def test_from_xml_to_xml(self):
# Get the raw xml thing
# TODO maybe get this via get_flow(), which would have to be refactored to allow getting only the xml dictionary
# TODO: no sklearn flows.
for flow_id in [
3,
5,
7,
9,
]:
flow_xml = _perform_api_call("flow/%d" % flow_id)
flow_dict = xmltodict.parse(flow_xml)
flow = openml.OpenMLFlow._from_dict(flow_dict)
new_xml = flow._to_xml()
flow_xml = flow_xml.replace(' ', '').replace(
'\t', '').strip().replace('\n\n', '\n').replace('"', '"')
flow_xml = re.sub(r'^$', '', flow_xml)
new_xml = new_xml.replace(' ',
'').replace('\t', '').strip().replace(
'\n\n', '\n').replace('"', '"')
new_xml = re.sub(r'^$', '', new_xml)
self.assertEqual(new_xml, flow_xml)
def _list_flows(api_call):
# TODO add proper error handling here!
xml_string = _perform_api_call(api_call)
flows_dict = xmltodict.parse(xml_string, force_list=('oml:flow', ))
# Minimalistic check if the XML is useful
assert type(flows_dict['oml:flows']['oml:flow']) == list, \
type(flows_dict['oml:flows'])
assert flows_dict['oml:flows']['@xmlns:oml'] == \
'http://openml.org/openml', flows_dict['oml:flows']['@xmlns:oml']
flows = dict()
for flow_ in flows_dict['oml:flows']['oml:flow']:
fid = int(flow_['oml:id'])
flow = {
'id': fid,
'full_name': flow_['oml:full_name'],
'name': flow_['oml:name'],
'version': flow_['oml:version'],
'external_version': flow_['oml:external_version'],
'uploader': flow_['oml:uploader']
}
flows[fid] = flow
return flows
def test_from_xml_to_xml(self):
# Get the raw xml thing
# TODO maybe get this via get_flow(), which would have to be refactored
# to allow getting only the xml dictionary
# TODO: no sklearn flows.
for flow_id in [
3,
5,
7,
9,
]:
flow_xml = _perform_api_call("flow/%d" % flow_id,
request_method="get")
flow_dict = xmltodict.parse(flow_xml)
flow = openml.OpenMLFlow._from_dict(flow_dict)
new_xml = flow._to_xml()
flow_xml = (flow_xml.replace(" ", "").replace(
"\t", "").strip().replace("\n\n", "\n").replace(""", '"'))
flow_xml = re.sub(r"^$", "", flow_xml)
new_xml = (new_xml.replace(" ", "").replace(
"\t", "").strip().replace("\n\n", "\n").replace(""", '"'))
new_xml = re.sub(r"^$", "", new_xml)
self.assertEqual(new_xml, flow_xml)
def flow_exists(name, external_version):
"""Retrieves the flow id of the flow uniquely identified by name + external_version.
Parameter
---------
name : string
Name of the flow
version : string
Version information associated with flow.
Returns
-------
flow_exist : int
flow id iff exists, False otherwise
Notes
-----
see http://www.openml.org/api_docs/#!/flow/get_flow_exists_name_version
"""
if not (isinstance(name, six.string_types) and len(name) > 0):
raise ValueError('Argument \'name\' should be a non-empty string')
if not (isinstance(name, six.string_types) and len(external_version) > 0):
raise ValueError('Argument \'version\' should be a non-empty string')
xml_response = _perform_api_call("flow/exists",
data={'name': name, 'external_version': external_version})
result_dict = xmltodict.parse(xml_response)
flow_id = int(result_dict['oml:flow_exists']['oml:id'])
if flow_id > 0:
return flow_id
else:
return False
def test_from_xml_to_xml(self):
# Get the raw xml thing
# TODO maybe get this via get_flow(), which would have to be refactored to allow getting only the xml dictionary
# TODO: no sklearn flows.
for flow_id in [3, 5, 7, 9, ]:
flow_xml = _perform_api_call("flow/%d" % flow_id)
flow_dict = xmltodict.parse(flow_xml)
flow = openml.OpenMLFlow._from_dict(flow_dict)
new_xml = flow._to_xml()
flow_xml = flow_xml.replace(' ', '').replace('\t', '').strip().replace('\n\n', '\n').replace('"', '"')
flow_xml = re.sub(r'^$', '', flow_xml)
new_xml = new_xml.replace(' ', '').replace('\t', '').strip().replace('\n\n', '\n').replace('"', '"')
new_xml = re.sub(r'^$', '', new_xml)
self.assertEqual(new_xml, flow_xml)
def get_flow(flow_id):
"""Download the OpenML flow for a given flow ID.
Parameters
----------
flow_id : int
The OpenML flow id.
"""
# TODO add caching here!
try:
flow_id = int(flow_id)
except:
raise ValueError("Flow ID must be an int, got %s." % str(flow_id))
flow_xml = _perform_api_call("flow/%d" % flow_id)
flow_dict = xmltodict.parse(flow_xml)
flow = OpenMLFlow._from_dict(flow_dict)
return flow
def get_flow(flow_id):
"""Download the OpenML flow for a given flow ID.
Parameters
----------
flow_id : int
The OpenML flow id.
"""
# TODO add caching here!
try:
flow_id = int(flow_id)
except:
raise ValueError("Flow ID must be an int, got %s." % str(flow_id))
flow_xml = _perform_api_call("flow/%d" % flow_id)
flow_dict = xmltodict.parse(flow_xml)
flow = OpenMLFlow._from_dict(flow_dict)
return flow
def _list_flows(api_call):
# TODO add proper error handling here!
xml_string = _perform_api_call(api_call)
flows_dict = xmltodict.parse(xml_string)
# Minimalistic check if the XML is useful
assert type(flows_dict['oml:flows']['oml:flow']) == list, \
type(flows_dict['oml:flows'])
assert flows_dict['oml:flows']['@xmlns:oml'] == \
'http://openml.org/openml', flows_dict['oml:flows']['@xmlns:oml']
flows = dict()
for flow_ in flows_dict['oml:flows']['oml:flow']:
fid = int(flow_['oml:id'])
flow = {'id': fid,
'full_name': flow_['oml:full_name'],
'name': flow_['oml:name'],
'version': flow_['oml:version'],
'external_version': flow_['oml:external_version'],
'uploader': flow_['oml:uploader']}
flows[fid] = flow
return flows
