def parseCSV(self, specfile):
csv_dictionary = ip.loadSpec(specfile, csv_flag=True)
file1 = csv_dictionary['csv1']
file2 = csv_dictionary['csv2']
key1 = csv_dictionary['csv1_key']
key2 = csv_dictionary['csv2_key']
join = csv_dictionary['join']
object_name = csv_dictionary['objects']
attribute_name = csv_dictionary['attributes']
dic = ip.constructDictFromCSVFiles(file1, file2, key1, key2,
object_name, attribute_name, join)
self.attributes = list(set(itertools.chain.from_iterable(
dic.values())))
self.num_attributes = len(self.attributes)
precontext = [(k, [self.attributes.index(e) for e in v])
for k, v in dic.items()]
self.objects, self.context = zip(*[(c[0], ''.join(
['1' if i in c[1] else '0' for i in range(self.num_attributes)]))
for c in precontext])
self.num_objects = len(self.objects)
return self
def parseQueryRes(
self, filepath,
query_res): #generates context object from json output of a query
#print('----------------------------------------')
#print('parsing query result json',filepath,query_res)
spec = ip.loadSpec(filepath, csv_flag=False)
query = spec['query']
obj_name = spec['objects']
att_name = (spec['attributes'].strip() if ',' not in spec['attributes']
else [s.strip() for s in spec['attributes'].split(',')])
query_content = json.load(open(query_res, 'r'))
dic = collections.defaultdict(list)
self.attributes = (list({e[att_name]
for e in query_content})
if type(att_name) == str else list({
str(a) + '#' + str(e[a])
for a in att_name for e in query_content
}))
self.num_attributes = len(self.attributes)
attributes_index = dict(
(self.attributes[i], i) for i in range(self.num_attributes))
if type(att_name) == str:
for e in query_content:
dic[e[obj_name]].append(attributes_index[e[att_name]])
elif type(att_name) == list:
for e in query_content:
dic[e[obj_name]].extend([
attributes_index[str(a) + '#' + str(e[a])]
for a in att_name
])
self.objects, self.context = zip(*[(k, ''.join(
['1' if i in v else '0' for i in range(self.num_attributes)]))
for k, v in dic.items()])
self.num_objects = len(self.objects)
#print('parseQueryRes done')
#print('----------------------------------------')
return self
def parseQuery(
self, filepath
): #generates context object from json query specification file
spec = ip.loadSpec(filepath, csv_flag=False)
query = spec['query']
obj_name = spec['objects']
att_name = (spec['attributes'].strip() if ',' not in spec['attributes']
else [s.strip() for s in spec['attributes'].split(',')])
if 'port' in spec.keys():
port = spec['port']
else:
port = 8080
query_res = ip.getQuery(query, port)
dic = collections.defaultdict(list)
self.attributes = (list({e[att_name]
for e in query_res})
if type(att_name) == str else list({
str(a) + '#' + str(e[a])
for a in att_name for e in query_res
}))
self.num_attributes = len(self.attributes)
attributes_index = dict(
(self.attributes[i], i) for i in range(self.num_attributes))
if type(att_name) == str:
for e in query_res:
dic[e[obj_name]].append(attributes_index[e[att_name]])
elif type(att_name) == list:
for e in query_res:
dic[e[obj_name]].extend([
attributes_index[str(a) + '#' + str(e[a])]
for a in att_name
])
self.objects, self.context = zip(*[(k, ''.join(
['1' if i in v else '0' for i in range(self.num_attributes)]))
for k, v in dic.items()])
self.num_objects = len(self.objects)
return self