diff --git a/swh/loader/package/archive/loader.py b/swh/loader/package/archive/loader.py index d13183d..8269556 100644 --- a/swh/loader/package/archive/loader.py +++ b/swh/loader/package/archive/loader.py @@ -1,148 +1,170 @@ # Copyright (C) 2019 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 logging from os import path -from typing import Any, Dict, Iterator, Mapping, Optional, Sequence, Tuple +from typing import Any, Dict, Iterator, Optional, Sequence, Tuple, Union import attr import iso8601 from swh.loader.package.loader import PackageLoader, BasePackageInfo -from swh.loader.package.utils import release_name, artifact_identity +from swh.loader.package.utils import release_name from swh.model.model import ( Sha1Git, Person, TimestampWithTimezone, Revision, RevisionType, ) logger = logging.getLogger(__name__) SWH_PERSON = Person( name=b"Software Heritage", fullname=b"Software Heritage", email=b"robot@softwareheritage.org", ) REVISION_MESSAGE = b"swh-loader-package: synthetic revision message" +@attr.s class ArchivePackageInfo(BasePackageInfo): raw = attr.ib(type=Dict[str, Any]) + length = attr.ib(type=int) + """Size of the archive file""" + time = attr.ib(type=Union[str, datetime.datetime]) + """Timestamp of the archive file on the server""" + version = attr.ib(type=str) + + # default keys for gnu + ID_KEYS = ["time", "url", "length", "version"] + + def artifact_identity(self, id_keys=None): + if id_keys is None: + id_keys = self.ID_KEYS + # TODO: use parsed attributes instead of self.raw + return [self.raw.get(k) for k in id_keys] + + @classmethod + def from_metadata(cls, a_metadata: Dict[str, Any]) -> "ArchivePackageInfo": + url = a_metadata["url"] + filename = a_metadata.get("filename") + return cls( + url=url, + filename=filename if filename else path.split(url)[-1], + raw=a_metadata, + length=a_metadata["length"], + time=a_metadata["time"], + version=a_metadata["version"], + ) class ArchiveLoader(PackageLoader[ArchivePackageInfo]): """Load archive origin's artifact files into swh archive """ visit_type = "tar" def __init__( self, url: str, - artifacts: Sequence[Mapping[str, Any]], + artifacts: Sequence[Dict[str, Any]], identity_artifact_keys: Optional[Sequence[str]] = None, ): """Loader constructor. For now, this is the lister's task output. Args: url: Origin url artifacts: List of artifact information with keys: - **time**: last modification time as either isoformat date string or timestamp - **url**: the artifact url to retrieve filename - **artifact's filename version**: artifact's version length - **length**: artifact's length identity_artifact_keys: Optional List of keys forming the "identity" of an artifact """ super().__init__(url=url) self.artifacts = artifacts # assume order is enforced in the lister - if not identity_artifact_keys: - # default keys for gnu - identity_artifact_keys = ["time", "url", "length", "version"] self.identity_artifact_keys = identity_artifact_keys def get_versions(self) -> Sequence[str]: versions = [] for archive in self.artifacts: v = archive.get("version") if v: versions.append(v) return versions def get_default_version(self) -> str: # It's the most recent, so for this loader, it's the last one return self.artifacts[-1]["version"] def get_package_info( self, version: str ) -> Iterator[Tuple[str, ArchivePackageInfo]]: for a_metadata in self.artifacts: - url = a_metadata["url"] - package_version = a_metadata["version"] - if version == package_version: - filename = a_metadata.get("filename") - p_info = ArchivePackageInfo( - url=url, - filename=filename if filename else path.split(url)[-1], - raw=a_metadata, - ) + p_info = ArchivePackageInfo.from_metadata(a_metadata) + if version == p_info.version: # FIXME: this code assumes we have only 1 artifact per # versioned package yield release_name(version), p_info def resolve_revision_from( - self, known_artifacts: Dict, artifact_metadata: Dict + self, known_artifacts: Dict, p_info: ArchivePackageInfo ) -> Optional[bytes]: - identity = artifact_identity( - artifact_metadata, id_keys=self.identity_artifact_keys - ) + identity = p_info.artifact_identity(id_keys=self.identity_artifact_keys) for rev_id, known_artifact in known_artifacts.items(): logging.debug("known_artifact: %s", known_artifact) reference_artifact = known_artifact["extrinsic"]["raw"] - known_identity = artifact_identity( - reference_artifact, id_keys=self.identity_artifact_keys + reference_artifact_info = ArchivePackageInfo.from_metadata( + reference_artifact + ) + known_identity = reference_artifact_info.artifact_identity( + id_keys=self.identity_artifact_keys ) if identity == known_identity: return rev_id return None def build_revision( - self, a_metadata: Mapping[str, Any], uncompressed_path: str, directory: Sha1Git + self, p_info: ArchivePackageInfo, uncompressed_path: str, directory: Sha1Git ) -> Optional[Revision]: - time = a_metadata["time"] # assume it's a timestamp + time = p_info.time # assume it's a timestamp if isinstance(time, str): # otherwise, assume it's a parsable date - time = iso8601.parse_date(time) - normalized_time = TimestampWithTimezone.from_datetime(time) + parsed_time = iso8601.parse_date(time) + else: + parsed_time = time + normalized_time = TimestampWithTimezone.from_datetime(parsed_time) return Revision( type=RevisionType.TAR, message=REVISION_MESSAGE, date=normalized_time, author=SWH_PERSON, committer=SWH_PERSON, committer_date=normalized_time, parents=(), directory=directory, synthetic=True, metadata={ "intrinsic": {}, "extrinsic": { "provider": self.url, "when": self.visit_date.isoformat(), - "raw": a_metadata, + "raw": p_info.raw, }, }, ) diff --git a/swh/loader/package/archive/tasks.py b/swh/loader/package/archive/tasks.py index a9cb9e4..1f1d2f5 100644 --- a/swh/loader/package/archive/tasks.py +++ b/swh/loader/package/archive/tasks.py @@ -1,16 +1,14 @@ # Copyright (C) 2019 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 from celery import shared_task from swh.loader.package.archive.loader import ArchiveLoader @shared_task(name=__name__ + ".LoadArchive") -def load_archive_files(*, url=None, artifacts=None, identity_artifact_keys=None): +def load_archive_files(*, url=None, artifacts=None): """Load archive's artifacts (e.g gnu, etc...)""" - return ArchiveLoader( - url, artifacts, identity_artifact_keys=identity_artifact_keys - ).load() + return ArchiveLoader(url, artifacts).load() diff --git a/swh/loader/package/archive/tests/test_archive.py b/swh/loader/package/archive/tests/test_archive.py index 200f99c..5388b6f 100644 --- a/swh/loader/package/archive/tests/test_archive.py +++ b/swh/loader/package/archive/tests/test_archive.py @@ -1,345 +1,379 @@ # 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 +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] assert revision is not None check_metadata_paths( 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"]: 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={**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/cran/loader.py b/swh/loader/package/cran/loader.py index 68a8e28..b2f5af2 100644 --- a/swh/loader/package/cran/loader.py +++ b/swh/loader/package/cran/loader.py @@ -1,194 +1,203 @@ # 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 dateutil.parser import datetime from datetime import timezone import os from os import path import logging import re from typing import Any, Dict, Iterator, List, Mapping, Optional, Tuple import attr from debian.deb822 import Deb822 from swh.loader.package.loader import BasePackageInfo, PackageLoader -from swh.loader.package.utils import release_name, artifact_identity +from swh.loader.package.utils import release_name from swh.model.model import ( Person, TimestampWithTimezone, Sha1Git, Revision, RevisionType, ) logger = logging.getLogger(__name__) DATE_PATTERN = re.compile(r"^(?P\d{4})-(?P\d{2})$") +@attr.s class CRANPackageInfo(BasePackageInfo): raw = attr.ib(type=Dict[str, Any]) + version = attr.ib(type=str) + + ID_KEYS = ["url", "version"] + + @classmethod + def from_metadata(cls, a_metadata: Dict[str, Any]) -> "CRANPackageInfo": + url = a_metadata["url"] + return CRANPackageInfo( + url=url, + filename=path.basename(url), + raw=a_metadata, + version=a_metadata["version"], + ) class CRANLoader(PackageLoader[CRANPackageInfo]): visit_type = "cran" def __init__(self, url: str, artifacts: List[Dict]): """Loader constructor. Args: url: Origin url to retrieve cran artifact(s) from artifacts: List of associated artifact for the origin url """ super().__init__(url=url) # explicit what we consider the artifact identity - self.id_keys = ["url", "version"] self.artifacts = artifacts def get_versions(self) -> List[str]: versions = [] for artifact in self.artifacts: versions.append(artifact["version"]) return versions def get_default_version(self) -> str: return self.artifacts[-1]["version"] def get_package_info(self, version: str) -> Iterator[Tuple[str, CRANPackageInfo]]: for a_metadata in self.artifacts: - url = a_metadata["url"] - package_version = a_metadata["version"] - if version == package_version: - p_info = CRANPackageInfo( - url=url, filename=path.basename(url), raw=a_metadata, - ) + p_info = CRANPackageInfo.from_metadata(a_metadata) + if version == p_info.version: yield release_name(version), p_info def resolve_revision_from( - self, - known_artifacts: Mapping[bytes, Mapping], - artifact_metadata: Mapping[str, Any], + self, known_artifacts: Mapping[bytes, Mapping], p_info: CRANPackageInfo, ) -> Optional[bytes]: """Given known_artifacts per revision, try to determine the revision for artifact_metadata """ - new_identity = artifact_identity(artifact_metadata, self.id_keys) + new_identity = p_info.artifact_identity() for rev_id, known_artifact_meta in known_artifacts.items(): logging.debug("known_artifact_meta: %s", known_artifact_meta) known_artifact = known_artifact_meta["extrinsic"]["raw"] - known_identity = artifact_identity(known_artifact, self.id_keys) + known_identity = CRANPackageInfo.from_metadata( + known_artifact + ).artifact_identity() if new_identity == known_identity: return rev_id return None def build_revision( - self, a_metadata: Mapping[str, Any], uncompressed_path: str, directory: Sha1Git + self, p_info: CRANPackageInfo, uncompressed_path: str, directory: Sha1Git ) -> Optional[Revision]: # a_metadata is empty metadata = extract_intrinsic_metadata(uncompressed_path) date = parse_date(metadata.get("Date")) author = Person.from_fullname(metadata.get("Maintainer", "").encode()) - version = metadata.get("Version", a_metadata["version"]) + version = metadata.get("Version", p_info.version) return Revision( message=version.encode("utf-8"), type=RevisionType.TAR, date=date, author=author, committer=author, committer_date=date, parents=(), directory=directory, synthetic=True, metadata={ "intrinsic": {"tool": "DESCRIPTION", "raw": metadata,}, "extrinsic": { "provider": self.url, "when": self.visit_date.isoformat(), - "raw": a_metadata, + "raw": p_info.raw, }, }, ) def parse_debian_control(filepath: str) -> Dict[str, Any]: """Parse debian control at filepath""" metadata: Dict = {} logger.debug("Debian control file %s", filepath) for paragraph in Deb822.iter_paragraphs(open(filepath, "rb")): logger.debug("paragraph: %s", paragraph) metadata.update(**paragraph) logger.debug("metadata parsed: %s", metadata) return metadata def extract_intrinsic_metadata(dir_path: str) -> Dict[str, Any]: """Given an uncompressed path holding the DESCRIPTION file, returns a DESCRIPTION parsed structure as a dict. Cran origins describes their intrinsic metadata within a DESCRIPTION file at the root tree of a tarball. This DESCRIPTION uses a simple file format called DCF, the Debian control format. The release artifact contains at their root one folder. For example: $ tar tvf zprint-0.0.6.tar.gz drwxr-xr-x root/root 0 2018-08-22 11:01 zprint-0.0.6/ ... Args: dir_path (str): Path to the uncompressed directory representing a release artifact from pypi. Returns: the DESCRIPTION parsed structure as a dict (or empty dict if missing) """ # Retrieve the root folder of the archive if not os.path.exists(dir_path): return {} lst = os.listdir(dir_path) if len(lst) != 1: return {} project_dirname = lst[0] description_path = os.path.join(dir_path, project_dirname, "DESCRIPTION") if not os.path.exists(description_path): return {} return parse_debian_control(description_path) def parse_date(date: Optional[str]) -> Optional[TimestampWithTimezone]: """Parse a date into a datetime """ assert not date or isinstance(date, str) dt: Optional[datetime.datetime] = None if not date: return None try: specific_date = DATE_PATTERN.match(date) if specific_date: year = int(specific_date.group("year")) month = int(specific_date.group("month")) dt = datetime.datetime(year, month, 1) else: dt = dateutil.parser.parse(date) if not dt.tzinfo: # up for discussion the timezone needs to be set or # normalize_timestamp is not happy: ValueError: normalize_timestamp # received datetime without timezone: 2001-06-08 00:00:00 dt = dt.replace(tzinfo=timezone.utc) except Exception as e: logger.warning("Fail to parse date %s. Reason: %s", (date, e)) if dt: return TimestampWithTimezone.from_datetime(dt) else: return None diff --git a/swh/loader/package/debian/loader.py b/swh/loader/package/debian/loader.py index 34d2276..7123dc2 100644 --- a/swh/loader/package/debian/loader.py +++ b/swh/loader/package/debian/loader.py @@ -1,420 +1,480 @@ # Copyright (C) 2017-2019 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 email.utils import logging from os import path import re import subprocess -from typing import Any, Dict, Iterator, List, Mapping, Optional, Sequence, Tuple +from typing import ( + Any, + Dict, + FrozenSet, + Iterator, + List, + Mapping, + Optional, + Sequence, + Tuple, +) import attr from dateutil.parser import parse as parse_date from debian.changelog import Changelog from debian.deb822 import Dsc from swh.loader.package.loader import BasePackageInfo, PackageLoader from swh.loader.package.utils import download, release_name from swh.model.model import ( Sha1Git, Person, Revision, RevisionType, TimestampWithTimezone, ) logger = logging.getLogger(__name__) UPLOADERS_SPLIT = re.compile(r"(?<=\>)\s*,\s*") +@attr.s +class DebianFileMetadata: + md5sum = attr.ib(type=str) + name = attr.ib(type=str) + """Filename""" + sha256 = attr.ib(type=str) + size = attr.ib(type=int) + uri = attr.ib(type=str) + """URL of this specific file""" + + +@attr.s +class DebianPackageChangelog: + person = attr.ib(type=Dict[str, str]) + """A dict with fields like, model.Person, except they are str instead + of bytes, and 'email' is optional.""" + date = attr.ib(type=str) + """Date of the changelog entry.""" + history = attr.ib(type=List[Tuple[str, str]]) + """List of tuples (package_name, version)""" + + +@attr.s class DebianPackageInfo(BasePackageInfo): raw = attr.ib(type=Dict[str, Any]) + files = attr.ib(type=Dict[str, DebianFileMetadata]) + """Metadata of the files (.deb, .dsc, ...) of the package.""" + name = attr.ib(type=str) + version = attr.ib(type=str) + + @classmethod + def from_metadata(cls, a_metadata: Dict[str, Any], url: str) -> "DebianPackageInfo": + return cls( + url=url, + filename=None, + raw=a_metadata, + files={ + file_name: DebianFileMetadata(**file_metadata) + for (file_name, file_metadata) in a_metadata.get("files", {}).items() + }, + name=a_metadata["name"], + version=a_metadata["version"], + ) + + +@attr.s +class IntrinsicPackageMetadata: + """Metadata extracted from a package's .dsc file.""" + + name = attr.ib(type=str) + version = attr.ib(type=str) + changelog = attr.ib(type=DebianPackageChangelog) + maintainers = attr.ib(type=List[Dict[str, str]]) + """A list of dicts with fields like, model.Person, except they are str instead + of bytes, and 'email' is optional.""" class DebianLoader(PackageLoader[DebianPackageInfo]): """Load debian origins into swh archive. """ visit_type = "deb" def __init__(self, url: str, date: str, packages: Mapping[str, Any]): """Debian Loader implementation. Args: url: Origin url (e.g. deb://Debian/packages/cicero) date: Ignored packages: versioned packages and associated artifacts, example:: { 'stretch/contrib/0.7.2-3': { 'name': 'cicero', 'version': '0.7.2-3' 'files': { 'cicero_0.7.2-3.diff.gz': { 'md5sum': 'a93661b6a48db48d59ba7d26796fc9ce', 'name': 'cicero_0.7.2-3.diff.gz', 'sha256': 'f039c9642fe15c75bed5254315e2a29f...', 'size': 3964, 'uri': 'http://d.d.o/cicero_0.7.2-3.diff.gz', }, 'cicero_0.7.2-3.dsc': { 'md5sum': 'd5dac83eb9cfc9bb52a15eb618b4670a', 'name': 'cicero_0.7.2-3.dsc', 'sha256': '35b7f1048010c67adfd8d70e4961aefb...', 'size': 1864, 'uri': 'http://d.d.o/cicero_0.7.2-3.dsc', }, 'cicero_0.7.2.orig.tar.gz': { 'md5sum': '4353dede07c5728319ba7f5595a7230a', 'name': 'cicero_0.7.2.orig.tar.gz', 'sha256': '63f40f2436ea9f67b44e2d4bd669dbab...', 'size': 96527, 'uri': 'http://d.d.o/cicero_0.7.2.orig.tar.gz', } }, }, # ... } """ super().__init__(url=url) self.packages = packages def get_versions(self) -> Sequence[str]: """Returns the keys of the packages input (e.g. stretch/contrib/0.7.2-3, etc...) """ return list(self.packages.keys()) def get_package_info(self, version: str) -> Iterator[Tuple[str, DebianPackageInfo]]: meta = self.packages[version] - p_info = DebianPackageInfo(url=self.url, filename=None, raw=meta,) + p_info = DebianPackageInfo.from_metadata(meta, url=self.url) yield release_name(version), p_info def resolve_revision_from( - self, known_package_artifacts: Mapping, artifact_metadata: Mapping + self, known_package_artifacts: Mapping, p_info: DebianPackageInfo ) -> Optional[bytes]: - return resolve_revision_from(known_package_artifacts, artifact_metadata) + return resolve_revision_from(known_package_artifacts, p_info) def download_package( self, p_info: DebianPackageInfo, tmpdir: str ) -> List[Tuple[str, Mapping]]: """Contrary to other package loaders (1 package, 1 artifact), `p_info.files` represents the package's datafiles set to fetch: - .orig.tar.gz - .dsc - .diff.gz This is delegated to the `download_package` function. """ - all_hashes = download_package(p_info.raw, tmpdir) + all_hashes = download_package(p_info, tmpdir) logger.debug("all_hashes: %s", all_hashes) res = [] for hashes in all_hashes.values(): res.append((tmpdir, hashes)) logger.debug("res: %s", res) return res def uncompress( self, dl_artifacts: List[Tuple[str, Mapping[str, Any]]], dest: str ) -> str: logger.debug("dl_artifacts: %s", dl_artifacts) return extract_package(dl_artifacts, dest=dest) def build_revision( - self, a_metadata: Mapping[str, Any], uncompressed_path: str, directory: Sha1Git + self, p_info: DebianPackageInfo, uncompressed_path: str, directory: Sha1Git ) -> Optional[Revision]: - dsc_url, dsc_name = dsc_information(a_metadata) + dsc_url, dsc_name = dsc_information(p_info) if not dsc_name: raise ValueError("dsc name for url %s should not be None" % dsc_url) dsc_path = path.join(path.dirname(uncompressed_path), dsc_name) - i_metadata = get_package_metadata(a_metadata, dsc_path, uncompressed_path) + intrinsic_metadata = get_intrinsic_package_metadata( + p_info, dsc_path, uncompressed_path + ) - logger.debug("i_metadata: %s", i_metadata) - logger.debug("a_metadata: %s", a_metadata) + logger.debug("intrinsic_metadata: %s", intrinsic_metadata) + logger.debug("p_info: %s", p_info) msg = "Synthetic revision for Debian source package %s version %s" % ( - a_metadata["name"], - a_metadata["version"], + p_info.name, + p_info.version, ) - date = TimestampWithTimezone.from_iso8601(i_metadata["changelog"]["date"]) - author = prepare_person(i_metadata["changelog"]["person"]) + author = prepare_person(intrinsic_metadata.changelog.person) + date = TimestampWithTimezone.from_iso8601(intrinsic_metadata.changelog.date) # inspired from swh.loader.debian.converters.package_metadata_to_revision # noqa return Revision( type=RevisionType.DSC, message=msg.encode("utf-8"), author=author, date=date, committer=author, committer_date=date, parents=(), directory=directory, synthetic=True, metadata={ - "intrinsic": {"tool": "dsc", "raw": i_metadata,}, + "intrinsic": {"tool": "dsc", "raw": attr.asdict(intrinsic_metadata),}, "extrinsic": { "provider": dsc_url, "when": self.visit_date.isoformat(), - "raw": a_metadata, + "raw": p_info.raw, }, }, ) def resolve_revision_from( - known_package_artifacts: Mapping, artifact_metadata: Mapping + known_package_artifacts: Mapping, p_info: DebianPackageInfo ) -> Optional[bytes]: """Given known package artifacts (resolved from the snapshot of previous visit) and the new artifact to fetch, try to solve the corresponding revision. """ - artifacts_to_fetch = artifact_metadata.get("files") + artifacts_to_fetch = p_info.files if not artifacts_to_fetch: return None - def to_set(data): + def to_set(data: DebianPackageInfo) -> FrozenSet[Tuple[str, str, int]]: return frozenset( - [ - (name, meta["sha256"], meta["size"]) - for name, meta in data["files"].items() - ] + (name, meta.sha256, meta.size) for name, meta in data.files.items() ) # what we want to avoid downloading back if we have them already - set_new_artifacts = to_set(artifact_metadata) + set_new_artifacts = to_set(p_info) known_artifacts_revision_id = {} for rev_id, known_artifacts in known_package_artifacts.items(): extrinsic = known_artifacts.get("extrinsic") if not extrinsic: continue - s = to_set(extrinsic["raw"]) + s = to_set(DebianPackageInfo.from_metadata(extrinsic["raw"], url=p_info.url)) known_artifacts_revision_id[s] = rev_id return known_artifacts_revision_id.get(set_new_artifacts) -def uid_to_person(uid: str) -> Mapping[str, str]: +def uid_to_person(uid: str) -> Dict[str, str]: """Convert an uid to a person suitable for insertion. Args: uid: an uid of the form "Name " Returns: a dictionary with the following keys: - name: the name associated to the uid - email: the mail associated to the uid - fullname: the actual uid input """ logger.debug("uid: %s", uid) ret = { "name": "", "email": "", "fullname": uid, } name, mail = email.utils.parseaddr(uid) if name and email: ret["name"] = name ret["email"] = mail else: ret["name"] = uid return ret def prepare_person(person: Mapping[str, str]) -> Person: """Prepare person for swh serialization... Args: A person dict Returns: A person ready for storage """ return Person.from_dict( {key: value.encode("utf-8") for (key, value) in person.items()} ) -def download_package(package: Mapping[str, Any], tmpdir: Any) -> Mapping[str, Any]: +def download_package(p_info: DebianPackageInfo, tmpdir: Any) -> Mapping[str, Any]: """Fetch a source package in a temporary directory and check the checksums for all files. Args: - package: Dict defining the set of files representing a debian package + p_info: Information on a package tmpdir: Where to download and extract the files to ingest Returns: Dict of swh hashes per filename key """ all_hashes = {} - for filename, fileinfo in package["files"].items(): - uri = fileinfo["uri"] + for filename, fileinfo in p_info.files.items(): + uri = fileinfo.uri logger.debug("fileinfo: %s", fileinfo) - extrinsic_hashes = {"sha256": fileinfo["sha256"]} + extrinsic_hashes = {"sha256": fileinfo.sha256} logger.debug("extrinsic_hashes(%s): %s", filename, extrinsic_hashes) filepath, hashes = download( uri, dest=tmpdir, filename=filename, hashes=extrinsic_hashes ) all_hashes[filename] = hashes logger.debug("all_hashes: %s", all_hashes) return all_hashes -def dsc_information(package: Mapping[str, Any]) -> Tuple[Optional[str], Optional[str]]: +def dsc_information(p_info: DebianPackageInfo) -> Tuple[Optional[str], Optional[str]]: """Retrieve dsc information from a package. Args: - package: Package metadata information + p_info: Package metadata information Returns: Tuple of dsc file's uri, dsc's full disk path """ dsc_name = None dsc_url = None - for filename, fileinfo in package["files"].items(): + for filename, fileinfo in p_info.files.items(): if filename.endswith(".dsc"): if dsc_name: raise ValueError( "Package %s_%s references several dsc files." - % (package["name"], package["version"]) + % (p_info.name, p_info.version) ) - dsc_url = fileinfo["uri"] + dsc_url = fileinfo.uri dsc_name = filename return dsc_url, dsc_name def extract_package(dl_artifacts: List[Tuple[str, Mapping]], dest: str) -> str: """Extract a Debian source package to a given directory. Note that after extraction the target directory will be the root of the extracted package, rather than containing it. Args: package: package information dictionary dest: directory where the package files are stored Returns: Package extraction directory """ a_path = dl_artifacts[0][0] logger.debug("dl_artifacts: %s", dl_artifacts) for _, hashes in dl_artifacts: logger.debug("hashes: %s", hashes) filename = hashes["filename"] if filename.endswith(".dsc"): dsc_name = filename break dsc_path = path.join(a_path, dsc_name) destdir = path.join(dest, "extracted") logfile = path.join(dest, "extract.log") logger.debug( "extract Debian source package %s in %s" % (dsc_path, destdir), extra={"swh_type": "deb_extract", "swh_dsc": dsc_path, "swh_destdir": destdir,}, ) cmd = [ "dpkg-source", "--no-copy", "--no-check", "--ignore-bad-version", "-x", dsc_path, destdir, ] try: with open(logfile, "w") as stdout: subprocess.check_call(cmd, stdout=stdout, stderr=subprocess.STDOUT) except subprocess.CalledProcessError as e: logdata = open(logfile, "r").read() raise ValueError( "dpkg-source exited with code %s: %s" % (e.returncode, logdata) ) from None return destdir -def get_package_metadata( - package: Mapping[str, Any], dsc_path: str, extracted_path: str -) -> Mapping[str, Any]: +def get_intrinsic_package_metadata( + p_info: DebianPackageInfo, dsc_path: str, extracted_path: str +) -> IntrinsicPackageMetadata: """Get the package metadata from the source package at dsc_path, extracted in extracted_path. Args: - package: the package dict (with a dsc_path key) + p_info: the package information dsc_path: path to the package's dsc file extracted_path: the path where the package got extracted Returns: dict: a dictionary with the following keys: - history: list of (package_name, package_version) tuples parsed from the package changelog """ with open(dsc_path, "rb") as dsc: parsed_dsc = Dsc(dsc) # Parse the changelog to retrieve the rest of the package information changelog_path = path.join(extracted_path, "debian/changelog") - with open(changelog_path, "rb") as changelog: + with open(changelog_path, "rb") as changelog_file: try: - parsed_changelog = Changelog(changelog) + parsed_changelog = Changelog(changelog_file) except UnicodeDecodeError: logger.warning( "Unknown encoding for changelog %s," " falling back to iso" % changelog_path, extra={ "swh_type": "deb_changelog_encoding", - "swh_name": package["name"], - "swh_version": str(package["version"]), + "swh_name": p_info.name, + "swh_version": str(p_info.version), "swh_changelog": changelog_path, }, ) # need to reset as Changelog scrolls to the end of the file - changelog.seek(0) - parsed_changelog = Changelog(changelog, encoding="iso-8859-15") - - package_info = { - "name": package["name"], - "version": str(package["version"]), - "changelog": { - "person": uid_to_person(parsed_changelog.author), - "date": parse_date(parsed_changelog.date).isoformat(), - "history": [ - (block.package, str(block.version)) for block in parsed_changelog - ][1:], - }, - } + changelog_file.seek(0) + parsed_changelog = Changelog(changelog_file, encoding="iso-8859-15") + + changelog = DebianPackageChangelog( + person=uid_to_person(parsed_changelog.author), + date=parse_date(parsed_changelog.date).isoformat(), + history=[(block.package, str(block.version)) for block in parsed_changelog][1:], + ) maintainers = [ uid_to_person(parsed_dsc["Maintainer"]), ] maintainers.extend( uid_to_person(person) for person in UPLOADERS_SPLIT.split(parsed_dsc.get("Uploaders", "")) ) - package_info["maintainers"] = maintainers - return package_info + return IntrinsicPackageMetadata( + name=p_info.name, + version=str(p_info.version), + changelog=changelog, + maintainers=maintainers, + ) diff --git a/swh/loader/package/debian/tests/test_debian.py b/swh/loader/package/debian/tests/test_debian.py index 6f1268d..6860e46 100644 --- a/swh/loader/package/debian/tests/test_debian.py +++ b/swh/loader/package/debian/tests/test_debian.py @@ -1,467 +1,487 @@ # 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 copy import logging import pytest import random from os import path from swh.loader.package.debian.loader import ( DebianLoader, + DebianPackageInfo, + DebianPackageChangelog, + IntrinsicPackageMetadata, download_package, dsc_information, uid_to_person, prepare_person, - get_package_metadata, + get_intrinsic_package_metadata, extract_package, ) from swh.loader.tests import ( assert_last_visit_matches, check_snapshot, get_stats, ) from swh.loader.package.debian.loader import resolve_revision_from from swh.model.hashutil import hash_to_bytes from swh.model.model import Person, Snapshot, SnapshotBranch, TargetType logger = logging.getLogger(__name__) +URL = "deb://Debian/packages/cicero" + PACKAGE_FILES = { "name": "cicero", "version": "0.7.2-3", "files": { "cicero_0.7.2-3.diff.gz": { "md5sum": "a93661b6a48db48d59ba7d26796fc9ce", "name": "cicero_0.7.2-3.diff.gz", "sha256": "f039c9642fe15c75bed5254315e2a29f9f2700da0e29d9b0729b3ffc46c8971c", # noqa "size": 3964, "uri": "http://deb.debian.org/debian/pool/contrib/c/cicero/cicero_0.7.2-3.diff.gz", # noqa }, "cicero_0.7.2-3.dsc": { "md5sum": "d5dac83eb9cfc9bb52a15eb618b4670a", "name": "cicero_0.7.2-3.dsc", "sha256": "35b7f1048010c67adfd8d70e4961aefd8800eb9a83a4d1cc68088da0009d9a03", # noqa "size": 1864, "uri": "http://deb.debian.org/debian/pool/contrib/c/cicero/cicero_0.7.2-3.dsc", # noqa }, # noqa "cicero_0.7.2.orig.tar.gz": { "md5sum": "4353dede07c5728319ba7f5595a7230a", "name": "cicero_0.7.2.orig.tar.gz", "sha256": "63f40f2436ea9f67b44e2d4bd669dbabe90e2635a204526c20e0b3c8ee957786", # noqa "size": 96527, "uri": "http://deb.debian.org/debian/pool/contrib/c/cicero/cicero_0.7.2.orig.tar.gz", # noqa }, }, } PACKAGE_FILES2 = { "name": "cicero", "version": "0.7.2-4", "files": { "cicero_0.7.2-4.diff.gz": { "md5sum": "1e7e6fc4a59d57c98082a3af78145734", "name": "cicero_0.7.2-4.diff.gz", "sha256": "2e6fa296ee7005473ff58d0971f4fd325617b445671480e9f2cfb738d5dbcd01", # noqa "size": 4038, "uri": "http://deb.debian.org/debian/pool/contrib/c/cicero/cicero_0.7.2-4.diff.gz", # noqa }, "cicero_0.7.2-4.dsc": { "md5sum": "1a6c8855a73b4282bb31d15518f18cde", "name": "cicero_0.7.2-4.dsc", "sha256": "913ee52f7093913420de5cbe95d63cfa817f1a1daf997961149501894e754f8b", # noqa "size": 1881, "uri": "http://deb.debian.org/debian/pool/contrib/c/cicero/cicero_0.7.2-4.dsc", # noqa }, # noqa "cicero_0.7.2.orig.tar.gz": { "md5sum": "4353dede07c5728319ba7f5595a7230a", "name": "cicero_0.7.2.orig.tar.gz", "sha256": "63f40f2436ea9f67b44e2d4bd669dbabe90e2635a204526c20e0b3c8ee957786", # noqa "size": 96527, "uri": "http://deb.debian.org/debian/pool/contrib/c/cicero/cicero_0.7.2.orig.tar.gz", # noqa }, }, } PACKAGE_PER_VERSION = { "stretch/contrib/0.7.2-3": PACKAGE_FILES, } PACKAGES_PER_VERSION = { "stretch/contrib/0.7.2-3": PACKAGE_FILES, "buster/contrib/0.7.2-4": PACKAGE_FILES2, } def test_debian_first_visit(swh_config, requests_mock_datadir): """With no prior visit, load a gnu project ends up with 1 snapshot """ - url = "deb://Debian/packages/cicero" loader = DebianLoader( - url=url, date="2019-10-12T05:58:09.165557+00:00", packages=PACKAGE_PER_VERSION, + url=URL, date="2019-10-12T05:58:09.165557+00:00", packages=PACKAGE_PER_VERSION, ) actual_load_status = loader.load() expected_snapshot_id = "3b6b66e6ee4e7d903a379a882684a2a50480c0b4" assert actual_load_status == { "status": "eventful", "snapshot_id": expected_snapshot_id, } - assert_last_visit_matches(loader.storage, url, status="full", type="deb") + assert_last_visit_matches(loader.storage, URL, status="full", type="deb") stats = get_stats(loader.storage) assert { "content": 42, "directory": 2, "origin": 1, "origin_visit": 1, "person": 1, "release": 0, "revision": 1, # all artifacts under 1 revision "skipped_content": 0, "snapshot": 1, } == stats expected_snapshot = Snapshot( id=hash_to_bytes(expected_snapshot_id), branches={ b"releases/stretch/contrib/0.7.2-3": SnapshotBranch( target_type=TargetType.REVISION, target=hash_to_bytes("2807f5b3f84368b4889a9ae827fe85854ffecf07"), ) }, ) # different than the previous loader as no release is done check_snapshot(expected_snapshot, loader.storage) def test_debian_first_visit_then_another_visit(swh_config, requests_mock_datadir): """With no prior visit, load a debian project ends up with 1 snapshot """ - url = "deb://Debian/packages/cicero" loader = DebianLoader( - url=url, date="2019-10-12T05:58:09.165557+00:00", packages=PACKAGE_PER_VERSION + url=URL, date="2019-10-12T05:58:09.165557+00:00", packages=PACKAGE_PER_VERSION ) actual_load_status = loader.load() expected_snapshot_id = "3b6b66e6ee4e7d903a379a882684a2a50480c0b4" assert actual_load_status == { "status": "eventful", "snapshot_id": expected_snapshot_id, } - assert_last_visit_matches(loader.storage, url, status="full", type="deb") + assert_last_visit_matches(loader.storage, URL, status="full", type="deb") stats = get_stats(loader.storage) assert { "content": 42, "directory": 2, "origin": 1, "origin_visit": 1, "person": 1, "release": 0, "revision": 1, # all artifacts under 1 revision "skipped_content": 0, "snapshot": 1, } == stats expected_snapshot = Snapshot( id=hash_to_bytes(expected_snapshot_id), branches={ b"releases/stretch/contrib/0.7.2-3": SnapshotBranch( target_type=TargetType.REVISION, target=hash_to_bytes("2807f5b3f84368b4889a9ae827fe85854ffecf07"), ) }, ) # different than the previous loader as no release is done check_snapshot(expected_snapshot, loader.storage) # No change in between load actual_load_status2 = loader.load() assert actual_load_status2["status"] == "uneventful" - assert_last_visit_matches(loader.storage, url, status="full", type="deb") + assert_last_visit_matches(loader.storage, URL, status="full", type="deb") stats2 = get_stats(loader.storage) assert { "content": 42 + 0, "directory": 2 + 0, "origin": 1, "origin_visit": 1 + 1, # a new visit occurred "person": 1, "release": 0, "revision": 1, "skipped_content": 0, "snapshot": 1, # same snapshot across 2 visits } == stats2 urls = [ m.url for m in requests_mock_datadir.request_history if m.url.startswith("http://deb.debian.org") ] # visited each package artifact twice across 2 visits assert len(urls) == len(set(urls)) def test_uid_to_person(): uid = "Someone Name " actual_person = uid_to_person(uid) assert actual_person == { "name": "Someone Name", "email": "someone@orga.org", "fullname": uid, } def test_prepare_person(): actual_author = prepare_person( { "name": "Someone Name", "email": "someone@orga.org", "fullname": "Someone Name ", } ) assert actual_author == Person( name=b"Someone Name", email=b"someone@orga.org", fullname=b"Someone Name ", ) def test_download_package(datadir, tmpdir, requests_mock_datadir): tmpdir = str(tmpdir) # py3.5 work around (LocalPath issue) - all_hashes = download_package(PACKAGE_FILES, tmpdir) + p_info = DebianPackageInfo.from_metadata(PACKAGE_FILES, url=URL) + all_hashes = download_package(p_info, tmpdir) assert all_hashes == { "cicero_0.7.2-3.diff.gz": { "checksums": { "sha1": "0815282053f21601b0ec4adf7a8fe47eace3c0bc", "sha256": "f039c9642fe15c75bed5254315e2a29f9f2700da0e29d9b0729b3ffc46c8971c", # noqa }, "filename": "cicero_0.7.2-3.diff.gz", "length": 3964, }, "cicero_0.7.2-3.dsc": { "checksums": { "sha1": "abbec4e8efbbc80278236e1dd136831eac08accd", "sha256": "35b7f1048010c67adfd8d70e4961aefd8800eb9a83a4d1cc68088da0009d9a03", # noqa }, "filename": "cicero_0.7.2-3.dsc", "length": 1864, }, "cicero_0.7.2.orig.tar.gz": { "checksums": { "sha1": "a286efd63fe2c9c9f7bb30255c3d6fcdcf390b43", "sha256": "63f40f2436ea9f67b44e2d4bd669dbabe90e2635a204526c20e0b3c8ee957786", # noqa }, "filename": "cicero_0.7.2.orig.tar.gz", "length": 96527, }, } def test_dsc_information_ok(): fname = "cicero_0.7.2-3.dsc" - dsc_url, dsc_name = dsc_information(PACKAGE_FILES) + p_info = DebianPackageInfo.from_metadata(PACKAGE_FILES, url=URL) + dsc_url, dsc_name = dsc_information(p_info) assert dsc_url == PACKAGE_FILES["files"][fname]["uri"] assert dsc_name == PACKAGE_FILES["files"][fname]["name"] def test_dsc_information_not_found(): fname = "cicero_0.7.2-3.dsc" - package_files = copy.deepcopy(PACKAGE_FILES) - package_files["files"].pop(fname) + p_info = DebianPackageInfo.from_metadata(PACKAGE_FILES, url=URL) + p_info.files.pop(fname) - dsc_url, dsc_name = dsc_information(package_files) + dsc_url, dsc_name = dsc_information(p_info) assert dsc_url is None assert dsc_name is None def test_dsc_information_too_many_dsc_entries(): # craft an extra dsc file fname = "cicero_0.7.2-3.dsc" - package_files = copy.deepcopy(PACKAGE_FILES) - data = package_files["files"][fname] + p_info = DebianPackageInfo.from_metadata(PACKAGE_FILES, url=URL) + data = p_info.files[fname] fname2 = fname.replace("cicero", "ciceroo") - package_files["files"][fname2] = data + p_info.files[fname2] = data with pytest.raises( ValueError, match="Package %s_%s references several dsc" - % (package_files["name"], package_files["version"]), + % (PACKAGE_FILES["name"], PACKAGE_FILES["version"]), ): - dsc_information(package_files) + dsc_information(p_info) -def test_get_package_metadata(requests_mock_datadir, datadir, tmp_path): +def test_get_intrinsic_package_metadata(requests_mock_datadir, datadir, tmp_path): tmp_path = str(tmp_path) # py3.5 compat. - package = PACKAGE_FILES + p_info = DebianPackageInfo.from_metadata(PACKAGE_FILES, url=URL) - logger.debug("package: %s", package) + logger.debug("p_info: %s", p_info) # download the packages - all_hashes = download_package(package, tmp_path) + all_hashes = download_package(p_info, tmp_path) # Retrieve information from package - _, dsc_name = dsc_information(package) + _, dsc_name = dsc_information(p_info) dl_artifacts = [(tmp_path, hashes) for hashes in all_hashes.values()] # Extract information from package extracted_path = extract_package(dl_artifacts, tmp_path) # Retrieve information on package dsc_path = path.join(path.dirname(extracted_path), dsc_name) - actual_package_info = get_package_metadata(package, dsc_path, extracted_path) + actual_package_info = get_intrinsic_package_metadata( + p_info, dsc_path, extracted_path + ) logger.debug("actual_package_info: %s", actual_package_info) - assert actual_package_info == { - "changelog": { - "date": "2014-10-19T16:52:35+02:00", - "history": [ + assert actual_package_info == IntrinsicPackageMetadata( + changelog=DebianPackageChangelog( + date="2014-10-19T16:52:35+02:00", + history=[ ("cicero", "0.7.2-2"), ("cicero", "0.7.2-1"), ("cicero", "0.7-1"), ], - "person": { + person={ "email": "sthibault@debian.org", "fullname": "Samuel Thibault ", "name": "Samuel Thibault", }, - }, - "maintainers": [ + ), + maintainers=[ { "email": "debian-accessibility@lists.debian.org", "fullname": "Debian Accessibility Team " "", "name": "Debian Accessibility Team", }, { "email": "sthibault@debian.org", "fullname": "Samuel Thibault ", "name": "Samuel Thibault", }, ], - "name": "cicero", - "version": "0.7.2-3", - } + name="cicero", + version="0.7.2-3", + ) def test_debian_multiple_packages(swh_config, requests_mock_datadir): - url = "deb://Debian/packages/cicero" loader = DebianLoader( - url=url, date="2019-10-12T05:58:09.165557+00:00", packages=PACKAGES_PER_VERSION + url=URL, date="2019-10-12T05:58:09.165557+00:00", packages=PACKAGES_PER_VERSION ) actual_load_status = loader.load() expected_snapshot_id = "defc19021187f3727293121fcf6c5c82cb923604" assert actual_load_status == { "status": "eventful", "snapshot_id": expected_snapshot_id, } - assert_last_visit_matches(loader.storage, url, status="full", type="deb") + assert_last_visit_matches(loader.storage, URL, status="full", type="deb") expected_snapshot = Snapshot( id=hash_to_bytes(expected_snapshot_id), branches={ b"releases/stretch/contrib/0.7.2-3": SnapshotBranch( target_type=TargetType.REVISION, target=hash_to_bytes("2807f5b3f84368b4889a9ae827fe85854ffecf07"), ), b"releases/buster/contrib/0.7.2-4": SnapshotBranch( target_type=TargetType.REVISION, target=hash_to_bytes("8224139c274c984147ef4b09aa0e462c55a10bd3"), ), }, ) check_snapshot(expected_snapshot, loader.storage) def test_resolve_revision_from_edge_cases(): """Solving revision with empty data will result in unknown revision """ - for package_artifacts in [{}, PACKAGE_FILES]: - actual_revision = resolve_revision_from({}, package_artifacts) + empty_artifact = { + "name": PACKAGE_FILES["name"], + "version": PACKAGE_FILES["version"], + } + for package_artifacts in [empty_artifact, PACKAGE_FILES]: + p_info = DebianPackageInfo.from_metadata(package_artifacts, url=URL) + actual_revision = resolve_revision_from({}, p_info) assert actual_revision is None for known_artifacts in [{}, PACKAGE_FILES]: - actual_revision = resolve_revision_from(known_artifacts, {}) + actual_revision = resolve_revision_from( + known_artifacts, DebianPackageInfo.from_metadata(empty_artifact, url=URL) + ) assert actual_revision is None known_package_artifacts = { b"(\x07\xf5\xb3\xf8Ch\xb4\x88\x9a\x9a\xe8'\xfe\x85\x85O\xfe\xcf\x07": { "extrinsic": { # empty }, # ... removed the unnecessary intermediary data } } - assert not resolve_revision_from(known_package_artifacts, PACKAGE_FILES) + assert not resolve_revision_from( + known_package_artifacts, DebianPackageInfo.from_metadata(PACKAGE_FILES, url=URL) + ) def test_resolve_revision_from_edge_cases_hit_and_miss(): """Solving revision with inconsistent data will result in unknown revision """ artifact_metadata = PACKAGE_FILES2 + p_info = DebianPackageInfo.from_metadata(artifact_metadata, url=URL) expected_revision_id = ( b"(\x08\xf5\xb3\xf8Ch\xb4\x88\x9a\x9a\xe8'\xff\x85\x85O\xfe\xcf\x07" # noqa ) known_package_artifacts = { expected_revision_id: { "extrinsic": {"raw": PACKAGE_FILES,}, # ... removed the unnecessary intermediary data } } - actual_revision = resolve_revision_from(known_package_artifacts, artifact_metadata) + actual_revision = resolve_revision_from(known_package_artifacts, p_info) assert actual_revision is None def test_resolve_revision_from(): """Solving revision with consistent data will solve the revision """ artifact_metadata = PACKAGE_FILES + p_info = DebianPackageInfo.from_metadata(artifact_metadata, url=URL) expected_revision_id = ( b"(\x07\xf5\xb3\xf8Ch\xb4\x88\x9a\x9a\xe8'\xfe\x85\x85O\xfe\xcf\x07" # noqa ) files = artifact_metadata["files"] # shuffling dict's keys keys = list(files.keys()) random.shuffle(keys) - package_files = {"files": {k: files[k] for k in keys}} + package_files = { + "name": PACKAGE_FILES["name"], + "version": PACKAGE_FILES["version"], + "files": {k: files[k] for k in keys}, + } known_package_artifacts = { expected_revision_id: { "extrinsic": {"raw": package_files,}, # ... removed the unnecessary intermediary data } } - actual_revision = resolve_revision_from(known_package_artifacts, artifact_metadata) + actual_revision = resolve_revision_from(known_package_artifacts, p_info) assert actual_revision == expected_revision_id diff --git a/swh/loader/package/deposit/loader.py b/swh/loader/package/deposit/loader.py index 767f226..78e0b96 100644 --- a/swh/loader/package/deposit/loader.py +++ b/swh/loader/package/deposit/loader.py @@ -1,295 +1,325 @@ # 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 types 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, MetadataTargetType, RawExtrinsicMetadata, ) from swh.loader.package.loader import PackageLoader, BasePackageInfo from swh.loader.package.utils import download logger = logging.getLogger(__name__) +@attr.s class DepositPackageInfo(BasePackageInfo): filename = attr.ib(type=str) # instead of Optional[str] raw = 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. + + 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=metadata, + ) + 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"]) self.metadata: Dict[str, Any] = {} def get_versions(self) -> Sequence[str]: # only 1 branch 'HEAD' with no alias since we only have 1 snapshot # branch return ["HEAD"] def get_package_info( self, version: str ) -> Iterator[Tuple[str, DepositPackageInfo]]: - p_info = DepositPackageInfo( - url=self.url, filename="archive.zip", raw=self.metadata, + 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, a_metadata: Dict, uncompressed_path: str, directory: Sha1Git + self, p_info: DepositPackageInfo, uncompressed_path: str, directory: Sha1Git ) -> Optional[Revision]: - depo = a_metadata.pop("deposit") - - # 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. - - # date: codemeta:dateCreated if any, deposit completed_date otherwise - date = TimestampWithTimezone.from_dict(depo["author_date"]) - # commit_date: codemeta:datePublished if any, deposit completed_date otherwise - commit_date = TimestampWithTimezone.from_dict(depo["committer_date"]) - - client, id, collection = [depo[k] for k in ["client", "id", "collection"]] - message = f"{client}: Deposit {id} in collection {collection}".encode("utf-8") - - author = parse_author(depo["author"]) - committer = parse_author(depo["committer"]) + 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=author, - date=date, - committer=committer, - committer_date=commit_date, - parents=tuple([hash_to_bytes(p) for p in depo["revision_parents"]]), + 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": a_metadata, # Actually the processed metadata instead of raw + "raw": p_info.raw, }, }, ) def load(self) -> Dict: # First making sure the deposit is known prior to trigger a loading try: self.metadata = self.client.metadata_get(self.deposit_id) except ValueError: logger.error(f"Unknown deposit {self.deposit_id}, ignoring") return {"status": "failed"} # Then usual loading r = super().load() success = r["status"] != "failed" if success: # Update archive with metadata information origin_metadata = self.metadata["origin_metadata"] logger.debug("origin_metadata: %s", origin_metadata) provider = origin_metadata["provider"] assert provider["provider_type"] == "deposit_client" authority = MetadataAuthority( type=MetadataAuthorityType.DEPOSIT_CLIENT, url=provider["provider_url"], metadata={ "name": provider["provider_name"], **(provider["metadata"] or {}), }, ) self.storage.metadata_authority_add([authority]) tool = origin_metadata["tool"] fetcher = MetadataFetcher( name=tool["name"], version=tool["version"], metadata=tool["configuration"], ) self.storage.metadata_fetcher_add([fetcher]) self.storage.object_metadata_add( [ RawExtrinsicMetadata( type=MetadataTargetType.ORIGIN, id=self.url, discovery_date=self.visit_date, authority=authority, fetcher=fetcher, format="sword-v2-atom-codemeta-v2-in-json", metadata=json.dumps(origin_metadata["metadata"]).encode(), ) ] ) # 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"]) branches = self.storage.snapshot_get(snapshot_id)["branches"] logger.debug("branches: %s", branches) if not branches: return r rev_id = branches[b"HEAD"]["target"] revisions = self.storage.revision_get([rev_id]) # FIXME: inconsistency between tests and production code if isinstance(revisions, types.GeneratorType): revisions = list(revisions) revision = revisions[0] # 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), 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/loader.py b/swh/loader/package/loader.py index ed0493c..1cea8d0 100644 --- a/swh/loader/package/loader.py +++ b/swh/loader/package/loader.py @@ -1,501 +1,520 @@ # 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 logging import tempfile import os from typing import ( Any, Dict, Iterator, Generic, List, Mapping, Optional, Sequence, Tuple, TypeVar, ) import attr import sentry_sdk 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, ) from swh.storage import get_storage 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__) @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]) raw = attr.ib(type=Any) + @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 = "" 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 = 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, a_metadata: Dict, uncompressed_path: str, directory: Sha1Git + 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: - a_metadata: Artifact metadata + p_info: Package information uncompressed_path: Artifact uncompressed path on disk Returns: SWH data dict """ 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]: """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 } def resolve_revision_from( - self, known_artifacts: Dict, artifact_metadata: Dict + 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 - artifact_metadata: Information dict + 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 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) return result # Prepare origin and origin_visit origin = Origin(url=self.url) try: self.storage.origin_add([origin]) visit = 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.raw) + revision_id = self.resolve_revision_from(known_artifacts, p_info) if revision_id is None: try: revision_id = self._load_revision(p_info, origin) 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 ) 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 ) 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" return finalize_visit() 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 = 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) # FIXME: This should be release. cf. D409 revision = self.build_revision( - p_info.raw, uncompressed_path, directory=directory.hash + p_info, uncompressed_path, directory=directory.hash ) if not revision: # Some artifacts are missing intrinsic metadata # skipping those return None extra_metadata: Tuple[str, Any] = ( "original_artifact", [hashes for _, hashes in dl_artifacts], ) if revision.metadata is not None: full_metadata = list(revision.metadata.items()) + [extra_metadata] else: full_metadata = [extra_metadata] revision = attr.evolve(revision, metadata=ImmutableDict(full_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 diff --git a/swh/loader/package/nixguix/loader.py b/swh/loader/package/nixguix/loader.py index 6bd252b..6357ef1 100644 --- a/swh/loader/package/nixguix/loader.py +++ b/swh/loader/package/nixguix/loader.py @@ -1,241 +1,253 @@ # 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 json import logging import requests from typing import Any, Dict, Iterator, Mapping, Optional, Tuple import attr from swh.model import hashutil from swh.model.model import ( Revision, RevisionType, TargetType, Snapshot, BaseModel, Sha1Git, ) from swh.loader.package.utils import EMPTY_AUTHOR from swh.loader.package.loader import BasePackageInfo, PackageLoader logger = logging.getLogger(__name__) +@attr.s class NixGuixPackageInfo(BasePackageInfo): raw = 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=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) raw = retrieve_sources(url) clean = clean_sources(raw) 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() # 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( - url=url, filename=None, raw={"url": url, "integrity": integrity}, - ) + p_info = NixGuixPackageInfo.from_metadata({"url": url, "integrity": integrity}) yield url, p_info def known_artifacts(self, snapshot: Optional[Snapshot]) -> Dict[Sha1Git, BaseModel]: """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"] return ret def resolve_revision_from( - self, known_artifacts: Dict, artifact_metadata: Dict + 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 artifact_metadata["integrity"] == known_integrity: + 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, a_metadata: Dict, uncompressed_path: str, directory: Sha1Git + 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": a_metadata, + "raw": p_info.raw, }, }, ) def retrieve_sources(url: str) -> Dict[str, Any]: response = requests.get(url, allow_redirects=True) if response.status_code != 200: raise ValueError("Got %d HTTP code on %s", response.status_code, url) return json.loads(response.content.decode("utf-8")) def clean_sources(sources: Dict[str, Any]) -> Dict[str, Any]: """Validate and clean the sources structure. First, it ensures all top level keys are presents. Then, it walks on the sources list and removes sources that don't contain required keys. Raises: ValueError: if a top level key is missing """ # 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( "sources structure invalid, missing: %s", ",".join(missing_keys) ) # Only the version 1 is currently supported version = int(sources["version"]) if version != 1: raise ValueError( "The sources structure version '%d' is not supported", sources["version"] ) # 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( "Skip source '%s' because key '%s' is missing", source, required_key ) valid = False if valid and source["type"] != "url": logger.info( "Skip source '%s' because the type %s is not supported", source, source["type"], ) valid = False if valid and not isinstance(source["urls"], list): logger.info( "Skip source '%s' because the urls attribute is not a list", source ) valid = False if valid: verified_sources.append(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 7d59584..242410f 100644 --- a/swh/loader/package/nixguix/tests/test_nixguix.py +++ b/swh/loader/package/nixguix/tests/test_nixguix.py @@ -1,532 +1,537 @@ # 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 import logging import pytest from json.decoder import JSONDecodeError from swh.storage.interface import StorageInterface from typing import Dict, Optional, Tuple from unittest.mock import patch from swh.model.model import Snapshot, SnapshotBranch, TargetType from swh.loader.package.archive.loader import ArchiveLoader from swh.loader.package.nixguix.loader import ( + NixGuixPackageInfo, NixGuixLoader, retrieve_sources, clean_sources, ) from swh.loader.package.utils import download from swh.model.hashutil import hash_to_bytes, hash_to_hex from swh.storage.exc import HashCollision 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" 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: metadata = rev["metadata"] raw = metadata["extrinsic"]["raw"] assert "url" in raw assert "integrity" in raw def test_retrieve_sources(swh_config, requests_mock_datadir): j = 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): sources = { "version": 1, "sources": [ # Valid source {"type": "url", "urls": ["my-url"], "integrity": "my-integrity"}, # integrity is missing {"type": "url", "urls": ["my-url"],}, # urls is not a list {"type": "url", "urls": "my-url", "integrity": "my-integrity"}, # type is not url {"type": "git", "urls": ["my-url"], "integrity": "my-integrity"}, # missing fields which got double-checked nonetheless... {"integrity": "my-integrity"}, ], "revision": "my-revision", } clean = clean_sources(sources) assert len(clean["sources"]) == 1 def test_loader_one_visit(swh_config, requests_mock_datadir): 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" ) 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): loader = NixGuixLoader(sources_url) known_artifacts = { "id1": {"extrinsic": {"raw": {"url": "url1", "integrity": "integrity1"}}}, "id2": {"extrinsic": {"raw": {"url": "url2", "integrity": "integrity2"}}}, } - metadata = {"url": "url1", "integrity": "integrity1"} - assert loader.resolve_revision_from(known_artifacts, metadata) == "id1" - metadata = {"url": "url3", "integrity": "integrity3"} - assert loader.resolve_revision_from(known_artifacts, metadata) == None # noqa + 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 = archive_loader.storage.snapshot_get( 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 = loader.storage.snapshot_get(hash_to_bytes(snapshot_id)) snapshot.pop("next_branch") # snapshot_get endpoint detail to drop # 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"]])) # assert that revision is not in the right format assert old_revision["metadata"]["extrinsic"]["raw"].get("integrity", {}) == {} # mutate snapshot to create a clash snapshot["branches"][artifact_url.encode("utf-8")] = { "target_type": "revision", "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.from_dict(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 = loader.storage.snapshot_get(hash_to_bytes(new_snapshot_id)) new_revision_branch = last_snapshot["branches"][artifact_url.encode("utf-8")] assert new_revision_branch["target_type"] == "revision" new_revision = next( loader.storage.revision_get([new_revision_branch["target"]]) ) # 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 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"]), "reason": "'integrity'", "known_artifact": old_revision["metadata"], } diff --git a/swh/loader/package/npm/loader.py b/swh/loader/package/npm/loader.py index bffb2f1..88009a6 100644 --- a/swh/loader/package/npm/loader.py +++ b/swh/loader/package/npm/loader.py @@ -1,300 +1,325 @@ # 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 logging import os from codecs import BOM_UTF8 from typing import Any, Dict, Iterator, List, Optional, Sequence, Tuple, Union import attr import chardet from urllib.parse import quote from swh.model.model import ( Person, RevisionType, Revision, TimestampWithTimezone, Sha1Git, ) from swh.loader.package.loader import BasePackageInfo, PackageLoader from swh.loader.package.utils import api_info, release_name logger = logging.getLogger(__name__) EMPTY_PERSON = Person(fullname=b"", name=None, email=None) +@attr.s class NpmPackageInfo(BasePackageInfo): raw = attr.ib(type=Dict[str, Any]) + date = attr.ib(type=Optional[str]) + shasum = attr.ib(type=str) + """sha1 checksum""" + version = attr.ib(type=str) + + @classmethod + def from_metadata( + cls, project_metadata: Dict[str, Any], version: str + ) -> "NpmPackageInfo": + package_metadata = project_metadata["versions"][version] + url = package_metadata["dist"]["tarball"] + + # No date available in intrinsic metadata: retrieve it from the API + # metadata, using the version number that the API claims this package + # has. + extrinsic_version = package_metadata["version"] + + if "time" in project_metadata: + date = project_metadata["time"][extrinsic_version] + elif "mtime" in package_metadata: + date = package_metadata["mtime"] + else: + date = None + + return cls( + url=url, + filename=os.path.basename(url), + date=date, + shasum=package_metadata["dist"]["shasum"], + version=extrinsic_version, + raw=package_metadata, # FIXME: we're losing some of the project metadata + ) + class NpmLoader(PackageLoader[NpmPackageInfo]): """Load npm origin's artifact releases into swh archive. """ visit_type = "npm" def __init__(self, url: str): """Constructor Args str: origin url (e.g. https://www.npmjs.com/package/) """ super().__init__(url=url) package_name = url.split("https://www.npmjs.com/package/")[1] safe_name = quote(package_name, safe="") self.provider_url = f"https://replicate.npmjs.com/{safe_name}/" self._info: Dict[str, Any] = {} self._versions = None @property def info(self) -> Dict[str, Any]: """Return the project metadata information (fetched from npm registry) """ if not self._info: self._info = api_info(self.provider_url) return self._info def get_versions(self) -> Sequence[str]: return sorted(list(self.info["versions"].keys())) def get_default_version(self) -> str: return self.info["dist-tags"].get("latest", "") - def get_package_info(self, version: str) -> Iterator[Tuple[str, NpmPackageInfo]]: - meta = self.info["versions"][version] - url = meta["dist"]["tarball"] - p_info = NpmPackageInfo(url=url, filename=os.path.basename(url), raw=meta,) + def get_package_info( + self, version: str + ) -> Iterator[Tuple[str, NpmPackageInfo]]: + p_info = NpmPackageInfo.from_metadata( + project_metadata=self.info, version=version + ) yield release_name(version), p_info def resolve_revision_from( - self, known_artifacts: Dict, artifact_metadata: Dict + self, known_artifacts: Dict, p_info: NpmPackageInfo ) -> Optional[bytes]: - return artifact_to_revision_id(known_artifacts, artifact_metadata) + return artifact_to_revision_id(known_artifacts, p_info) def build_revision( - self, a_metadata: Dict, uncompressed_path: str, directory: Sha1Git + self, p_info: NpmPackageInfo, uncompressed_path: str, directory: Sha1Git ) -> Optional[Revision]: i_metadata = extract_intrinsic_metadata(uncompressed_path) if not i_metadata: return None - # from intrinsic metadata author = extract_npm_package_author(i_metadata) message = i_metadata["version"].encode("ascii") - # from extrinsic metadata - - # No date available in intrinsic metadata: retrieve it from the API - # metadata, using the version number that the API claims this package - # has. - extrinsic_version = a_metadata["version"] - - if "time" in self.info: - date = self.info["time"][extrinsic_version] - elif "mtime" in a_metadata: - date = a_metadata["mtime"] - else: - artifact_name = os.path.basename(a_metadata["dist"]["tarball"]) + if p_info.date is None: + url = p_info.url + artifact_name = os.path.basename(url) raise ValueError( "Origin %s: Cannot determine upload time for artifact %s." - % (self.url, artifact_name) + % (p_info.url, artifact_name) ) - date = TimestampWithTimezone.from_iso8601(date) + date = TimestampWithTimezone.from_iso8601(p_info.date) # FIXME: this is to remain bug-compatible with earlier versions: date = attr.evolve(date, timestamp=attr.evolve(date.timestamp, microseconds=0)) r = Revision( type=RevisionType.TAR, message=message, author=author, date=date, committer=author, committer_date=date, parents=(), directory=directory, synthetic=True, metadata={ "intrinsic": {"tool": "package.json", "raw": i_metadata,}, "extrinsic": { "provider": self.provider_url, "when": self.visit_date.isoformat(), - "raw": a_metadata, + "raw": p_info.raw, }, }, ) return r def artifact_to_revision_id( - known_artifacts: Dict, artifact_metadata: Dict + known_artifacts: Dict, p_info: NpmPackageInfo ) -> Optional[bytes]: """Given metadata artifact, solves the associated revision id. The following code allows to deal with 2 metadata formats: - old format sample:: { 'package_source': { 'sha1': '05181c12cd8c22035dd31155656826b85745da37', } } - new format sample:: { 'original_artifact': [{ 'checksums': { 'sha256': '6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec', # noqa ... }, }], ... } """ - shasum = artifact_metadata["dist"]["shasum"] + shasum = p_info.shasum for rev_id, known_artifact in known_artifacts.items(): known_original_artifact = known_artifact.get("original_artifact") if not known_original_artifact: # previous loader-npm version kept original artifact elsewhere known_original_artifact = known_artifact.get("package_source") if not known_original_artifact: continue original_hash = known_original_artifact["sha1"] else: assert isinstance(known_original_artifact, list) original_hash = known_original_artifact[0]["checksums"]["sha1"] if shasum == original_hash: return rev_id return None def _author_str(author_data: Union[Dict, List, str]) -> str: """Parse author from package.json author fields """ if isinstance(author_data, dict): author_str = "" name = author_data.get("name") if name is not None: if isinstance(name, str): author_str += name elif isinstance(name, list): author_str += _author_str(name[0]) if len(name) > 0 else "" email = author_data.get("email") if email is not None: author_str += f" <{email}>" result = author_str elif isinstance(author_data, list): result = _author_str(author_data[0]) if len(author_data) > 0 else "" else: result = author_data return result def extract_npm_package_author(package_json: Dict[str, Any]) -> Person: """ Extract package author from a ``package.json`` file content and return it in swh format. Args: package_json: Dict holding the content of parsed ``package.json`` file Returns: Person """ for author_key in ("author", "authors"): if author_key in package_json: author_data = package_json[author_key] if author_data is None: return EMPTY_PERSON author_str = _author_str(author_data) return Person.from_fullname(author_str.encode()) return EMPTY_PERSON def _lstrip_bom(s, bom=BOM_UTF8): if s.startswith(bom): return s[len(bom) :] else: return s def load_json(json_bytes): """ Try to load JSON from bytes and return a dictionary. First try to decode from utf-8. If the decoding failed, try to detect the encoding and decode again with replace error handling. If JSON is malformed, an empty dictionary will be returned. Args: json_bytes (bytes): binary content of a JSON file Returns: dict: JSON data loaded in a dictionary """ json_data = {} try: json_str = _lstrip_bom(json_bytes).decode("utf-8") except UnicodeDecodeError: encoding = chardet.detect(json_bytes)["encoding"] if encoding: json_str = json_bytes.decode(encoding, "replace") try: json_data = json.loads(json_str) except json.decoder.JSONDecodeError: pass return json_data def extract_intrinsic_metadata(dir_path: str) -> Dict: """Given an uncompressed path holding the pkginfo file, returns a pkginfo parsed structure as a dict. The release artifact contains at their root one folder. For example: $ tar tvf zprint-0.0.6.tar.gz drwxr-xr-x root/root 0 2018-08-22 11:01 zprint-0.0.6/ ... Args: dir_path (str): Path to the uncompressed directory representing a release artifact from npm. Returns: the pkginfo parsed structure as a dict if any or None if none was present. """ # Retrieve the root folder of the archive if not os.path.exists(dir_path): return {} lst = os.listdir(dir_path) if len(lst) == 0: return {} project_dirname = lst[0] package_json_path = os.path.join(dir_path, project_dirname, "package.json") if not os.path.exists(package_json_path): return {} with open(package_json_path, "rb") as package_json_file: package_json_bytes = package_json_file.read() return load_json(package_json_bytes) diff --git a/swh/loader/package/npm/tests/test_npm.py b/swh/loader/package/npm/tests/test_npm.py index 6e02b73..1607bee 100644 --- a/swh/loader/package/npm/tests/test_npm.py +++ b/swh/loader/package/npm/tests/test_npm.py @@ -1,659 +1,659 @@ # 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 from swh.model.model import Person, Snapshot, SnapshotBranch, TargetType 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, ) 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] assert revision is not None check_metadata_paths( 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"]: check_metadata_paths( original_artifact, paths=[("filename", str), ("length", int), ("checksums", dict),], ) def test_npm_loader_first_visit(swh_config, requests_mock_datadir): 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 assert len( list(loader.storage.content_get(_expected_new_contents_first_visit)) ) == 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) 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 loader._info = None # reset loader internal state 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 """ - artifact_metadata = { - "dist": {"shasum": "05181c12cd8c22035dd31155656826b85745da37",}, - } + + 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 """ - artifact_metadata = { - "dist": {"shasum": "05181c12cd8c22035dd31155656826b85745da37",} - } + + 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 """ - artifact_metadata = { - "dist": {"shasum": "05181c12cd8c22035dd31155656826b85745da37",} - } + + 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/loader.py b/swh/loader/package/pypi/loader.py index 7cb8cdd..c3e4680 100644 --- a/swh/loader/package/pypi/loader.py +++ b/swh/loader/package/pypi/loader.py @@ -1,249 +1,264 @@ # Copyright (C) 2019 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 logging from typing import Any, Dict, Iterator, Optional, Sequence, Tuple from urllib.parse import urlparse import attr from pkginfo import UnpackedSDist from swh.model.model import ( Person, Sha1Git, TimestampWithTimezone, Revision, RevisionType, ) from swh.loader.package.loader import BasePackageInfo, PackageLoader from swh.loader.package.utils import api_info, release_name, EMPTY_AUTHOR logger = logging.getLogger(__name__) +@attr.s class PyPIPackageInfo(BasePackageInfo): raw = attr.ib(type=Dict[str, Any]) + comment_text = attr.ib(type=Optional[str]) + sha256 = attr.ib(type=str) + upload_time = attr.ib(type=str) + + @classmethod + def from_metadata(cls, metadata: Dict[str, Any]) -> "PyPIPackageInfo": + return cls( + url=metadata["url"], + filename=metadata["filename"], + raw=metadata, + comment_text=metadata.get("comment_text"), + sha256=metadata["digests"]["sha256"], + upload_time=metadata["upload_time"], + ) + class PyPILoader(PackageLoader[PyPIPackageInfo]): """Load pypi origin's artifact releases into swh archive. """ visit_type = "pypi" def __init__(self, url): super().__init__(url=url) self._info = None self.provider_url = pypi_api_url(self.url) @property def info(self) -> Dict: """Return the project metadata information (fetched from pypi registry) """ if not self._info: self._info = api_info(self.provider_url) return self._info def get_versions(self) -> Sequence[str]: return self.info["releases"].keys() def get_default_version(self) -> str: return self.info["info"]["version"] def get_package_info(self, version: str) -> Iterator[Tuple[str, PyPIPackageInfo]]: res = [] for meta in self.info["releases"][version]: if meta["packagetype"] != "sdist": continue - filename = meta["filename"] - p_info = PyPIPackageInfo(url=meta["url"], filename=filename, raw=meta,) + p_info = PyPIPackageInfo.from_metadata(meta) res.append((version, p_info)) if len(res) == 1: version, p_info = res[0] yield release_name(version), p_info else: for version, p_info in res: yield release_name(version, p_info.filename), p_info def resolve_revision_from( - self, known_artifacts: Dict, artifact_metadata: Dict + self, known_artifacts: Dict, p_info: PyPIPackageInfo ) -> Optional[bytes]: - return artifact_to_revision_id(known_artifacts, artifact_metadata) + return artifact_to_revision_id(known_artifacts, p_info) def build_revision( - self, a_metadata: Dict, uncompressed_path: str, directory: Sha1Git + self, p_info: PyPIPackageInfo, uncompressed_path: str, directory: Sha1Git ) -> Optional[Revision]: i_metadata = extract_intrinsic_metadata(uncompressed_path) if not i_metadata: return None # from intrinsic metadata name = i_metadata["version"] _author = author(i_metadata) # from extrinsic metadata - message = a_metadata.get("comment_text", "") + message = p_info.comment_text or "" message = "%s: %s" % (name, message) if message else name - date = TimestampWithTimezone.from_iso8601(a_metadata["upload_time"]) + date = TimestampWithTimezone.from_iso8601(p_info.upload_time) return Revision( type=RevisionType.TAR, message=message.encode("utf-8"), author=_author, date=date, committer=_author, committer_date=date, parents=(), directory=directory, synthetic=True, metadata={ "intrinsic": {"tool": "PKG-INFO", "raw": i_metadata,}, "extrinsic": { "provider": self.provider_url, "when": self.visit_date.isoformat(), - "raw": a_metadata, + "raw": p_info.raw, }, }, ) def artifact_to_revision_id( - known_artifacts: Dict, artifact_metadata: Dict + known_artifacts: Dict, p_info: PyPIPackageInfo ) -> Optional[bytes]: """Given metadata artifact, solves the associated revision id. The following code allows to deal with 2 metadata formats (column metadata in 'revision') - old format sample:: { 'original_artifact': { 'sha256': '6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec', # noqa ... }, ... } - new format sample:: { 'original_artifact': [{ 'checksums': { 'sha256': '6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec', # noqa ... }, }], ... } """ - sha256 = artifact_metadata["digests"]["sha256"] + sha256 = p_info.sha256 for rev_id, known_artifact in known_artifacts.items(): original_artifact = known_artifact["original_artifact"] if isinstance(original_artifact, dict): # previous loader-pypi version stored metadata as dict original_sha256 = original_artifact["sha256"] if sha256 == original_sha256: return rev_id continue # new pypi loader actually store metadata dict differently... assert isinstance(original_artifact, list) # current loader-pypi stores metadata as list of dict for original_artifact in known_artifact["original_artifact"]: if sha256 == original_artifact["checksums"]["sha256"]: return rev_id return None def pypi_api_url(url: str) -> str: """Compute api url from a project url Args: url (str): PyPI instance's url (e.g: https://pypi.org/project/requests) This deals with correctly transforming the project's api url (e.g https://pypi.org/pypi/requests/json) Returns: api url """ p_url = urlparse(url) project_name = p_url.path.rstrip("/").split("/")[-1] url = "%s://%s/pypi/%s/json" % (p_url.scheme, p_url.netloc, project_name) return url def extract_intrinsic_metadata(dir_path: str) -> Dict: """Given an uncompressed path holding the pkginfo file, returns a pkginfo parsed structure as a dict. The release artifact contains at their root one folder. For example: $ tar tvf zprint-0.0.6.tar.gz drwxr-xr-x root/root 0 2018-08-22 11:01 zprint-0.0.6/ ... Args: dir_path (str): Path to the uncompressed directory representing a release artifact from pypi. Returns: the pkginfo parsed structure as a dict if any or None if none was present. """ # Retrieve the root folder of the archive if not os.path.exists(dir_path): return {} lst = os.listdir(dir_path) if len(lst) != 1: return {} project_dirname = lst[0] pkginfo_path = os.path.join(dir_path, project_dirname, "PKG-INFO") if not os.path.exists(pkginfo_path): return {} pkginfo = UnpackedSDist(pkginfo_path) raw = pkginfo.__dict__ raw.pop("filename") # this gets added with the ondisk location return raw def author(data: Dict) -> Person: """Given a dict of project/release artifact information (coming from PyPI), returns an author subset. Args: data (dict): Representing either artifact information or release information. Returns: swh-model dict representing a person. """ name = data.get("author") email = data.get("author_email") fullname = None # type: Optional[str] if email: fullname = "%s <%s>" % (name, email) else: fullname = name if not fullname: return EMPTY_AUTHOR if name is not None: name = name.encode("utf-8") if email is not None: email = email.encode("utf-8") return Person(fullname=fullname.encode("utf-8"), name=name, email=email) diff --git a/swh/loader/package/pypi/tests/test_pypi.py b/swh/loader/package/pypi/tests/test_pypi.py index 73e92e0..a35f2dc 100644 --- a/swh/loader/package/pypi/tests/test_pypi.py +++ b/swh/loader/package/pypi/tests/test_pypi.py @@ -1,849 +1,829 @@ # 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 from swh.model.model import Person, Snapshot, SnapshotBranch, TargetType 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, ) 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] assert revision is not None check_metadata_paths( 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"]: check_metadata_paths( original_artifact, paths=[("filename", str), ("length", int), ("checksums", dict),], ) 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 loader._info = None visit2_actual_load_status = loader.load() visit2_stats = get_stats(loader.storage) assert visit2_actual_load_status["status"] == "eventful" 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 """ - artifact_metadata = { - "digests": { - "sha256": "6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec", # noqa - }, - } + + 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 """ - artifact_metadata = { - "digests": { - "sha256": "6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec", # noqa - } - } + + 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 """ - artifact_metadata = { - "digests": { - "sha256": "6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec", # noqa - } - } + + 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_to_revision_id_failures(): - with pytest.raises(KeyError, match="sha256"): - artifact_metadata = { - "digests": {}, - } - assert artifact_to_revision_id({}, artifact_metadata) - - with pytest.raises(KeyError, match="digests"): - artifact_metadata = { - "something": "wrong", - } - assert artifact_to_revision_id({}, artifact_metadata) - - 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/package/tests/test_loader.py b/swh/loader/package/tests/test_loader.py index 8cf0de3..5aa7ef1 100644 --- a/swh/loader/package/tests/test_loader.py +++ b/swh/loader/package/tests/test_loader.py @@ -1,38 +1,69 @@ # 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 -from swh.loader.package.loader import PackageLoader +import attr + +from swh.loader.package.loader import BasePackageInfo, PackageLoader class FakeStorage: def origin_add(self, origins): raise ValueError("We refuse to add an origin") def origin_visit_get_latest(self, origin): return None class FakeStorage2(FakeStorage): def origin_add(self, origins): pass def origin_visit_add(self, visits): raise ValueError("We refuse to add an origin visit") def test_loader_origin_visit_failure(swh_config): """Failure to add origin or origin visit should failed immediately """ loader = PackageLoader("some-url") loader.storage = FakeStorage() actual_load_status = loader.load() assert actual_load_status == {"status": "failed"} loader.storage = FakeStorage2() actual_load_status2 = loader.load() assert actual_load_status2 == {"status": "failed"} + + +def test_artifact_identity(): + """Compute primary key should return the right identity + + """ + + @attr.s + class TestPackageInfo(BasePackageInfo): + a = attr.ib() + b = attr.ib() + length = attr.ib() + filename = attr.ib() + version = attr.ib() + + ID_KEYS = ["a", "b"] + + p_info = TestPackageInfo( + url="http://example.org/", + raw={}, + a=1, + b=2, + length=221837, + filename="8sync-0.1.0.tar.gz", + version="0.1.0", + ) + + actual_id = p_info.artifact_identity() + assert actual_id == [1, 2] diff --git a/swh/loader/package/tests/test_utils.py b/swh/loader/package/tests/test_utils.py index 5116ba9..f4827f3 100644 --- a/swh/loader/package/tests/test_utils.py +++ b/swh/loader/package/tests/test_utils.py @@ -1,179 +1,158 @@ # Copyright (C) 2019 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 pytest import swh.loader.package -from swh.loader.package.utils import download, api_info, release_name, artifact_identity +from swh.loader.package.utils import download, api_info, release_name def test_version_generation(): assert ( swh.loader.package.__version__ != "devel" ), "Make sure swh.loader.core is installed (e.g. pip install -e .)" @pytest.mark.fs def test_download_fail_to_download(tmp_path, requests_mock): url = "https://pypi.org/pypi/arrow/json" status_code = 404 requests_mock.get(url, status_code=status_code) with pytest.raises(ValueError) as e: download(url, tmp_path) assert e.value.args[0] == "Fail to query '%s'. Reason: %s" % (url, status_code) @pytest.mark.fs def test_download_ok(tmp_path, requests_mock): """Download without issue should provide filename and hashes""" filename = "requests-0.0.1.tar.gz" url = "https://pypi.org/pypi/requests/%s" % filename data = "this is something" requests_mock.get(url, text=data, headers={"content-length": str(len(data))}) actual_filepath, actual_hashes = download(url, dest=str(tmp_path)) actual_filename = os.path.basename(actual_filepath) assert actual_filename == filename assert actual_hashes["length"] == len(data) assert ( actual_hashes["checksums"]["sha1"] == "fdd1ce606a904b08c816ba84f3125f2af44d92b2" ) # noqa assert ( actual_hashes["checksums"]["sha256"] == "1d9224378d77925d612c9f926eb9fb92850e6551def8328011b6a972323298d5" ) @pytest.mark.fs def test_download_ok_no_header(tmp_path, requests_mock): """Download without issue should provide filename and hashes""" filename = "requests-0.0.1.tar.gz" url = "https://pypi.org/pypi/requests/%s" % filename data = "this is something" requests_mock.get(url, text=data) # no header information actual_filepath, actual_hashes = download(url, dest=str(tmp_path)) actual_filename = os.path.basename(actual_filepath) assert actual_filename == filename assert actual_hashes["length"] == len(data) assert ( actual_hashes["checksums"]["sha1"] == "fdd1ce606a904b08c816ba84f3125f2af44d92b2" ) # noqa assert ( actual_hashes["checksums"]["sha256"] == "1d9224378d77925d612c9f926eb9fb92850e6551def8328011b6a972323298d5" ) @pytest.mark.fs def test_download_ok_with_hashes(tmp_path, requests_mock): """Download without issue should provide filename and hashes""" filename = "requests-0.0.1.tar.gz" url = "https://pypi.org/pypi/requests/%s" % filename data = "this is something" requests_mock.get(url, text=data, headers={"content-length": str(len(data))}) # good hashes for such file good = { "sha1": "fdd1ce606a904b08c816ba84f3125f2af44d92b2", "sha256": "1d9224378d77925d612c9f926eb9fb92850e6551def8328011b6a972323298d5", # noqa } actual_filepath, actual_hashes = download(url, dest=str(tmp_path), hashes=good) actual_filename = os.path.basename(actual_filepath) assert actual_filename == filename assert actual_hashes["length"] == len(data) assert actual_hashes["checksums"]["sha1"] == good["sha1"] assert actual_hashes["checksums"]["sha256"] == good["sha256"] @pytest.mark.fs def test_download_fail_hashes_mismatch(tmp_path, requests_mock): """Mismatch hash after download should raise """ filename = "requests-0.0.1.tar.gz" url = "https://pypi.org/pypi/requests/%s" % filename data = "this is something" requests_mock.get(url, text=data, headers={"content-length": str(len(data))}) # good hashes for such file good = { "sha1": "fdd1ce606a904b08c816ba84f3125f2af44d92b2", "sha256": "1d9224378d77925d612c9f926eb9fb92850e6551def8328011b6a972323298d5", # noqa } for hash_algo in good.keys(): wrong_hash = good[hash_algo].replace("1", "0") expected_hashes = good.copy() expected_hashes[hash_algo] = wrong_hash # set the wrong hash expected_msg = "Failure when fetching %s. " "Checksum mismatched: %s != %s" % ( url, wrong_hash, good[hash_algo], ) with pytest.raises(ValueError, match=expected_msg): download(url, dest=str(tmp_path), hashes=expected_hashes) def test_api_info_failure(requests_mock): """Failure to fetch info/release information should raise""" url = "https://pypi.org/pypi/requests/json" status_code = 400 requests_mock.get(url, status_code=status_code) with pytest.raises(ValueError) as e0: api_info(url) assert e0.value.args[0] == "Fail to query '%s'. Reason: %s" % (url, status_code) def test_api_info(requests_mock): """Fetching json info from pypi project should be ok""" url = "https://pypi.org/pypi/requests/json" requests_mock.get(url, text='{"version": "0.0.1"}') actual_info = api_info(url) assert actual_info == { "version": "0.0.1", } def test_release_name(): for version, filename, expected_release in [ ("0.0.1", None, "releases/0.0.1"), ("0.0.2", "something", "releases/0.0.2/something"), ]: assert release_name(version, filename) == expected_release - - -def test_artifact_identity(): - """Compute primary key should return the right identity - - """ - data = { - "a": 1, - "b": 2, - "length": 221837, - "filename": "8sync-0.1.0.tar.gz", - "version": "0.1.0", - } - - for id_keys, expected_id in [ - (["a", "b"], [1, 2]), - ([], []), - (["a", "key-that-does-not-exist"], [1, None]), - ]: - actual_id = artifact_identity(data, id_keys=id_keys) - assert actual_id == expected_id diff --git a/swh/loader/package/utils.py b/swh/loader/package/utils.py index 949e621..3e69fd8 100644 --- a/swh/loader/package/utils.py +++ b/swh/loader/package/utils.py @@ -1,138 +1,123 @@ # Copyright (C) 2019 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 copy import logging import os import requests -from typing import Any, Dict, List, Mapping, Optional, Sequence, Tuple +from typing import Dict, Optional, Tuple from swh.model.hashutil import MultiHash, HASH_BLOCK_SIZE from swh.model.model import Person from swh.loader.package import DEFAULT_PARAMS logger = logging.getLogger(__name__) DOWNLOAD_HASHES = set(["sha1", "sha256", "length"]) EMPTY_AUTHOR = Person(fullname=b"", name=None, email=None,) def api_info(url: str) -> Dict: """Basic api client to retrieve information on project. This deals with fetching json metadata about pypi projects. Args: url (str): The api url (e.g PyPI, npm, etc...) Raises: ValueError in case of query failures (for some reasons: 404, ...) Returns: The associated response's information dict """ response = requests.get(url, **DEFAULT_PARAMS) if response.status_code != 200: raise ValueError("Fail to query '%s'. Reason: %s" % (url, response.status_code)) return response.json() def download( url: str, dest: str, hashes: Dict = {}, filename: Optional[str] = None, auth: Optional[Tuple[str, str]] = None, ) -> Tuple[str, Dict]: """Download a remote tarball from url, uncompresses and computes swh hashes on it. Args: url: Artifact uri to fetch, uncompress and hash dest: Directory to write the archive to hashes: Dict of expected hashes (key is the hash algo) for the artifact to download (those hashes are expected to be hex string) auth: Optional tuple of login/password (for http authentication service, e.g. deposit) Raises: ValueError in case of any error when fetching/computing (length, checksums mismatched...) Returns: Tuple of local (filepath, hashes of filepath) """ params = copy.deepcopy(DEFAULT_PARAMS) if auth is not None: params["auth"] = auth # so the connection does not hang indefinitely (read/connection timeout) timeout = params.get("timeout", 60) response = requests.get(url, **params, timeout=timeout, stream=True) if response.status_code != 200: raise ValueError("Fail to query '%s'. Reason: %s" % (url, response.status_code)) filename = filename if filename else os.path.basename(url) logger.debug("filename: %s", filename) filepath = os.path.join(dest, filename) logger.debug("filepath: %s", filepath) h = MultiHash(hash_names=DOWNLOAD_HASHES) with open(filepath, "wb") as f: for chunk in response.iter_content(chunk_size=HASH_BLOCK_SIZE): h.update(chunk) f.write(chunk) # Also check the expected hashes if provided if hashes: actual_hashes = h.hexdigest() for algo_hash in hashes.keys(): actual_digest = actual_hashes[algo_hash] expected_digest = hashes[algo_hash] if actual_digest != expected_digest: raise ValueError( "Failure when fetching %s. " "Checksum mismatched: %s != %s" % (url, expected_digest, actual_digest) ) computed_hashes = h.hexdigest() length = computed_hashes.pop("length") extrinsic_metadata = { "length": length, "filename": filename, "checksums": computed_hashes, } logger.debug("extrinsic_metadata", extrinsic_metadata) return filepath, extrinsic_metadata def release_name(version: str, filename: Optional[str] = None) -> str: if filename: return "releases/%s/%s" % (version, filename) return "releases/%s" % version - - -def artifact_identity(d: Mapping[str, Any], id_keys: Sequence[str]) -> List[Any]: - """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 - - """ - return [d.get(k) for k in id_keys]