def test_polygon_lenght(self):
params = {'distance': 3000}
link_strategy = Polygon(params)
ds_a = App.get_ds(self.gf1.get_id())
ds_b = App.get_ds(self.gf2.get_id())
dataset = link_strategy.link(ds_a, ds_b)
self.assertEqual(len(dataset.index), 9)
def test_empty_polygon(self):
params = {'distance': 3}
link_strategy = Polygon(params)
ds_a = App.get_ds(self.gf1.get_id())
ds_b = App.get_ds(self.gf2.get_id())
dataset = link_strategy.link(ds_a, ds_b)
self.assertEqual(dataset.size, 0)
def test_closest_point_lenght_no_filter(self):
params = {'filter': False, 'distance': 0}
link_strategy = ClosestPoint(params)
ds_a = App.get_ds(self.gf1.get_id())
ds_b = App.get_ds(self.gf2.get_id())
dataset = link_strategy.link(ds_a, ds_b)
self.assertEqual(len(dataset.index), 3)
def test_empty_closest_point(self):
params = {'filter': True, 'distance': 3}
link_strategy = ClosestPoint(params)
ds_a = App.get_ds(self.gf1.get_id())
ds_b = App.get_ds(self.gf2.get_id())
dataset = link_strategy.link(ds_a, ds_b)
self.assertEqual(dataset.size, 0)
def layers_configuration(request, pk_a, pk_b):
popup_data = request.data['popup_data']
colours = request.data['colours']
name = request.data['name']
description = request.data['description']
App.save_layers_configuration(pk_a, pk_b, popup_data, colours, name,
description)
return Response(data={"result": "ok"}, status=status.HTTP_200_OK)
def link_similarity_preview(request, pk_a, pk_b):
rules = request.data['rules']
results = App.link_files_similarity_preview(pk_a, pk_b, {'rules': rules})
[data, cols] = App.make_response_link(results)
return Response(data={
"data": data,
"cols": cols
},
status=status.HTTP_200_OK)
def link_polygon_preview(request, pk_a, pk_b, max_distance):
params = {'distance': max_distance / 1000}
results = App.link_files_polygon_preview(pk_a, pk_b, params)
[data, cols] = App.make_response_link(results)
return Response(data={
"data": data,
"cols": cols
},
status=status.HTTP_200_OK)
def link_closest_point_preview(request, pk_a, pk_b):
params = {'distance': 0, 'filter': False}
results = App.link_files_closest_point_preview(pk_a, pk_b, params)
[data, cols] = App.make_response_link(results)
return Response(data={
"data": data,
"cols": cols
},
status=status.HTTP_200_OK)
def test_polygon(self):
params = {'distance': 3000}
link_strategy = Polygon(params)
ds_a = App.get_ds(self.gf1.get_id())
ds_b = App.get_ds(self.gf2.get_id())
dataset = link_strategy.link(ds_a, ds_b)
linked_file = App.link_polygon(self.gf1.get_id(), self.gf2.get_id(), "name", "description", params)
dataset_linked = linked_file.get_data()
pd.testing.assert_frame_equal(dataset, dataset_linked)
def test_closest_point(self):
params = {'distance': 0, 'filter': False}
link_strategy = ClosestPoint(params)
ds_a = App.get_ds(self.gf1.get_id())
ds_b = App.get_ds(self.gf2.get_id())
dataset = link_strategy.link(ds_a, ds_b)
linked_file = App.link_closest_points(self.gf1.get_id(), self.gf2.get_id(), "name", "description", params)
dataset_linked = linked_file.get_data()
pd.testing.assert_frame_equal(dataset, dataset_linked)
def test_link_similarity_same_y_column(self):
r_1 = Rule('verificado', 'JURISDICCI', [{"Verificado": "PROV"}])
r_2 = Rule('valorVinchuca', 'JURISDICCI', [{"patagonica": "PROV"}])
rules = [r_1, r_2]
params = {'rules': rules}
link_strategy = Similarity(params)
ds_a = App.get_ds(self.gf1.get_id())
ds_b = App.get_ds(self.gf2.get_id())
dataset = link_strategy.link(ds_a, ds_b)
self.assertTrue(len(dataset.index) == 1)
def link_polygon(request, pk_a, pk_b, max_distance, name, description):
params = {'distance': max_distance / 1000}
linked_file = App.link_polygon(pk_a, pk_b, name, description, params)
[data, cols] = App.make_response_link(linked_file.get_data())
return Response(data={
"data": data,
"id": linked_file.get_id(),
"cols": cols
},
status=status.HTTP_200_OK)
def link_similarity(request, pk_a, pk_b, name, description):
rules = request.data['rules']
linked_file = App.link_similarity(pk_a, pk_b, name, description,
{'rules': rules})
[data, cols] = App.make_response_link(linked_file.get_data())
return Response(data={
"data": data,
"id": linked_file.get_id(),
"cols": cols
},
status=status.HTTP_200_OK)
def link_closest_point(request, pk_a, pk_b, name, description):
params = {'distance': 0, 'filter': False}
linked_file = App.link_closest_points(pk_a, pk_b, name, description,
params)
[data, cols] = App.make_response_link(linked_file.get_data())
return Response(data={
"data": data,
"id": linked_file.get_id(),
"cols": cols
},
status=status.HTTP_200_OK)
def link_closest_point_filter(request, pk_a, pk_b, max_distance, name,
description):
# create the strategy
params = {'distance': max_distance / 1000, 'filter': True}
linked_file = App.link_closest_points(pk_a, pk_b, name, description,
params)
[data, cols] = App.make_response_link(linked_file.get_data())
return Response(data={
"data": data,
"id": linked_file.get_id(),
"cols": cols
},
status=status.HTTP_200_OK)
def clusterize_meanshift_preview(request, pk_ids, col_a, col_b):
results = App.clusterize_meanshift_preview(pk_ids, col_a, col_b)
[centroids, labels, data, cluster_size, cats,
cols] = App.make_response_cluster(results)
return Response(data={
"centroids": centroids,
"labels": labels,
"data": data,
"cluster_size": cluster_size,
"cats": cats,
"cols": cols
},
status=status.HTTP_200_OK)
def test_categorize_uncategorize_values(self):
dataset = App.get_ds(self.gf2.get_id()).get_data()
cat_col = Categorizer.categorize_column(dataset, 'p1')
dataset['p1_cat'] = cat_col
cat_value = dataset.iloc[0]['p1_cat']
real_value = dataset.iloc[0]['p1']
uncat_value = Categorizer.uncategorize_value(dataset, 'p1', cat_value)
self.assertEqual(real_value, uncat_value)
def get_clusterization(request, pk_conf):
[
name, description, col_a, col_b, cols, data, strategy, labels,
centroids, lat_col, lon_col, cats
] = App.get_cluster_configuration(pk_conf)
return Response(data={
"name": name,
"description": description,
"col_a": col_a,
"col_b": col_b,
"data": data,
"strategy": strategy,
"cols": cols,
"cats": cats,
"cluster_size": len(centroids),
"centroids": centroids,
"labels": labels,
"lat": lat_col,
"lon": lon_col
},
status=status.HTTP_200_OK)
def get_clusterizations(request):
confs = App.get_clusterizations()
serializer = ConfigurationSerializer(confs, many=True)
return JsonResponse(serializer.data, safe=False)
def test_get_details(self):
ds_a = App.get_ds(self.gf.get_id())
details = ds_a.get_details()
self.assertEqual(details['name'], 'geofile test')
self.assertEqual(details['description'], 'description')
self.assertEqual(details['doc'], 'test_files/GeoVin_sample.csv')
def test_categorize_column_long(self):
dataset = App.get_ds(self.gf2.get_id()).get_data()
cat_col = Categorizer.categorize_column(dataset, 'p1')
uncat_col = dataset[['p1']]
self.assertEqual(len(cat_col), len(uncat_col))
def clusterize_kmeans(request, pk_ids, name, description, col_a, col_b, k=3):
results = App.clusterize_kmeans(name, description, pk_ids, col_a, col_b, k)
return Response(data={"results": "OK"}, status=status.HTTP_200_OK)
def get_ds_details(request, pk_ids):
details = App.get_details(pk_ids)
return Response(data={"details": details}, status=status.HTTP_200_OK)
def get_configuration(request, pk_conf):
conf = App.get_configuration(pk_conf)
return Response(data={"conf": conf}, status=status.HTTP_200_OK)
def data_files(request):
files = App.get_data_files()
serializer = IDataSourceSerializer(files, many=True)
return JsonResponse(serializer.data, safe=False)
