def test_abs_symlink(tmpdir):
cas_dir = os.path.join(str(tmpdir), "cas")
cas_cache = CASCache(cas_dir,
log_directory=os.path.join(str(tmpdir), "logs"))
try:
d = CasBasedDirectory(cas_cache)
Content_to_check = "two step file"
test_dir = os.path.join(str(tmpdir), "importfrom")
filesys_discription = [
("a", "D", ""),
("a/l", "S", "/target"),
("target", "D", ""),
("target/file", "F", Content_to_check),
]
generate_import_root(test_dir, filesys_discription)
d.import_files(test_dir)
digest = d.descend("a", "l",
follow_symlinks=True).index["file"].get_digest()
with open(cas_cache.objpath(digest)) as fp:
content = fp.read()
assert Content_to_check == content
finally:
cas_cache.release_resources()
def test_bad_symlinks(tmpdir):
cas_dir = os.path.join(str(tmpdir), "cas")
cas_cache = CASCache(cas_dir,
log_directory=os.path.join(str(tmpdir), "logs"))
try:
d = CasBasedDirectory(cas_cache)
test_dir = os.path.join(str(tmpdir), "importfrom")
filesys_discription = [("a", "D", ""), ("a/l", "S", "../target"),
("target", "F", "You got me")]
generate_import_root(test_dir, filesys_discription)
d.import_files(test_dir)
exp_reason = "not-a-directory"
with pytest.raises(VirtualDirectoryError) as error:
d.descend("a", "l", follow_symlinks=True)
assert error.reason == exp_reason
with pytest.raises(VirtualDirectoryError) as error:
d.descend("a", "l")
assert error.reason == exp_reason
with pytest.raises(VirtualDirectoryError) as error:
d.descend("a", "f")
assert error.reason == exp_reason
finally:
cas_cache.release_resources()
def setup_backend(backend_class, tmpdir):
if backend_class == FileBasedDirectory:
path = os.path.join(tmpdir, "vdir")
os.mkdir(path)
yield backend_class(path)
else:
cas_cache = CASCache(os.path.join(tmpdir, "cas"),
log_directory=os.path.join(tmpdir, "logs"))
try:
yield backend_class(cas_cache)
finally:
cas_cache.release_resources()
def get_cache_usage(directory):
cas_cache = CASCache(directory, log_directory=os.path.dirname(directory))
try:
wait = 0.1
for _ in range(0, int(5 / wait)):
used_size = cas_cache.get_cache_usage().used_size
if used_size is not None:
return used_size
time.sleep(wait)
assert False, "Unable to retrieve cache usage"
return None
finally:
cas_cache.release_resources()
def _listing_test(tmpdir, root, generator_function):
cas_cache = CASCache(tmpdir, log_directory=os.path.join(tmpdir, "logs"))
try:
# Create some fake content
generator_function(root, tmpdir)
d = create_new_filedir(root, tmpdir)
filelist = list(d.list_relative_paths())
d2 = create_new_casdir(root, cas_cache, tmpdir)
filelist2 = list(d2.list_relative_paths())
assert filelist == filelist2
finally:
cas_cache.release_resources()
def test_descend(tmpdir):
cas_dir = os.path.join(str(tmpdir), "cas")
cas_cache = CASCache(cas_dir,
log_directory=os.path.join(str(tmpdir), "logs"))
try:
d = CasBasedDirectory(cas_cache)
Content_to_check = "You got me"
test_dir = os.path.join(str(tmpdir), "importfrom")
filesys_discription = [("a", "D", ""), ("a/l", "D", ""),
("a/l/g", "F", Content_to_check)]
generate_import_root(test_dir, filesys_discription)
d.import_files(test_dir)
digest = d.descend("a", "l").index["g"].get_digest()
with open(cas_cache.objpath(digest)) as fp:
content = fp.read()
assert Content_to_check == content
finally:
cas_cache.release_resources()
def test_bad_sym_escape(tmpdir):
cas_dir = os.path.join(str(tmpdir), "cas")
cas_cache = CASCache(cas_dir,
log_directory=os.path.join(str(tmpdir), "logs"))
try:
d = CasBasedDirectory(cas_cache)
test_dir = os.path.join(str(tmpdir), "importfrom")
filesys_discription = [
("jail", "D", ""),
("jail/a", "D", ""),
("jail/a/l", "S", "../../target"),
("target", "D", ""),
("target/file", "F", "two step file"),
]
generate_import_root(test_dir, filesys_discription)
d.import_files(os.path.join(test_dir, "jail"))
with pytest.raises(VirtualDirectoryError) as error:
d.descend("a", "l", follow_symlinks=True)
assert error.reason == "directory-not-found"
finally:
cas_cache.release_resources()
class ArtifactShare(BaseArtifactShare):
def __init__(self, directory, *, quota=None, casd=False, index_only=False):
# The working directory for the artifact share (in case it
# needs to do something outside of its backend's storage folder).
#
self.directory = os.path.abspath(directory)
# The directory the actual repo will be stored in.
#
# Unless this gets more complicated, just use this directly
# in tests as a remote artifact push/pull configuration
#
self.repodir = os.path.join(self.directory, "repo")
os.makedirs(self.repodir)
logdir = os.path.join(self.directory, "logs") if casd else None
self.cas = CASCache(self.repodir, casd=casd, log_directory=logdir)
self.quota = quota
self.index_only = index_only
super().__init__()
def _create_server(self):
return create_server(
self.repodir,
quota=self.quota,
enable_push=True,
index_only=self.index_only,
)
# has_object():
#
# Checks whether the object is present in the share
#
# Args:
# digest (str): The object's digest
#
# Returns:
# (bool): True if the object exists in the share, otherwise false.
def has_object(self, digest):
assert isinstance(digest, remote_execution_pb2.Digest)
object_path = self.cas.objpath(digest)
return os.path.exists(object_path)
def get_artifact_proto(self, artifact_name):
url = urlparse(self.repo)
channel = grpc.insecure_channel("{}:{}".format(url.hostname, url.port))
try:
fetch_service = remote_asset_pb2_grpc.FetchStub(channel)
uri = REMOTE_ASSET_ARTIFACT_URN_TEMPLATE.format(artifact_name)
request = remote_asset_pb2.FetchBlobRequest()
request.uris.append(uri)
try:
response = fetch_service.FetchBlob(request)
except grpc.RpcError as e:
if e.code() == grpc.StatusCode.NOT_FOUND:
return None
raise
if response.status.code != code_pb2.OK:
return None
return response.blob_digest
finally:
channel.close()
def get_source_proto(self, source_name):
url = urlparse(self.repo)
channel = grpc.insecure_channel("{}:{}".format(url.hostname, url.port))
try:
fetch_service = remote_asset_pb2_grpc.FetchStub(channel)
uri = REMOTE_ASSET_SOURCE_URN_TEMPLATE.format(source_name)
request = remote_asset_pb2.FetchDirectoryRequest()
request.uris.append(uri)
try:
response = fetch_service.FetchDirectory(request)
except grpc.RpcError as e:
if e.code() == grpc.StatusCode.NOT_FOUND:
return None
raise
if response.status.code != code_pb2.OK:
return None
return response.root_directory_digest
finally:
channel.close()
def get_cas_files(self, artifact_proto_digest):
reachable = set()
def reachable_dir(digest):
self.cas._reachable_refs_dir(reachable,
digest,
update_mtime=False,
check_exists=True)
try:
artifact_proto_path = self.cas.objpath(artifact_proto_digest)
if not os.path.exists(artifact_proto_path):
return None
artifact_proto = artifact_pb2.Artifact()
try:
with open(artifact_proto_path, "rb") as f:
artifact_proto.ParseFromString(f.read())
except FileNotFoundError:
return None
if str(artifact_proto.files):
reachable_dir(artifact_proto.files)
if str(artifact_proto.buildtree):
reachable_dir(artifact_proto.buildtree)
if str(artifact_proto.public_data):
if not os.path.exists(
self.cas.objpath(artifact_proto.public_data)):
return None
for log_file in artifact_proto.logs:
if not os.path.exists(self.cas.objpath(log_file.digest)):
return None
return artifact_proto.files
except CASError:
return None
except FileNotFoundError:
return None
# has_artifact():
#
# Checks whether the artifact is present in the share
#
# Args:
# artifact_name (str): The composed complete artifact name
#
# Returns:
# (ArtifactProto): artifact digest if the artifact exists in the share, otherwise None.
def get_artifact(self, artifact_name):
artifact_proto = self.get_artifact_proto(artifact_name)
if not artifact_proto:
return None
return self.get_cas_files(artifact_proto)
# close():
#
# Remove the artifact share.
#
def close(self):
super().close()
self.cas.release_resources()
shutil.rmtree(self.directory)
def _import_test(tmpdir,
original,
overlay,
generator_function,
verify_contents=False):
# Skip this test if we do not have support for subsecond precision mtimes
#
if not have_subsecond_mtime(str(tmpdir)):
pytest.skip(
"Filesystem does not support subsecond mtime precision: {}".format(
str(tmpdir)))
cas_cache = CASCache(tmpdir, log_directory=os.path.join(tmpdir, "logs"))
try:
# Create some fake content
generator_function(original, tmpdir)
if original != overlay:
generator_function(overlay, tmpdir)
d = create_new_casdir(original, cas_cache, tmpdir)
duplicate_cas = create_new_casdir(original, cas_cache, tmpdir)
assert duplicate_cas._get_digest().hash == d._get_digest().hash
d2 = create_new_casdir(overlay, cas_cache, tmpdir)
d.import_files(d2, properties=["mtime"])
export_dir = os.path.join(tmpdir,
"output-{}-{}".format(original, overlay))
roundtrip_dir = os.path.join(
tmpdir, "roundtrip-{}-{}".format(original, overlay))
d2.export_files(roundtrip_dir)
d.export_files(export_dir)
if verify_contents:
for item in root_filesets[overlay - 1]:
(path, typename, content) = item
realpath = resolve_symlinks(path, export_dir)
if typename == "F":
if os.path.isdir(realpath) and directory_not_empty(
realpath):
# The file should not have overwritten the directory in this case.
pass
else:
assert os.path.isfile(
realpath
), "{} did not exist in the combined virtual directory".format(
path)
assert file_contents_are(realpath, content)
roundtrip = os.path.join(roundtrip_dir, path)
assert os.path.getmtime(roundtrip) == MTIME
assert os.path.getmtime(realpath) == MTIME
elif typename == "S":
if os.path.isdir(realpath) and directory_not_empty(
realpath):
# The symlink should not have overwritten the directory in this case.
pass
else:
assert os.path.islink(realpath)
assert os.readlink(realpath) == content
elif typename == "D":
# We can't do any more tests than this because it
# depends on things present in the original. Blank
# directories here will be ignored and the original
# left in place.
assert os.path.lexists(realpath)
# Now do the same thing with filebaseddirectories and check the contents match
duplicate_cas.import_files(roundtrip_dir, properties=["mtime"])
assert duplicate_cas._get_digest().hash == d._get_digest().hash
finally:
cas_cache.release_resources()
class ArtifactShare(BaseArtifactShare):
def __init__(self, directory, *, quota=None, casd=False, index_only=False):
# The working directory for the artifact share (in case it
# needs to do something outside of its backend's storage folder).
#
self.directory = os.path.abspath(directory)
# The directory the actual repo will be stored in.
#
# Unless this gets more complicated, just use this directly
# in tests as a remote artifact push/pull configuration
#
self.repodir = os.path.join(self.directory, "repo")
os.makedirs(self.repodir)
self.artifactdir = os.path.join(self.repodir, "artifacts", "refs")
os.makedirs(self.artifactdir)
self.sourcedir = os.path.join(self.repodir, "source_protos")
os.makedirs(self.sourcedir)
logdir = os.path.join(self.directory, "logs") if casd else None
self.cas = CASCache(self.repodir, casd=casd, log_directory=logdir)
self.quota = quota
self.index_only = index_only
super().__init__()
def _create_server(self):
return create_server(self.repodir, quota=self.quota, enable_push=True, index_only=self.index_only,)
# has_object():
#
# Checks whether the object is present in the share
#
# Args:
# digest (str): The object's digest
#
# Returns:
# (bool): True if the object exists in the share, otherwise false.
def has_object(self, digest):
assert isinstance(digest, remote_execution_pb2.Digest)
object_path = self.cas.objpath(digest)
return os.path.exists(object_path)
def get_artifact_proto(self, artifact_name):
artifact_proto = artifact_pb2.Artifact()
artifact_path = os.path.join(self.artifactdir, artifact_name)
try:
with open(artifact_path, "rb") as f:
artifact_proto.ParseFromString(f.read())
except FileNotFoundError:
return None
return artifact_proto
def get_source_proto(self, source_name):
source_proto = source_pb2.Source()
source_path = os.path.join(self.sourcedir, source_name)
try:
with open(source_path, "rb") as f:
source_proto.ParseFromString(f.read())
except FileNotFoundError:
return None
return source_proto
def get_cas_files(self, artifact_proto):
reachable = set()
def reachable_dir(digest):
self.cas._reachable_refs_dir(reachable, digest, update_mtime=False, check_exists=True)
try:
if str(artifact_proto.files):
reachable_dir(artifact_proto.files)
if str(artifact_proto.buildtree):
reachable_dir(artifact_proto.buildtree)
if str(artifact_proto.public_data):
if not os.path.exists(self.cas.objpath(artifact_proto.public_data)):
return None
for log_file in artifact_proto.logs:
if not os.path.exists(self.cas.objpath(log_file.digest)):
return None
return artifact_proto.files
except CASError:
return None
except FileNotFoundError:
return None
# has_artifact():
#
# Checks whether the artifact is present in the share
#
# Args:
# artifact_name (str): The composed complete artifact name
#
# Returns:
# (ArtifactProto): artifact digest if the artifact exists in the share, otherwise None.
def get_artifact(self, artifact_name):
artifact_proto = self.get_artifact_proto(artifact_name)
if not artifact_proto:
return None
return self.get_cas_files(artifact_proto)
# close():
#
# Remove the artifact share.
#
def close(self):
super().close()
self.cas.release_resources()
shutil.rmtree(self.directory)