def test_config_wrong_consistency_should_raise(self):
storage_config = dict(
cls="cassandra",
hosts=["first"],
port=9999,
keyspace="any",
consistency_level="fake",
journal_writer={"cls": "memory"},
objstorage={"cls": "memory"},
)
with pytest.raises(ValueError, match="Unknown consistency"):
get_storage(**storage_config)
def __init__(self):
self.config: Dict[str, Any] = config.load_from_envvar(DEFAULT_CONFIG)
self.scheduler: SchedulerInterface = get_scheduler(
**self.config["scheduler"])
self.tool = {
"name": "swh-deposit",
"version": __version__,
"configuration": {
"sword_version": "2"
},
}
self.storage: StorageInterface = get_storage(**self.config["storage"])
self.storage_metadata: StorageInterface = get_storage(
**self.config["storage_metadata"])
def test_storage_direct_writer_anonymized(
kafka_prefix: str, kafka_server, consumer: Consumer
):
writer_config = {
"cls": "kafka",
"brokers": [kafka_server],
"client_id": "kafka_writer",
"prefix": kafka_prefix,
"anonymize": True,
}
storage_config: Dict[str, Any] = {
"cls": "pipeline",
"steps": [
{"cls": "memory", "journal_writer": writer_config},
],
}
storage = get_storage(**storage_config)
expected_messages = 0
for obj_type, objs in TEST_OBJECTS.items():
if obj_type == "origin_visit":
# these have non-consistent API and are unrelated with what we
# want to test here
continue
method = getattr(storage, obj_type + "_add")
method(objs)
expected_messages += len(objs)
existing_topics = set(
topic
for topic in consumer.list_topics(timeout=10).topics.keys()
if topic.startswith(kafka_prefix)
)
assert existing_topics == {
f"{kafka_prefix}.{obj_type}"
for obj_type in (
"content",
"directory",
"extid",
"metadata_authority",
"metadata_fetcher",
"origin",
"origin_visit",
"origin_visit_status",
"raw_extrinsic_metadata",
"release",
"revision",
"snapshot",
"skipped_content",
)
} | {
f"{kafka_prefix}_privileged.{obj_type}"
for obj_type in (
"release",
"revision",
)
}
def storage():
"""An instance of in-memory storage that gets injected
into the CLI functions."""
storage = get_storage(cls="memory")
with patch("swh.storage.get_storage") as get_storage_mock:
get_storage_mock.return_value = storage
yield storage
def schedule_origin_metadata_index(
ctx, type, options, storage_url, origin_batch_size, page_token, limit, dry_run
):
"""Schedules tasks for origins that are already known.
The first argument is the name of the task type, further ones are
keyword argument(s) of the task in the form key=value, where value is
in YAML format.
Usage sample:
swh-scheduler --database 'service=swh-scheduler' \
task schedule_origins index-origin-metadata
"""
from itertools import islice
from swh.storage import get_storage
from .utils import parse_options, schedule_origin_batches
scheduler = ctx.obj["scheduler"]
storage = get_storage("remote", url=storage_url)
if dry_run:
scheduler = None
(args, kw) = parse_options(options)
if args:
raise click.ClickException("Only keywords arguments are allowed.")
origins = iter_origins(storage, page_token=page_token)
if limit:
origins = islice(origins, limit)
origin_urls = (origin.url for origin in origins)
schedule_origin_batches(scheduler, type, origin_urls, origin_batch_size, kw)
def get_config(config_file="web/web"):
"""Read the configuration file `config_file`.
If an environment variable SWH_CONFIG_FILENAME is defined, this
takes precedence over the config_file parameter.
In any case, update the app with parameters (secret_key, conf)
and return the parsed configuration as a dict.
If no configuration file is provided, return a default
configuration.
"""
if not swhweb_config:
config_filename = os.environ.get("SWH_CONFIG_FILENAME")
if config_filename:
config_file = config_filename
cfg = config.load_named_config(config_file, DEFAULT_CONFIG)
swhweb_config.update(cfg)
config.prepare_folders(swhweb_config, "log_dir")
if swhweb_config.get("search"):
swhweb_config["search"] = get_search(**swhweb_config["search"])
else:
swhweb_config["search"] = None
swhweb_config["storage"] = get_storage(**swhweb_config["storage"])
swhweb_config["vault"] = get_vault(**swhweb_config["vault"])
swhweb_config["indexer_storage"] = get_indexer_storage(
**swhweb_config["indexer_storage"])
swhweb_config["scheduler"] = get_scheduler(
**swhweb_config["scheduler"])
return swhweb_config
def test_revision_extra_header_in_metadata(swh_storage_backend_config, sample_data):
storage = get_storage(**swh_storage_backend_config)
rev = sample_data.revision
md_w_extra = dict(
rev.metadata.items(),
extra_headers=headers_to_db(
[
["gpgsig", b"test123"],
["mergetag", b"foo\\bar"],
["mergetag", b"\x22\xaf\x89\x80\x01\x00"],
]
),
)
bw_rev = attr.evolve(rev, extra_headers=())
object.__setattr__(bw_rev, "metadata", md_w_extra)
assert bw_rev.extra_headers == ()
assert storage.revision_add([bw_rev]) == {"revision:add": 1}
# check data in the db are old format
with db_transaction(storage) as (_, cur):
cur.execute("SELECT metadata, extra_headers FROM revision")
metadata, extra_headers = cur.fetchone()
assert extra_headers == []
assert metadata == bw_rev.metadata
# check the Revision build from revision_get is the original, "new style", Revision
assert storage.revision_get([rev.id]) == [rev]
def test_resolve_object_from_extids_missing_target() -> None:
storage = get_storage("memory")
target = b"\x01" * 20
rel = Release(
name=b"aaaa",
message=b"aaaa",
target=target,
target_type=ModelObjectType.DIRECTORY,
synthetic=False,
)
loader = StubPackageLoader(storage, "http://example.org/")
p_info = Mock(wraps=BasePackageInfo(None, None, None)) # type: ignore
known_extids = {("extid-type", 0, b"extid-of-aaaa"): [rel.swhid()]}
p_info.extid.return_value = ("extid-type", 0, b"extid-of-aaaa")
whitelist = {rel.id}
# Targeted release is missing from the storage
assert loader.resolve_object_from_extids(known_extids, p_info,
whitelist) is None
storage.release_add([rel])
# Targeted release now exists
assert (loader.resolve_object_from_extids(known_extids, p_info,
whitelist) == rel.swhid())
def _partial_copy_storage(old_storage, origin_url: str, mechanism: str,
copy_revisions: bool):
"""Create a new storage, and only copy ExtIDs or head revisions to it."""
new_storage = get_storage(cls="memory")
snapshot = snapshot_get_latest(old_storage, origin_url)
assert snapshot
heads = [branch.target for branch in snapshot.branches.values()]
if mechanism == "extid":
extids = old_storage.extid_get_from_target(ObjectType.REVISION, heads)
new_storage.extid_add(extids)
if copy_revisions:
# copy revisions, but erase their metadata to make sure the loader doesn't
# fallback to revision.metadata["nodeid"]
revisions = [
attr.evolve(rev, metadata={})
for rev in old_storage.revision_get(heads) if rev
]
new_storage.revision_add(revisions)
else:
assert mechanism == "same storage"
return old_storage
# copy origin, visit, status
new_storage.origin_add(old_storage.origin_get([origin_url]))
visit = old_storage.origin_visit_get_latest(origin_url)
new_storage.origin_visit_add([visit])
statuses = old_storage.origin_visit_status_get(origin_url,
visit.visit).results
new_storage.origin_visit_status_add(statuses)
new_storage.snapshot_add([snapshot])
return new_storage
def get_config(config_file='web/web'):
"""Read the configuration file `config_file`.
If an environment variable SWH_CONFIG_FILENAME is defined, this
takes precedence over the config_file parameter.
In any case, update the app with parameters (secret_key, conf)
and return the parsed configuration as a dict.
If no configuration file is provided, return a default
configuration.
"""
if not swhweb_config:
config_filename = os.environ.get('SWH_CONFIG_FILENAME')
if config_filename:
config_file = config_filename
cfg = config.load_named_config(config_file, DEFAULT_CONFIG)
swhweb_config.update(cfg)
config.prepare_folders(swhweb_config, 'log_dir')
swhweb_config['storage'] = get_storage(**swhweb_config['storage'])
swhweb_config['vault'] = get_vault(**swhweb_config['vault'])
swhweb_config['indexer_storage'] = \
get_indexer_storage(**swhweb_config['indexer_storage'])
swhweb_config['scheduler'] = get_scheduler(
**swhweb_config['scheduler'])
return swhweb_config
def __init__(self, config):
self.config = config
self.storage = get_storage(config['storage_class'],
config['storage_args'])
self.log = logging.getLogger(
'swh.antelink.loader.AntelinkSesiInjecter')
def __init__(self, config):
self.config = config
self.storage = get_storage(config['storage_class'],
config['storage_args'])
self.log = logging.getLogger(
'swh.antelink.loader.AntelinkSesiInjecter')
def storage(swh_indexer_config):
"""An instance of in-memory storage that gets injected into all indexers
classes.
"""
storage = get_storage(**swh_indexer_config["storage"])
fill_storage(storage)
return storage
def test_config_consistency_used(self, swh_storage_backend_config):
config_with_consistency = dict(swh_storage_backend_config,
**{"consistency_level": "THREE"})
storage = get_storage(**config_with_consistency)
with pytest.raises(NoHostAvailable):
storage.content_get_random()
def get_storage_with_buffer_config(**buffer_config) -> BufferingProxyStorage:
steps = [
{"cls": "buffer", **buffer_config},
{"cls": "memory"},
]
ret = get_storage("pipeline", steps=steps)
assert isinstance(ret, BufferingProxyStorage)
return ret
def __init__(self, storage):
self.storage: StorageInterface = get_storage(**storage)
for attribute_name in dir(StorageInterface):
if attribute_name.startswith("_"):
continue
attribute = getattr(self.storage, attribute_name)
if hasattr(attribute, "__call__"):
setattr(self, attribute_name,
retry_function(self.storage, attribute_name))
def test_resolve_object_from_extids() -> None:
storage = get_storage("memory")
target = b"\x01" * 20
rel1 = Release(
name=b"aaaa",
message=b"aaaa",
target=target,
target_type=ModelObjectType.DIRECTORY,
synthetic=False,
)
rel2 = Release(
name=b"bbbb",
message=b"bbbb",
target=target,
target_type=ModelObjectType.DIRECTORY,
synthetic=False,
)
storage.release_add([rel1, rel2])
loader = StubPackageLoader(storage, "http://example.org/")
p_info = Mock(wraps=BasePackageInfo(None, None, None)) # type: ignore
# The PackageInfo does not support extids
p_info.extid.return_value = None
known_extids = {("extid-type", 0, b"extid-of-aaaa"): [rel1.swhid()]}
whitelist = {b"unused"}
assert loader.resolve_object_from_extids(known_extids, p_info,
whitelist) is None
# Some known extid, and the PackageInfo is not one of them (ie. cache miss)
p_info.extid.return_value = ("extid-type", 0, b"extid-of-cccc")
assert loader.resolve_object_from_extids(known_extids, p_info,
whitelist) is None
# Some known extid, and the PackageInfo is one of them (ie. cache hit),
# but the target release was not in the previous snapshot
p_info.extid.return_value = ("extid-type", 0, b"extid-of-aaaa")
assert loader.resolve_object_from_extids(known_extids, p_info,
whitelist) is None
# Some known extid, and the PackageInfo is one of them (ie. cache hit),
# and the target release was in the previous snapshot
whitelist = {rel1.id}
assert (loader.resolve_object_from_extids(known_extids, p_info,
whitelist) == rel1.swhid())
# Same as before, but there is more than one extid, and only one is an allowed
# release
whitelist = {rel1.id}
known_extids = {
("extid-type", 0, b"extid-of-aaaa"): [rel2.swhid(),
rel1.swhid()]
}
assert (loader.resolve_object_from_extids(known_extids, p_info,
whitelist) == rel1.swhid())
def main(origin_url: str, incremental: bool) -> Dict[str, Any]:
from swh.storage import get_storage
storage = get_storage(cls="memory")
loader = GitLoader(
storage,
origin_url,
incremental=incremental,
)
return loader.load()
def swh_storage():
storage_config = {
"cls": "pipeline",
"steps": [
{"cls": "validate"},
{"cls": "memory"},
],
}
return get_storage(**storage_config)
def __init__(self, config):
self.config = config
s3_folder = self.config['s3_folder']
if not s3_folder.endswith('/'):
self.config['s3_folder'] = s3_folder + '/'
self.storage = get_storage(config['storage_class'],
config['storage_args'])
self.log = logging.getLogger('swh.antelink.loader.AntelinkS3Injecter')
def get_tenacious_storage(**config):
storage_config = {
"cls": "pipeline",
"steps": [
{"cls": "validate"},
{"cls": "tenacious", **config},
{"cls": "memory"},
],
}
return get_storage(**storage_config)
def __init__(self, config):
self.config = config
s3_folder = self.config['s3_folder']
if not s3_folder.endswith('/'):
self.config['s3_folder'] = s3_folder + '/'
self.storage = get_storage(config['storage_class'],
config['storage_args'])
self.log = logging.getLogger(
'swh.antelink.loader.AntelinkS3Injecter')
def __init__(self) -> None:
self.config = load_from_envvar(DEFAULT_CONFIG)
self.storage = get_storage(**self.config["storage"])
self.objstorage = get_objstorage(**self.config["objstorage"])
self.compute_checksums = self.config["compute_checksums"]
self.recompute_checksums = self.config["recompute_checksums"]
self.batch_size_retrieve_content = self.config[
"batch_size_retrieve_content"]
self.batch_size_update = self.config["batch_size_update"]
self.log = logging.getLogger("swh.indexer.rehash")
if not self.compute_checksums:
raise ValueError("Checksums list should not be empty.")
def test_pypi_origin_from_project_name(mocker):
origin_url = "https://pypi.org/project/ProjectName/"
storage = get_storage("memory")
revision_id = b"41" * 10
snapshot_id = b"42" * 10
storage.origin_add([Origin(url=origin_url)])
storage.origin_visit_add(
[OriginVisit(origin=origin_url, visit=1, date=now(), type="pypi")])
storage.origin_visit_status_add([
OriginVisitStatus(
origin=origin_url,
visit=1,
date=now(),
status="partial",
snapshot=snapshot_id,
)
])
storage.snapshot_add([
Snapshot(
id=snapshot_id,
branches={
b"foo":
SnapshotBranch(
target_type=TargetType.REVISION,
target=revision_id,
)
},
)
])
class response:
code = 200
def read(self):
return b'{"info": {"name": "ProjectName"}}'
mock_urlopen = mocker.patch(
"swh.storage.migrate_extrinsic_metadata.urlopen",
return_value=response(),
)
assert (pypi_origin_from_filename(
storage, revision_id, "ProjectName-1.0.0.tar.gz") == origin_url)
mock_urlopen.assert_not_called()
assert (pypi_origin_from_filename(
storage, revision_id, "projectname-1.0.0.tar.gz") == origin_url)
mock_urlopen.assert_called_once_with(
"https://pypi.org/pypi/projectname/json/")
def get_cooker(bundle_type: str, swhid: CoreSWHID):
"""Instantiate a cooker class of type bundle_type.
Returns:
Cooker class in charge of cooking the bundle_type with id swhid.
Raises:
ValueError in case of a missing top-level vault key configuration or a storage
key.
EnvironmentError in case the vault configuration reference a non remote class.
"""
if "SWH_CONFIG_FILENAME" in os.environ:
cfg = read_config(os.environ["SWH_CONFIG_FILENAME"], DEFAULT_CONFIG)
else:
cfg = load_named_config(DEFAULT_CONFIG_PATH, DEFAULT_CONFIG)
cooker_cls = get_cooker_cls(bundle_type, swhid.object_type)
cfg = check_config(cfg)
vcfg = cfg["vault"]
storage = get_storage(**vcfg.pop("storage"))
backend = get_vault(**vcfg)
try:
from swh.graph.client import RemoteGraphClient # optional dependency
graph = RemoteGraphClient(
**vcfg["graph"]) if vcfg.get("graph") else None
except ModuleNotFoundError:
if vcfg.get("graph"):
raise EnvironmentError(
"Graph configuration required but module is not installed.")
else:
graph = None
kwargs = {
k: v
for (k, v) in cfg.items()
if k in ("max_bundle_size", "thread_pool_size")
}
return cooker_cls(
swhid,
backend=backend,
storage=storage,
graph=graph,
**kwargs,
)
def __init__(self, **config):
self.config = config
self.cache = VaultCache(**config["cache"])
self.scheduler = get_scheduler(**config["scheduler"])
self.storage = get_storage(**config["storage"])
self.smtp_server = smtplib.SMTP(**config.get("smtp", {}))
db_conn = config["db"]
self._pool = psycopg2.pool.ThreadedConnectionPool(
config.get("min_pool_conns", 1),
config.get("max_pool_conns", 10),
db_conn,
cursor_factory=psycopg2.extras.RealDictCursor,
)
self._db = None
def __init__(
self,
storage,
error_rate_limit: Optional[Dict[str, int]] = None,
retries: int = 3,
):
self.storage = get_storage(**storage)
if error_rate_limit is None:
error_rate_limit = {"errors": 10, "window_size": 1000}
assert "errors" in error_rate_limit
assert "window_size" in error_rate_limit
self.rate_queue = RateQueue(
size=error_rate_limit["window_size"],
max_errors=error_rate_limit["errors"],
)
self._single_object_retries: int = retries
def test_storage_replayer_with_validation_nok_raises(
replayer_storage_and_client, caplog, redisdb):
"""Replayer scenario with invalid objects
with raise_on_error set to True
This:
- writes both valid & invalid objects to a source storage
- a StorageArgumentException should be raised while replayer consumes
objects from the topic and replays them
"""
src, replayer = replayer_storage_and_client
replayer.value_deserializer = ModelObjectDeserializer(
validate=True, reporter=redisdb.set, raise_on_error=True).convert
caplog.set_level(logging.ERROR, "swh.journal.replay")
# Fill Kafka using a source storage
nb_sent = 0
for object_type, objects in TEST_OBJECTS.items():
method = getattr(src, object_type + "_add")
method(objects)
if object_type == "origin_visit":
nb_sent += len(
objects) # origin-visit-add adds origin-visit-status as well
nb_sent += len(objects)
# insert invalid objects
for object_type in ("revision", "directory", "release", "snapshot"):
method = getattr(src, object_type + "_add")
method([attr.evolve(TEST_OBJECTS[object_type][0], id=b"\x00" * 20)])
nb_sent += 1
# Fill the destination storage from Kafka
dst = get_storage(cls="memory")
worker_fn = functools.partial(process_replay_objects, storage=dst)
with pytest.raises(StorageArgumentException):
replayer.process(worker_fn)
# check we do have invalid objects reported
invalid = 0
for record in caplog.records:
logtext = record.getMessage()
if WRONG_ID_REG.match(logtext):
invalid += 1
assert invalid == 1, "One invalid objects should be detected"
assert len(redisdb.keys()) == 1
def __init__(self, storage: Mapping, min_batch_size: Mapping = {}):
self.storage: StorageInterface = get_storage(**storage)
self._buffer_thresholds = {
**DEFAULT_BUFFER_THRESHOLDS,
**min_batch_size
}
self._objects: Dict[LObjectType,
Dict[Tuple[str, ...],
BaseModel]] = {k: {}
for k in OBJECT_TYPES}
self._contents_size: int = 0
self._directory_entries: int = 0
self._revision_parents: int = 0
self._revision_size: int = 0
self._release_size: int = 0
def swh_storage():
storage_config = {
"cls": "pipeline",
"steps": [
{"cls": "tenacious"},
{
"cls": "memory",
"journal_writer": {
"cls": "memory",
},
},
],
}
storage = get_storage(**storage_config)
storage.journal_writer = storage.storage.journal_writer
return storage
def test_load_get_known_extids() -> None:
"""Checks PackageLoader.load() fetches known extids efficiently"""
storage = Mock(wraps=get_storage("memory"))
loader = StubPackageLoader(storage, "http://example.org")
loader.load()
# Calls should be grouped by extid type
storage.extid_get_from_extid.assert_has_calls(
[
call("extid-type1", [b"extid-of-v1.0", b"extid-of-v2.0"],
version=0),
call("extid-type2", [b"extid-of-v3.0", b"extid-of-v4.0"],
version=0),
],
any_order=True,
)
def swh_storage():
storage_config = {
"cls":
"pipeline",
"steps": [
{
"cls": "counter",
"counters": {
"cls": "memory"
}
},
{
"cls": "memory"
},
],
}
return get_storage(**storage_config)
def send_jobs(db_url, block_size, block_max_files, limit, dry_run, huge,
storage_class, storage_args):
"""Send paths for worker to retrieve from sesi machine.
"""
from swh.scheduler.celery_backend.config import app
from swh.loader.antelink import tasks # noqa
# right inputs
if isinstance(block_size, str):
block_size = int(block_size)
if isinstance(block_max_files, str):
block_max_files = int(block_max_files)
if limit and isinstance(limit, str):
limit = int(limit)
if dry_run:
print('** DRY RUN **')
if db_url:
store = storage.Storage(db_url)
gen_data = store.read_content_sesi_not_in_swh(huge, limit)
else:
gen_data = utils.gen_path_length_from_stdin()
if huge:
task_name = 'swh.loader.antelink.tasks.AntelinkSesiInjecterHugeTsk'
else:
task_name = 'swh.loader.antelink.tasks.AntelinkSesiInjecterTsk'
swhstorage = get_storage(storage_class, storage_args.split(','))
gen_data = retrieve_unknown_sha1s(swhstorage, gen_data)
nb_total_blocks = 0
for paths, size in utils.split_data_per_size(gen_data, block_size,
block_max_files):
nb_total_blocks += 1
print('%s paths (%s bytes) sent.' % (len(paths), size))
if dry_run:
continue
app.tasks[task_name].delay(paths)
print('Number of jobs: %s' % nb_total_blocks)
def compute_s3_jobs(db_url, block_size, block_max_files, limit, dry_run,
final, huge, storage_class, storage_args):
from swh.scheduler.celery_backend.config import app
from swh.loader.antelink import tasks # noqa
# right inputs
if isinstance(block_size, str):
block_size = int(block_size)
if isinstance(block_max_files, str):
block_max_files = int(block_max_files)
if limit and isinstance(limit, str):
limit = int(limit)
if dry_run:
print('** DRY RUN **')
swhstorage = get_storage(storage_class, storage_args.split(','))
if db_url:
store = storage.Storage(db_url)
gen_data = retrieve_unknown_sha1s(
swhstorage,
store.read_content_s3_not_in_sesi_nor_in_swh(huge, final, limit))
else:
gen_data = retrieve_unknown_sha1s(swhstorage,
utils.gen_path_length_from_stdin())
nb_total_blocks = 0
for paths, size in utils.split_data_per_size(gen_data, block_size,
block_max_files):
nb_total_blocks += 1
print('%s paths (%s bytes) sent.' % (len(paths), size))
if dry_run:
continue
app.tasks[task_name].delay(paths)
print('Number of jobs: %s' % nb_total_blocks)
def run(self, *package_names):
"""Load the history of the given package from snapshot.debian.org"""
config = self.config
snapshot = SnapshotDebianOrg(
connstr=config['snapshot_connstr'],
basedir=config['snapshot_basedir'],
)
storage = get_storage(
config['storage_class'],
config['storage_args'],
)
swh_authority_dt = open(
os.path.join(config['snapshot_basedir'], 'TIMESTAMP')
).read()
swh_authority = {
'authority': '5f4d4c51-498a-4e28-88b3-b3e4e8396cba',
'validity': dateutil.parser.parse(swh_authority_dt),
}
tmpdir = tempfile.mkdtemp()
os.makedirs(os.path.join(tmpdir, 'source'))
pkgs = snapshot.prepare_packages(
package_names,
os.path.join(tmpdir, 'source'),
log=self.log,
)
origins = snapshot.prepare_origins(package_names, storage)
closed = False
fetch_histories = {}
for origin in origins.values():
id = origin['id']
fetch_histories[id] = storage.fetch_history_start(id)
try:
sorted_pkgs = []
for p in pkgs.values():
p['origin_id'] = origins[p['name']]['id']
sorted_pkgs.append(p)
sorted_pkgs.sort(key=lambda p: (p['name'], p['version']))
partial = {}
for partial in process_source_packages(
sorted_pkgs,
config['keyrings'],
tmpdir,
log=self.log,
):
try_flush_partial(
storage, partial,
content_packet_size=config['content_packet_size'],
content_packet_length=config['content_packet_length'],
content_max_length_one=config['content_max_length_one'],
directory_packet_size=config['directory_packet_size'],
log=self.log,
)
if partial:
try_flush_partial(
storage, partial,
content_packet_size=config['content_packet_size'],
content_packet_length=config['content_packet_length'],
content_max_length_one=config['content_max_length_one'],
directory_packet_size=config['directory_packet_size'],
force=True,
log=self.log,
)
packages = flush_revision(storage, partial, log=self.log)
packages_w_revs = flush_release(
storage,
packages,
log=self.log
)
flush_occurrences(storage, packages_w_revs, [swh_authority],
log=self.log)
for fh in fetch_histories.values():
storage.fetch_history_end(fh, {'status': True})
closed = True
finally:
shutil.rmtree(tmpdir)
if not closed:
data = {
'status': False,
'stderr': traceback.format_exc(),
}
for fh in fetch_histories.values():
storage.fetch_history_end(fh, data)
