def dataverse(action: str, doi: str = None, fileid: str = None):
api = NativeApi(BASE_URL, API_TOKEN)
PID = 'doi:10.5072/FK2/3OZLV6'
if doi:
PID = doi
if action == 'showfiles':
files = api.get_datafiles(PID, ':latest').json()
if not fileid:
return files
if action == 'getfile':
if not fileid:
df = pd.DataFrame(files['data'])
filesindex = {}
for i in df.index:
filesindex[df.iloc[i].label] = df.iloc[i].dataFile
pdfiles = pd.DataFrame(filesindex)
FILEID = pdfiles.loc['id'][0]
return pdfiles
else:
FILEID = fileid
fileURL = "%s/api/access/datafile/%s" % (BASE_URL, FILEID)
df = pd.read_csv(fileURL)
data = {}
datapoints = json.loads(df.to_json(orient='records'))
data['data'] = datapoints
return data
def test_token_right_create_dataset_rights(self):
BASE_URL = os.getenv("BASE_URL")
api_su = NativeApi(BASE_URL, os.getenv("API_TOKEN_SUPERUSER"))
api_nru = NativeApi(BASE_URL,
os.getenv("API_TOKEN_TEST_NO_RIGHTS"))
resp = api_su.get_info_version()
assert resp.json()["data"]["version"] == "4.15.1"
assert resp.json()["data"]["build"] == "1377-701b56b"
resp = api_nru.get_info_version()
assert resp.json()["data"]["version"] == "4.15.1"
assert resp.json()["data"]["build"] == "1377-701b56b"
ds = Dataset()
ds.from_json(
read_file(
os.path.join(
BASE_DIR,
"tests/data/dataset_upload_min_default.json")))
resp = api_su.create_dataset(":root", ds.json())
pid = resp.json()["data"]["persistentId"]
assert resp.json()["status"] == "OK"
with pytest.raises(ApiAuthorizationError):
resp = api_nru.get_dataset(pid)
resp = api_su.delete_dataset(pid)
assert resp.json()["status"] == "OK"
def test_token_empty_string(self):
BASE_URL = os.getenv("BASE_URL")
api = NativeApi(BASE_URL, "")
resp = api.get_info_version()
assert resp.json()["data"]["version"] == "4.18.1"
assert resp.json()["data"]["build"] == "267-a91d370"
with pytest.raises(ApiAuthorizationError):
ds = Dataset()
ds.from_json(
read_file(
os.path.join(
BASE_DIR,
"tests/data/dataset_upload_min_default.json")))
api.create_dataset(":root", ds.json())
def test_token_no_rights(self):
BASE_URL = os.getenv("BASE_URL")
API_TOKEN = os.getenv("API_TOKEN_NO_RIGHTS")
api = NativeApi(BASE_URL, API_TOKEN)
resp = api.get_info_version()
assert resp.json()["data"]["version"] == "4.15.1"
assert resp.json()["data"]["build"] == "1377-701b56b"
with pytest.raises(ApiAuthorizationError):
ds = Dataset()
ds.from_json(
read_file(
os.path.join(
BASE_DIR,
"tests/data/dataset_upload_min_default.json")))
api.create_dataset(":root", ds.json())
def native_api(monkeypatch):
"""Fixture, so set up an Api connection.
Returns
-------
Api
Api object.
"""
monkeypatch.setenv("BASE_URL", "https://demo.dataverse.org")
return NativeApi(os.getenv("BASE_URL"))
def collect_data(
user_handle: str,
parent: str,
data_types: List[str],
filename: str,
create_json: bool,
) -> None:
"""Collect data of a Dataverse installation.
Collect data from a data node down the Dataverse
tree-like data structure.
Collects the complete data of a Dataverse instance in
a tree structure (`tree.json`), containing all
Dataverses, Datasets and Datafiles. The file is
stored in your instance directory (e. g.
`utils/data/instances/dataverse_production`).
"""
if user_handle == "public":
api = NativeApi(config.BASE_URL)
else:
users = read_json(config.USER_FILENAME)
api = NativeApi(config.BASE_URL, users[user_handle]["api-token"])
tree = api.get_children(parent, children_types=data_types)
if not os.path.isdir(os.path.join(ROOT_DIR, "data")):
os.makedirs(os.path.join(ROOT_DIR, "data"))
if not os.path.isdir(os.path.join(ROOT_DIR, "data", "utils")):
os.makedirs(os.path.join(ROOT_DIR, "data", "utils"))
if not os.path.isdir(
os.path.join(ROOT_DIR, "data", "utils", user_handle)):
os.makedirs(
os.path.join(ROOT_DIR, "data", "utils", user_handle))
write_json(os.path.join(UTILS_DATA_DIR, user_handle, filename), tree)
if create_json:
generate_data(tree, user_handle, filename)
def datasync(self):
native_api = NativeApi(BASE_URL, API_TOKEN)
self.ds_id = str(
int(self.make_dataset_id(self.REPO).hexdigest(),
16))[:6] ## turn the md5 string into a 6 digits integer
metadata = self.make_dataset_metadata(self.REPO)
print(metadata)
self.ds = Dataset()
self.ds.set(metadata)
self.ds.displayName = metadata['title']
self.ds.json = metadata
print(self.ds.get())
if self.DEBUG:
print("[datasync]")
print(self.ds)
print(self.ds_id)
print(self.ds.displayName)
self.create_dataset(native_api, self.ds, DV_ALIAS, self.ds_id,
BASE_URL)
if self.DEBUG:
print(metadata)
self.upload_files_to_dataverse(self.ds_id, self.urls_found)
return True
def remove_testdata(
user_handle: str,
parent: str,
data_types: List[str] = ["dataverses", "datasets"],
force: bool = False,
parent_data_type: str = "dataverse",
remove_parent: bool = False,
) -> None:
"""Remove testdata.
Removes all data created by `create-testdata`.
It recursively collects all Dataverses and Datasets
from a passed Dataverse down (by default =
`science`). If `PRODUCTION` is `true`, this function
will not execute, as long as you not add `--force`
to the function call. This is to protect from
unwanted changes on a production instance.
"""
if config.PRODUCTION and not force:
print(
"Delete testdata on a PRODUCTION instance not allowed. Use --force to force it."
)
sys.exit()
user = read_json(config.USER_FILENAME)[user_handle]
api = NativeApi(config.BASE_URL, user["api-token"])
# Clean up
data = api.get_children(parent, children_types=data_types)
dataverses, datasets, = dataverse_tree_walker(data)
if parent_data_type == "dataverse" and remove_parent:
dataverses.append({"dataverse_alias": parent})
for ds in datasets:
api.destroy_dataset(ds["pid"])
for dv in dataverses:
api.delete_dataverse(dv["dataverse_alias"])
def __init__(self, host, api_token=None):
self._host = host
self.api_token = api_token
self.api = Api(host, api_token=api_token)
self.native_api = NativeApi(host, api_token=api_token)
self.data_access_api = DataAccessApi(host, api_token=api_token)
class DataverseClient(object):
def __init__(self, host, api_token=None):
self._host = host
self.api_token = api_token
self.api = Api(host, api_token=api_token)
self.native_api = NativeApi(host, api_token=api_token)
self.data_access_api = DataAccessApi(host, api_token=api_token)
def get_ddi(self, doi, format=dv_static.EXPORTER_FORMAT_DDI):
"""
Get DDI metadata file
"""
response = self.native_api.get_dataset_export(doi, format)
return DDI(response.content)
def get_user_info(self, user_api_token=None):
"""
Placeholder until pyDataverse API is updated
"""
api_token = user_api_token if user_api_token else self.api_token
# remove any trailing "/"
ye_host = RegisteredDataverse.format_dv_url(self._host)
#while ye_host.endswith('/'):
# ye_host = ye_host[:-1]
# format url
dv_url = f'{ye_host}/api/v1/users/:me'
# make the request
headers = {'X-Dataverse-key': api_token}
try:
response = requests.get(dv_url, headers=headers)
except ConnectionError as err_obj:
return err_resp(f'Failed to connect. {err_obj}')
if response.status_code == 200:
if not response.content:
# In this instance the response content is an empty string or None -- shouldn't happen...
#
return err_resp(f"Dataverse returned an HTTP 200 status code but failed to return a response.")
resp_json = response.json()
dv_status = resp_json.get(dv_static.DV_KEY_STATUS)
if not dv_status:
return err_resp(f"Dataverse response failed to return a 'status'.")
if dv_status == dv_static.STATUS_VAL_ERROR:
user_msg = resp_json.get(dv_static.DV_KEY_MESSAGE,
'(No message from Dataverse)')
return err_resp(f"Dataverse error: {user_msg}")
return ok_resp(response.json())
try:
json_resp = response.json()
if 'message' in json_resp:
return err_resp(json_resp['message'])
except ValueError:
pass
return err_resp(f'Status code: {response.status_code} {response.text}')
def get_schema_org(self, doi):
"""
Get schema.org data
"""
return self.get_dataset_export_json(doi, dv_static.EXPORTER_FORMAT_SCHEMA_ORG)
def get_dataset_export_json(self, doi, format_type):
"""
Get dataset export
"""
try:
response = self.native_api.get_dataset_export(doi, format_type)
except ConnectionError as err_obj:
return err_resp(f'Failed to connect. {err_obj}')
return response
def create_testdata(config_file: str, force: bool) -> None:
"""Create testdata defined in a config file.
Creates a pre-defined set of testdata on your
instance. By default, the function uses the
AUSSDA test data repository, which is so far not
publicly available. If `PRODUCTION` is `true`,
this function will not execute, as long as you
not add `--force` to the function call. This is
to protect from unwanted changes on a production
instance.
"""
# Init
if config.PRODUCTION and not force:
print(
"Create testdata on a PRODUCTION instance not allowed. Use --force to force it."
)
sys.exit()
pid_idx = []
users = read_json(config.USER_FILENAME)
workflow = read_json(os.path.join(ROOT_DIR, config_file))
# Dataverses
for dv_conf in workflow["dataverses"]:
dv_alias = None
if "create" in dv_conf:
api = NativeApi(
config.BASE_URL,
users[dv_conf["create"]["user-handle"]]["api-token"])
dv = Dataverse()
dv_filename = os.path.join(ROOT_DIR,
dv_conf["create"]["metadata-filename"])
dv.from_json(read_file(dv_filename))
if "update" in dv_conf["create"]:
for key, val in dv_conf["create"]["update"].items():
kwargs = {key: val}
dv.set(kwargs)
dv_alias = dv.get()["alias"]
resp = api.create_dataverse(dv_conf["create"]["parent"], dv.json())
if "publish" in dv_conf:
api = NativeApi(
config.BASE_URL,
users[dv_conf["publish"]["user-handle"]]["api-token"])
if not dv_alias and "alias" in dv_conf["publish"]:
dv_alias = dv_conf["publish"]["alias"]
resp = api.publish_dataverse(dv_alias)
# Datasets
for ds_conf in workflow["datasets"]:
pid = None
if "create" in ds_conf:
api = NativeApi(
config.BASE_URL,
users[ds_conf["create"]["user-handle"]]["api-token"])
ds = Dataset()
ds_filename = os.path.join(ROOT_DIR,
ds_conf["create"]["metadata-filename"])
ds.from_json(read_file(ds_filename))
if "update" in ds_conf["create"]:
for key, val in ds_conf["create"]["update"].items():
kwargs = {key: val}
ds.set(kwargs)
resp = api.create_dataset(dv_alias, ds.json())
pid = resp.json()["data"]["persistentId"]
pid_idx.append(pid)
if "publish" in ds_conf:
if not pid:
print("ERROR: PID missing!")
sys.exit()
api = NativeApi(
config.BASE_URL,
users[ds_conf["publish"]["user-handle"]]["api-token"])
resp = api.publish_dataset(pid, release_type="major")
# Datafiles
for dataset_id, ds_datafiles in workflow["datafiles"].items():
if int(dataset_id) == workflow["datasets"][int(dataset_id)]["id"]:
pid = pid_idx[int(dataset_id)]
else:
print("ERROR: Dataset ID not matching.")
sys.exit()
for df_conf in ds_datafiles:
if "upload" in df_conf:
api = NativeApi(
config.BASE_URL,
users[df_conf["upload"]["user-handle"]]["api-token"],
)
metadata = read_json(df_conf["upload"]["metadata-filename"])
df = Datafile()
df.set(metadata)
if "update" in df_conf["upload"]:
for key, val in df_conf["upload"]["update"].items():
kwargs = {key: val}
df.set(kwargs)
df.set({"pid": pid})
filename = df_conf["upload"]["filename"]
resp = api.upload_datafile(pid, filename, df.json())
if filename[-4:] == ".sav" or filename[-4:] == ".dta":
sleep(30)
else:
sleep(3)
if "publish-dataset" in df_conf:
api = NativeApi(
config.BASE_URL,
users[df_conf["publish-dataset"]["user-handle"]]["api-token"],
)
if df_conf["publish-dataset"]:
resp = api.publish_dataset(pid, release_type="major")
def native_api(config):
"""Initialize pyDataverse Native Api object."""
yield NativeApi(config.BASE_URL)
def test_api_connect_base_url_wrong(self):
"""Test native_api connection with wrong `base_url`."""
# None
with pytest.raises(ApiUrlError):
NativeApi(None)
print("uploading ",
os.path.basename(os.path.normpath(folder)) + os.sep + file)
# make POST request to target URL
r = requests.post(url_persistent_id, data=payload, files=files)
# store upload status
status = r.json()["status"]
return status
# %% run every 5 seconds until upload is successful (circumventing denied URL requests)
# create a NativeApi instance
api = NativeApi(dataverse_server, api_key)
# declare a flag to track upload status
flag = None
while flag != "OK":
try:
flag = upload_files_to_dataverse(args.InputFolder, file_list,
file_description)
except Exception as e:
print(e)
print("... starting again")
time.sleep(5)
# make a new release
