def test_data_fetcher_shared_types(self, mock_ws):
df = DataFetcher(self.cfg["workspace-url"],
self.cfg["auth-service-url"],
self.get_context()["token"])
# 1. shared data, default options, include Narratives, but only return KBaseModule.SomeType
shared_data = df.fetch_accessible_data({
"data_set":
"shared",
"ignore_narratives":
0,
"types": ["KBaseModule.SomeType"]
})
self.assertEqual(len(shared_data["objects"]), 36)
for obj in shared_data["objects"]:
self._validate_obj(obj, "KBaseModule.SomeType")
self._validate_ws_display(shared_data["workspace_display"], 9)
self.assertNotIn("type_counts", shared_data)
# 2. shared data, default options, include Narratives, but only return KBaseNarrative.Narrative
shared_data = df.fetch_accessible_data({
"data_set":
"shared",
"ignore_narratives":
0,
"types": ["KBaseNarrative.Narrative"]
})
self.assertEqual(len(shared_data["objects"]), 4)
for obj in shared_data["objects"]:
self._validate_obj(obj, "KBaseNarrative.Narrative-4.0")
self._validate_ws_display(shared_data["workspace_display"], 1)
self.assertNotIn("type_counts", shared_data)
def list_all_data(self, ctx, params):
"""
This is intended to support the Narrative front end. It returns all data a user
owns, or is shared with them, excluding global data.
:param params: instance of type "ListAllDataParams" (data_set -
should be one of "mine", "shared" - other values with throw an
error include_type_counts - (default 0) if 1, will populate the
list of types with the count of each data type simple_types -
(default 0) if 1, will "simplify" types to just their subtype
(KBaseGenomes.Genome -> Genome) ignore_narratives - (default 1) if
1, won't return any KBaseNarrative.* objects include_metadata -
(default 0) if 1, includes object metadata ignore_workspaces -
(optional) list of workspace ids - if present, will ignore any
workspace ids given (useful for skipping the currently loaded
Narrative)) -> structure: parameter "data_set" of String,
parameter "include_type_counts" of type "boolean" (@range [0,1]),
parameter "simple_types" of type "boolean" (@range [0,1]),
parameter "ignore_narratives" of type "boolean" (@range [0,1]),
parameter "include_metadata" of type "boolean" (@range [0,1]),
parameter "ignore_workspaces" of list of Long
:returns: instance of type "ListDataResult" (objects - list of
objects returned by this function type_counts - mapping of type ->
count in this function call. If simple_types was 1, these types
are all the "simple" format (Genome vs KBaseGenomes.Genome)
workspace_display - handy thing for quickly displaying Narrative
info.) -> structure: parameter "objects" of list of type
"DataObjectView" (upa - the UPA for the most recent version of the
object (wsid/objid/ver format) name - the string name for the
object narr_name - the name of the Narrative that the object came
from type - the type of object this is (if simple_types was used,
then this will be the simple type) savedate - the timestamp this
object was saved saved_by - the user id who saved this object) ->
structure: parameter "upa" of String, parameter "name" of String,
parameter "narr_name" of String, parameter "type" of String,
parameter "savedate" of Long, parameter "saved_by" of String,
parameter "type_counts" of mapping from String to Long, parameter
"workspace_display" of mapping from Long to type "WorkspaceStats"
(display - the display name for the workspace (typically the
Narrative name) count - the number of objects found in the
workspace (excluding Narratives, if requested)) -> structure:
parameter "display" of String, parameter "count" of Long
"""
# ctx is the context object
# return variables are: result
#BEGIN list_all_data
auth_url = self.config["auth-service-url"]
fetcher = DataFetcher(self.workspaceURL, auth_url, ctx["token"])
result = fetcher.fetch_accessible_data(params)
#END list_all_data
# At some point might do deeper type checking...
if not isinstance(result, dict):
raise ValueError('Method list_all_data return value ' +
'result is not type dict as required.')
# return the results
return [result]
def test_data_fetcher_limit(self, mock_ws):
df = DataFetcher(self.cfg["workspace-url"],
self.cfg["auth-service-url"],
self.get_context()["token"])
# Get my data, but limit it.
limit = 19
data = df.fetch_accessible_data({"data_set": "mine", "limit": limit})
self.assertEqual(data["limit_reached"], 1)
self.assertEqual(len(data["objects"]), limit)
def list_workspace_data(self, ctx, params):
"""
Also intended to support the Narrative front end. It returns data from a list of
workspaces. If the authenticated user doesn't have access to any of workspaces, it raises
an exception. Otherwise, it returns the same structure of results as list_all_data.
:param params: instance of type "ListWorkspaceDataParams"
(workspace_ids - list of workspace ids - will only return info
from these workspaces. include_type_counts - (default 0) if 1,
will populate the list of types with the count of each data type
simple_types - (default 0) if 1, will "simplify" types to just
their subtype (KBaseGenomes.Genome -> Genome) ignore_narratives -
(default 1) if 1, won't return any KBaseNarrative.* objects
include_metadata - (default 0) if 1, includes object metadata) ->
structure: parameter "workspace_ids" of list of Long, parameter
"include_type_counts" of type "boolean" (@range [0,1]), parameter
"simple_types" of type "boolean" (@range [0,1]), parameter
"ignore_narratives" of type "boolean" (@range [0,1]), parameter
"include_metadata" of type "boolean" (@range [0,1])
:returns: instance of type "ListDataResult" (objects - list of
objects returned by this function type_counts - mapping of type ->
count in this function call. If simple_types was 1, these types
are all the "simple" format (Genome vs KBaseGenomes.Genome)
workspace_display - handy thing for quickly displaying Narrative
info.) -> structure: parameter "objects" of list of type
"DataObjectView" (upa - the UPA for the most recent version of the
object (wsid/objid/ver format) name - the string name for the
object narr_name - the name of the Narrative that the object came
from type - the type of object this is (if simple_types was used,
then this will be the simple type) savedate - the timestamp this
object was saved saved_by - the user id who saved this object) ->
structure: parameter "upa" of String, parameter "name" of String,
parameter "narr_name" of String, parameter "type" of String,
parameter "savedate" of Long, parameter "saved_by" of String,
parameter "type_counts" of mapping from String to Long, parameter
"workspace_display" of mapping from Long to type "WorkspaceStats"
(display - the display name for the workspace (typically the
Narrative name) count - the number of objects found in the
workspace (excluding Narratives, if requested)) -> structure:
parameter "display" of String, parameter "count" of Long
"""
# ctx is the context object
# return variables are: result
#BEGIN list_workspace_data
auth_url = self.config["auth-service-url"]
fetcher = DataFetcher(self.workspaceURL, auth_url, ctx["token"])
result = fetcher.fetch_specific_workspace_data(params)
#END list_workspace_data
# At some point might do deeper type checking...
if not isinstance(result, dict):
raise ValueError('Method list_workspace_data return value ' +
'result is not type dict as required.')
# return the results
return [result]
def test_data_fetcher_specific_types(self, mock_ws):
df = DataFetcher(self.cfg["workspace-url"],
self.cfg["auth-service-url"],
self.get_context()["token"])
# 1. ws 1,2,3,5, default options, include Narratives, but only return KBaseModule.SomeType
data = df.fetch_specific_workspace_data({
"workspace_ids": [1, 2, 3, 5],
"ignore_narratives":
0,
"types": ["KBaseModule.SomeType"]
})
self.assertEqual(len(data["objects"]), 36)
for obj in data["objects"]:
self._validate_obj(obj, "KBaseModule.SomeType")
self._validate_ws_display(data["workspace_display"], 9)
self.assertNotIn("type_counts", data)
# 2. ws 1,2,3,5, default options, include Narratives, but only return KBaseNarrative.Narrative
data = df.fetch_specific_workspace_data({
"workspace_ids": [1, 2, 3, 5],
"ignore_narratives":
0,
"types": ["KBaseNarrative.Narrative"]
})
self.assertEqual(len(data["objects"]), 4)
for obj in data["objects"]:
self._validate_obj(obj, "KBaseNarrative.Narrative-4.0")
self._validate_ws_display(data["workspace_display"], 1)
self.assertNotIn("type_counts", data)
# 3. ws 1,2,3,5, default options, include Narratives and SomeType, so return everything
data = df.fetch_specific_workspace_data({
"workspace_ids": [1, 2, 3, 5],
"ignore_narratives":
0,
"types": ["KBaseNarrative.Narrative", "KBaseModule.SomeType"]
})
self.assertEqual(len(data["objects"]), 40)
for obj in data["objects"]:
if obj["obj_id"] == 1:
self._validate_obj(obj, "KBaseNarrative.Narrative-4.0")
else:
self._validate_obj(obj, "KBaseModule.SomeType")
self._validate_ws_display(data["workspace_display"], 10)
self.assertNotIn("type_counts", data)
def test_fetch_specific_bad_params(self):
df = DataFetcher(self.cfg["workspace-url"],
self.cfg["auth-service-url"],
self.get_context()["token"])
with self.assertRaises(ValueError) as err:
df.fetch_specific_workspace_data({"workspace_ids": "foo"})
self.assertIn("Parameter 'workspace_ids' must be a list of integers.",
str(err.exception))
with self.assertRaises(ValueError) as err:
df.fetch_specific_workspace_data({"workspace_ids": []})
self.assertIn("Parameter 'workspace_ids' must be a list of integers.",
str(err.exception))
with self.assertRaises(ValueError) as err:
df.fetch_specific_workspace_data({"workspace_ids": ["foo"]})
self.assertIn("Parameter 'workspace_ids' must be a list of integers.",
str(err.exception))
with self.assertRaises(ValueError) as err:
df.fetch_specific_workspace_data(
{"workspace_ids": [1, 2, 3, "foo", 5]})
self.assertIn("Parameter 'workspace_ids' must be a list of integers.",
str(err.exception))
optional_params = [
"include_type_counts", "simple_types", "ignore_narratives"
]
for opt in optional_params:
with self.assertRaises(ValueError) as err:
df.fetch_specific_workspace_data({
"workspace_ids": [1, 2],
opt: "wat"
})
self.assertIn(
"Parameter '{}' must be 0 or 1, not 'wat'".format(opt),
str(err.exception))
bad_limits = [0, -5, "a", "foo", ["foo", "bar"], {"no": "wai"}]
for bad in bad_limits:
with self.assertRaises(ValueError) as err:
df.fetch_specific_workspace_data({
"workspace_ids": [1],
"limit": bad
})
self.assertIn("Parameter 'limit' must be an integer > 0",
str(err.exception))
with self.assertRaises(ValueError) as err:
df.fetch_specific_workspace_data({
"workspace_ids": [1],
"types": "wat"
})
self.assertIn("Parameter 'types' must be a list if present.",
str(err.exception))
def test_fetch_accessible_bad_params(self):
df = DataFetcher(self.cfg["workspace-url"],
self.cfg["auth-service-url"],
self.get_context()["token"])
with self.assertRaises(ValueError) as err:
df.fetch_accessible_data({"data_set": "foo"})
self.assertIn(
"Parameter 'data_set' must be either 'mine' or 'shared', not 'foo'",
str(err.exception))
optional_params = [
"include_type_counts", "simple_types", "ignore_narratives"
]
for opt in optional_params:
with self.assertRaises(ValueError) as err:
df.fetch_accessible_data({"data_set": "mine", opt: "wat"})
self.assertIn(
"Parameter '{}' must be 0 or 1, not 'wat'".format(opt),
str(err.exception))
with self.assertRaises(ValueError) as err:
df.fetch_accessible_data({
"data_set": "mine",
"ignore_workspaces": "wat"
})
self.assertIn(
"Parameter 'ignore_workspaces' must be a list if present",
str(err.exception))
bad_limits = [0, -5, "a", "foo", ["foo", "bar"], {"no": "wai"}]
for bad in bad_limits:
with self.assertRaises(ValueError) as err:
df.fetch_accessible_data({"data_set": "mine", "limit": bad})
self.assertIn("Parameter 'limit' must be an integer > 0",
str(err.exception))
with self.assertRaises(ValueError) as err:
df.fetch_accessible_data({"data_set": "mine", "types": "wat"})
self.assertIn("Parameter 'types' must be a list if present.",
str(err.exception))
def test_data_fetcher_shared(self, mock_ws):
df = DataFetcher(self.cfg["workspace-url"],
self.cfg["auth-service-url"],
self.get_context()["token"])
# 1. shared data, default options
shared_data = df.fetch_accessible_data({"data_set": "shared"})
self.assertEqual(len(shared_data["objects"]), 36)
for obj in shared_data["objects"]:
self._validate_obj(obj, "KBaseModule.SomeType")
self._validate_ws_display(shared_data["workspace_display"], 9)
self.assertNotIn("type_counts", shared_data)
# 2. shared data, with type counts
shared_data = df.fetch_accessible_data({
"data_set": "shared",
"include_type_counts": 1
})
self.assertEqual(len(shared_data["objects"]), 36)
for obj in shared_data["objects"]:
self._validate_obj(obj, "KBaseModule.SomeType")
self._validate_ws_display(shared_data["workspace_display"], 9)
self.assertIn("type_counts", shared_data)
self.assertEqual(len(shared_data["type_counts"]),
9) # one for each version of SomeType
self.assertIn("KBaseModule.SomeType-1.0", shared_data["type_counts"])
self.assertEqual(
shared_data["type_counts"]["KBaseModule.SomeType-1.0"], 4)
# 3. shared data, with simple types, with type counts
shared_data = df.fetch_accessible_data({
"data_set": "shared",
"include_type_counts": 1,
"simple_types": 1
})
self.assertEqual(len(shared_data["objects"]), 36)
for obj in shared_data["objects"]:
self._validate_obj(obj, "SomeType")
self._validate_ws_display(shared_data["workspace_display"], 9)
self.assertIn("type_counts", shared_data)
self.assertEqual(len(shared_data["type_counts"]), 1)
self.assertIn("SomeType", shared_data["type_counts"])
self.assertEqual(shared_data["type_counts"]["SomeType"], 36)
# 4. shared data, with simple types, and type counts, don't ignore narratives
shared_data = df.fetch_accessible_data({
"data_set": "shared",
"include_type_counts": 1,
"simple_types": 1,
"ignore_narratives": 0
})
self.assertEqual(len(shared_data["objects"]), 40)
for obj in shared_data["objects"]:
if obj["obj_id"] == 1:
self._validate_obj(obj, "Narrative")
else:
self._validate_obj(obj, "SomeType")
self._validate_ws_display(shared_data["workspace_display"], 10)
self.assertIn("type_counts", shared_data)
self.assertEqual(len(shared_data["type_counts"]), 2)
self.assertIn("SomeType", shared_data["type_counts"])
self.assertEqual(shared_data["type_counts"]["SomeType"], 36)
self.assertIn("Narrative", shared_data["type_counts"])
self.assertEqual(shared_data["type_counts"]["Narrative"], 4)