def __init__(self,
source_agregated_scores,
out_file=None,
n_desirable_complete_classes=1000,
n_instances_already_counted=False):
self._in_ag_scores = source_agregated_scores
self._out_file = out_file
self._n_desirable = n_desirable_complete_classes
self._summary_list = []
self._n_instances_already_counted = n_instances_already_counted
# Communications
self._api_reader = WikidataApiReader()
class AliasesPropertiesCommand(object):
def __init__(self, source_file, out_file, json_input=False):
self._in_file = source_file
self._out_file = out_file
self._is_json_input = json_input
self._api_reader = WikidataApiReader()
def exec_command(self, string_return=False):
sorted_result_list = []
for a_property in self._read_target_properties():
try:
sorted_result_list.append(self._get_complete_property(a_property))
except:
print "Error with property " + str(a_property)
return JsonPropertyDumper(
out_file=self._out_file,
indent=4,
strict_mode=True,
string_return=string_return,
needed_fields=[P_ID, P_LABEL, P_APPEARANCES, P_DESC],
).persist_properties(sorted_result_list)
def _read_target_properties(self):
if not self._is_json_input:
for a_prop in Ccv01PropertiesParser(source_file=self._in_file).yield_properties():
yield a_prop
else:
for a_prop in Json05PropertiesParser(source_file=self._in_file).yield_properties():
yield a_prop
def _get_complete_property(self, anemic_property):
tmp = self._api_reader.get_property(anemic_property.id)
tmp.n_appearances = anemic_property.n_appearances
return tmp
def __init__(self, source_agregated_scores, out_file=None, n_desirable_complete_classes=1000,
n_instances_already_counted=False):
self._in_ag_scores = source_agregated_scores
self._out_file = out_file
self._n_desirable = n_desirable_complete_classes
self._summary_list = []
self._n_instances_already_counted = n_instances_already_counted
# Communications
self._api_reader = WikidataApiReader()
class AliasesPropertiesCommand(object):
def __init__(self, source_file, out_file, json_input=False):
self._in_file = source_file
self._out_file = out_file
self._is_json_input = json_input
self._api_reader = WikidataApiReader()
def exec_command(self, string_return=False):
sorted_result_list = []
for a_property in self._read_target_properties():
try:
sorted_result_list.append(
self._get_complete_property(a_property))
except:
print "Error with property " + str(a_property)
return JsonPropertyDumper(out_file=self._out_file,
indent=4,
strict_mode=True,
string_return=string_return,
needed_fields=[P_ID, P_LABEL, P_APPEARANCES, P_DESC]). \
persist_properties(sorted_result_list)
def _read_target_properties(self):
if not self._is_json_input:
for a_prop in Ccv01PropertiesParser(
source_file=self._in_file).yield_properties():
yield a_prop
else:
for a_prop in Json05PropertiesParser(
source_file=self._in_file).yield_properties():
yield a_prop
def _get_complete_property(self, anemic_property):
tmp = self._api_reader.get_property(anemic_property.id)
tmp.n_appearances = anemic_property.n_appearances
return tmp
def __init__(self, source_file, out_file, json_input=False):
self._in_file = source_file
self._out_file = out_file
self._is_json_input = json_input
self._api_reader = WikidataApiReader()
def __init__(self, source_file, out_file, json_input=False):
self._in_file = source_file
self._out_file = out_file
self._is_json_input = json_input
self._api_reader = WikidataApiReader()
class AgregatedClassSummaryCommand(object):
def __init__(self,
source_agregated_scores,
out_file=None,
n_desirable_complete_classes=1000,
n_instances_already_counted=False):
self._in_ag_scores = source_agregated_scores
self._out_file = out_file
self._n_desirable = n_desirable_complete_classes
self._summary_list = []
self._n_instances_already_counted = n_instances_already_counted
# Communications
self._api_reader = WikidataApiReader()
def exec_command(self, string_return=False):
tracked_counter = 0
raw_classes_list = self._read_raw_classes()
for a_class_dict in raw_classes_list:
self._summary_list.append(
self._get_summary_dict(a_class_dict, tracked_counter))
tracked_counter += 1
print "Total: ", tracked_counter
self._serialize_results(string_return)
def _get_summary_dict(self, raw_class_dict, counter):
result = {
KEY_ID: raw_class_dict[KEY_ID],
KEY_ACCUMULATED: raw_class_dict[KEY_ACCUMULATED],
KEY_LABEL: None,
KEY_DESC: None
}
if KEY_POS_INSTANCE_COUNTING in raw_class_dict:
result[KEY_POS_INSTANCE_COUNTING] = raw_class_dict[
KEY_POS_INSTANCE_COUNTING]
if KEY_POS_CLASSRANK in raw_class_dict:
result[KEY_POS_CLASSRANK] = raw_class_dict[KEY_POS_CLASSRANK]
if not self._n_instances_already_counted:
result[KEY_N_INSTANCES] = self._count_dict_instances(
raw_class_dict)
else:
result[KEY_N_INSTANCES] = raw_class_dict[KEY_N_INSTANCES]
if counter < self._n_desirable:
try:
tracked_entity = self._api_reader.get_entity(
raw_class_dict[KEY_ID])
result[KEY_LABEL] = tracked_entity.label
result[KEY_DESC] = tracked_entity.description
except:
print "Hubo problemas con ", raw_class_dict[KEY_ID]
print counter
return result
def _count_dict_instances(self, raw_class_dict):
resultset = set()
for a_prop_key in raw_class_dict[KEY_INSTANCES]:
for an_entity_id in raw_class_dict[KEY_INSTANCES][a_prop_key]:
resultset.add(an_entity_id)
return len(resultset)
def _read_raw_classes(self):
# return [{"id": "Q31",
# KEY_ACCUMULATED: 8,
# KEY_INSTANCES: {"P1" : ["Q5", "Q3"] # 2 instances
# }
# },
# {"id": "Q35",
# KEY_ACCUMULATED: 8,
# KEY_INSTANCES:{"P1" : ["Q5", "Q3"],
# "P2" : ["Q3", "Q2", "Q1"] # 4 instances (Q3 repeated)
# }}]
return read_json_object(self._in_ag_scores)
def _serialize_results(self, string_return):
# TODO: implement string_return mode. Now, we are assuming string_return=False
write_json_object(json_object=self._summary_list,
path=self._out_file,
indent=4)
class AgregatedClassSummaryCommand(object):
def __init__(self, source_agregated_scores, out_file=None, n_desirable_complete_classes=1000,
n_instances_already_counted=False):
self._in_ag_scores = source_agregated_scores
self._out_file = out_file
self._n_desirable = n_desirable_complete_classes
self._summary_list = []
self._n_instances_already_counted = n_instances_already_counted
# Communications
self._api_reader = WikidataApiReader()
def exec_command(self, string_return=False):
tracked_counter = 0
raw_classes_list = self._read_raw_classes()
for a_class_dict in raw_classes_list:
self._summary_list.append(self._get_summary_dict(a_class_dict, tracked_counter))
tracked_counter += 1
print "Total: ", tracked_counter
self._serialize_results(string_return)
def _get_summary_dict(self, raw_class_dict, counter):
result = {KEY_ID: raw_class_dict[KEY_ID],
KEY_ACCUMULATED: raw_class_dict[KEY_ACCUMULATED],
KEY_LABEL: None,
KEY_DESC: None}
if KEY_POS_INSTANCE_COUNTING in raw_class_dict:
result[KEY_POS_INSTANCE_COUNTING] = raw_class_dict[KEY_POS_INSTANCE_COUNTING]
if KEY_POS_CLASSRANK in raw_class_dict:
result[KEY_POS_CLASSRANK] = raw_class_dict[KEY_POS_CLASSRANK]
if not self._n_instances_already_counted:
result[KEY_N_INSTANCES] = self._count_dict_instances(raw_class_dict)
else:
result[KEY_N_INSTANCES] = raw_class_dict[KEY_N_INSTANCES]
if counter < self._n_desirable:
try:
tracked_entity = self._api_reader.get_entity(raw_class_dict[KEY_ID])
result[KEY_LABEL] = tracked_entity.label
result[KEY_DESC] = tracked_entity.description
except:
print "Hubo problemas con ", raw_class_dict[KEY_ID]
print counter
return result
def _count_dict_instances(self, raw_class_dict):
resultset = set()
for a_prop_key in raw_class_dict[KEY_INSTANCES]:
for an_entity_id in raw_class_dict[KEY_INSTANCES][a_prop_key]:
resultset.add(an_entity_id)
return len(resultset)
def _read_raw_classes(self):
# return [{"id": "Q31",
# KEY_ACCUMULATED: 8,
# KEY_INSTANCES: {"P1" : ["Q5", "Q3"] # 2 instances
# }
# },
# {"id": "Q35",
# KEY_ACCUMULATED: 8,
# KEY_INSTANCES:{"P1" : ["Q5", "Q3"],
# "P2" : ["Q3", "Q2", "Q1"] # 4 instances (Q3 repeated)
# }}]
return read_json_object(self._in_ag_scores)
def _serialize_results(self, string_return):
# TODO: implement string_return mode. Now, we are assuming string_return=False
write_json_object(json_object=self._summary_list,
path=self._out_file,
indent=4)
