sha1(expected_contents)) == [] expected_dirs = map(hash_to_bytes, _expected_new_directories_first_visit) assert list(loader.storage.directory_missing(expected_dirs)) == [] expected_revs = map(hash_to_bytes, _expected_new_revisions_first_visit) assert list(loader.storage.revision_missing(expected_revs)) == [] expected_snapshot = Snapshot( id=expected_snapshot_first_visit_id, branches={ b"HEAD": SnapshotBranch( target_type=TargetType.ALIAS, target=b"releases/0.1.0", ), b"releases/0.1.0": SnapshotBranch( target_type=TargetType.REVISION, target=hash_to_bytes("44183488c0774ce3c957fa19ba695cf18a4a42b3"), ), }, ) check_snapshot(expected_snapshot, loader.storage) def test_2_visits_without_change(swh_config, requests_mock_datadir): """With no prior visit, load a gnu project ends up with 1 snapshot """ url = URL loader = ArchiveLoader(url, artifacts=GNU_ARTIFACTS) actual_load_status = loader.load() assert actual_load_status["status"] == "eventful" assert actual_load_status["snapshot_id"] is not None assert_last_visit_matches(loader.storage, url, status="full", type="tar") actual_load_status2 = loader.load() assert actual_load_status2["status"] == "uneventful" assert actual_load_status2["snapshot_id"] is not None assert actual_load_status["snapshot_id"] == actual_load_status2["snapshot_id"] assert_last_visit_matches(loader.storage, url, status="full", type="tar") urls = [ m.url for m in requests_mock_datadir.request_history if m.url.startswith("https://ftp.gnu.org") ] assert len(urls) == 1 def test_2_visits_with_new_artifact(swh_config, requests_mock_datadir): """With no prior visit, load a gnu project ends up with 1 snapshot """ url = URL artifact1 = GNU_ARTIFACTS[0] loader = ArchiveLoader(url, [artifact1]) actual_load_status = loader.load() assert actual_load_status["status"] == "eventful" assert actual_load_status["snapshot_id"] is not None assert_last_visit_matches(loader.storage, url, status="full", type="tar") stats = get_stats(loader.storage) assert { "content": len(_expected_new_contents_first_visit), "directory": len(_expected_new_directories_first_visit), "origin": 1, "origin_visit": 1, "person": 1, "release": 0, "revision": len(_expected_new_revisions_first_visit), "skipped_content": 0, "snapshot": 1, } == stats urls = [ m.url for m in requests_mock_datadir.request_history if m.url.startswith("https://ftp.gnu.org") ] assert len(urls) == 1 artifact2 = { "time": 1480991830, "url": "https://ftp.gnu.org/gnu/8sync/8sync-0.2.0.tar.gz", "length": 238466, "filename": "8sync-0.2.0.tar.gz", "version": "0.2.0", } loader2 = ArchiveLoader(url, [artifact1, artifact2]) # implementation detail: share the storage in between visits loader2.storage = loader.storage stats2 = get_stats(loader2.storage) assert stats == stats2 # ensure we share the storage actual_load_status2 = loader2.load() assert actual_load_status2["status"] == "eventful" assert actual_load_status2["snapshot_id"] is not None stats2 = get_stats(loader.storage) assert { "content": len(_expected_new_contents_first_visit) + 14, "directory": len(_expected_new_directories_first_visit) + 8, "origin": 1, "origin_visit": 1 + 1, "person": 1, "release": 0, "revision": len(_expected_new_revisions_first_visit) + 1, "skipped_content": 0, "snapshot": 1 + 1, } == stats2 assert_last_visit_matches(loader.storage, url, status="full", type="tar") urls = [ m.url for m in requests_mock_datadir.request_history if m.url.startswith("https://ftp.gnu.org") ] # 1 artifact (2nd time no modification) + 1 new artifact assert len(urls) == 2 def test_2_visits_without_change_not_gnu(swh_config, requests_mock_datadir): """Load a project archive (not gnu) ends up with 1 snapshot """ url = "https://something.else.org/8sync/" artifacts = [ # this is not a gnu artifact { "time": "1999-12-09T09:53:30+00:00", # it's also not a timestamp "sha256": "d5d1051e59b2be6f065a9fc6aedd3a391e44d0274b78b9bb4e2b57a09134dbe4", # noqa # keep a gnu artifact reference to avoid adding other test files "url": "https://ftp.gnu.org/gnu/8sync/8sync-0.2.0.tar.gz", "length": 238466, "filename": "8sync-0.2.0.tar.gz", "version": "0.2.0", } ] # Here the loader defines the id_keys to use for existence in the snapshot # It's not the default archive loader which loader = ArchiveLoader( url, artifacts=artifacts, identity_artifact_keys=["sha256", "length", "url"] ) actual_load_status = loader.load() assert actual_load_status["status"] == "eventful" assert actual_load_status["snapshot_id"] is not None assert_last_visit_matches(loader.storage, url, status="full", type="tar") actual_load_status2 = loader.load() assert actual_load_status2["status"] == "uneventful" assert actual_load_status2["snapshot_id"] == actual_load_status["snapshot_id"] assert_last_visit_matches(loader.storage, url, status="full", type="tar") urls = [ m.url for m in requests_mock_datadir.request_history if m.url.startswith("https://ftp.gnu.org") ] assert len(urls) == 1 def test_artifact_identity(): """Compute primary key should return the right identity """ @attr.s class TestPackageInfo(ArchivePackageInfo): a = attr.ib() b = attr.ib() metadata = GNU_ARTIFACTS[0] p_info = TestPackageInfo( raw_info={**metadata, "a": 1, "b": 2}, a=1, b=2, **metadata, ) for id_keys, expected_id in [ (["a", "b"], [1, 2]), ([], []), (["a", "key-that-does-not-exist"], [1, None]), ( None, [ metadata["time"], metadata["url"], metadata["length"], metadata["version"], ], ), ]: actual_id = p_info.artifact_identity(id_keys=id_keys) assert actual_id == expected_id diff --git a/swh/loader/package/deposit/loader.py b/swh/loader/package/deposit/loader.py index 621d5d0..53d6c3c 100644 --- a/swh/loader/package/deposit/loader.py +++ b/swh/loader/package/deposit/loader.py @@ -1,333 +1,326 @@ # Copyright (C) 2019-2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information import datetime import json import logging import requests from typing import Any, Dict, Iterator, List, Mapping, Optional, Sequence, Tuple, Union import attr from swh.model.hashutil import hash_to_hex, hash_to_bytes from swh.model.model import ( Person, Revision, RevisionType, TimestampWithTimezone, Sha1Git, MetadataAuthority, MetadataAuthorityType, MetadataFetcher, ) from swh.loader.package.loader import ( BasePackageInfo, PackageLoader, RawExtrinsicMetadataCore, ) from swh.loader.package.utils import cached_method, download from swh.storage.algos.snapshot import snapshot_get_all_branches logger = logging.getLogger(__name__) @attr.s class DepositPackageInfo(BasePackageInfo): filename = attr.ib(type=str) # instead of Optional[str] raw_info = attr.ib(type=Dict[str, Any]) author_date = attr.ib(type=datetime.datetime) """codemeta:dateCreated if any, deposit completed_date otherwise""" commit_date = attr.ib(type=datetime.datetime) """codemeta:datePublished if any, deposit completed_date otherwise""" client = attr.ib(type=str) id = attr.ib(type=int) """Internal ID of the deposit in the deposit DB""" collection = attr.ib(type=str) """The collection in the deposit; see SWORD specification.""" author = attr.ib(type=Person) committer = attr.ib(type=Person) revision_parents = attr.ib(type=Tuple[Sha1Git, ...]) """Revisions created from previous deposits, that will be used as parents of the revision created for this deposit.""" @classmethod def from_metadata( cls, metadata: Dict[str, Any], url: str, filename: str ) -> "DepositPackageInfo": # Note: # `date` and `committer_date` are always transmitted by the deposit read api # which computes itself the values. The loader needs to use those to create the # revision. raw_metadata_from_origin = json.dumps( metadata["origin_metadata"]["metadata"] ).encode() metadata = metadata.copy() # FIXME: this removes information from 'raw' metadata depo = metadata.pop("deposit") return cls( url=url, filename=filename, author_date=depo["author_date"], commit_date=depo["committer_date"], client=depo["client"], id=depo["id"], collection=depo["collection"], author=parse_author(depo["author"]), committer=parse_author(depo["committer"]), revision_parents=tuple(hash_to_bytes(p) for p in depo["revision_parents"]), raw_info=metadata, revision_extrinsic_metadata=[ RawExtrinsicMetadataCore( format="sword-v2-atom-codemeta-v2-in-json", metadata=raw_metadata_from_origin, ), ], ) class DepositLoader(PackageLoader[DepositPackageInfo]): """Load pypi origin's artifact releases into swh archive. """ visit_type = "deposit" def __init__(self, url: str, deposit_id: str): """Constructor Args: url: Origin url to associate the artifacts/metadata to deposit_id: Deposit identity """ super().__init__(url=url) config_deposit = self.config["deposit"] self.deposit_id = deposit_id self.client = ApiClient(url=config_deposit["url"], auth=config_deposit["auth"]) def get_versions(self) -> Sequence[str]: # only 1 branch 'HEAD' with no alias since we only have 1 snapshot # branch return ["HEAD"] def get_metadata_authority(self) -> MetadataAuthority: provider = self.metadata()["origin_metadata"]["provider"] assert provider["provider_type"] == "deposit_client" return MetadataAuthority( type=MetadataAuthorityType.DEPOSIT_CLIENT, url=provider["provider_url"], metadata={ "name": provider["provider_name"], **(provider["metadata"] or {}), }, ) def get_metadata_fetcher(self) -> MetadataFetcher: tool = self.metadata()["origin_metadata"]["tool"] return MetadataFetcher( name=tool["name"], version=tool["version"], metadata=tool["configuration"], ) def get_package_info( self, version: str ) -> Iterator[Tuple[str, DepositPackageInfo]]: p_info = DepositPackageInfo.from_metadata( self.metadata(), url=self.url, filename="archive.zip", ) yield "HEAD", p_info def download_package( self, p_info: DepositPackageInfo, tmpdir: str ) -> List[Tuple[str, Mapping]]: """Override to allow use of the dedicated deposit client """ return [self.client.archive_get(self.deposit_id, tmpdir, p_info.filename)] def build_revision( self, p_info: DepositPackageInfo, uncompressed_path: str, directory: Sha1Git ) -> Optional[Revision]: message = ( f"{p_info.client}: Deposit {p_info.id} in collection {p_info.collection}" ).encode("utf-8") return Revision( type=RevisionType.TAR, message=message, author=p_info.author, date=TimestampWithTimezone.from_dict(p_info.author_date), committer=p_info.committer, committer_date=TimestampWithTimezone.from_dict(p_info.commit_date), parents=p_info.revision_parents, directory=directory, synthetic=True, metadata={ "extrinsic": { "provider": self.client.metadata_url(self.deposit_id), "when": self.visit_date.isoformat(), "raw": p_info.raw_info, }, }, ) def get_extrinsic_origin_metadata(self) -> List[RawExtrinsicMetadataCore]: origin_metadata = self.metadata()["origin_metadata"] return [ RawExtrinsicMetadataCore( format="sword-v2-atom-codemeta-v2-in-json", metadata=json.dumps(origin_metadata["metadata"]).encode(), ) ] @cached_method def metadata(self): """Returns metadata from the deposit server""" return self.client.metadata_get(self.deposit_id) def load(self) -> Dict: # First making sure the deposit is known prior to trigger a loading try: self.metadata() except ValueError: logger.error(f"Unknown deposit {self.deposit_id}, ignoring") return {"status": "failed"} # Then usual loading r = super().load() success = r["status"] != "failed" # Update deposit status try: if not success: self.client.status_update(self.deposit_id, status="failed") return r snapshot_id = hash_to_bytes(r["snapshot_id"]) snapshot = snapshot_get_all_branches(self.storage, snapshot_id) if not snapshot: return r branches = snapshot.branches logger.debug("branches: %s", branches) if not branches: return r rev_id = branches[b"HEAD"].target - revisions = list(self.storage.revision_get([rev_id])) - if not revisions: - return r - - revision = revisions[0] + revision = self.storage.revision_get([rev_id])[0] if not revision: return r - # Retrieve the revision identifier - dir_id = revision["directory"] - # update the deposit's status to success with its # revision-id and directory-id self.client.status_update( self.deposit_id, status="done", revision_id=hash_to_hex(rev_id), - directory_id=hash_to_hex(dir_id), + directory_id=hash_to_hex(revision.directory), snapshot_id=r["snapshot_id"], origin_url=self.url, ) except Exception: logger.exception("Problem when trying to update the deposit's status") return {"status": "failed"} return r def parse_author(author) -> Person: """See prior fixme """ return Person( fullname=author["fullname"].encode("utf-8"), name=author["name"].encode("utf-8"), email=author["email"].encode("utf-8"), ) class ApiClient: """Private Deposit Api client """ def __init__(self, url, auth: Optional[Mapping[str, str]]): self.base_url = url.rstrip("/") self.auth = None if not auth else (auth["username"], auth["password"]) def do(self, method: str, url: str, *args, **kwargs): """Internal method to deal with requests, possibly with basic http authentication. Args: method (str): supported http methods as in get/post/put Returns: The request's execution output """ method_fn = getattr(requests, method) if self.auth: kwargs["auth"] = self.auth return method_fn(url, *args, **kwargs) def archive_get( self, deposit_id: Union[int, str], tmpdir: str, filename: str ) -> Tuple[str, Dict]: """Retrieve deposit's archive artifact locally """ url = f"{self.base_url}/{deposit_id}/raw/" return download(url, dest=tmpdir, filename=filename, auth=self.auth) def metadata_url(self, deposit_id: Union[int, str]) -> str: return f"{self.base_url}/{deposit_id}/meta/" def metadata_get(self, deposit_id: Union[int, str]) -> Dict[str, Any]: """Retrieve deposit's metadata artifact as json """ url = self.metadata_url(deposit_id) r = self.do("get", url) if r.ok: return r.json() msg = f"Problem when retrieving deposit metadata at {url}" logger.error(msg) raise ValueError(msg) def status_update( self, deposit_id: Union[int, str], status: str, revision_id: Optional[str] = None, directory_id: Optional[str] = None, snapshot_id: Optional[str] = None, origin_url: Optional[str] = None, ): """Update deposit's information including status, and persistent identifiers result of the loading. """ url = f"{self.base_url}/{deposit_id}/update/" payload = {"status": status} if revision_id: payload["revision_id"] = revision_id if directory_id: payload["directory_id"] = directory_id if snapshot_id: payload["snapshot_id"] = snapshot_id if origin_url: payload["origin_url"] = origin_url self.do("put", url, json=payload) diff --git a/swh/loader/package/deposit/tests/test_deposit.py b/swh/loader/package/deposit/tests/test_deposit.py index e2cde63..a105f99 100644 --- a/swh/loader/package/deposit/tests/test_deposit.py +++ b/swh/loader/package/deposit/tests/test_deposit.py @@ -1,409 +1,408 @@ # Copyright (C) 2019-2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information import json import re import attr import pytest from swh.model.hashutil import hash_to_bytes, hash_to_hex from swh.model.identifiers import SWHID from swh.model.model import ( Snapshot, SnapshotBranch, TargetType, MetadataAuthority, MetadataAuthorityType, MetadataFetcher, MetadataTargetType, RawExtrinsicMetadata, ) from swh.loader.package.deposit.loader import DepositLoader from swh.loader.package.tests.common import check_metadata_paths from swh.loader.tests import ( assert_last_visit_matches, check_snapshot, get_stats, ) from swh.core.pytest_plugin import requests_mock_datadir_factory DEPOSIT_URL = "https://deposit.softwareheritage.org/1/private" @pytest.fixture def requests_mock_datadir(requests_mock_datadir): """Enhance default mock data to mock put requests as the loader does some internal update queries there. """ requests_mock_datadir.put(re.compile("https")) return requests_mock_datadir def test_deposit_init_ok(swh_config, swh_loader_config): url = "some-url" deposit_id = 999 loader = DepositLoader(url, deposit_id) # Something that does not exist assert loader.url == url assert loader.client is not None assert loader.client.base_url == swh_loader_config["deposit"]["url"] def test_deposit_loading_unknown_deposit(swh_config, requests_mock_datadir): """Loading an unknown deposit should fail no origin, no visit, no snapshot """ # private api url form: 'https://deposit.s.o/1/private/hal/666/raw/' url = "some-url" unknown_deposit_id = 667 loader = DepositLoader(url, unknown_deposit_id) # does not exist actual_load_status = loader.load() assert actual_load_status == {"status": "failed"} stats = get_stats(loader.storage) assert { "content": 0, "directory": 0, "origin": 0, "origin_visit": 0, "person": 0, "release": 0, "revision": 0, "skipped_content": 0, "snapshot": 0, } == stats requests_mock_datadir_missing_one = requests_mock_datadir_factory( ignore_urls=[f"{DEPOSIT_URL}/666/raw/",] ) def test_deposit_loading_failure_to_retrieve_1_artifact( swh_config, requests_mock_datadir_missing_one ): """Deposit with missing artifact ends up with an uneventful/partial visit """ # private api url form: 'https://deposit.s.o/1/private/hal/666/raw/' url = "some-url-2" deposit_id = 666 loader = DepositLoader(url, deposit_id) actual_load_status = loader.load() assert actual_load_status["status"] == "uneventful" assert actual_load_status["snapshot_id"] is not None assert_last_visit_matches(loader.storage, url, status="partial", type="deposit") stats = get_stats(loader.storage) assert { "content": 0, "directory": 0, "origin": 1, "origin_visit": 1, "person": 0, "release": 0, "revision": 0, "skipped_content": 0, "snapshot": 1, } == stats def test_revision_metadata_structure(swh_config, requests_mock_datadir): url = "https://hal-test.archives-ouvertes.fr/some-external-id" deposit_id = 666 loader = DepositLoader(url, deposit_id) actual_load_status = loader.load() assert actual_load_status["status"] == "eventful" assert actual_load_status["snapshot_id"] is not None expected_revision_id = hash_to_bytes("637318680351f5d78856d13264faebbd91efe9bb") - revision = list(loader.storage.revision_get([expected_revision_id]))[0] - + revision = loader.storage.revision_get([expected_revision_id])[0] assert revision is not None check_metadata_paths( - revision["metadata"], + revision.metadata, paths=[ ("extrinsic.provider", str), ("extrinsic.when", str), ("extrinsic.raw", dict), ("original_artifact", list), ], ) # Only 2 top-level keys now - assert set(revision["metadata"].keys()) == {"extrinsic", "original_artifact"} + assert set(revision.metadata.keys()) == {"extrinsic", "original_artifact"} - for original_artifact in revision["metadata"]["original_artifact"]: + for original_artifact in revision.metadata["original_artifact"]: check_metadata_paths( original_artifact, paths=[("filename", str), ("length", int), ("checksums", dict),], ) def test_deposit_loading_ok(swh_config, requests_mock_datadir): url = "https://hal-test.archives-ouvertes.fr/some-external-id" deposit_id = 666 loader = DepositLoader(url, deposit_id) actual_load_status = loader.load() expected_snapshot_id = "b2b327b33dc85818bd23c3ccda8b7e675a66ecbd" assert actual_load_status == { "status": "eventful", "snapshot_id": expected_snapshot_id, } assert_last_visit_matches(loader.storage, url, status="full", type="deposit") stats = get_stats(loader.storage) assert { "content": 303, "directory": 12, "origin": 1, "origin_visit": 1, "person": 1, "release": 0, "revision": 1, "skipped_content": 0, "snapshot": 1, } == stats revision_id_hex = "637318680351f5d78856d13264faebbd91efe9bb" revision_id = hash_to_bytes(revision_id_hex) expected_snapshot = Snapshot( id=hash_to_bytes(expected_snapshot_id), branches={ b"HEAD": SnapshotBranch( target=revision_id, target_type=TargetType.REVISION, ), }, ) check_snapshot(expected_snapshot, storage=loader.storage) - revision = next(loader.storage.revision_get([revision_id])) - assert revision + revision = loader.storage.revision_get([revision_id])[0] + assert revision is not None # check metadata fetcher = MetadataFetcher(name="swh-deposit", version="0.0.1",) authority = MetadataAuthority( type=MetadataAuthorityType.DEPOSIT_CLIENT, url="https://hal-test.archives-ouvertes.fr/", ) # Check origin metadata orig_meta = loader.storage.raw_extrinsic_metadata_get( MetadataTargetType.ORIGIN, url, authority ) assert orig_meta.next_page_token is None assert len(orig_meta.results) == 1 orig_meta0 = orig_meta.results[0] assert orig_meta0.authority == authority assert orig_meta0.fetcher == fetcher # Check revision metadata revision_swhid = SWHID(object_type="revision", object_id=revision_id) rev_meta = loader.storage.raw_extrinsic_metadata_get( MetadataTargetType.REVISION, revision_swhid, authority ) assert rev_meta.next_page_token is None assert len(rev_meta.results) == 1 rev_meta0 = rev_meta.results[0] assert rev_meta0.authority == authority assert rev_meta0.fetcher == fetcher # Retrieve the information for deposit status update query to the deposit urls = [ m for m in requests_mock_datadir.request_history if m.url == f"{DEPOSIT_URL}/{deposit_id}/update/" ] assert len(urls) == 1 update_query = urls[0] body = update_query.json() expected_body = { "status": "done", "revision_id": revision_id_hex, - "directory_id": hash_to_hex(revision["directory"]), + "directory_id": hash_to_hex(revision.directory), "snapshot_id": expected_snapshot_id, "origin_url": url, } assert body == expected_body def test_deposit_loading_ok_2(swh_config, requests_mock_datadir): """Field dates should be se appropriately """ external_id = "some-external-id" url = f"https://hal-test.archives-ouvertes.fr/{external_id}" deposit_id = 777 loader = DepositLoader(url, deposit_id) actual_load_status = loader.load() expected_snapshot_id = "3e68440fdd7c81d283f8f3aebb6f0c8657864192" assert actual_load_status == { "status": "eventful", "snapshot_id": expected_snapshot_id, } assert_last_visit_matches(loader.storage, url, status="full", type="deposit") revision_id = "564d18943d71be80d0d73b43a77cfb205bcde96c" expected_snapshot = Snapshot( id=hash_to_bytes(expected_snapshot_id), branches={ b"HEAD": SnapshotBranch( target=hash_to_bytes(revision_id), target_type=TargetType.REVISION ) }, ) check_snapshot(expected_snapshot, storage=loader.storage) raw_meta = loader.client.metadata_get(deposit_id) # Ensure the date fields are set appropriately in the revision # Retrieve the revision - revision = next(loader.storage.revision_get([hash_to_bytes(revision_id)])) + revision = loader.storage.revision_get([hash_to_bytes(revision_id)])[0] assert revision - assert revision["date"] == raw_meta["deposit"]["author_date"] - assert revision["committer_date"] == raw_meta["deposit"]["committer_date"] + assert revision.date.to_dict() == raw_meta["deposit"]["author_date"] + assert revision.committer_date.to_dict() == raw_meta["deposit"]["committer_date"] read_api = f"{DEPOSIT_URL}/{deposit_id}/meta/" - assert revision["metadata"] == { + assert revision.metadata == { "extrinsic": { "provider": read_api, "raw": { "origin": {"type": "deposit", "url": url,}, "origin_metadata": { "metadata": { "@xmlns": ["http://www.w3.org/2005/Atom"], "author": ["some awesome author", "another one", "no one",], "codemeta:dateCreated": "2017-10-07T15:17:08Z", "codemeta:datePublished": "2017-10-08T15:00:00Z", "external_identifier": "some-external-id", "url": url, }, "provider": { "metadata": None, "provider_name": "hal", "provider_type": "deposit_client", "provider_url": "https://hal-test.archives-ouvertes.fr/", }, "tool": { "configuration": {"sword_version": "2"}, "name": "swh-deposit", "version": "0.0.1", }, }, }, - "when": revision["metadata"]["extrinsic"]["when"], # dynamic + "when": revision.metadata["extrinsic"]["when"], # dynamic }, "original_artifact": [ { "checksums": { "sha1": "f8c63d7c890a7453498e6cf9fef215d85ec6801d", "sha256": "474bf646aeeff6d945eb752b1a9f8a40f3d81a88909ee7bd2d08cc822aa361e6", # noqa }, "filename": "archive.zip", "length": 956830, "url": "https://deposit.softwareheritage.org/1/private/777/raw/", } ], } fetcher = MetadataFetcher(name="swh-deposit", version="0.0.1",) authority = MetadataAuthority( type=MetadataAuthorityType.DEPOSIT_CLIENT, url="https://hal-test.archives-ouvertes.fr/", ) # Check the origin metadata swh side orig_meta = loader.storage.raw_extrinsic_metadata_get( MetadataTargetType.ORIGIN, url, authority ) assert orig_meta.next_page_token is None assert len(orig_meta.results) == 1 orig_meta0 = orig_meta.results[0] expected_metadata = RawExtrinsicMetadata( type=MetadataTargetType.ORIGIN, id=url, discovery_date=orig_meta0.discovery_date, metadata=json.dumps( { "@xmlns": ["http://www.w3.org/2005/Atom"], "author": ["some awesome author", "another one", "no one"], "codemeta:dateCreated": "2017-10-07T15:17:08Z", "codemeta:datePublished": "2017-10-08T15:00:00Z", "external_identifier": "some-external-id", "url": "https://hal-test.archives-ouvertes.fr/some-external-id", } ).encode(), format="sword-v2-atom-codemeta-v2-in-json", authority=authority, fetcher=fetcher, ) assert orig_meta0 == expected_metadata # Check the revision metadata swh side revision_swhid = SWHID(object_type="revision", object_id=revision_id) rev_meta = loader.storage.raw_extrinsic_metadata_get( MetadataTargetType.REVISION, revision_swhid, authority ) assert rev_meta.next_page_token is None assert len(rev_meta.results) == 1 rev_meta0 = rev_meta.results[0] assert rev_meta0 == attr.evolve( expected_metadata, type=MetadataTargetType.REVISION, id=revision_swhid, origin=url, ) # Retrieve the information for deposit status update query to the deposit urls = [ m for m in requests_mock_datadir.request_history if m.url == f"{DEPOSIT_URL}/{deposit_id}/update/" ] assert len(urls) == 1 update_query = urls[0] body = update_query.json() expected_body = { "status": "done", "revision_id": revision_id, - "directory_id": hash_to_hex(revision["directory"]), + "directory_id": hash_to_hex(revision.directory), "snapshot_id": expected_snapshot_id, "origin_url": url, } assert body == expected_body diff --git a/swh/loader/package/loader.py b/swh/loader/package/loader.py index 988d8f4..41a8c07 100644 --- a/swh/loader/package/loader.py +++ b/swh/loader/package/loader.py @@ -1,791 +1,789 @@ # Copyright (C) 2019-2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information import datetime import json import logging import tempfile import os import sys from typing import ( Any, Dict, Iterator, Generic, Iterable, List, Mapping, Optional, Sequence, Tuple, TypeVar, ) import attr import sentry_sdk from itertools import islice from swh.core.tarball import uncompress from swh.core.config import SWHConfig from swh.model import from_disk from swh.model.collections import ImmutableDict from swh.model.hashutil import hash_to_hex from swh.model.model import ( - BaseModel, Sha1Git, Revision, TargetType, Snapshot, Origin, OriginVisit, OriginVisitStatus, MetadataAuthority, MetadataAuthorityType, MetadataFetcher, MetadataTargetType, RawExtrinsicMetadata, ) from swh.model.identifiers import SWHID from swh.storage import get_storage from swh.storage.interface import StorageInterface from swh.storage.utils import now from swh.storage.algos.snapshot import snapshot_get_latest from swh.loader.package.utils import download logger = logging.getLogger(__name__) SWH_METADATA_AUTHORITY = MetadataAuthority( type=MetadataAuthorityType.REGISTRY, url="https://softwareheritage.org/", metadata={}, ) """Metadata authority for extrinsic metadata generated by Software Heritage. Used for metadata on "original artifacts", ie. length, filename, and checksums of downloaded archive files.""" @attr.s class RawExtrinsicMetadataCore: """Contains the core of the metadata extracted by a loader, that will be used to build a full RawExtrinsicMetadata object by adding object identifier, context, and provenance information.""" format = attr.ib(type=str) metadata = attr.ib(type=bytes) discovery_date = attr.ib(type=Optional[datetime.datetime], default=None) """Defaults to the visit date.""" @attr.s class BasePackageInfo: """Compute the primary key for a dict using the id_keys as primary key composite. Args: d: A dict entry to compute the primary key on id_keys: Sequence of keys to use as primary key Returns: The identity for that dict entry """ url = attr.ib(type=str) filename = attr.ib(type=Optional[str]) # The following attribute has kw_only=True in order to allow subclasses # to add attributes. Without kw_only, attributes without default values cannot # go after attributes with default values. # See revision_extrinsic_metadata = attr.ib( type=List[RawExtrinsicMetadataCore], default=[], kw_only=True, ) # TODO: add support for metadata for directories and contents @property def ID_KEYS(self): raise NotImplementedError(f"{self.__class__.__name__} is missing ID_KEYS") def artifact_identity(self): return [getattr(self, k) for k in self.ID_KEYS] TPackageInfo = TypeVar("TPackageInfo", bound=BasePackageInfo) class PackageLoader(Generic[TPackageInfo]): # Origin visit type (str) set by the loader visit_type = "" DEFAULT_CONFIG = { "create_authorities": ("bool", True), "create_fetchers": ("bool", True), } def __init__(self, url): """Loader's constructor. This raises exception if the minimal required configuration is missing (cf. fn:`check` method). Args: url (str): Origin url to load data from """ # This expects to use the environment variable SWH_CONFIG_FILENAME self.config = SWHConfig.parse_config_file() self._check_configuration() self.storage: StorageInterface = get_storage(**self.config["storage"]) self.url = url self.visit_date = datetime.datetime.now(tz=datetime.timezone.utc) self.max_content_size = self.config["max_content_size"] def _check_configuration(self): """Checks the minimal configuration required is set for the loader. If some required configuration is missing, exception detailing the issue is raised. """ if "storage" not in self.config: raise ValueError("Misconfiguration, at least the storage key should be set") def get_versions(self) -> Sequence[str]: """Return the list of all published package versions. Returns: Sequence of published versions """ return [] def get_package_info(self, version: str) -> Iterator[Tuple[str, TPackageInfo]]: """Given a release version of a package, retrieve the associated package information for such version. Args: version: Package version Returns: (branch name, package metadata) """ yield from {} def build_revision( self, p_info: TPackageInfo, uncompressed_path: str, directory: Sha1Git ) -> Optional[Revision]: """Build the revision from the archive metadata (extrinsic artifact metadata) and the intrinsic metadata. Args: p_info: Package information uncompressed_path: Artifact uncompressed path on disk Returns: Revision object """ raise NotImplementedError("build_revision") def get_default_version(self) -> str: """Retrieve the latest release version if any. Returns: Latest version """ return "" def last_snapshot(self) -> Optional[Snapshot]: """Retrieve the last snapshot out of the last visit. """ return snapshot_get_latest(self.storage, self.url) - def known_artifacts(self, snapshot: Optional[Snapshot]) -> Dict[Sha1Git, BaseModel]: + def known_artifacts( + self, snapshot: Optional[Snapshot] + ) -> Dict[Sha1Git, Optional[ImmutableDict[str, object]]]: """Retrieve the known releases/artifact for the origin. Args snapshot: snapshot for the visit Returns: Dict of keys revision id (bytes), values a metadata Dict. """ if not snapshot: return {} # retrieve only revisions (e.g the alias we do not want here) revs = [ rev.target for rev in snapshot.branches.values() if rev and rev.target_type == TargetType.REVISION ] known_revisions = self.storage.revision_get(revs) - return { - revision["id"]: revision["metadata"] - for revision in known_revisions - if revision + revision.id: revision.metadata for revision in known_revisions if revision } def resolve_revision_from( self, known_artifacts: Dict, p_info: TPackageInfo, ) -> Optional[bytes]: """Resolve the revision from a snapshot and an artifact metadata dict. If the artifact has already been downloaded, this will return the existing revision targeting that uncompressed artifact directory. Otherwise, this returns None. Args: snapshot: Snapshot p_info: Package information Returns: None or revision identifier """ return None def download_package( self, p_info: TPackageInfo, tmpdir: str ) -> List[Tuple[str, Mapping]]: """Download artifacts for a specific package. All downloads happen in in the tmpdir folder. Default implementation expects the artifacts package info to be about one artifact per package. Note that most implementation have 1 artifact per package. But some implementation have multiple artifacts per package (debian), some have none, the package is the artifact (gnu). Args: artifacts_package_info: Information on the package artifacts to download (url, filename, etc...) tmpdir: Location to retrieve such artifacts Returns: List of (path, computed hashes) """ return [download(p_info.url, dest=tmpdir, filename=p_info.filename)] def uncompress( self, dl_artifacts: List[Tuple[str, Mapping[str, Any]]], dest: str ) -> str: """Uncompress the artifact(s) in the destination folder dest. Optionally, this could need to use the p_info dict for some more information (debian). """ uncompressed_path = os.path.join(dest, "src") for a_path, _ in dl_artifacts: uncompress(a_path, dest=uncompressed_path) return uncompressed_path def extra_branches(self) -> Dict[bytes, Mapping[str, Any]]: """Return an extra dict of branches that are used to update the set of branches. """ return {} def load(self) -> Dict: """Load for a specific origin the associated contents. for each package version of the origin 1. Fetch the files for one package version By default, this can be implemented as a simple HTTP request. Loaders with more specific requirements can override this, e.g.: the PyPI loader checks the integrity of the downloaded files; the Debian loader has to download and check several files for one package version. 2. Extract the downloaded files By default, this would be a universal archive/tarball extraction. Loaders for specific formats can override this method (for instance, the Debian loader uses dpkg-source -x). 3. Convert the extracted directory to a set of Software Heritage objects Using swh.model.from_disk. 4. Extract the metadata from the unpacked directories This would only be applicable for "smart" loaders like npm (parsing the package.json), PyPI (parsing the PKG-INFO file) or Debian (parsing debian/changelog and debian/control). On "minimal-metadata" sources such as the GNU archive, the lister should provide the minimal set of metadata needed to populate the revision/release objects (authors, dates) as an argument to the task. 5. Generate the revision/release objects for the given version. From the data generated at steps 3 and 4. end for each 6. Generate and load the snapshot for the visit Using the revisions/releases collected at step 5., and the branch information from step 0., generate a snapshot and load it into the Software Heritage archive """ status_load = "uneventful" # either: eventful, uneventful, failed status_visit = "full" # either: partial, full tmp_revisions = {} # type: Dict[str, List] snapshot = None failed_branches: List[str] = [] def finalize_visit() -> Dict[str, Any]: """Finalize the visit: - flush eventual unflushed data to storage - update origin visit's status - return the task's status """ self.storage.flush() snapshot_id: Optional[bytes] = None if snapshot and snapshot.id: # to prevent the snapshot.id to b"" snapshot_id = snapshot.id assert visit.visit visit_status = OriginVisitStatus( origin=self.url, visit=visit.visit, date=now(), status=status_visit, snapshot=snapshot_id, ) self.storage.origin_visit_status_add([visit_status]) result: Dict[str, Any] = { "status": status_load, } if snapshot_id: result["snapshot_id"] = hash_to_hex(snapshot_id) if failed_branches: logger.warning("%d failed branches", len(failed_branches)) for i, urls in enumerate(islice(failed_branches, 50)): prefix_url = "Failed branches: " if i == 0 else "" logger.warning("%s%s", prefix_url, urls) return result # Prepare origin and origin_visit origin = Origin(url=self.url) try: self.storage.origin_add([origin]) visit = list( self.storage.origin_visit_add( [ OriginVisit( origin=self.url, date=self.visit_date, type=self.visit_type, ) ] ) )[0] except Exception as e: logger.exception("Failed to initialize origin_visit for %s", self.url) sentry_sdk.capture_exception(e) return {"status": "failed"} try: last_snapshot = self.last_snapshot() logger.debug("last snapshot: %s", last_snapshot) known_artifacts = self.known_artifacts(last_snapshot) logger.debug("known artifacts: %s", known_artifacts) except Exception as e: logger.exception("Failed to get previous state for %s", self.url) sentry_sdk.capture_exception(e) status_visit = "partial" status_load = "failed" return finalize_visit() load_exceptions: List[Exception] = [] for version in self.get_versions(): # for each logger.debug("version: %s", version) tmp_revisions[version] = [] # `p_` stands for `package_` for branch_name, p_info in self.get_package_info(version): logger.debug("package_info: %s", p_info) revision_id = self.resolve_revision_from(known_artifacts, p_info) if revision_id is None: try: revision_id = self._load_revision(p_info, origin) if revision_id: self._load_extrinsic_revision_metadata(p_info, revision_id) self.storage.flush() status_load = "eventful" except Exception as e: self.storage.clear_buffers() load_exceptions.append(e) sentry_sdk.capture_exception(e) logger.exception( "Failed loading branch %s for %s", branch_name, self.url ) failed_branches.append(branch_name) continue if revision_id is None: continue tmp_revisions[version].append((branch_name, revision_id)) if load_exceptions: status_visit = "partial" if not tmp_revisions: # We could not load any revisions; fail completely status_visit = "partial" status_load = "failed" return finalize_visit() try: # Retrieve the default release version (the "latest" one) default_version = self.get_default_version() logger.debug("default version: %s", default_version) # Retrieve extra branches extra_branches = self.extra_branches() logger.debug("extra branches: %s", extra_branches) snapshot = self._load_snapshot( default_version, tmp_revisions, extra_branches ) self.storage.flush() except Exception as e: logger.exception("Failed to build snapshot for origin %s", self.url) sentry_sdk.capture_exception(e) status_visit = "partial" status_load = "failed" if snapshot: try: metadata_objects = self.build_extrinsic_snapshot_metadata(snapshot.id) self._load_metadata_objects(metadata_objects) except Exception as e: logger.exception( "Failed to load extrinsic snapshot metadata for %s", self.url ) sentry_sdk.capture_exception(e) status_visit = "partial" status_load = "failed" try: metadata_objects = self.build_extrinsic_origin_metadata() self._load_metadata_objects(metadata_objects) except Exception as e: logger.exception( "Failed to load extrinsic origin metadata for %s", self.url ) sentry_sdk.capture_exception(e) status_visit = "partial" status_load = "failed" return finalize_visit() def _load_directory( self, dl_artifacts: List[Tuple[str, Mapping[str, Any]]], tmpdir: str ) -> Tuple[str, from_disk.Directory]: uncompressed_path = self.uncompress(dl_artifacts, dest=tmpdir) logger.debug("uncompressed_path: %s", uncompressed_path) directory = from_disk.Directory.from_disk( path=uncompressed_path.encode("utf-8"), max_content_length=self.max_content_size, ) contents, skipped_contents, directories = from_disk.iter_directory(directory) logger.debug("Number of skipped contents: %s", len(skipped_contents)) self.storage.skipped_content_add(skipped_contents) logger.debug("Number of contents: %s", len(contents)) self.storage.content_add(contents) logger.debug("Number of directories: %s", len(directories)) self.storage.directory_add(directories) return (uncompressed_path, directory) def _load_revision(self, p_info: TPackageInfo, origin) -> Optional[Sha1Git]: """Does all the loading of a revision itself: * downloads a package and uncompresses it * loads it from disk * adds contents, directories, and revision to self.storage * returns (revision_id, loaded) Raises exception when unable to download or uncompress artifacts """ with tempfile.TemporaryDirectory() as tmpdir: dl_artifacts = self.download_package(p_info, tmpdir) (uncompressed_path, directory) = self._load_directory(dl_artifacts, tmpdir) # FIXME: This should be release. cf. D409 revision = self.build_revision( p_info, uncompressed_path, directory=directory.hash ) if not revision: # Some artifacts are missing intrinsic metadata # skipping those return None metadata = [metadata for (filepath, metadata) in dl_artifacts] extra_metadata: Tuple[str, Any] = ( "original_artifact", metadata, ) if revision.metadata is not None: full_metadata = list(revision.metadata.items()) + [extra_metadata] else: full_metadata = [extra_metadata] # TODO: don't add these extrinsic metadata to the revision. revision = attr.evolve(revision, metadata=ImmutableDict(full_metadata)) original_artifact_metadata = RawExtrinsicMetadata( type=MetadataTargetType.REVISION, id=SWHID(object_type="revision", object_id=revision.id), discovery_date=self.visit_date, authority=SWH_METADATA_AUTHORITY, fetcher=self.get_metadata_fetcher(), format="original-artifacts-json", metadata=json.dumps(metadata).encode(), origin=self.url, ) self._load_metadata_objects([original_artifact_metadata]) logger.debug("Revision: %s", revision) self.storage.revision_add([revision]) return revision.id def _load_snapshot( self, default_version: str, revisions: Dict[str, List[Tuple[str, bytes]]], extra_branches: Dict[bytes, Mapping[str, Any]], ) -> Optional[Snapshot]: """Build snapshot out of the current revisions stored and extra branches. Then load it in the storage. """ logger.debug("revisions: %s", revisions) # Build and load the snapshot branches = {} # type: Dict[bytes, Mapping[str, Any]] for version, branch_name_revisions in revisions.items(): if version == default_version and len(branch_name_revisions) == 1: # only 1 branch (no ambiguity), we can create an alias # branch 'HEAD' branch_name, _ = branch_name_revisions[0] # except for some corner case (deposit) if branch_name != "HEAD": branches[b"HEAD"] = { "target_type": "alias", "target": branch_name.encode("utf-8"), } for branch_name, target in branch_name_revisions: branches[branch_name.encode("utf-8")] = { "target_type": "revision", "target": target, } # Deal with extra-branches for name, branch_target in extra_branches.items(): if name in branches: logger.error("Extra branch '%s' has been ignored", name) else: branches[name] = branch_target snapshot_data = {"branches": branches} logger.debug("snapshot: %s", snapshot_data) snapshot = Snapshot.from_dict(snapshot_data) logger.debug("snapshot: %s", snapshot) self.storage.snapshot_add([snapshot]) return snapshot def get_loader_name(self) -> str: """Returns a fully qualified name of this loader.""" return f"{self.__class__.__module__}.{self.__class__.__name__}" def get_loader_version(self) -> str: """Returns the version of the current loader.""" module_name = self.__class__.__module__ or "" module_name_parts = module_name.split(".") # Iterate rootward through the package hierarchy until we find a parent of this # loader's module with a __version__ attribute. for prefix_size in range(len(module_name_parts), 0, -1): package_name = ".".join(module_name_parts[0:prefix_size]) module = sys.modules[package_name] if hasattr(module, "__version__"): return module.__version__ # type: ignore # If this loader's class has no parent package with a __version__, # it should implement it itself. raise NotImplementedError( f"Could not dynamically find the version of {self.get_loader_name()}." ) def get_metadata_fetcher(self) -> MetadataFetcher: """Returns a MetadataFetcher instance representing this package loader; which is used to for adding provenance information to extracted extrinsic metadata, if any.""" return MetadataFetcher( name=self.get_loader_name(), version=self.get_loader_version(), metadata={}, ) def get_metadata_authority(self) -> MetadataAuthority: """For package loaders that get extrinsic metadata, returns the authority the metadata are coming from. """ raise NotImplementedError("get_metadata_authority") def get_extrinsic_origin_metadata(self) -> List[RawExtrinsicMetadataCore]: """Returns metadata items, used by build_extrinsic_origin_metadata.""" return [] def build_extrinsic_origin_metadata(self) -> List[RawExtrinsicMetadata]: """Builds a list of full RawExtrinsicMetadata objects, using metadata returned by get_extrinsic_origin_metadata.""" metadata_items = self.get_extrinsic_origin_metadata() if not metadata_items: # If this package loader doesn't write metadata, no need to require # an implementation for get_metadata_authority. return [] authority = self.get_metadata_authority() fetcher = self.get_metadata_fetcher() metadata_objects = [] for item in metadata_items: metadata_objects.append( RawExtrinsicMetadata( type=MetadataTargetType.ORIGIN, id=self.url, discovery_date=item.discovery_date or self.visit_date, authority=authority, fetcher=fetcher, format=item.format, metadata=item.metadata, ) ) return metadata_objects def get_extrinsic_snapshot_metadata(self) -> List[RawExtrinsicMetadataCore]: """Returns metadata items, used by build_extrinsic_snapshot_metadata.""" return [] def build_extrinsic_snapshot_metadata( self, snapshot_id: Sha1Git ) -> List[RawExtrinsicMetadata]: """Builds a list of full RawExtrinsicMetadata objects, using metadata returned by get_extrinsic_snapshot_metadata.""" metadata_items = self.get_extrinsic_snapshot_metadata() if not metadata_items: # If this package loader doesn't write metadata, no need to require # an implementation for get_metadata_authority. return [] authority = self.get_metadata_authority() fetcher = self.get_metadata_fetcher() metadata_objects = [] for item in metadata_items: metadata_objects.append( RawExtrinsicMetadata( type=MetadataTargetType.SNAPSHOT, id=SWHID(object_type="snapshot", object_id=snapshot_id), discovery_date=item.discovery_date or self.visit_date, authority=authority, fetcher=fetcher, format=item.format, metadata=item.metadata, origin=self.url, ) ) return metadata_objects def build_extrinsic_revision_metadata( self, p_info: TPackageInfo, revision_id: Sha1Git ) -> List[RawExtrinsicMetadata]: if not p_info.revision_extrinsic_metadata: # If this package loader doesn't write metadata, no need to require # an implementation for get_metadata_authority. return [] authority = self.get_metadata_authority() fetcher = self.get_metadata_fetcher() metadata_objects = [] for item in p_info.revision_extrinsic_metadata: metadata_objects.append( RawExtrinsicMetadata( type=MetadataTargetType.REVISION, id=SWHID(object_type="revision", object_id=revision_id), discovery_date=item.discovery_date or self.visit_date, authority=authority, fetcher=fetcher, format=item.format, metadata=item.metadata, origin=self.url, ) ) return metadata_objects def _load_extrinsic_revision_metadata( self, p_info: TPackageInfo, revision_id: Sha1Git ) -> None: metadata_objects = self.build_extrinsic_revision_metadata(p_info, revision_id) self._load_metadata_objects(metadata_objects) def _load_metadata_objects( self, metadata_objects: List[RawExtrinsicMetadata] ) -> None: if not metadata_objects: # If this package loader doesn't write metadata, no need to require # an implementation for get_metadata_authority. return self._create_authorities(mo.authority for mo in metadata_objects) self._create_fetchers(mo.fetcher for mo in metadata_objects) self.storage.raw_extrinsic_metadata_add(metadata_objects) def _create_authorities(self, authorities: Iterable[MetadataAuthority]) -> None: deduplicated_authorities = { (authority.type, authority.url): authority for authority in authorities } if authorities: self.storage.metadata_authority_add(list(deduplicated_authorities.values())) def _create_fetchers(self, fetchers: Iterable[MetadataFetcher]) -> None: deduplicated_fetchers = { (fetcher.name, fetcher.version): fetcher for fetcher in fetchers } if fetchers: self.storage.metadata_fetcher_add(list(deduplicated_fetchers.values())) diff --git a/swh/loader/package/nixguix/loader.py b/swh/loader/package/nixguix/loader.py index f865541..0f1f17c 100644 --- a/swh/loader/package/nixguix/loader.py +++ b/swh/loader/package/nixguix/loader.py @@ -1,321 +1,323 @@ # Copyright (C) 2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information import attr import copy import json import logging import re from typing import Any, Dict, List, Iterator, Mapping, Optional, Tuple from swh.model import hashutil +from swh.model.collections import ImmutableDict from swh.model.model import ( MetadataAuthority, MetadataAuthorityType, Revision, RevisionType, TargetType, Snapshot, - BaseModel, Sha1Git, ) from swh.loader.package.utils import EMPTY_AUTHOR, api_info from swh.loader.package.loader import ( BasePackageInfo, PackageLoader, RawExtrinsicMetadataCore, ) logger = logging.getLogger(__name__) @attr.s class NixGuixPackageInfo(BasePackageInfo): raw_info = attr.ib(type=Dict[str, Any]) integrity = attr.ib(type=str) """Hash of the archive, formatted as in the Subresource Integrity specification.""" @classmethod def from_metadata(cls, metadata: Dict[str, Any]) -> "NixGuixPackageInfo": return cls( url=metadata["url"], filename=None, integrity=metadata["integrity"], raw_info=metadata, ) class NixGuixLoader(PackageLoader[NixGuixPackageInfo]): """Load sources from a sources.json file. This loader is used to load sources used by functional package manager (eg. Nix and Guix). """ visit_type = "nixguix" def __init__(self, url): super().__init__(url=url) unsupported_file_extensions = self.config.get("unsupported_file_extensions", []) self.raw_sources = retrieve_sources(url) clean = clean_sources( parse_sources(self.raw_sources), unsupported_file_extensions ) self.sources = clean["sources"] self.provider_url = url self._integrityByUrl = {s["urls"][0]: s["integrity"] for s in self.sources} # The revision used to create the sources.json file. For Nix, # this revision belongs to the github.com/nixos/nixpkgs # repository self.revision = clean["revision"] # Note: this could be renamed get_artifacts in the PackageLoader # base class. def get_versions(self): """The first mirror of the mirror list is used as branch name in the snapshot. """ return self._integrityByUrl.keys() def get_metadata_authority(self): return MetadataAuthority( type=MetadataAuthorityType.FORGE, url=self.url, metadata={}, ) def get_extrinsic_snapshot_metadata(self): return [ RawExtrinsicMetadataCore( format="nixguix-sources-json", metadata=self.raw_sources, ), ] # Note: this could be renamed get_artifact_info in the PackageLoader # base class. def get_package_info(self, url) -> Iterator[Tuple[str, NixGuixPackageInfo]]: # TODO: try all mirrors and not only the first one. A source # can be fetched from several urls, called mirrors. We # currently only use the first one, but if the first one # fails, we should try the second one and so on. integrity = self._integrityByUrl[url] p_info = NixGuixPackageInfo.from_metadata({"url": url, "integrity": integrity}) yield url, p_info - def known_artifacts(self, snapshot: Optional[Snapshot]) -> Dict[Sha1Git, BaseModel]: + def known_artifacts( + self, snapshot: Optional[Snapshot] + ) -> Dict[Sha1Git, Optional[ImmutableDict[str, object]]]: """Almost same implementation as the default one except it filters out the extra "evaluation" branch which does not have the right metadata structure. """ if not snapshot: return {} # Skip evaluation revision which has no metadata revs = [ rev.target for branch_name, rev in snapshot.branches.items() if ( rev and rev.target_type == TargetType.REVISION and branch_name != b"evaluation" ) ] known_revisions = self.storage.revision_get(revs) ret = {} for revision in known_revisions: if not revision: # revision_get can return None continue - ret[revision["id"]] = revision["metadata"] + ret[revision.id] = revision.metadata return ret def resolve_revision_from( self, known_artifacts: Dict, p_info: NixGuixPackageInfo, ) -> Optional[bytes]: for rev_id, known_artifact in known_artifacts.items(): try: known_integrity = known_artifact["extrinsic"]["raw"]["integrity"] except KeyError as e: logger.exception( "Unexpected metadata revision structure detected: %(context)s", { "context": { "revision": hashutil.hash_to_hex(rev_id), "reason": str(e), "known_artifact": known_artifact, } }, ) # metadata field for the revision is not as expected by the loader # nixguix. We consider this not the right revision and continue checking # the other revisions continue else: if p_info.integrity == known_integrity: return rev_id return None def extra_branches(self) -> Dict[bytes, Mapping[str, Any]]: """We add a branch to the snapshot called 'evaluation' pointing to the revision used to generate the sources.json file. This revision is specified in the sources.json file itself. For the nixpkgs origin, this revision is coming from the github.com/nixos/nixpkgs repository. Note this repository is not loaded explicitly. So, this pointer can target a nonexistent revision for a time. However, the github and gnu loaders are supposed to load this revision and should create the revision pointed by this branch. This branch can be used to identify the snapshot associated to a Nix/Guix evaluation. """ return { b"evaluation": { "target_type": "revision", "target": hashutil.hash_to_bytes(self.revision), } } def build_revision( self, p_info: NixGuixPackageInfo, uncompressed_path: str, directory: Sha1Git ) -> Optional[Revision]: return Revision( type=RevisionType.TAR, message=b"", author=EMPTY_AUTHOR, date=None, committer=EMPTY_AUTHOR, committer_date=None, parents=(), directory=directory, synthetic=True, metadata={ "extrinsic": { "provider": self.provider_url, "when": self.visit_date.isoformat(), "raw": p_info.raw_info, }, }, ) def retrieve_sources(url: str) -> bytes: return api_info(url, allow_redirects=True) def parse_sources(raw_sources: bytes) -> Dict[str, Any]: return json.loads(raw_sources.decode("utf-8")) def make_pattern_unsupported_file_extension(unsupported_file_extensions: List[str],): """Make a regexp pattern for unsupported file extension out of a list of unsupported archive extension list. """ return re.compile( rf".*\.({'|'.join(map(re.escape, unsupported_file_extensions))})$", re.DOTALL ) def clean_sources( sources: Dict[str, Any], unsupported_file_extensions=[] ) -> Dict[str, Any]: """Validate and clean the sources structure. First, ensure all top level keys are present. Then, walk the sources list and remove sources that do not contain required keys. Filter out source entries whose: - required keys are missing - source type is not supported - urls attribute type is not a list - extension is known not to be supported by the loader Raises: ValueError if: - a required top level key is missing - top-level version is not 1 Returns: source Dict cleaned up """ pattern_unsupported_file = make_pattern_unsupported_file_extension( unsupported_file_extensions ) # Required top level keys required_keys = ["version", "revision", "sources"] missing_keys = [] for required_key in required_keys: if required_key not in sources: missing_keys.append(required_key) if missing_keys != []: raise ValueError( f"sources structure invalid, missing: {','.join(missing_keys)}" ) # Only the version 1 is currently supported version = int(sources["version"]) if version != 1: raise ValueError( f"The sources structure version '{sources['version']}' is not supported" ) # If a source doesn't contain required attributes, this source is # skipped but others could still be archived. verified_sources = [] for source in sources["sources"]: valid = True required_keys = ["urls", "integrity", "type"] for required_key in required_keys: if required_key not in source: logger.info( f"Skip source '{source}' because key '{required_key}' is missing", ) valid = False if valid and source["type"] != "url": logger.info( f"Skip source '{source}' because the type {source['type']} " "is not supported", ) valid = False if valid and not isinstance(source["urls"], list): logger.info( f"Skip source {source} because the urls attribute is not a list" ) valid = False if valid and len(source["urls"]) > 0: # Filter out unsupported archives supported_sources: List[str] = [] for source_url in source["urls"]: if pattern_unsupported_file.match(source_url): logger.info(f"Skip unsupported artifact url {source_url}") continue supported_sources.append(source_url) if len(supported_sources) == 0: logger.info( f"Skip source {source} because urls only reference " "unsupported artifacts. Unsupported " f"artifacts so far: {pattern_unsupported_file}" ) continue new_source = copy.deepcopy(source) new_source["urls"] = supported_sources verified_sources.append(new_source) sources["sources"] = verified_sources return sources diff --git a/swh/loader/package/nixguix/tests/test_nixguix.py b/swh/loader/package/nixguix/tests/test_nixguix.py index 7a9e9a7..2a9169d 100644 --- a/swh/loader/package/nixguix/tests/test_nixguix.py +++ b/swh/loader/package/nixguix/tests/test_nixguix.py @@ -1,681 +1,682 @@ # Copyright (C) 2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information import os import json from json.decoder import JSONDecodeError import logging from typing import Dict, Optional, Tuple from unittest.mock import patch import attr import pytest from swh.loader.package.archive.loader import ArchiveLoader from swh.loader.package.nixguix.loader import ( NixGuixPackageInfo, NixGuixLoader, parse_sources, retrieve_sources, clean_sources, make_pattern_unsupported_file_extension, ) from swh.loader.package.utils import download from swh.model.identifiers import SWHID from swh.model.model import ( MetadataAuthority, MetadataAuthorityType, MetadataFetcher, MetadataTargetType, RawExtrinsicMetadata, Snapshot, SnapshotBranch, TargetType, ) from swh.model.hashutil import hash_to_bytes, hash_to_hex from swh.storage.exc import HashCollision from swh.storage.algos.origin import origin_get_latest_visit_status from swh.storage.algos.snapshot import snapshot_get_all_branches from swh.storage.interface import PagedResult, StorageInterface from swh.loader.package import __version__ from swh.loader.tests import ( assert_last_visit_matches, get_stats, check_snapshot as check_snapshot_full, ) sources_url = "https://nix-community.github.io/nixpkgs-swh/sources.json" @pytest.fixture def raw_sources(datadir) -> bytes: with open( os.path.join( datadir, "https_nix-community.github.io", "nixpkgs-swh_sources.json" ), "rb", ) as f: return f.read() SNAPSHOT1 = Snapshot( id=hash_to_bytes("0c5881c74283793ebe9a09a105a9381e41380383"), branches={ b"evaluation": SnapshotBranch( target=hash_to_bytes("cc4e04c26672dd74e5fd0fecb78b435fb55368f7"), target_type=TargetType.REVISION, ), b"https://github.com/owner-1/repository-1/revision-1.tgz": SnapshotBranch( target=hash_to_bytes("488ad4e7b8e2511258725063cf43a2b897c503b4"), target_type=TargetType.REVISION, ), }, ) def check_snapshot(snapshot: Snapshot, storage: StorageInterface): # The `evaluation` branch is allowed to be unresolvable. It's possible at current # nixguix visit time, it is not yet visited (the git loader is in charge of its # visit for now). For more details, check the # swh.loader.package.nixguix.NixGuixLoader.extra_branches docstring. check_snapshot_full( snapshot, storage, allowed_empty=[(TargetType.REVISION, b"evaluation")] ) assert isinstance(snapshot, Snapshot) # then ensure the snapshot revisions are structurally as expected revision_ids = [] for name, branch in snapshot.branches.items(): if name == b"evaluation": continue # skipping that particular branch (cf. previous comment) if branch.target_type == TargetType.REVISION: revision_ids.append(branch.target) revisions = storage.revision_get(revision_ids) for rev in revisions: assert rev is not None - metadata = rev["metadata"] + metadata = rev.metadata + assert metadata is not None raw = metadata["extrinsic"]["raw"] assert "url" in raw assert "integrity" in raw def test_retrieve_sources(swh_config, requests_mock_datadir): j = parse_sources(retrieve_sources(sources_url)) assert "sources" in j.keys() assert len(j["sources"]) == 2 def test_retrieve_non_existing(swh_config, requests_mock_datadir): with pytest.raises(ValueError): NixGuixLoader("https://non-existing-url") def test_retrieve_non_json(swh_config, requests_mock_datadir): with pytest.raises(JSONDecodeError): NixGuixLoader("https://example.com/file.txt") def test_clean_sources_invalid_schema(swh_config, requests_mock_datadir): sources = {} with pytest.raises(ValueError, match="sources structure invalid, missing: .*"): clean_sources(sources) def test_clean_sources_invalid_version(swh_config, requests_mock_datadir): for version_ok in [1, "1"]: # Check those versions are fine clean_sources({"version": version_ok, "sources": [], "revision": "my-revision"}) for version_ko in [0, "0", 2, "2"]: # Check version != 1 raise an error with pytest.raises( ValueError, match="sources structure version .* is not supported" ): clean_sources( {"version": version_ko, "sources": [], "revision": "my-revision"} ) def test_clean_sources_invalid_sources(swh_config, requests_mock_datadir): valid_sources = [ # 1 valid source {"type": "url", "urls": ["my-url.tar.gz"], "integrity": "my-integrity"}, ] sources = { "version": 1, "sources": valid_sources + [ # integrity is missing {"type": "url", "urls": ["my-url.tgz"],}, # urls is not a list {"type": "url", "urls": "my-url.zip", "integrity": "my-integrity"}, # type is not url {"type": "git", "urls": ["my-url.zip"], "integrity": "my-integrity"}, # missing fields which got double-checked nonetheless... {"integrity": "my-integrity"}, ], "revision": "my-revision", } clean = clean_sources(sources) assert len(clean["sources"]) == len(valid_sources) def test_make_pattern_unsupported_file_extension(): unsupported_extensions = ["el", "c", "txt"] supported_extensions = ["Z", "7z"] # for test actual_unsupported_pattern = make_pattern_unsupported_file_extension( unsupported_extensions ) for supported_ext in supported_extensions: assert supported_ext not in unsupported_extensions supported_filepath = f"anything.{supported_ext}" actual_match = actual_unsupported_pattern.match(supported_filepath) assert not actual_match for unsupported_ext in unsupported_extensions: unsupported_filepath = f"something.{unsupported_ext}" actual_match = actual_unsupported_pattern.match(unsupported_filepath) assert actual_match def test_clean_sources_unsupported_artifacts(swh_config, requests_mock_datadir): unsupported_file_extensions = [ "iso", "whl", "gem", "pom", "msi", "pod", "png", "rock", "ttf", "jar", "c", "el", "rpm", "diff", "patch", ] supported_sources = [ { "type": "url", "urls": [f"https://server.org/my-url.{ext}"], "integrity": "my-integrity", } for ext in [ "known-unknown-but-ok", # this is fine as well with the current approach "zip", "tar.gz", "tgz", "tar.bz2", "tbz", "tbz2", "tar.xz", "tar", "zip", "7z", "Z", ] ] unsupported_sources = [ { "type": "url", "urls": [f"https://server.org/my-url.{ext}"], "integrity": "my-integrity", } for ext in unsupported_file_extensions ] sources = { "version": 1, "sources": supported_sources + unsupported_sources, "revision": "my-revision", } clean = clean_sources(sources, unsupported_file_extensions) assert len(clean["sources"]) == len(supported_sources) def test_loader_one_visit(swh_config, requests_mock_datadir, raw_sources): loader = NixGuixLoader(sources_url) res = loader.load() assert res["status"] == "eventful" stats = get_stats(loader.storage) assert { "content": 1, "directory": 3, "origin": 1, "origin_visit": 1, "person": 1, "release": 0, "revision": 1, "skipped_content": 0, "snapshot": 1, } == stats # The visit is partial because urls pointing to non tarball file # are not handled yet assert_last_visit_matches( loader.storage, sources_url, status="partial", type="nixguix" ) (_, visit_status) = origin_get_latest_visit_status(loader.storage, sources_url) snapshot_swhid = SWHID( object_type="snapshot", object_id=hash_to_hex(visit_status.snapshot) ) metadata_authority = MetadataAuthority( type=MetadataAuthorityType.FORGE, url=sources_url, ) expected_metadata = [ RawExtrinsicMetadata( type=MetadataTargetType.SNAPSHOT, id=snapshot_swhid, authority=metadata_authority, fetcher=MetadataFetcher( name="swh.loader.package.nixguix.loader.NixGuixLoader", version=__version__, ), discovery_date=loader.visit_date, format="nixguix-sources-json", metadata=raw_sources, origin=sources_url, ) ] assert loader.storage.raw_extrinsic_metadata_get( type=MetadataTargetType.SNAPSHOT, id=snapshot_swhid, authority=metadata_authority, ) == PagedResult(next_page_token=None, results=expected_metadata,) def test_uncompress_failure(swh_config, requests_mock_datadir): """Non tarball files are currently not supported and the uncompress function fails on such kind of files. However, even in this case of failure (because of the url https://example.com/file.txt), a snapshot and a visit has to be created (with a status partial since all files are not archived). """ loader = NixGuixLoader(sources_url) loader_status = loader.load() urls = [s["urls"][0] for s in loader.sources] assert "https://example.com/file.txt" in urls assert loader_status["status"] == "eventful" # The visit is partial because urls pointing to non tarball files # are not handled yet assert_last_visit_matches( loader.storage, sources_url, status="partial", type="nixguix" ) def test_loader_incremental(swh_config, requests_mock_datadir): """Ensure a second visit do not download artifact already downloaded by the previous visit. """ loader = NixGuixLoader(sources_url) load_status = loader.load() loader.load() assert load_status == {"status": "eventful", "snapshot_id": SNAPSHOT1.id.hex()} assert_last_visit_matches( loader.storage, sources_url, status="partial", type="nixguix", snapshot=SNAPSHOT1.id, ) check_snapshot(SNAPSHOT1, storage=loader.storage) urls = [ m.url for m in requests_mock_datadir.request_history if m.url == ("https://github.com/owner-1/repository-1/revision-1.tgz") ] # The artifact # 'https://github.com/owner-1/repository-1/revision-1.tgz' is only # visited one time assert len(urls) == 1 def test_loader_two_visits(swh_config, requests_mock_datadir_visits): """To ensure there is only one origin, but two visits, two revisions and two snapshots are created. The first visit creates a snapshot containing one tarball. The second visit creates a snapshot containing the same tarball and another tarball. """ loader = NixGuixLoader(sources_url) load_status = loader.load() assert load_status == {"status": "eventful", "snapshot_id": SNAPSHOT1.id.hex()} assert_last_visit_matches( loader.storage, sources_url, status="partial", type="nixguix", snapshot=SNAPSHOT1.id, ) check_snapshot(SNAPSHOT1, storage=loader.storage) stats = get_stats(loader.storage) assert { "content": 1, "directory": 3, "origin": 1, "origin_visit": 1, "person": 1, "release": 0, "revision": 1, "skipped_content": 0, "snapshot": 1, } == stats loader = NixGuixLoader(sources_url) load_status = loader.load() expected_snapshot_id_hex = "b0bfa75cbd0cc90aac3b9e95fb0f59c731176d97" expected_snapshot_id = hash_to_bytes(expected_snapshot_id_hex) assert load_status == { "status": "eventful", "snapshot_id": expected_snapshot_id_hex, } assert_last_visit_matches( loader.storage, sources_url, status="partial", type="nixguix", snapshot=expected_snapshot_id, ) # This ensures visits are incremental. Indeed, if we request a # second time an url, because of the requests_mock_datadir_visits # fixture, the file has to end with `_visit1`. expected_snapshot = Snapshot( id=expected_snapshot_id, branches={ b"evaluation": SnapshotBranch( target=hash_to_bytes("602140776b2ce6c9159bcf52ada73a297c063d5e"), target_type=TargetType.REVISION, ), b"https://github.com/owner-1/repository-1/revision-1.tgz": SnapshotBranch( target=hash_to_bytes("488ad4e7b8e2511258725063cf43a2b897c503b4"), target_type=TargetType.REVISION, ), b"https://github.com/owner-2/repository-1/revision-1.tgz": SnapshotBranch( target=hash_to_bytes("85e0bad74e33e390aaeb74f139853ae3863ee544"), target_type=TargetType.REVISION, ), }, ) check_snapshot(expected_snapshot, storage=loader.storage) stats = get_stats(loader.storage) assert { "content": 2, "directory": 5, "origin": 1, "origin_visit": 2, "person": 1, "release": 0, "revision": 2, "skipped_content": 0, "snapshot": 2, } == stats def test_resolve_revision_from(swh_config, requests_mock_datadir, datadir): loader = NixGuixLoader(sources_url) known_artifacts = { "id1": {"extrinsic": {"raw": {"url": "url1", "integrity": "integrity1"}}}, "id2": {"extrinsic": {"raw": {"url": "url2", "integrity": "integrity2"}}}, } p_info = NixGuixPackageInfo.from_metadata( {"url": "url1", "integrity": "integrity1"} ) assert loader.resolve_revision_from(known_artifacts, p_info) == "id1" p_info = NixGuixPackageInfo.from_metadata( {"url": "url3", "integrity": "integrity3"} ) assert loader.resolve_revision_from(known_artifacts, p_info) == None # noqa def test_evaluation_branch(swh_config, requests_mock_datadir): loader = NixGuixLoader(sources_url) res = loader.load() assert res["status"] == "eventful" assert_last_visit_matches( loader.storage, sources_url, status="partial", type="nixguix", snapshot=SNAPSHOT1.id, ) check_snapshot(SNAPSHOT1, storage=loader.storage) def test_eoferror(swh_config, requests_mock_datadir): """Load a truncated archive which is invalid to make the uncompress function raising the exception EOFError. We then check if a snapshot is created, meaning this error is well managed. """ sources = ( "https://nix-community.github.io/nixpkgs-swh/sources-EOFError.json" # noqa ) loader = NixGuixLoader(sources) loader.load() expected_snapshot = Snapshot( id=hash_to_bytes("4257fa2350168c6bfec726a06452ea27a2c0cb33"), branches={ b"evaluation": SnapshotBranch( target=hash_to_bytes("cc4e04c26672dd74e5fd0fecb78b435fb55368f7"), target_type=TargetType.REVISION, ), }, ) check_snapshot(expected_snapshot, storage=loader.storage) def fake_download( url: str, dest: str, hashes: Dict = {}, filename: Optional[str] = None, auth: Optional[Tuple[str, str]] = None, ) -> Tuple[str, Dict]: """Fake download which raises HashCollision (for the sake of test simpliciy, let's accept that makes sense) For tests purpose only. """ if url == "https://example.com/file.txt": # instead of failing because it's a file not dealt with by the nix guix # loader, make it raise a hash collision raise HashCollision("sha1", "f92d74e3874587aaf443d1db961d4e26dde13e9c", []) return download(url, dest, hashes, filename, auth) def test_raise_exception(swh_config, requests_mock_datadir, mocker): mock_download = mocker.patch("swh.loader.package.loader.download") mock_download.side_effect = fake_download loader = NixGuixLoader(sources_url) res = loader.load() assert res == { "status": "eventful", "snapshot_id": SNAPSHOT1.id.hex(), } check_snapshot(SNAPSHOT1, storage=loader.storage) assert len(mock_download.mock_calls) == 2 # The visit is partial because some artifact downloads failed assert_last_visit_matches( loader.storage, sources_url, status="partial", type="nixguix" ) def test_load_nixguix_one_common_artifact_from_other_loader( swh_config, datadir, requests_mock_datadir_visits, caplog ): """Misformatted revision should be caught and logged, then loading continues """ caplog.set_level(logging.ERROR, "swh.loader.package.nixguix.loader") # 1. first ingest with for example the archive loader gnu_url = "https://ftp.gnu.org/gnu/8sync/" release = "0.1.0" artifact_url = f"https://ftp.gnu.org/gnu/8sync/8sync-{release}.tar.gz" gnu_artifacts = [ { "time": 944729610, "url": artifact_url, "length": 221837, "filename": f"8sync-{release}.tar.gz", "version": release, } ] archive_loader = ArchiveLoader(url=gnu_url, artifacts=gnu_artifacts) actual_load_status = archive_loader.load() expected_snapshot_id = "c419397fd912039825ebdbea378bc6283f006bf5" assert actual_load_status["status"] == "eventful" assert actual_load_status["snapshot_id"] == expected_snapshot_id # noqa assert_last_visit_matches( archive_loader.storage, gnu_url, status="full", type="tar" ) gnu_snapshot: Snapshot = snapshot_get_all_branches( archive_loader.storage, hash_to_bytes(expected_snapshot_id) ) first_revision = gnu_snapshot.branches[f"releases/{release}".encode("utf-8")] # 2. Then ingest with the nixguix loader which lists the same artifact within its # sources.json # ensure test setup is ok data_sources = os.path.join( datadir, "https_nix-community.github.io", "nixpkgs-swh_sources_special.json" ) all_sources = json.loads(open(data_sources).read()) found = False for source in all_sources["sources"]: if source["urls"][0] == artifact_url: found = True assert ( found is True ), f"test setup error: {artifact_url} must be in {data_sources}" # first visit with a snapshot, ok sources_url = "https://nix-community.github.io/nixpkgs-swh/sources_special.json" loader = NixGuixLoader(sources_url) actual_load_status2 = loader.load() assert actual_load_status2["status"] == "eventful" assert_last_visit_matches( loader.storage, sources_url, status="full", type="nixguix" ) snapshot_id = actual_load_status2["snapshot_id"] snapshot = snapshot_get_all_branches(loader.storage, hash_to_bytes(snapshot_id)) assert snapshot # simulate a snapshot already seen with a revision with the wrong metadata structure # This revision should be skipped, thus making the artifact being ingested again. with patch( "swh.loader.package.loader.PackageLoader.last_snapshot" ) as last_snapshot: # mutate the snapshot to target a revision with the wrong metadata structure # snapshot["branches"][artifact_url.encode("utf-8")] = first_revision - old_revision = next(loader.storage.revision_get([first_revision.target])) + old_revision = loader.storage.revision_get([first_revision.target])[0] # assert that revision is not in the right format - assert old_revision["metadata"]["extrinsic"]["raw"].get("integrity", {}) == {} + assert old_revision.metadata["extrinsic"]["raw"].get("integrity", {}) == {} # mutate snapshot to create a clash snapshot = attr.evolve( snapshot, branches={ **snapshot.branches, artifact_url.encode("utf-8"): SnapshotBranch( target_type=TargetType.REVISION, - target=hash_to_bytes(old_revision["id"]), + target=hash_to_bytes(old_revision.id), ), }, ) # modify snapshot to actually change revision metadata structure so we simulate # a revision written by somebody else (structure different) last_snapshot.return_value = snapshot loader = NixGuixLoader(sources_url) actual_load_status3 = loader.load() assert last_snapshot.called assert actual_load_status3["status"] == "eventful" assert_last_visit_matches( loader.storage, sources_url, status="full", type="nixguix" ) new_snapshot_id = "32ff641e510aceefc3a6d0dcbf208b2854d2e965" assert actual_load_status3["snapshot_id"] == new_snapshot_id last_snapshot = snapshot_get_all_branches( loader.storage, hash_to_bytes(new_snapshot_id) ) new_revision_branch = last_snapshot.branches[artifact_url.encode("utf-8")] assert new_revision_branch.target_type == TargetType.REVISION - new_revision = next(loader.storage.revision_get([new_revision_branch.target])) + new_revision = loader.storage.revision_get([new_revision_branch.target])[0] # the new revision has the correct structure, so it got ingested alright by the # new run - assert new_revision["metadata"]["extrinsic"]["raw"]["integrity"] is not None + assert new_revision.metadata["extrinsic"]["raw"]["integrity"] is not None nb_detections = 0 actual_detection: Dict for record in caplog.records: logtext = record.getMessage() if "Unexpected metadata revision structure detected:" in logtext: nb_detections += 1 actual_detection = record.args["context"] assert actual_detection # as many calls as there are sources listed in the sources.json assert nb_detections == len(all_sources["sources"]) assert actual_detection == { - "revision": hash_to_hex(old_revision["id"]), + "revision": hash_to_hex(old_revision.id), "reason": "'integrity'", - "known_artifact": old_revision["metadata"], + "known_artifact": old_revision.metadata, } diff --git a/swh/loader/package/npm/tests/test_npm.py b/swh/loader/package/npm/tests/test_npm.py index 78b770c..d982392 100644 --- a/swh/loader/package/npm/tests/test_npm.py +++ b/swh/loader/package/npm/tests/test_npm.py @@ -1,708 +1,707 @@ # Copyright (C) 2019-2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information import json import os import pytest from swh.model.hashutil import hash_to_bytes, hash_to_hex from swh.model.identifiers import SWHID from swh.model.model import ( MetadataAuthority, MetadataAuthorityType, MetadataFetcher, MetadataTargetType, Person, RawExtrinsicMetadata, Snapshot, SnapshotBranch, TargetType, ) from swh.storage.interface import PagedResult from swh.loader.package import __version__ from swh.loader.package.npm.loader import ( _author_str, NpmLoader, extract_npm_package_author, artifact_to_revision_id, ) from swh.loader.package.tests.common import check_metadata_paths from swh.loader.tests import ( assert_last_visit_matches, check_snapshot, get_stats, ) @pytest.fixture def org_api_info(datadir) -> bytes: with open(os.path.join(datadir, "https_replicate.npmjs.com", "org"), "rb",) as f: return f.read() def test_npm_author_str(): for author, expected_author in [ ("author", "author"), ( ["Al from quantum leap", "hal from 2001 space odyssey"], "Al from quantum leap", ), ([], ""), ({"name": "groot", "email": "groot@galaxy.org",}, "groot "), ({"name": "somebody",}, "somebody"), ({"email": "no@one.org"}, " "), # note first elt is an extra blank ({"name": "no one", "email": None,}, "no one"), ({"email": None,}, ""), ({"name": None}, ""), ({"name": None, "email": None,}, ""), ({}, ""), (None, None), ({"name": []}, "",), ( {"name": ["Susan McSween", "William H. Bonney", "Doc Scurlock",]}, "Susan McSween", ), (None, None), ]: assert _author_str(author) == expected_author def test_extract_npm_package_author(datadir): package_metadata_filepath = os.path.join( datadir, "https_replicate.npmjs.com", "org_visit1" ) with open(package_metadata_filepath) as json_file: package_metadata = json.load(json_file) extract_npm_package_author(package_metadata["versions"]["0.0.2"]) == Person( fullname=b"mooz ", name=b"mooz", email=b"stillpedant@gmail.com", ) assert extract_npm_package_author(package_metadata["versions"]["0.0.3"]) == Person( fullname=b"Masafumi Oyamada ", name=b"Masafumi Oyamada", email=b"stillpedant@gmail.com", ) package_json = json.loads( """ { "name": "highlightjs-line-numbers.js", "version": "2.7.0", "description": "Highlight.js line numbers plugin.", "main": "src/highlightjs-line-numbers.js", "dependencies": {}, "devDependencies": { "gulp": "^4.0.0", "gulp-rename": "^1.4.0", "gulp-replace": "^0.6.1", "gulp-uglify": "^1.2.0" }, "repository": { "type": "git", "url": "https://github.com/wcoder/highlightjs-line-numbers.js.git" }, "author": "Yauheni Pakala ", "license": "MIT", "bugs": { "url": "https://github.com/wcoder/highlightjs-line-numbers.js/issues" }, "homepage": "http://wcoder.github.io/highlightjs-line-numbers.js/" }""" ) assert extract_npm_package_author(package_json) == Person( fullname=b"Yauheni Pakala ", name=b"Yauheni Pakala", email=b"evgeniy.pakalo@gmail.com", ) package_json = json.loads( """ { "name": "3-way-diff", "version": "0.0.1", "description": "3-way diffing of JavaScript objects", "main": "index.js", "authors": [ { "name": "Shawn Walsh", "url": "https://github.com/shawnpwalsh" }, { "name": "Markham F Rollins IV", "url": "https://github.com/mrollinsiv" } ], "keywords": [ "3-way diff", "3 way diff", "three-way diff", "three way diff" ], "devDependencies": { "babel-core": "^6.20.0", "babel-preset-es2015": "^6.18.0", "mocha": "^3.0.2" }, "dependencies": { "lodash": "^4.15.0" } }""" ) assert extract_npm_package_author(package_json) == Person( fullname=b"Shawn Walsh", name=b"Shawn Walsh", email=None ) package_json = json.loads( """ { "name": "yfe-ynpm", "version": "1.0.0", "homepage": "http://gitlab.ywwl.com/yfe/yfe-ynpm", "repository": { "type": "git", "url": "git@gitlab.ywwl.com:yfe/yfe-ynpm.git" }, "author": [ "fengmk2 (https://fengmk2.com)", "xufuzi (https://7993.org)" ], "license": "MIT" }""" ) assert extract_npm_package_author(package_json) == Person( fullname=b"fengmk2 (https://fengmk2.com)", name=b"fengmk2", email=b"fengmk2@gmail.com", ) package_json = json.loads( """ { "name": "umi-plugin-whale", "version": "0.0.8", "description": "Internal contract component", "authors": { "name": "xiaohuoni", "email": "448627663@qq.com" }, "repository": "alitajs/whale", "devDependencies": { "np": "^3.0.4", "umi-tools": "*" }, "license": "MIT" }""" ) assert extract_npm_package_author(package_json) == Person( fullname=b"xiaohuoni <448627663@qq.com>", name=b"xiaohuoni", email=b"448627663@qq.com", ) package_json_no_authors = json.loads( """{ "authors": null, "license": "MIT" }""" ) assert extract_npm_package_author(package_json_no_authors) == Person( fullname=b"", name=None, email=None ) def normalize_hashes(hashes): if isinstance(hashes, str): return hash_to_bytes(hashes) if isinstance(hashes, list): return [hash_to_bytes(x) for x in hashes] return {hash_to_bytes(k): hash_to_bytes(v) for k, v in hashes.items()} _expected_new_contents_first_visit = normalize_hashes( [ "4ce3058e16ab3d7e077f65aabf855c34895bf17c", "858c3ceee84c8311adc808f8cdb30d233ddc9d18", "0fa33b4f5a4e0496da6843a38ff1af8b61541996", "85a410f8ef8eb8920f2c384a9555566ad4a2e21b", "9163ac8025923d5a45aaac482262893955c9b37b", "692cf623b8dd2c5df2c2998fd95ae4ec99882fb4", "18c03aac6d3e910efb20039c15d70ab5e0297101", "41265c42446aac17ca769e67d1704f99e5a1394d", "783ff33f5882813dca9239452c4a7cadd4dba778", "b029cfb85107aee4590c2434a3329bfcf36f8fa1", "112d1900b4c2e3e9351050d1b542c9744f9793f3", "5439bbc4bd9a996f1a38244e6892b71850bc98fd", "d83097a2f994b503185adf4e719d154123150159", "d0939b4898e83090ee55fd9d8a60e312cfadfbaf", "b3523a26f7147e4af40d9d462adaae6d49eda13e", "cd065fb435d6fb204a8871bcd623d0d0e673088c", "2854a40855ad839a54f4b08f5cff0cf52fca4399", "b8a53bbaac34ebb8c6169d11a4b9f13b05c583fe", "0f73d56e1cf480bded8a1ecf20ec6fc53c574713", "0d9882b2dfafdce31f4e77fe307d41a44a74cefe", "585fc5caab9ead178a327d3660d35851db713df1", "e8cd41a48d79101977e3036a87aeb1aac730686f", "5414efaef33cceb9f3c9eb5c4cc1682cd62d14f7", "9c3cc2763bf9e9e37067d3607302c4776502df98", "3649a68410e354c83cd4a38b66bd314de4c8f5c9", "e96ed0c091de1ebdf587104eaf63400d1974a1fe", "078ca03d2f99e4e6eab16f7b75fbb7afb699c86c", "38de737da99514de6559ff163c988198bc91367a", ] ) _expected_new_directories_first_visit = normalize_hashes( [ "3370d20d6f96dc1c9e50f083e2134881db110f4f", "42753c0c2ab00c4501b552ac4671c68f3cf5aece", "d7895533ef5edbcffdea3f057d9fef3a1ef845ce", "80579be563e2ef3e385226fe7a3f079b377f142c", "3b0ddc6a9e58b4b53c222da4e27b280b6cda591c", "bcad03ce58ac136f26f000990fc9064e559fe1c0", "5fc7e82a1bc72e074665c6078c6d3fad2f13d7ca", "e3cd26beba9b1e02f6762ef54bd9ac80cc5f25fd", "584b5b4b6cf7f038095e820b99386a9c232de931", "184c8d6d0d242f2b1792ef9d3bf396a5434b7f7a", "bb5f4ee143c970367eb409f2e4c1104898048b9d", "1b95491047add1103db0dfdfa84a9735dcb11e88", "a00c6de13471a2d66e64aca140ddb21ef5521e62", "5ce6c1cd5cda2d546db513aaad8c72a44c7771e2", "c337091e349b6ac10d38a49cdf8c2401ef9bb0f2", "202fafcd7c0f8230e89d5496ad7f44ab12b807bf", "775cc516543be86c15c1dc172f49c0d4e6e78235", "ff3d1ead85a14f891e8b3fa3a89de39db1b8de2e", ] ) _expected_new_revisions_first_visit = normalize_hashes( { "d8a1c7474d2956ac598a19f0f27d52f7015f117e": ( "42753c0c2ab00c4501b552ac4671c68f3cf5aece" ), "5f9eb78af37ffd12949f235e86fac04898f9f72a": ( "3370d20d6f96dc1c9e50f083e2134881db110f4f" ), "ba019b192bdb94bd0b5bd68b3a5f92b5acc2239a": ( "d7895533ef5edbcffdea3f057d9fef3a1ef845ce" ), } ) def package_url(package): return "https://www.npmjs.com/package/%s" % package def package_metadata_url(package): return "https://replicate.npmjs.com/%s/" % package def test_revision_metadata_structure(swh_config, requests_mock_datadir): package = "org" loader = NpmLoader(package_url(package)) actual_load_status = loader.load() assert actual_load_status["status"] == "eventful" assert actual_load_status["snapshot_id"] is not None expected_revision_id = hash_to_bytes("d8a1c7474d2956ac598a19f0f27d52f7015f117e") - revision = list(loader.storage.revision_get([expected_revision_id]))[0] - + revision = loader.storage.revision_get([expected_revision_id])[0] assert revision is not None check_metadata_paths( - revision["metadata"], + revision.metadata, paths=[ ("intrinsic.tool", str), ("intrinsic.raw", dict), ("extrinsic.provider", str), ("extrinsic.when", str), ("extrinsic.raw", dict), ("original_artifact", list), ], ) - for original_artifact in revision["metadata"]["original_artifact"]: + for original_artifact in revision.metadata["original_artifact"]: check_metadata_paths( original_artifact, paths=[("filename", str), ("length", int), ("checksums", dict),], ) def test_npm_loader_first_visit(swh_config, requests_mock_datadir, org_api_info): package = "org" url = package_url(package) loader = NpmLoader(url) actual_load_status = loader.load() expected_snapshot_id = hash_to_bytes("d0587e1195aed5a8800411a008f2f2d627f18e2d") assert actual_load_status == { "status": "eventful", "snapshot_id": expected_snapshot_id.hex(), } assert_last_visit_matches( loader.storage, url, status="full", type="npm", snapshot=expected_snapshot_id ) stats = get_stats(loader.storage) assert { "content": len(_expected_new_contents_first_visit), "directory": len(_expected_new_directories_first_visit), "origin": 1, "origin_visit": 1, "person": 2, "release": 0, "revision": len(_expected_new_revisions_first_visit), "skipped_content": 0, "snapshot": 1, } == stats contents = loader.storage.content_get(_expected_new_contents_first_visit) count = sum(0 if content is None else 1 for content in contents) assert count == len(_expected_new_contents_first_visit) assert ( list(loader.storage.directory_missing(_expected_new_directories_first_visit)) == [] ) assert ( list(loader.storage.revision_missing(_expected_new_revisions_first_visit)) == [] ) expected_snapshot = Snapshot( id=expected_snapshot_id, branches={ b"HEAD": SnapshotBranch( target=b"releases/0.0.4", target_type=TargetType.ALIAS ), b"releases/0.0.2": SnapshotBranch( target=hash_to_bytes("d8a1c7474d2956ac598a19f0f27d52f7015f117e"), target_type=TargetType.REVISION, ), b"releases/0.0.3": SnapshotBranch( target=hash_to_bytes("5f9eb78af37ffd12949f235e86fac04898f9f72a"), target_type=TargetType.REVISION, ), b"releases/0.0.4": SnapshotBranch( target=hash_to_bytes("ba019b192bdb94bd0b5bd68b3a5f92b5acc2239a"), target_type=TargetType.REVISION, ), }, ) check_snapshot(expected_snapshot, loader.storage) snapshot_swhid = SWHID( object_type="snapshot", object_id=hash_to_hex(expected_snapshot_id) ) metadata_authority = MetadataAuthority( type=MetadataAuthorityType.FORGE, url="https://npmjs.com/", ) expected_metadata = [ RawExtrinsicMetadata( type=MetadataTargetType.SNAPSHOT, id=snapshot_swhid, authority=metadata_authority, fetcher=MetadataFetcher( name="swh.loader.package.npm.loader.NpmLoader", version=__version__, ), discovery_date=loader.visit_date, format="replicate-npm-package-json", metadata=org_api_info, origin="https://www.npmjs.com/package/org", ) ] assert loader.storage.raw_extrinsic_metadata_get( type=MetadataTargetType.SNAPSHOT, id=snapshot_swhid, authority=metadata_authority, ) == PagedResult(next_page_token=None, results=expected_metadata,) def test_npm_loader_incremental_visit(swh_config, requests_mock_datadir_visits): package = "org" url = package_url(package) loader = NpmLoader(url) expected_snapshot_id = hash_to_bytes("d0587e1195aed5a8800411a008f2f2d627f18e2d") actual_load_status = loader.load() assert actual_load_status == { "status": "eventful", "snapshot_id": expected_snapshot_id.hex(), } assert_last_visit_matches( loader.storage, url, status="full", type="npm", snapshot=expected_snapshot_id ) stats = get_stats(loader.storage) assert { "content": len(_expected_new_contents_first_visit), "directory": len(_expected_new_directories_first_visit), "origin": 1, "origin_visit": 1, "person": 2, "release": 0, "revision": len(_expected_new_revisions_first_visit), "skipped_content": 0, "snapshot": 1, } == stats # reset loader internal state del loader._cached_info del loader._cached__raw_info actual_load_status2 = loader.load() assert actual_load_status2["status"] == "eventful" snap_id2 = actual_load_status2["snapshot_id"] assert snap_id2 is not None assert snap_id2 != actual_load_status["snapshot_id"] assert_last_visit_matches(loader.storage, url, status="full", type="npm") stats = get_stats(loader.storage) assert { # 3 new releases artifacts "content": len(_expected_new_contents_first_visit) + 14, "directory": len(_expected_new_directories_first_visit) + 15, "origin": 1, "origin_visit": 2, "person": 2, "release": 0, "revision": len(_expected_new_revisions_first_visit) + 3, "skipped_content": 0, "snapshot": 2, } == stats urls = [ m.url for m in requests_mock_datadir_visits.request_history if m.url.startswith("https://registry.npmjs.org") ] assert len(urls) == len(set(urls)) # we visited each artifact once across @pytest.mark.usefixtures("requests_mock_datadir") def test_npm_loader_version_divergence(swh_config): package = "@aller_shared" url = package_url(package) loader = NpmLoader(url) actual_load_status = loader.load() expected_snapshot_id = hash_to_bytes("b11ebac8c9d0c9e5063a2df693a18e3aba4b2f92") assert actual_load_status == { "status": "eventful", "snapshot_id": expected_snapshot_id.hex(), } assert_last_visit_matches( loader.storage, url, status="full", type="npm", snapshot=expected_snapshot_id ) stats = get_stats(loader.storage) assert { # 1 new releases artifacts "content": 534, "directory": 153, "origin": 1, "origin_visit": 1, "person": 1, "release": 0, "revision": 2, "skipped_content": 0, "snapshot": 1, } == stats expected_snapshot = Snapshot( id=expected_snapshot_id, branches={ b"HEAD": SnapshotBranch( target_type=TargetType.ALIAS, target=b"releases/0.1.0" ), b"releases/0.1.0": SnapshotBranch( target_type=TargetType.REVISION, target=hash_to_bytes("845673bfe8cbd31b1eaf757745a964137e6f9116"), ), b"releases/0.1.1-alpha.14": SnapshotBranch( target_type=TargetType.REVISION, target=hash_to_bytes("05181c12cd8c22035dd31155656826b85745da37"), ), }, ) check_snapshot(expected_snapshot, loader.storage) def test_npm_artifact_to_revision_id_none(): """Current loader version should stop soon if nothing can be found """ class artifact_metadata: shasum = "05181c12cd8c22035dd31155656826b85745da37" known_artifacts = { "b11ebac8c9d0c9e5063a2df693a18e3aba4b2f92": {}, } assert artifact_to_revision_id(known_artifacts, artifact_metadata) is None def test_npm_artifact_to_revision_id_old_loader_version(): """Current loader version should solve old metadata scheme """ class artifact_metadata: shasum = "05181c12cd8c22035dd31155656826b85745da37" known_artifacts = { hash_to_bytes("b11ebac8c9d0c9e5063a2df693a18e3aba4b2f92"): { "package_source": {"sha1": "something-wrong"} }, hash_to_bytes("845673bfe8cbd31b1eaf757745a964137e6f9116"): { "package_source": {"sha1": "05181c12cd8c22035dd31155656826b85745da37",} }, } assert artifact_to_revision_id(known_artifacts, artifact_metadata) == hash_to_bytes( "845673bfe8cbd31b1eaf757745a964137e6f9116" ) def test_npm_artifact_to_revision_id_current_loader_version(): """Current loader version should be able to solve current metadata scheme """ class artifact_metadata: shasum = "05181c12cd8c22035dd31155656826b85745da37" known_artifacts = { hash_to_bytes("b11ebac8c9d0c9e5063a2df693a18e3aba4b2f92"): { "original_artifact": [ {"checksums": {"sha1": "05181c12cd8c22035dd31155656826b85745da37"},} ], }, hash_to_bytes("845673bfe8cbd31b1eaf757745a964137e6f9116"): { "original_artifact": [{"checksums": {"sha1": "something-wrong"},}], }, } assert artifact_to_revision_id(known_artifacts, artifact_metadata) == hash_to_bytes( "b11ebac8c9d0c9e5063a2df693a18e3aba4b2f92" ) def test_npm_artifact_with_no_intrinsic_metadata(swh_config, requests_mock_datadir): """Skip artifact with no intrinsic metadata during ingestion """ package = "nativescript-telerik-analytics" url = package_url(package) loader = NpmLoader(url) actual_load_status = loader.load() # no branch as one artifact without any intrinsic metadata expected_snapshot = Snapshot( id=hash_to_bytes("1a8893e6a86f444e8be8e7bda6cb34fb1735a00e"), branches={}, ) assert actual_load_status == { "status": "eventful", "snapshot_id": expected_snapshot.id.hex(), } check_snapshot(expected_snapshot, loader.storage) assert_last_visit_matches( loader.storage, url, status="full", type="npm", snapshot=expected_snapshot.id ) def test_npm_artifact_with_no_upload_time(swh_config, requests_mock_datadir): """With no time upload, artifact is skipped """ package = "jammit-no-time" url = package_url(package) loader = NpmLoader(url) actual_load_status = loader.load() # no branch as one artifact without any intrinsic metadata expected_snapshot = Snapshot( id=hash_to_bytes("1a8893e6a86f444e8be8e7bda6cb34fb1735a00e"), branches={}, ) assert actual_load_status == { "status": "uneventful", "snapshot_id": expected_snapshot.id.hex(), } check_snapshot(expected_snapshot, loader.storage) assert_last_visit_matches( loader.storage, url, status="partial", type="npm", snapshot=expected_snapshot.id ) def test_npm_artifact_use_mtime_if_no_time(swh_config, requests_mock_datadir): """With no time upload, artifact is skipped """ package = "jammit-express" url = package_url(package) loader = NpmLoader(url) actual_load_status = loader.load() expected_snapshot_id = hash_to_bytes("d6e08e19159f77983242877c373c75222d5ae9dd") assert actual_load_status == { "status": "eventful", "snapshot_id": expected_snapshot_id.hex(), } # artifact is used expected_snapshot = Snapshot( id=expected_snapshot_id, branches={ b"HEAD": SnapshotBranch( target_type=TargetType.ALIAS, target=b"releases/0.0.1" ), b"releases/0.0.1": SnapshotBranch( target_type=TargetType.REVISION, target=hash_to_bytes("9e4dd2b40d1b46b70917c0949aa2195c823a648e"), ), }, ) check_snapshot(expected_snapshot, loader.storage) assert_last_visit_matches( loader.storage, url, status="full", type="npm", snapshot=expected_snapshot.id ) def test_npm_no_artifact(swh_config, requests_mock_datadir): """If no artifacts at all is found for origin, the visit fails completely """ package = "catify" url = package_url(package) loader = NpmLoader(url) actual_load_status = loader.load() assert actual_load_status == { "status": "failed", } assert_last_visit_matches(loader.storage, url, status="partial", type="npm") diff --git a/swh/loader/package/pypi/tests/test_pypi.py b/swh/loader/package/pypi/tests/test_pypi.py index 71f0363..a3a3a69 100644 --- a/swh/loader/package/pypi/tests/test_pypi.py +++ b/swh/loader/package/pypi/tests/test_pypi.py @@ -1,886 +1,885 @@ # Copyright (C) 2019-2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information import os from os import path import pytest from unittest.mock import patch from swh.core.tarball import uncompress from swh.core.pytest_plugin import requests_mock_datadir_factory from swh.model.hashutil import hash_to_bytes, hash_to_hex from swh.model.identifiers import SWHID from swh.model.model import ( MetadataAuthority, MetadataAuthorityType, MetadataFetcher, MetadataTargetType, Person, RawExtrinsicMetadata, Snapshot, SnapshotBranch, TargetType, ) from swh.storage.interface import PagedResult from swh.loader.package import __version__ from swh.loader.package.pypi.loader import ( PyPILoader, pypi_api_url, author, extract_intrinsic_metadata, artifact_to_revision_id, ) from swh.loader.package.tests.common import check_metadata_paths from swh.loader.tests import ( assert_last_visit_matches, check_snapshot, get_stats, ) @pytest.fixture def _0805nexter_api_info(datadir) -> bytes: with open( os.path.join(datadir, "https_pypi.org", "pypi_0805nexter_json"), "rb", ) as f: return f.read() def test_author_basic(): data = { "author": "i-am-groot", "author_email": "iam@groot.org", } actual_author = author(data) expected_author = Person( fullname=b"i-am-groot ", name=b"i-am-groot", email=b"iam@groot.org", ) assert actual_author == expected_author def test_author_empty_email(): data = { "author": "i-am-groot", "author_email": "", } actual_author = author(data) expected_author = Person(fullname=b"i-am-groot", name=b"i-am-groot", email=b"",) assert actual_author == expected_author def test_author_empty_name(): data = { "author": "", "author_email": "iam@groot.org", } actual_author = author(data) expected_author = Person( fullname=b" ", name=b"", email=b"iam@groot.org", ) assert actual_author == expected_author def test_author_malformed(): data = { "author": "['pierre', 'paul', 'jacques']", "author_email": None, } actual_author = author(data) expected_author = Person( fullname=b"['pierre', 'paul', 'jacques']", name=b"['pierre', 'paul', 'jacques']", email=None, ) assert actual_author == expected_author def test_author_malformed_2(): data = { "author": "[marie, jeanne]", "author_email": "[marie@some, jeanne@thing]", } actual_author = author(data) expected_author = Person( fullname=b"[marie, jeanne] <[marie@some, jeanne@thing]>", name=b"[marie, jeanne]", email=b"[marie@some, jeanne@thing]", ) assert actual_author == expected_author def test_author_malformed_3(): data = { "author": "[marie, jeanne, pierre]", "author_email": "[marie@somewhere.org, jeanne@somewhere.org]", } actual_author = author(data) expected_author = Person( fullname=( b"[marie, jeanne, pierre] " b"<[marie@somewhere.org, jeanne@somewhere.org]>" ), name=b"[marie, jeanne, pierre]", email=b"[marie@somewhere.org, jeanne@somewhere.org]", ) actual_author == expected_author # configuration error # def test_badly_configured_loader_raise(monkeypatch): """Badly configured loader should raise""" monkeypatch.delenv("SWH_CONFIG_FILENAME", raising=False) with pytest.raises(ValueError) as e: PyPILoader(url="some-url") assert "Misconfiguration" in e.value.args[0] def test_pypi_api_url(): """Compute pypi api url from the pypi project url should be ok""" url = pypi_api_url("https://pypi.org/project/requests") assert url == "https://pypi.org/pypi/requests/json" def test_pypi_api_url_with_slash(): """Compute pypi api url from the pypi project url should be ok""" url = pypi_api_url("https://pypi.org/project/requests/") assert url == "https://pypi.org/pypi/requests/json" @pytest.mark.fs def test_extract_intrinsic_metadata(tmp_path, datadir): """Parsing existing archive's PKG-INFO should yield results""" uncompressed_archive_path = str(tmp_path) archive_path = path.join( datadir, "https_files.pythonhosted.org", "0805nexter-1.1.0.zip" ) uncompress(archive_path, dest=uncompressed_archive_path) actual_metadata = extract_intrinsic_metadata(uncompressed_archive_path) expected_metadata = { "metadata_version": "1.0", "name": "0805nexter", "version": "1.1.0", "summary": "a simple printer of nested lest", "home_page": "http://www.hp.com", "author": "hgtkpython", "author_email": "2868989685@qq.com", "platforms": ["UNKNOWN"], } assert actual_metadata == expected_metadata @pytest.mark.fs def test_extract_intrinsic_metadata_failures(tmp_path): """Parsing inexistent path/archive/PKG-INFO yield None""" tmp_path = str(tmp_path) # py3.5 work around (PosixPath issue) # inexistent first level path assert extract_intrinsic_metadata("/something-inexistent") == {} # inexistent second level path (as expected by pypi archives) assert extract_intrinsic_metadata(tmp_path) == {} # inexistent PKG-INFO within second level path existing_path_no_pkginfo = path.join(tmp_path, "something") os.mkdir(existing_path_no_pkginfo) assert extract_intrinsic_metadata(tmp_path) == {} # LOADER SCENARIO # # "edge" cases (for the same origin) # # no release artifact: # {visit full, status: uneventful, no contents, etc...} requests_mock_datadir_missing_all = requests_mock_datadir_factory( ignore_urls=[ "https://files.pythonhosted.org/packages/ec/65/c0116953c9a3f47de89e71964d6c7b0c783b01f29fa3390584dbf3046b4d/0805nexter-1.1.0.zip", # noqa "https://files.pythonhosted.org/packages/c4/a0/4562cda161dc4ecbbe9e2a11eb365400c0461845c5be70d73869786809c4/0805nexter-1.2.0.zip", # noqa ] ) def test_no_release_artifact(swh_config, 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(url) actual_load_status = loader.load() assert actual_load_status["status"] == "uneventful" assert actual_load_status["snapshot_id"] is not None stats = get_stats(loader.storage) assert { "content": 0, "directory": 0, "origin": 1, "origin_visit": 1, "person": 0, "release": 0, "revision": 0, "skipped_content": 0, "snapshot": 1, } == stats assert_last_visit_matches(loader.storage, url, status="partial", type="pypi") # problem during loading: # {visit: partial, status: uneventful, no snapshot} def test_release_with_traceback(swh_config, 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(url) actual_load_status = loader.load() assert actual_load_status == {"status": "failed"} stats = get_stats(loader.storage) assert { "content": 0, "directory": 0, "origin": 1, "origin_visit": 1, "person": 0, "release": 0, "revision": 0, "skipped_content": 0, "snapshot": 0, } == stats assert_last_visit_matches(loader.storage, url, status="partial", type="pypi") # problem during loading: failure early enough in between swh contents... # some contents (contents, directories, etc...) have been written in storage # {visit: partial, status: eventful, no snapshot} # problem during loading: failure late enough we can have snapshots (some # revisions are written in storage already) # {visit: partial, status: eventful, snapshot} # "normal" cases (for the same origin) # requests_mock_datadir_missing_one = requests_mock_datadir_factory( ignore_urls=[ "https://files.pythonhosted.org/packages/ec/65/c0116953c9a3f47de89e71964d6c7b0c783b01f29fa3390584dbf3046b4d/0805nexter-1.1.0.zip", # noqa ] ) # some missing release artifacts: # {visit partial, status: eventful, 1 snapshot} def test_revision_metadata_structure(swh_config, requests_mock_datadir): url = "https://pypi.org/project/0805nexter" loader = PyPILoader(url) actual_load_status = loader.load() assert actual_load_status["status"] == "eventful" assert actual_load_status["snapshot_id"] is not None expected_revision_id = hash_to_bytes("e445da4da22b31bfebb6ffc4383dbf839a074d21") - revision = list(loader.storage.revision_get([expected_revision_id]))[0] - + revision = loader.storage.revision_get([expected_revision_id])[0] assert revision is not None check_metadata_paths( - revision["metadata"], + revision.metadata, paths=[ ("intrinsic.tool", str), ("intrinsic.raw", dict), ("extrinsic.provider", str), ("extrinsic.when", str), ("extrinsic.raw", dict), ("original_artifact", list), ], ) - for original_artifact in revision["metadata"]["original_artifact"]: + for original_artifact in revision.metadata["original_artifact"]: check_metadata_paths( original_artifact, paths=[("filename", str), ("length", int), ("checksums", dict),], ) def test_snapshot_metadata(swh_config, requests_mock_datadir, _0805nexter_api_info): url = "https://pypi.org/project/0805nexter" loader = PyPILoader(url) actual_load_status = loader.load() assert actual_load_status["status"] == "eventful" assert actual_load_status["snapshot_id"] is not None snapshot_swhid = SWHID( object_type="snapshot", object_id=hash_to_hex(actual_load_status["snapshot_id"]) ) metadata_authority = MetadataAuthority( type=MetadataAuthorityType.FORGE, url="https://pypi.org/", ) expected_metadata = [ RawExtrinsicMetadata( type=MetadataTargetType.SNAPSHOT, id=snapshot_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=_0805nexter_api_info, origin=url, ) ] assert loader.storage.raw_extrinsic_metadata_get( type=MetadataTargetType.SNAPSHOT, id=snapshot_swhid, authority=metadata_authority, ) == PagedResult(next_page_token=None, results=expected_metadata,) def test_visit_with_missing_artifact(swh_config, 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(url) actual_load_status = loader.load() expected_snapshot_id = hash_to_bytes("dd0e4201a232b1c104433741dbf45895b8ac9355") assert actual_load_status == { "status": "eventful", "snapshot_id": expected_snapshot_id.hex(), } stats = get_stats(loader.storage) assert { "content": 3, "directory": 2, "origin": 1, "origin_visit": 1, "person": 1, "release": 0, "revision": 1, "skipped_content": 0, "snapshot": 1, } == stats expected_contents = map( hash_to_bytes, [ "405859113963cb7a797642b45f171d6360425d16", "e5686aa568fdb1d19d7f1329267082fe40482d31", "83ecf6ec1114fd260ca7a833a2d165e71258c338", ], ) assert list(loader.storage.content_missing_per_sha1(expected_contents)) == [] expected_dirs = map( hash_to_bytes, [ "b178b66bd22383d5f16f4f5c923d39ca798861b4", "c3a58f8b57433a4b56caaa5033ae2e0931405338", ], ) assert list(loader.storage.directory_missing(expected_dirs)) == [] # {revision hash: directory hash} expected_revs = { hash_to_bytes("e445da4da22b31bfebb6ffc4383dbf839a074d21"): hash_to_bytes( "b178b66bd22383d5f16f4f5c923d39ca798861b4" ), # noqa } assert list(loader.storage.revision_missing(expected_revs)) == [] expected_snapshot = Snapshot( id=hash_to_bytes(expected_snapshot_id), branches={ b"releases/1.2.0": SnapshotBranch( target=hash_to_bytes("e445da4da22b31bfebb6ffc4383dbf839a074d21"), target_type=TargetType.REVISION, ), b"HEAD": SnapshotBranch( target=b"releases/1.2.0", target_type=TargetType.ALIAS, ), }, ) check_snapshot(expected_snapshot, storage=loader.storage) assert_last_visit_matches( loader.storage, url, status="partial", type="pypi", snapshot=expected_snapshot_id, ) def test_visit_with_1_release_artifact(swh_config, requests_mock_datadir): """With no prior visit, load a pypi project ends up with 1 snapshot """ url = "https://pypi.org/project/0805nexter" loader = PyPILoader(url) actual_load_status = loader.load() expected_snapshot_id = hash_to_bytes("ba6e158ada75d0b3cfb209ffdf6daa4ed34a227a") assert actual_load_status == { "status": "eventful", "snapshot_id": expected_snapshot_id.hex(), } stats = get_stats(loader.storage) assert { "content": 6, "directory": 4, "origin": 1, "origin_visit": 1, "person": 1, "release": 0, "revision": 2, "skipped_content": 0, "snapshot": 1, } == stats expected_contents = map( hash_to_bytes, [ "a61e24cdfdab3bb7817f6be85d37a3e666b34566", "938c33483285fd8ad57f15497f538320df82aeb8", "a27576d60e08c94a05006d2e6d540c0fdb5f38c8", "405859113963cb7a797642b45f171d6360425d16", "e5686aa568fdb1d19d7f1329267082fe40482d31", "83ecf6ec1114fd260ca7a833a2d165e71258c338", ], ) assert list(loader.storage.content_missing_per_sha1(expected_contents)) == [] expected_dirs = map( hash_to_bytes, [ "05219ba38bc542d4345d5638af1ed56c7d43ca7d", "cf019eb456cf6f78d8c4674596f1c9a97ece8f44", "b178b66bd22383d5f16f4f5c923d39ca798861b4", "c3a58f8b57433a4b56caaa5033ae2e0931405338", ], ) assert list(loader.storage.directory_missing(expected_dirs)) == [] # {revision hash: directory hash} expected_revs = { hash_to_bytes("4c99891f93b81450385777235a37b5e966dd1571"): hash_to_bytes( "05219ba38bc542d4345d5638af1ed56c7d43ca7d" ), # noqa hash_to_bytes("e445da4da22b31bfebb6ffc4383dbf839a074d21"): hash_to_bytes( "b178b66bd22383d5f16f4f5c923d39ca798861b4" ), # noqa } assert list(loader.storage.revision_missing(expected_revs)) == [] expected_snapshot = Snapshot( id=expected_snapshot_id, branches={ b"releases/1.1.0": SnapshotBranch( target=hash_to_bytes("4c99891f93b81450385777235a37b5e966dd1571"), target_type=TargetType.REVISION, ), b"releases/1.2.0": SnapshotBranch( target=hash_to_bytes("e445da4da22b31bfebb6ffc4383dbf839a074d21"), target_type=TargetType.REVISION, ), b"HEAD": SnapshotBranch( target=b"releases/1.2.0", target_type=TargetType.ALIAS, ), }, ) check_snapshot(expected_snapshot, loader.storage) assert_last_visit_matches( loader.storage, url, status="full", type="pypi", snapshot=expected_snapshot_id ) def test_multiple_visits_with_no_change(swh_config, requests_mock_datadir): """Multiple visits with no changes results in 1 same snapshot """ url = "https://pypi.org/project/0805nexter" loader = PyPILoader(url) actual_load_status = loader.load() snapshot_id = hash_to_bytes("ba6e158ada75d0b3cfb209ffdf6daa4ed34a227a") assert actual_load_status == { "status": "eventful", "snapshot_id": snapshot_id.hex(), } assert_last_visit_matches( loader.storage, url, status="full", type="pypi", snapshot=snapshot_id ) stats = get_stats(loader.storage) assert { "content": 6, "directory": 4, "origin": 1, "origin_visit": 1, "person": 1, "release": 0, "revision": 2, "skipped_content": 0, "snapshot": 1, } == stats expected_snapshot = Snapshot( id=snapshot_id, branches={ b"releases/1.1.0": SnapshotBranch( target=hash_to_bytes("4c99891f93b81450385777235a37b5e966dd1571"), target_type=TargetType.REVISION, ), b"releases/1.2.0": SnapshotBranch( target=hash_to_bytes("e445da4da22b31bfebb6ffc4383dbf839a074d21"), target_type=TargetType.REVISION, ), b"HEAD": SnapshotBranch( target=b"releases/1.2.0", target_type=TargetType.ALIAS, ), }, ) check_snapshot(expected_snapshot, loader.storage) actual_load_status2 = loader.load() assert actual_load_status2 == { "status": "uneventful", "snapshot_id": actual_load_status2["snapshot_id"], } visit_status2 = assert_last_visit_matches( loader.storage, url, status="full", type="pypi" ) stats2 = get_stats(loader.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_incremental_visit(swh_config, requests_mock_datadir_visits): """With prior visit, 2nd load will result with a different snapshot """ url = "https://pypi.org/project/0805nexter" loader = PyPILoader(url) visit1_actual_load_status = loader.load() visit1_stats = get_stats(loader.storage) expected_snapshot_id = hash_to_bytes("ba6e158ada75d0b3cfb209ffdf6daa4ed34a227a") assert visit1_actual_load_status == { "status": "eventful", "snapshot_id": expected_snapshot_id.hex(), } assert_last_visit_matches( loader.storage, url, status="full", type="pypi", snapshot=expected_snapshot_id ) assert { "content": 6, "directory": 4, "origin": 1, "origin_visit": 1, "person": 1, "release": 0, "revision": 2, "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(loader.storage) assert visit2_actual_load_status["status"] == "eventful", visit2_actual_load_status expected_snapshot_id2 = hash_to_bytes("2e5149a7b0725d18231a37b342e9b7c4e121f283") assert visit2_actual_load_status == { "status": "eventful", "snapshot_id": expected_snapshot_id2.hex(), } assert_last_visit_matches( loader.storage, url, status="full", type="pypi", snapshot=expected_snapshot_id2 ) assert { "content": 6 + 1, # 1 more content "directory": 4 + 2, # 2 more directories "origin": 1, "origin_visit": 1 + 1, "person": 1, "release": 0, "revision": 2 + 1, # 1 more revision "skipped_content": 0, "snapshot": 1 + 1, # 1 more snapshot } == visit2_stats expected_contents = map( hash_to_bytes, [ "a61e24cdfdab3bb7817f6be85d37a3e666b34566", "938c33483285fd8ad57f15497f538320df82aeb8", "a27576d60e08c94a05006d2e6d540c0fdb5f38c8", "405859113963cb7a797642b45f171d6360425d16", "e5686aa568fdb1d19d7f1329267082fe40482d31", "83ecf6ec1114fd260ca7a833a2d165e71258c338", "92689fa2b7fb4d4fc6fb195bf73a50c87c030639", ], ) assert list(loader.storage.content_missing_per_sha1(expected_contents)) == [] expected_dirs = map( hash_to_bytes, [ "05219ba38bc542d4345d5638af1ed56c7d43ca7d", "cf019eb456cf6f78d8c4674596f1c9a97ece8f44", "b178b66bd22383d5f16f4f5c923d39ca798861b4", "c3a58f8b57433a4b56caaa5033ae2e0931405338", "e226e7e4ad03b4fc1403d69a18ebdd6f2edd2b3a", "52604d46843b898f5a43208045d09fcf8731631b", ], ) assert list(loader.storage.directory_missing(expected_dirs)) == [] # {revision hash: directory hash} expected_revs = { hash_to_bytes("4c99891f93b81450385777235a37b5e966dd1571"): hash_to_bytes( "05219ba38bc542d4345d5638af1ed56c7d43ca7d" ), # noqa hash_to_bytes("e445da4da22b31bfebb6ffc4383dbf839a074d21"): hash_to_bytes( "b178b66bd22383d5f16f4f5c923d39ca798861b4" ), # noqa hash_to_bytes("51247143b01445c9348afa9edfae31bf7c5d86b1"): hash_to_bytes( "e226e7e4ad03b4fc1403d69a18ebdd6f2edd2b3a" ), # noqa } assert list(loader.storage.revision_missing(expected_revs)) == [] expected_snapshot = Snapshot( id=expected_snapshot_id2, branches={ b"releases/1.1.0": SnapshotBranch( target=hash_to_bytes("4c99891f93b81450385777235a37b5e966dd1571"), target_type=TargetType.REVISION, ), b"releases/1.2.0": SnapshotBranch( target=hash_to_bytes("e445da4da22b31bfebb6ffc4383dbf839a074d21"), target_type=TargetType.REVISION, ), b"releases/1.3.0": SnapshotBranch( target=hash_to_bytes("51247143b01445c9348afa9edfae31bf7c5d86b1"), target_type=TargetType.REVISION, ), b"HEAD": SnapshotBranch( target=b"releases/1.3.0", target_type=TargetType.ALIAS, ), }, ) check_snapshot(expected_snapshot, loader.storage) assert_last_visit_matches( loader.storage, url, status="full", type="pypi", snapshot=expected_snapshot.id ) 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)) # release artifact, no new artifact # {visit full, status uneventful, same snapshot as before} # release artifact, old artifact with different checksums # {visit full, status full, new snapshot with shared history and some new # different history} # release with multiple sdist artifacts per pypi "version" # snapshot branch output is different def test_visit_1_release_with_2_artifacts(swh_config, requests_mock_datadir): """With no prior visit, load a pypi project ends up with 1 snapshot """ url = "https://pypi.org/project/nexter" loader = PyPILoader(url) actual_load_status = loader.load() expected_snapshot_id = hash_to_bytes("a27e638a4dad6fbfa273c6ebec1c4bf320fb84c6") assert actual_load_status == { "status": "eventful", "snapshot_id": expected_snapshot_id.hex(), } expected_snapshot = Snapshot( id=expected_snapshot_id, branches={ b"releases/1.1.0/nexter-1.1.0.zip": SnapshotBranch( target=hash_to_bytes("4c99891f93b81450385777235a37b5e966dd1571"), target_type=TargetType.REVISION, ), b"releases/1.1.0/nexter-1.1.0.tar.gz": SnapshotBranch( target=hash_to_bytes("0bf88f5760cca7665d0af4d6575d9301134fe11a"), target_type=TargetType.REVISION, ), }, ) check_snapshot(expected_snapshot, loader.storage) assert_last_visit_matches( loader.storage, url, status="full", type="pypi", snapshot=expected_snapshot.id ) def test_pypi_artifact_to_revision_id_none(): """Current loader version should stop soon if nothing can be found """ class artifact_metadata: sha256 = "6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec" assert artifact_to_revision_id({}, artifact_metadata) is None known_artifacts = { "b11ebac8c9d0c9e5063a2df693a18e3aba4b2f92": { "original_artifact": {"sha256": "something-irrelevant",}, }, } assert artifact_to_revision_id(known_artifacts, artifact_metadata) is None def test_pypi_artifact_to_revision_id_old_loader_version(): """Current loader version should solve old metadata scheme """ class artifact_metadata: sha256 = "6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec" known_artifacts = { hash_to_bytes("b11ebac8c9d0c9e5063a2df693a18e3aba4b2f92"): { "original_artifact": {"sha256": "something-wrong",}, }, hash_to_bytes("845673bfe8cbd31b1eaf757745a964137e6f9116"): { "original_artifact": { "sha256": "6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec", # noqa }, }, } assert artifact_to_revision_id(known_artifacts, artifact_metadata) == hash_to_bytes( "845673bfe8cbd31b1eaf757745a964137e6f9116" ) def test_pypi_artifact_to_revision_id_current_loader_version(): """Current loader version should be able to solve current metadata scheme """ class artifact_metadata: sha256 = "6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec" known_artifacts = { hash_to_bytes("b11ebac8c9d0c9e5063a2df693a18e3aba4b2f92"): { "original_artifact": [ { "checksums": { "sha256": "6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec", # noqa }, } ], }, hash_to_bytes("845673bfe8cbd31b1eaf757745a964137e6f9116"): { "original_artifact": [{"checksums": {"sha256": "something-wrong"},}], }, } assert artifact_to_revision_id(known_artifacts, artifact_metadata) == hash_to_bytes( "b11ebac8c9d0c9e5063a2df693a18e3aba4b2f92" ) def test_pypi_artifact_with_no_intrinsic_metadata(swh_config, requests_mock_datadir): """Skip artifact with no intrinsic metadata during ingestion """ url = "https://pypi.org/project/upymenu" loader = PyPILoader(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={}) check_snapshot(expected_snapshot, loader.storage) assert_last_visit_matches( loader.storage, url, status="full", type="pypi", snapshot=expected_snapshot.id ) diff --git a/swh/loader/tests/__init__.py b/swh/loader/tests/__init__.py index e91b290..dbaf3f3 100644 --- a/swh/loader/tests/__init__.py +++ b/swh/loader/tests/__init__.py @@ -1,263 +1,263 @@ # Copyright (C) 2018-2020 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information import os import subprocess from collections import defaultdict from pathlib import PosixPath from typing import Dict, Iterable, List, Optional, Tuple, Union from swh.model.model import OriginVisitStatus, Snapshot, TargetType from swh.model.hashutil import hash_to_bytes from swh.storage.interface import StorageInterface from swh.storage.algos.origin import origin_get_latest_visit_status from swh.storage.algos.snapshot import snapshot_get_all_branches def assert_last_visit_matches( storage, url: str, status: str, type: Optional[str] = None, snapshot: Optional[bytes] = None, ) -> OriginVisitStatus: """This retrieves the last visit and visit_status which are expected to exist. This also checks that the {visit|visit_status} have their respective properties correctly set. This returns the last visit_status for that given origin. Args: url: Origin url status: Check that the visit status has the given status type: Check that the returned visit has the given type snapshot: Check that the visit status points to the given snapshot Raises: AssertionError in case visit or visit status is not found, or any of the type, status and snapshot mismatch Returns: the visit status for further check during the remaining part of the test. """ visit_and_status = origin_get_latest_visit_status(storage, url) assert visit_and_status is not None, f"Origin {url} has no visits" visit, visit_status = visit_and_status if type: assert visit.type == type, f"Visit has type {visit.type} instead of {type}" assert ( visit_status.status == status ), f"Visit_status has status {visit_status.status} instead of {status}" if snapshot is not None: assert visit_status.snapshot is not None assert visit_status.snapshot == snapshot, ( f"Visit_status points to snapshot {visit_status.snapshot.hex()} " f"instead of {snapshot.hex()}" ) return visit_status def prepare_repository_from_archive( archive_path: str, filename: Optional[str] = None, tmp_path: Union[PosixPath, str] = "/tmp", ) -> str: """Given an existing archive_path, uncompress it. Returns a file repo url which can be used as origin url. This does not deal with the case where the archive passed along does not exist. """ if not isinstance(tmp_path, str): tmp_path = str(tmp_path) # uncompress folder/repositories/dump for the loader to ingest subprocess.check_output(["tar", "xf", archive_path, "-C", tmp_path]) # build the origin url (or some derivative form) _fname = filename if filename else os.path.basename(archive_path) repo_url = f"file://{tmp_path}/{_fname}" return repo_url def encode_target(target: Dict) -> Dict: """Test helper to ease readability in test """ if not target: return target target_type = target["target_type"] target_data = target["target"] if target_type == "alias" and isinstance(target_data, str): encoded_target = target_data.encode("utf-8") elif isinstance(target_data, str): encoded_target = hash_to_bytes(target_data) else: encoded_target = target_data return {"target": encoded_target, "target_type": target_type} class InconsistentAliasBranchError(AssertionError): """When an alias branch targets an inexistent branch.""" pass class InexistentObjectsError(AssertionError): """When a targeted branch reference does not exist in the storage""" pass def check_snapshot( snapshot: Snapshot, storage: StorageInterface, allowed_empty: Iterable[Tuple[TargetType, bytes]] = [], ) -> Snapshot: """Check that: - snapshot exists in the storage and match - each object reference up to the revision/release targets exists Args: snapshot: full snapshot to check for existence and consistency storage: storage to lookup information into allowed_empty: Iterable of branch we allow to be empty (some edge case loaders allows this case to happen, nixguix for example allows the branch evaluation" to target the nixpkgs git commit reference, which may not yet be resolvable at loading time) Returns: the snapshot stored in the storage for further test assertion if any is needed. """ if not isinstance(snapshot, Snapshot): raise AssertionError(f"variable 'snapshot' must be a snapshot: {snapshot!r}") expected_snapshot = snapshot_get_all_branches(storage, snapshot.id) if expected_snapshot is None: raise AssertionError(f"Snapshot {snapshot.id.hex()} is not found") assert snapshot == expected_snapshot objects_by_target_type = defaultdict(list) object_to_branch = {} for branch, target in expected_snapshot.branches.items(): if (target.target_type, branch) in allowed_empty: # safe for those elements to not be checked for existence continue objects_by_target_type[target.target_type].append(target.target) object_to_branch[target.target] = branch # check that alias references target something that exists, otherwise raise aliases: List[bytes] = objects_by_target_type.get(TargetType.ALIAS, []) for alias in aliases: if alias not in expected_snapshot.branches: raise InconsistentAliasBranchError( f"Alias branch {alias.decode('utf-8')} " f"should be in {list(expected_snapshot.branches)}" ) revs = objects_by_target_type.get(TargetType.REVISION) if revs: - revisions = list(storage.revision_get(revs)) + revisions = storage.revision_get(revs) not_found = [rev_id for rev_id, rev in zip(revs, revisions) if rev is None] if not_found: missing_objs = ", ".join( str((object_to_branch[rev], rev.hex())) for rev in not_found ) raise InexistentObjectsError( f"Branch/Revision(s) {missing_objs} should exist in storage" ) # retrieve information from revision - for rev in revisions: - assert rev is not None - objects_by_target_type[TargetType.DIRECTORY].append(rev["directory"]) - object_to_branch[rev["directory"]] = rev["id"] + for revision in revisions: + assert revision is not None + objects_by_target_type[TargetType.DIRECTORY].append(revision.directory) + object_to_branch[revision.directory] = revision.id rels = objects_by_target_type.get(TargetType.RELEASE) if rels: not_found = list(storage.release_missing(rels)) if not_found: missing_objs = ", ".join( str((object_to_branch[rel], rel.hex())) for rel in not_found ) raise InexistentObjectsError( f"Branch/Release(s) {missing_objs} should exist in storage" ) # first level dirs exist? dirs = objects_by_target_type.get(TargetType.DIRECTORY) if dirs: not_found = list(storage.directory_missing(dirs)) if not_found: missing_objs = ", ".join( str((object_to_branch[dir_].hex(), dir_.hex())) for dir_ in not_found ) raise InexistentObjectsError( f"Missing directories {missing_objs}: " "(revision exists, directory target does not)" ) for dir_ in dirs: # retrieve new objects to check for existence paths = storage.directory_ls(dir_, recursive=True) for path in paths: if path["type"] == "dir": target_type = TargetType.DIRECTORY else: target_type = TargetType.CONTENT target = path["target"] objects_by_target_type[target_type].append(target) object_to_branch[target] = dir_ # check nested directories dirs = objects_by_target_type.get(TargetType.DIRECTORY) if dirs: not_found = list(storage.directory_missing(dirs)) if not_found: missing_objs = ", ".join( str((object_to_branch[dir_].hex(), dir_.hex())) for dir_ in not_found ) raise InexistentObjectsError( f"Missing directories {missing_objs}: " "(revision exists, directory target does not)" ) # check contents directories cnts = objects_by_target_type.get(TargetType.CONTENT) if cnts: not_found = list(storage.content_missing_per_sha1_git(cnts)) if not_found: missing_objs = ", ".join( str((object_to_branch[cnt].hex(), cnt.hex())) for cnt in not_found ) raise InexistentObjectsError(f"Missing contents {missing_objs}") return snapshot def get_stats(storage) -> Dict: """Adaptation utils to unify the stats counters across storage implementation. """ storage.refresh_stat_counters() stats = storage.stat_counters() keys = [ "content", "directory", "origin", "origin_visit", "person", "release", "revision", "skipped_content", "snapshot", ] return {k: stats.get(k) for k in keys}