diff --git a/swh/loader/package/pypi/loader.py b/swh/loader/package/pypi/loader.py
index 650ff7a..a7b6066 100644
--- a/swh/loader/package/pypi/loader.py
+++ b/swh/loader/package/pypi/loader.py
@@ -1,265 +1,287 @@
 # 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 logging
 import json
 import os
 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 (
+    MetadataAuthority,
+    MetadataAuthorityType,
     Person,
     Sha1Git,
     TimestampWithTimezone,
     Revision,
     RevisionType,
 )
 
-from swh.loader.package.loader import BasePackageInfo, PackageLoader
+from swh.loader.package.loader import (
+    BasePackageInfo,
+    PackageLoader,
+    RawExtrinsicMetadataCore,
+)
 from swh.loader.package.utils import api_info, release_name, EMPTY_AUTHOR
 
 logger = logging.getLogger(__name__)
 
 
 @attr.s
 class PyPIPackageInfo(BasePackageInfo):
     raw_info = 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_info=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 = json.loads(api_info(self.provider_url))
+            self._raw_info = api_info(self.provider_url)
+            self._info = json.loads(self._raw_info)
         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_metadata_authority(self):
+        p_url = urlparse(self.url)
+        return MetadataAuthority(
+            type=MetadataAuthorityType.FORGE,
+            url=f"{p_url.scheme}://{p_url.netloc}/",
+            metadata={},
+        )
+
+    def get_extrinsic_snapshot_metadata(self):
+        return [
+            RawExtrinsicMetadataCore(
+                format="pypi-project-json", metadata=self._raw_info,
+            ),
+        ]
+
     def get_package_info(self, version: str) -> Iterator[Tuple[str, PyPIPackageInfo]]:
         res = []
         for meta in self.info["releases"][version]:
             if meta["packagetype"] != "sdist":
                 continue
             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, p_info: PyPIPackageInfo
     ) -> Optional[bytes]:
         return artifact_to_revision_id(known_artifacts, p_info)
 
     def build_revision(
         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 = p_info.comment_text or ""
         message = "%s: %s" % (name, message) if message else name
         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": p_info.raw_info,
                 },
             },
         )
 
 
 def artifact_to_revision_id(
     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 = 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 a35f2dc..c070aab 100644
--- a/swh/loader/package/pypi/tests/test_pypi.py
+++ b/swh/loader/package/pypi/tests/test_pypi.py
@@ -1,829 +1,885 @@
 # Copyright (C) 2019-2020 The Software Heritage developers
 # See the AUTHORS file at the top-level directory of this distribution
 # License: GNU General Public License version 3, or any later version
 # See top-level LICENSE file for more information
 
 import os
 
 from os import path
 
 import pytest
 
 from unittest.mock import patch
 
 from swh.core.tarball import uncompress
 from swh.core.pytest_plugin import requests_mock_datadir_factory
-from swh.model.hashutil import hash_to_bytes
-from swh.model.model import Person, Snapshot, SnapshotBranch, TargetType
+from swh.model.hashutil import hash_to_bytes, hash_to_hex
+from swh.model.identifiers import SWHID
+from swh.model.model import (
+    MetadataAuthority,
+    MetadataAuthorityType,
+    MetadataFetcher,
+    MetadataTargetType,
+    Person,
+    RawExtrinsicMetadata,
+    Snapshot,
+    SnapshotBranch,
+    TargetType,
+)
+from swh.storage.interface import PagedResult
 
+from swh.loader.package import __version__
 from swh.loader.package.pypi.loader import (
     PyPILoader,
     pypi_api_url,
     author,
     extract_intrinsic_metadata,
     artifact_to_revision_id,
 )
 from swh.loader.package.tests.common import check_metadata_paths
 from swh.loader.tests import (
     assert_last_visit_matches,
     check_snapshot,
     get_stats,
 )
 
 
+@pytest.fixture
+def _0805nexter_api_info(datadir) -> bytes:
+    with open(
+        os.path.join(datadir, "https_pypi.org", "pypi_0805nexter_json"), "rb",
+    ) as f:
+        return f.read()
+
+
 def test_author_basic():
     data = {
         "author": "i-am-groot",
         "author_email": "iam@groot.org",
     }
     actual_author = author(data)
 
     expected_author = Person(
         fullname=b"i-am-groot <iam@groot.org>",
         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" <iam@groot.org>", 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_snapshot_metadata(swh_config, requests_mock_datadir, _0805nexter_api_info):
+    url = "https://pypi.org/project/0805nexter"
+    loader = PyPILoader(url)
+
+    actual_load_status = loader.load()
+    assert actual_load_status["status"] == "eventful"
+    assert actual_load_status["snapshot_id"] is not None
+
+    snapshot_swhid = SWHID(
+        object_type="snapshot", object_id=hash_to_hex(actual_load_status["snapshot_id"])
+    )
+    metadata_authority = MetadataAuthority(
+        type=MetadataAuthorityType.FORGE, url="https://pypi.org/",
+    )
+    expected_metadata = [
+        RawExtrinsicMetadata(
+            type=MetadataTargetType.SNAPSHOT,
+            id=snapshot_swhid,
+            authority=metadata_authority,
+            fetcher=MetadataFetcher(
+                name="swh.loader.package.pypi.loader.PyPILoader", version=__version__,
+            ),
+            discovery_date=loader.visit_date,
+            format="pypi-project-json",
+            metadata=_0805nexter_api_info,
+            origin=url,
+        )
+    ]
+    assert loader.storage.raw_extrinsic_metadata_get(
+        type=MetadataTargetType.SNAPSHOT,
+        id=snapshot_swhid,
+        authority=metadata_authority,
+    ) == PagedResult(next_page_token=None, results=expected_metadata,)
+
+
 def test_visit_with_missing_artifact(swh_config, requests_mock_datadir_missing_one):
     """Load a pypi project with some missing artifacts ends up with 1 snapshot
 
     """
     url = "https://pypi.org/project/0805nexter"
     loader = PyPILoader(url)
 
     actual_load_status = loader.load()
     expected_snapshot_id = hash_to_bytes("dd0e4201a232b1c104433741dbf45895b8ac9355")
     assert actual_load_status == {
         "status": "eventful",
         "snapshot_id": expected_snapshot_id.hex(),
     }
 
     stats = get_stats(loader.storage)
 
     assert {
         "content": 3,
         "directory": 2,
         "origin": 1,
         "origin_visit": 1,
         "person": 1,
         "release": 0,
         "revision": 1,
         "skipped_content": 0,
         "snapshot": 1,
     } == stats
 
     expected_contents = map(
         hash_to_bytes,
         [
             "405859113963cb7a797642b45f171d6360425d16",
             "e5686aa568fdb1d19d7f1329267082fe40482d31",
             "83ecf6ec1114fd260ca7a833a2d165e71258c338",
         ],
     )
 
     assert list(loader.storage.content_missing_per_sha1(expected_contents)) == []
 
     expected_dirs = map(
         hash_to_bytes,
         [
             "b178b66bd22383d5f16f4f5c923d39ca798861b4",
             "c3a58f8b57433a4b56caaa5033ae2e0931405338",
         ],
     )
 
     assert list(loader.storage.directory_missing(expected_dirs)) == []
 
     # {revision hash: directory hash}
     expected_revs = {
         hash_to_bytes("e445da4da22b31bfebb6ffc4383dbf839a074d21"): hash_to_bytes(
             "b178b66bd22383d5f16f4f5c923d39ca798861b4"
         ),  # noqa
     }
     assert list(loader.storage.revision_missing(expected_revs)) == []
 
     expected_snapshot = Snapshot(
         id=hash_to_bytes(expected_snapshot_id),
         branches={
             b"releases/1.2.0": SnapshotBranch(
                 target=hash_to_bytes("e445da4da22b31bfebb6ffc4383dbf839a074d21"),
                 target_type=TargetType.REVISION,
             ),
             b"HEAD": SnapshotBranch(
                 target=b"releases/1.2.0", target_type=TargetType.ALIAS,
             ),
         },
     )
     check_snapshot(expected_snapshot, storage=loader.storage)
 
     assert_last_visit_matches(
         loader.storage,
         url,
         status="partial",
         type="pypi",
         snapshot=expected_snapshot_id,
     )
 
 
 def test_visit_with_1_release_artifact(swh_config, requests_mock_datadir):
     """With no prior visit, load a pypi project ends up with 1 snapshot
 
     """
     url = "https://pypi.org/project/0805nexter"
     loader = PyPILoader(url)
 
     actual_load_status = loader.load()
     expected_snapshot_id = hash_to_bytes("ba6e158ada75d0b3cfb209ffdf6daa4ed34a227a")
     assert actual_load_status == {
         "status": "eventful",
         "snapshot_id": expected_snapshot_id.hex(),
     }
 
     stats = get_stats(loader.storage)
     assert {
         "content": 6,
         "directory": 4,
         "origin": 1,
         "origin_visit": 1,
         "person": 1,
         "release": 0,
         "revision": 2,
         "skipped_content": 0,
         "snapshot": 1,
     } == stats
 
     expected_contents = map(
         hash_to_bytes,
         [
             "a61e24cdfdab3bb7817f6be85d37a3e666b34566",
             "938c33483285fd8ad57f15497f538320df82aeb8",
             "a27576d60e08c94a05006d2e6d540c0fdb5f38c8",
             "405859113963cb7a797642b45f171d6360425d16",
             "e5686aa568fdb1d19d7f1329267082fe40482d31",
             "83ecf6ec1114fd260ca7a833a2d165e71258c338",
         ],
     )
 
     assert list(loader.storage.content_missing_per_sha1(expected_contents)) == []
 
     expected_dirs = map(
         hash_to_bytes,
         [
             "05219ba38bc542d4345d5638af1ed56c7d43ca7d",
             "cf019eb456cf6f78d8c4674596f1c9a97ece8f44",
             "b178b66bd22383d5f16f4f5c923d39ca798861b4",
             "c3a58f8b57433a4b56caaa5033ae2e0931405338",
         ],
     )
 
     assert list(loader.storage.directory_missing(expected_dirs)) == []
 
     # {revision hash: directory hash}
     expected_revs = {
         hash_to_bytes("4c99891f93b81450385777235a37b5e966dd1571"): hash_to_bytes(
             "05219ba38bc542d4345d5638af1ed56c7d43ca7d"
         ),  # noqa
         hash_to_bytes("e445da4da22b31bfebb6ffc4383dbf839a074d21"): hash_to_bytes(
             "b178b66bd22383d5f16f4f5c923d39ca798861b4"
         ),  # noqa
     }
     assert list(loader.storage.revision_missing(expected_revs)) == []
 
     expected_snapshot = Snapshot(
         id=expected_snapshot_id,
         branches={
             b"releases/1.1.0": SnapshotBranch(
                 target=hash_to_bytes("4c99891f93b81450385777235a37b5e966dd1571"),
                 target_type=TargetType.REVISION,
             ),
             b"releases/1.2.0": SnapshotBranch(
                 target=hash_to_bytes("e445da4da22b31bfebb6ffc4383dbf839a074d21"),
                 target_type=TargetType.REVISION,
             ),
             b"HEAD": SnapshotBranch(
                 target=b"releases/1.2.0", target_type=TargetType.ALIAS,
             ),
         },
     )
     check_snapshot(expected_snapshot, loader.storage)
 
     assert_last_visit_matches(
         loader.storage, url, status="full", type="pypi", snapshot=expected_snapshot_id
     )
 
 
 def test_multiple_visits_with_no_change(swh_config, requests_mock_datadir):
     """Multiple visits with no changes results in 1 same snapshot
 
     """
     url = "https://pypi.org/project/0805nexter"
     loader = PyPILoader(url)
 
     actual_load_status = loader.load()
     snapshot_id = hash_to_bytes("ba6e158ada75d0b3cfb209ffdf6daa4ed34a227a")
     assert actual_load_status == {
         "status": "eventful",
         "snapshot_id": snapshot_id.hex(),
     }
     assert_last_visit_matches(
         loader.storage, url, status="full", type="pypi", snapshot=snapshot_id
     )
 
     stats = get_stats(loader.storage)
 
     assert {
         "content": 6,
         "directory": 4,
         "origin": 1,
         "origin_visit": 1,
         "person": 1,
         "release": 0,
         "revision": 2,
         "skipped_content": 0,
         "snapshot": 1,
     } == stats
 
     expected_snapshot = Snapshot(
         id=snapshot_id,
         branches={
             b"releases/1.1.0": SnapshotBranch(
                 target=hash_to_bytes("4c99891f93b81450385777235a37b5e966dd1571"),
                 target_type=TargetType.REVISION,
             ),
             b"releases/1.2.0": SnapshotBranch(
                 target=hash_to_bytes("e445da4da22b31bfebb6ffc4383dbf839a074d21"),
                 target_type=TargetType.REVISION,
             ),
             b"HEAD": SnapshotBranch(
                 target=b"releases/1.2.0", target_type=TargetType.ALIAS,
             ),
         },
     )
     check_snapshot(expected_snapshot, loader.storage)
 
     actual_load_status2 = loader.load()
     assert actual_load_status2 == {
         "status": "uneventful",
         "snapshot_id": actual_load_status2["snapshot_id"],
     }
 
     visit_status2 = assert_last_visit_matches(
         loader.storage, url, status="full", type="pypi"
     )
 
     stats2 = get_stats(loader.storage)
     expected_stats2 = stats.copy()
     expected_stats2["origin_visit"] = 1 + 1
     assert expected_stats2 == stats2
 
     # same snapshot
     assert visit_status2.snapshot == snapshot_id
 
 
 def test_incremental_visit(swh_config, requests_mock_datadir_visits):
     """With prior visit, 2nd load will result with a different snapshot
 
     """
     url = "https://pypi.org/project/0805nexter"
     loader = PyPILoader(url)
 
     visit1_actual_load_status = loader.load()
     visit1_stats = get_stats(loader.storage)
     expected_snapshot_id = hash_to_bytes("ba6e158ada75d0b3cfb209ffdf6daa4ed34a227a")
     assert visit1_actual_load_status == {
         "status": "eventful",
         "snapshot_id": expected_snapshot_id.hex(),
     }
 
     assert_last_visit_matches(
         loader.storage, url, status="full", type="pypi", snapshot=expected_snapshot_id
     )
 
     assert {
         "content": 6,
         "directory": 4,
         "origin": 1,
         "origin_visit": 1,
         "person": 1,
         "release": 0,
         "revision": 2,
         "skipped_content": 0,
         "snapshot": 1,
     } == visit1_stats
 
     # Reset internal state
     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
 
     """
 
     class artifact_metadata:
         sha256 = "6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec"
 
     assert artifact_to_revision_id({}, artifact_metadata) is None
 
     known_artifacts = {
         "b11ebac8c9d0c9e5063a2df693a18e3aba4b2f92": {
             "original_artifact": {"sha256": "something-irrelevant",},
         },
     }
 
     assert artifact_to_revision_id(known_artifacts, artifact_metadata) is None
 
 
 def test_pypi_artifact_to_revision_id_old_loader_version():
     """Current loader version should solve old metadata scheme
 
     """
 
     class artifact_metadata:
         sha256 = "6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec"
 
     known_artifacts = {
         hash_to_bytes("b11ebac8c9d0c9e5063a2df693a18e3aba4b2f92"): {
             "original_artifact": {"sha256": "something-wrong",},
         },
         hash_to_bytes("845673bfe8cbd31b1eaf757745a964137e6f9116"): {
             "original_artifact": {
                 "sha256": "6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec",  # noqa
             },
         },
     }
 
     assert artifact_to_revision_id(known_artifacts, artifact_metadata) == hash_to_bytes(
         "845673bfe8cbd31b1eaf757745a964137e6f9116"
     )
 
 
 def test_pypi_artifact_to_revision_id_current_loader_version():
     """Current loader version should be able to solve current metadata scheme
 
     """
 
     class artifact_metadata:
         sha256 = "6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec"
 
     known_artifacts = {
         hash_to_bytes("b11ebac8c9d0c9e5063a2df693a18e3aba4b2f92"): {
             "original_artifact": [
                 {
                     "checksums": {
                         "sha256": "6975816f2c5ad4046acc676ba112f2fff945b01522d63948531f11f11e0892ec",  # noqa
                     },
                 }
             ],
         },
         hash_to_bytes("845673bfe8cbd31b1eaf757745a964137e6f9116"): {
             "original_artifact": [{"checksums": {"sha256": "something-wrong"},}],
         },
     }
 
     assert artifact_to_revision_id(known_artifacts, artifact_metadata) == hash_to_bytes(
         "b11ebac8c9d0c9e5063a2df693a18e3aba4b2f92"
     )
 
 
 def test_pypi_artifact_with_no_intrinsic_metadata(swh_config, requests_mock_datadir):
     """Skip artifact with no intrinsic metadata during ingestion
 
     """
     url = "https://pypi.org/project/upymenu"
     loader = PyPILoader(url)
 
     actual_load_status = loader.load()
     expected_snapshot_id = hash_to_bytes("1a8893e6a86f444e8be8e7bda6cb34fb1735a00e")
     assert actual_load_status == {
         "status": "eventful",
         "snapshot_id": expected_snapshot_id.hex(),
     }
 
     # no branch as one artifact without any intrinsic metadata
     expected_snapshot = Snapshot(id=expected_snapshot_id, branches={})
     check_snapshot(expected_snapshot, loader.storage)
 
     assert_last_visit_matches(
         loader.storage, url, status="full", type="pypi", snapshot=expected_snapshot.id
     )