Newer
Older
request = {
"method": "getExperiments",
"params": [
self.token,
[ search_request ],
fetchopts
],
}
resp = self._post_request(self.as_v3, request)
if len(resp) == 0:
raise ValueError("No such experiment: %s" % expId)
Swen Vermeul
committed
return Experiment(self,
self.get_experiment_type(resp[expId]["type"]["code"]),
resp[expId]
)
def new_experiment(self, type, **kwargs):
""" Creates a new experiment of a given experiment type.
"""
return Experiment(self, self.get_experiment_type(type), None, **kwargs)
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
def update_experiment(self, experimentId, properties=None, tagIds=None, attachments=None):
params = {
"experimentId": {
"permId": experimentId,
"@type": "as.dto.experiment.id.ExperimentPermId"
},
"@type": "as.dto.experiment.update.ExperimentUpdate"
}
if properties is not None:
params["properties"]= properties
if tagIds is not None:
params["tagIds"] = tagIds
if attachments is not None:
params["attachments"] = attachments
request = {
"method": "updateExperiments",
"params": [
self.token,
[ params ]
]
}
self._post_request(self.as_v3, request)
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
def create_sample(self, space_ident, code, type,
project_ident=None, experiment_ident=None, properties=None, attachments=None, tags=None):
tagIds = _create_tagIds(tags)
typeId = _create_typeId(type)
projectId = _create_projectId(project_ident)
experimentId = _create_experimentId(experiment_ident)
if properties is None:
properties = {}
request = {
"method": "createSamples",
"params": [
self.token,
[
{
"properties": properties,
"code": code,
"typeId" : typeId,
"projectId": projectId,
"experimentId": experimentId,
"tagIds": tagIds,
"attachments": attachments,
"@type": "as.dto.sample.create.SampleCreation",
}
]
],
}
resp = self._post_request(self.as_v3, request)
return self.get_sample(resp[0]['permId'])
def update_sample(self, sampleId, space=None, project=None, experiment=None,
parents=None, children=None, components=None, properties=None, tagIds=None, attachments=None):
params = {
"sampleId": {
"permId": sampleId,
"@type": "as.dto.sample.id.SamplePermId"
},
"@type": "as.dto.sample.update.SampleUpdate"
}
if space is not None:
params['spaceId'] = space
if project is not None:
params['projectId'] = project
if properties is not None:
params["properties"]= properties
if tagIds is not None:
params["tagIds"] = tagIds
if attachments is not None:
params["attachments"] = attachments
request = {
"method": "updateSamples",
"params": [
self.token,
[ params ]
]
}
self._post_request(self.as_v3, request)
def delete_entity(self, entity, permid, reason):
"""Deletes Spaces, Projects, Experiments, Samples and DataSets
"""
entity_type = "as.dto.{}.id.{}PermId".format(entity.lower(), entity.capitalize())
request = {
"method": "delete" + entity.capitalize() + 's',
"params": [
self.token,
[
{
"permId": permid,
"@type": entity_type
}
],
{
"reason": reason,
"@type": "as.dto.{}.delete.{}DeletionOptions".format(entity.lower(), entity.capitalize())
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
}
]
}
self._post_request(self.as_v3, request)
def get_deletions(self):
request = {
"method": "searchDeletions",
"params": [
self.token,
{},
{
"deletedObjects": {
"@type": "as.dto.deletion.fetchoptions.DeletedObjectFetchOptions"
}
}
]
}
resp = self._post_request(self.as_v3, request)
objects = resp['objects']
parse_jackson(objects)
new_objs = []
for value in objects:
del_objs = extract_deletion(value)
if len(del_objs) > 0:
new_objs.append(*del_objs)
return DataFrame(new_objs)
def new_project(self, space, code, description=None, **kwargs):
return Project(self, None, space=space, code=code, description=description, **kwargs)
def _gen_fetchoptions(self, options):
fo = {}
for option in options:
fo[option] = fetch_option[option]
return fo
def get_project(self, projectId):
options = ['space', 'registrator', 'modifier', 'attachments']
if is_identifier(projectId):
request = self._create_get_request(
'getProjects', 'project', projectId, options
)
resp = self._post_request(self.as_v3, request)
return Project(self, resp[projectId])
else:
search_criteria = _gen_search_criteria({
'project': 'Project',
'operator': 'AND',
'code': projectId
})
fo = self._gen_fetchoptions(options)
request = {
"method": "searchProjects",
"params": [self.token, search_criteria, fo]
}
resp = self._post_request(self.as_v3, request)
return Project(self, resp['objects'][0])
def get_projects(self, space=None):
""" Get a list of all available projects (DataFrame object).
"""
sub_criteria = []
if space:
sub_criteria.append(_subcriteria_for_code(space, 'space'))
criteria = {
"criteria": sub_criteria,
"@type": "as.dto.project.search.ProjectSearchCriteria",
"operator": "AND"
}
fetchopts = { "@type": "as.dto.project.fetchoptions.ProjectFetchOptions" }
for option in ['registrator', 'modifier', 'leader' ]:
fetchopts[option] = fetch_option[option]
request = {
"method": "searchProjects",
"params": [ self.token,
criteria,
],
}
resp = self._post_request(self.as_v3, request)
if resp is not None:
objects = resp['objects']
parse_jackson(objects)
projects = DataFrame(objects)
if len(projects) is 0:
raise ValueError("No projects found!")
projects['registrationDate']= projects['registrationDate'].map(format_timestamp)
projects['modificationDate']= projects['modificationDate'].map(format_timestamp)
projects['leader'] = projects['leader'].map(extract_person)
projects['registrator'] = projects['registrator'].map(extract_person)
projects['modifier'] = projects['modifier'].map(extract_person)
projects['permId'] = projects['permId'].map(extract_permid)
projects['identifier'] = projects['identifier'].map(extract_identifier)
pros=projects[['identifier', 'permId', 'leader', 'registrator', 'registrationDate',
'modifier', 'modificationDate']]
return Things(self, 'project', pros, 'identifier')
else:
raise ValueError("No projects found!")
def _create_get_request(self, method_name, entity, permids, options):
if not isinstance(permids, list):
permids = [permids]
type = "as.dto.{}.id.{}".format(entity.lower(), entity.capitalize())
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
search_params = []
for permid in permids:
# decide if we got a permId or an identifier
match = re.match('/', permid)
if match:
search_params.append(
{ "identifier" : permid, "@type" : type + 'Identifier' }
)
else:
search_params.append(
{ "permId" : permid, "@type": type + 'PermId' }
)
fo = {}
for option in options:
fo[option] = fetch_option[option]
request = {
"method": method_name,
"params": [
self.token,
search_params,
fo
],
}
return request
def get_terms(self, vocabulary=None):
""" Returns information about vocabulary, including its controlled vocabulary
search_request = {}
if vocabulary is not None:
search_request = _gen_search_criteria( {
"vocabulary": "VocabularyTerm",
"criteria" : [{
"vocabulary": "Vocabulary",
"code": vocabulary
}]
})
fetch_options = {
"vocabulary" : { "@type" : "as.dto.vocabulary.fetchoptions.VocabularyFetchOptions" },
"@type": "as.dto.vocabulary.fetchoptions.VocabularyTermFetchOptions"
}
request = {
"method": "searchVocabularyTerms",
"params": [ self.token, search_request, fetch_options ]
}
resp = self._post_request(self.as_v3, request)
parse_jackson(resp)
return Vocabulary(resp)
def get_tags(self):
""" Returns a DataFrame of all
"""
request = {
"method": "searchTags",
"params": [ self.token, {}, {} ]
}
resp = self._post_request(self.as_v3, request)
parse_jackson(resp)
objects = DataFrame(resp['objects'])
objects['registrationDate'] = objects['registrationDate'].map(format_timestamp)
return objects[['code', 'registrationDate']]
Swen Vermeul
committed
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
def get_sample_types(self, type=None):
""" Returns a list of all available sample types
"""
return self._get_types_of(
"searchSampleTypes",
"Sample",
type,
["generatedCodePrefix"]
)
def get_sample_type(self, type):
try:
return self._get_types_of(
"searchSampleTypes",
"Sample",
type,
["generatedCodePrefix"]
)
except Exception:
raise ValueError("no such sample type: {}".format(type))
def get_experiment_types(self, type=None):
""" Returns a list of all available experiment types
"""
return self._get_types_of(
"searchExperimentTypes",
"Experiment",
type
)
def get_experiment_type(self, type):
try:
return self._get_types_of(
"searchExperimentTypes",
"Experiment",
type
)
except Exception:
raise ValueError("No such experiment type: {}".format(type))
def get_material_types(self, type=None):
""" Returns a list of all available material types
"""
return self._get_types_of("searchMaterialTypes", "Material", type)
def get_material_type(self, type):
try:
return self._get_types_of("searchMaterialTypes", "Material", type)
except Exception:
raise ValueError("No such material type: {}".format(type))
def get_dataset_types(self, type=None):
""" Returns a list (DataFrame object) of all currently available dataset types
"""
return self._get_types_of("searchDataSetTypes", "DataSet", type, ['kind'] )
Swen Vermeul
committed
def get_dataset_type(self, type):
try:
return self._get_types_of("searchDataSetTypes", "DataSet", type, ['kind'])
except Exception:
raise ValueError("No such dataSet type: {}".format(type))
def _get_types_of(self, method_name, entity, type_name=None, additional_attributes=None):
""" Returns a list of all available types of an entity.
If the name of the entity-type is given, it returns a PropertyAssignments object
if additional_attributes is None:
additional_attributes = []
Swen Vermeul
committed
attributes = ['code', 'description'] + additional_attributes + ['modificationDate']
search_request = {}
fetch_options = {}
if type_name is not None:
search_request = _gen_search_criteria({
"operator": "AND",
})
fetch_options = {
Swen Vermeul
committed
"@type": "as.dto.{}.fetchoptions.{}TypeFetchOptions".format(
Swen Vermeul
committed
)
}
Swen Vermeul
committed
fetch_options['propertyAssignments'] = fetch_option['propertyAssignments']
attributes.append('propertyAssignments')
"params": [ self.token, search_request, fetch_options ],
}
resp = self._post_request(self.as_v3, request)
parse_jackson(resp)
if type_name is not None and len(resp['objects']) == 1:
return PropertyAssignments(self, resp['objects'][0])
if len(resp['objects']) >= 1:
types = DataFrame(resp['objects'])
Swen Vermeul
committed
types['modificationDate'] = types['modificationDate'].map(format_timestamp)
return Things(self, entity.lower()+'_type', types[attributes])
Swen Vermeul
committed
Swen Vermeul
committed
def is_session_active(self):
""" checks whether a session is still active. Returns true or false.
"""
Swen Vermeul
committed
return self.is_token_valid(self.token)
def is_token_valid(self, token=None):
Chandrasekhar Ramakrishnan
committed
"""Check if the connection to openBIS is valid.
This method is useful to check if a token is still valid or if it has timed out,
requiring the user to login again.
Chandrasekhar Ramakrishnan
committed
:return: Return True if the token is valid, False if it is not valid.
"""
if token is None:
token = self.token
if token is None:
return False
request = {
"method": "isSessionActive",
"params": [ token ],
resp = self._post_request(self.as_v1, request)
"""fetch a dataset and some metadata attached to it:
- properties
- sample
- parents
- children
- containers
- dataStore
- physicalData
- linkedData
:return: a DataSet object
"""
criteria = [{
"permId": permid,
"@type": "as.dto.dataset.id.DataSetPermId"
}]
fetchopts = {
"parents": { "@type": "as.dto.dataset.fetchoptions.DataSetFetchOptions" },
"children": { "@type": "as.dto.dataset.fetchoptions.DataSetFetchOptions" },
"containers": { "@type": "as.dto.dataset.fetchoptions.DataSetFetchOptions" },
"type": { "@type": "as.dto.dataset.fetchoptions.DataSetTypeFetchOptions" },
}
for option in ['tags', 'properties', 'dataStore', 'physicalData', 'linkedData',
'experiment', 'sample']:
fetchopts[option] = fetch_option[option]
request = {
"params": [ self.token,
criteria,
fetchopts,
resp = self._post_request(self.as_v3, request)
if resp is None or len(resp) == 0:
raise ValueError('no such dataset found: '+permid)
if resp is not None:
for permid in resp:
#return resp[permid]
return DataSet(self, self.get_dataset_type(resp[permid]["type"]["code"]), resp[permid])
Chandrasekhar Ramakrishnan
committed
Swen Vermeul
committed
def get_sample(self, sample_ident, only_data=False, withAttachments=False):
Chandrasekhar Ramakrishnan
committed
"""Retrieve metadata for the sample.
Get metadata for the sample and any directly connected parents of the sample to allow access
to the same information visible in the ELN UI. The metadata will be on the file system.
:param sample_identifiers: A list of sample identifiers to retrieve.
"""
Swen Vermeul
committed
fetchopts = { "type": { "@type": "as.dto.sample.fetchoptions.SampleTypeFetchOptions" } }
search_request = search_request_for_identifier(sample_ident, 'sample')
Swen Vermeul
committed
for option in ['tags', 'properties', 'attachments', 'space', 'experiment', 'registrator', 'dataSets']:
fetchopts[option] = fetch_option[option]
if withAttachments:
fetchopts['attachments'] = fetch_option['attachmentsWithContent']
#fetchopts["parents"] = { "@type": "as.dto.sample.fetchoptions.SampleFetchOptions" }
#fetchopts["children"] = { "@type": "as.dto.sample.fetchoptions.SampleFetchOptions" }
sample_request = {
"method": "getSamples",
"params": [
self.token,
[ search_request ],
Swen Vermeul
committed
fetchopts
resp = self._post_request(self.as_v3, sample_request)
parse_jackson(resp)
if resp is None or len(resp) == 0:
raise ValueError('no such sample found: '+sample_ident)
else:
for sample_ident in resp:
if only_data:
return resp[sample_ident]
else:
return Sample(self, self.get_sample_type(resp[sample_ident]["type"]["code"]), resp[sample_ident])
def new_space(self, name, description=None):
request = {
"method": "createSpaces",
"params": [
self.token,
[ {
"code": name,
"description": description,
"@type": "as.dto.space.create.SpaceCreation"
} ]
],
}
resp = self._post_request(self.as_v3, request)
Swen Vermeul
committed
def new_analysis(self, name, description=None, sample=None, dss_code=None, result_files=None,
Swen Vermeul
committed
notebook_files=None, parents=None):
Swen Vermeul
committed
""" An analysis contains the Jupyter notebook file(s) and some result files.
Technically this method involves uploading files to the session workspace
and activating the dropbox aka dataset ingestion service "jupyter-uploader-api"
Swen Vermeul
committed
"""
if dss_code is None:
dss_code = self.get_datastores()['code'][0]
# if a sample identifier was given, use it as a string.
# if a sample object was given, take its identifier
Swen Vermeul
committed
sampleId = None
Swen Vermeul
committed
if isinstance(sample, str):
Swen Vermeul
committed
if (is_identifier(sample)):
sampleId = {
"identifier": sample,
"@type": "as.dto.sample.id.SampleIdentifier"
}
else:
sampleId = {
"permId": sample,
"@type": "as.dto.sample.id.SamplePermId"
}
Swen Vermeul
committed
else:
Swen Vermeul
committed
sampleId = {
"identifier": sample.identifier,
"@type": "as.dto.sample.id.SampleIdentifier"
}
parentIds = []
if parents is not None:
if not isinstance(parents, list):
parants = [parents]
for parent in parents:
parentIds.append(parent.permId)
Swen Vermeul
committed
Swen Vermeul
committed
folder = time.strftime('%Y-%m-%d_%H-%M-%S')
Swen Vermeul
committed
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
data_sets = []
if notebook_files is not None:
notebooks_folder = os.path.join(folder, 'notebook_files')
self.upload_files(
datastore_url = datastore_url,
files=notebook_files,
folder= notebooks_folder,
wait_until_finished=True
)
data_sets.append({
"dataSetType" : "JUPYTER_NOTEBOOk",
"sessionWorkspaceFolder": notebooks_folder,
"fileNames" : notebook_files,
"properties" : {}
})
if result_files is not None:
results_folder = os.path.join(folder, 'result_files')
self.upload_files(
datastore_url = datastore_url,
files=result_files,
folder=results_folder,
wait_until_finished=True
)
data_sets.append({
"dataSetType" : "JUPYTER_RESULT",
"sessionWorkspaceFolder" : results_folder,
"fileNames" : result_files,
"properties" : {}
})
request = {
"method": "createReportFromAggregationService",
"params": [
self.token,
dss_code,
Swen Vermeul
committed
DROPBOX_PLUGIN,
Chandrasekhar Ramakrishnan
committed
"sample" : { "identifier" : sampleId['identifier'] },
Swen Vermeul
committed
"sampleId": sampleId,
"parentIds": parentIds,
"containers" : [ {
"dataSetType" : "JUPYTER_CONTAINER",
"properties" : {
"NAME" : name,
"DESCRIPTION" : description
Swen Vermeul
committed
"dataSets" : data_sets,
resp = self._post_request(self.reg_v1, request)
try:
if resp['rows'][0][0]['value'] == 'OK':
return resp['rows'][0][1]['value']
except:
return resp
Swen Vermeul
committed
def new_sample(self, type, **kwargs):
""" Creates a new sample of a given sample type.
return Sample(self, self.get_sample_type(type), None, **kwargs)
def new_dataset(self, type, **kwargs):
""" Creates a new dataset of a given sample type.
"""
return DataSet(self, self.get_dataset_type(type.upper()), None, **kwargs)
def _get_dss_url(self, dss_code=None):
""" internal method to get the downloadURL of a datastore.
"""
Swen Vermeul
committed
dss = self.get_datastores()
if dss_code is None:
return dss['downloadUrl'][0]
else:
Swen Vermeul
committed
return dss[dss['code'] == dss_code]['downloadUrl'][0]
Swen Vermeul
committed
def upload_files(self, datastore_url=None, files=None, folder=None, wait_until_finished=False):
Swen Vermeul
committed
if datastore_url is None:
if files is None:
raise ValueError("Please provide a filename.")
Swen Vermeul
committed
if folder is None:
# create a unique foldername
folder = time.strftime('%Y-%m-%d_%H-%M-%S')
if isinstance(files, str):
files = [files]
self.files = files
self.startByte = 0
self.endByte = 0
# define a queue to handle the upload threads
queue = DataSetUploadQueue()
real_files = []
for filename in files:
if os.path.isdir(filename):
real_files.extend([os.path.join(dp, f) for dp, dn, fn in os.walk(os.path.expanduser(filename)) for f in fn])
else:
real_files.append(os.path.join(filename))
# compose the upload-URL and put URL and filename in the upload queue
for filename in real_files:
file_in_wsp = os.path.join(folder, filename)
Swen Vermeul
committed
self.files_in_wsp.append(file_in_wsp)
upload_url = (
Swen Vermeul
committed
datastore_url + '/session_workspace_file_upload'
+ '?filename=' + os.path.join(folder,filename)
+ '&id=1'
+ '&startByte=0&endByte=0'
+ '&sessionID=' + self.token
)
queue.put([upload_url, filename, self.verify_certificates])
# wait until all files have uploaded
if wait_until_finished:
queue.join()
# return files with full path in session workspace
Swen Vermeul
committed
return self.files_in_wsp
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
def __init__(self, workers=20):
# maximum files to be uploaded at once
self.upload_queue = Queue()
# define number of threads and start them
for t in range(workers):
t = Thread(target=self.upload_file)
t.daemon = True
t.start()
def put(self, things):
""" expects a list [url, filename] which is put into the upload queue
"""
self.upload_queue.put(things)
def join(self):
""" needs to be called if you want to wait for all uploads to be finished
"""
self.upload_queue.join()
def upload_file(self):
while True:
# get the next item in the queue
upload_url, filename, verify_certificates = self.upload_queue.get()
filesize = os.path.getsize(filename)
# upload the file to our DSS session workspace
with open(filename, 'rb') as f:
resp = requests.post(upload_url, data=f, verify=verify_certificates)
resp.raise_for_status()
data = resp.json()
assert filesize == int(data['size'])
# Tell the queue that we are done
self.upload_queue.task_done()
Swen Vermeul
committed
def __init__(self, workers=20):
# maximum files to be downloaded at once
self.download_queue = Queue()
# define number of threads
for t in range(workers):
t = Thread(target=self.download_file)
t.daemon = True
t.start()
def put(self, things):
""" expects a list [url, filename] which is put into the download queue
"""
self.download_queue.put(things)
Swen Vermeul
committed
def join(self):
""" needs to be called if you want to wait for all downloads to be finished
"""
self.download_queue.join()
def download_file(self):
while True:
url, filename, file_size, verify_certificates = self.download_queue.get()
Swen Vermeul
committed
# create the necessary directory structure if they don't exist yet
os.makedirs(os.path.dirname(filename), exist_ok=True)
# request the file in streaming mode
r = requests.get(url, stream=True, verify=verify_certificates)
Swen Vermeul
committed
with open(filename, 'wb') as f:
for chunk in r.iter_content(chunk_size=1024):
if chunk: # filter out keep-alive new chunks
f.write(chunk)
assert os.path.getsize(filename) == int(file_size)
Swen Vermeul
committed
self.download_queue.task_done()
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
def __init__(self, openbis_obj, type, data=None, **kwargs):
self.__dict__['openbis'] = openbis_obj
self.__dict__['type'] = type
self.__dict__['p'] = PropertyHolder(openbis_obj, type)
self.__dict__['a'] = AttrHolder(openbis_obj, 'DataSet', type)
# existing OpenBIS object
if data is not None:
self._set_data(data)
if kwargs is not None:
for key in kwargs:
setattr(self, key, kwargs[key])
def __eq__(self, other):
return str(self) == str(other)
def __ne__(self, other):
return str(self) != str(other)
def _set_data(self, data):
# assign the attribute data to self.a by calling it
# (invoking the AttrHolder.__call__ function)
self.a(data)
self.__dict__['data'] = data
# put the properties in the self.p namespace (without checking them)
for key, value in data['properties'].items():
self.p.__dict__[key.lower()] = value
try:
return self.openbis.get_project(self._project['identifier'])
except Exception:
pass
try:
return self.openbis.get_experiment(self._experiment['identifier'])
except Exception:
pass
try:
return self.openbis.get_sample(self._sample['permId']['permId'])
except Exception:
pass
def __getattr__(self, name):
return getattr(self.__dict__['a'], name)
def __setattr__(self, name, value):
if name in ['set_properties', 'set_tags', 'add_tags']:
raise ValueError("These are methods which should not be overwritten")
setattr(self.__dict__['a'], name, value)
def _repr_html_(self):
"""Print all the assigned attributes (identifier, tags, etc.) in a nicely formatted table. See
AttributeHolder class.
"""
return self.a._repr_html_()
def __repr__(self):
"""same thing as _repr_html_() but for IPython
"""
return self.a.__repr__()
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
class DataSet(OpenBisObject):
""" DataSet are openBIS objects that contain the actual files.
"""
def __init__(self, openbis_obj, type, data=None, **kwargs):
super(DataSet, self).__init__(openbis_obj, type, data, **kwargs)
# existing DataSet
if data is not None:
if data['physicalData'] is None:
self.__dict__['shareId'] = None
self.__dict__['location'] = None
else:
self.__dict__['shareId'] = data['physicalData']['shareId']
self.__dict__['location'] = data['physicalData']['location']
def __str__(self):
return self.data['code']
def __dir__(self):
return [
'props', 'get_parents()', 'get_children()',
'sample', 'experiment',
'tags', 'set_tags()', 'add_tags()', 'del_tags()',
'add_attachment()', 'get_attachments()', 'download_attachments()',
'data'
]
@property
def type(self):
return self.__dict__['type']
@type.setter
def type(self, type_name):
dataset_type = self.openbis.get_dataset_type(type_name.upper())
self.p.__dict__['_type'] = dataset_type
self.a.__dict__['_type'] = dataset_type
def set_properties(self, properties):
self.openbis.update_dataset(self.permId, properties=properties)
Swen Vermeul
committed
def download(self, files=None, destination=None, wait_until_finished=True, workers=10):
""" download the actual files and put them by default in the following folder:
__current_dir__/destination/dataset_permId/
If no files are specified, all files of a given dataset are downloaded.
If no destination is specified, the hostname is chosen instead.
Files are usually downloaded in parallel, using 10 workers by default. If you want to wait until
all the files are downloaded, set the wait_until_finished option to True.
Swen Vermeul
committed
"""
if files == None:
files = self.file_list()
elif isinstance(files, str):
files = [files]
if destination is None:
destination = self.openbis.hostname
base_url = self.data['dataStore']['downloadUrl'] + '/datastore_server/' + self.permId + '/'
Swen Vermeul
committed
queue = DataSetDownloadQueue(workers=workers)
# get file list and start download
for filename in files:
file_info = self.get_file_list(start_folder=filename)
file_size = file_info[0]['fileSize']
download_url = base_url + filename + '?sessionID=' + self.openbis.token
filename_dest = os.path.join(destination, self.permId, filename)
queue.put([download_url, filename_dest, file_size, self.openbis.verify_certificates])
Swen Vermeul
committed
# wait until all files have downloaded
if wait_until_finished:
queue.join()
print("Files downloaded to: %s" % os.path.join(destination, self.permId))
def get_parents(self):
return self.openbis.get_datasets(withChildren=self.permId)
def get_children(self):
return self.openbis.get_datasets(withParents=self.permId)
def file_list(self):
"""returns the list of files including their directories as an array of strings. Just folders are not
listed.
"""
files = []
for file in self.get_file_list(recursive=True):
if file['isDirectory']: