Mercurial > hg > GlobalNeighbors
changeset 3:49aae0c0293b
improved test coverage
author | Jeff Hammel <k0scist@gmail.com> |
---|---|
date | Sat, 24 Jun 2017 14:48:31 -0700 |
parents | 50ee13cddf58 |
children | 8e130b7bfed9 |
files | globalneighbors/distance.py tests/test_distance.py tests/test_grid.py tests/test_read.py tox.ini |
diffstat | 5 files changed, 57 insertions(+), 4 deletions(-) [+] |
line wrap: on
line diff
--- a/globalneighbors/distance.py Sat Jun 24 14:04:00 2017 -0700 +++ b/globalneighbors/distance.py Sat Jun 24 14:48:31 2017 -0700 @@ -57,19 +57,21 @@ k=10, lat_tol=1., lon_tol=1., - output=1000, + output=None, neighbors=None): """ calculate `k` nearest neighbors for each location + + locations -- dict of `geoid: (lat, lon)` """ neighbors = neighbors or {} - items = locations.items() + items = locations.items() # copy index = 0 n_items = len(items) start = int(time.time()) while items: index += 1 - if not index % output: + if output and not index % output: # output status counter now = int(time.time()) duration = now - start
--- a/tests/test_distance.py Sat Jun 24 14:04:00 2017 -0700 +++ b/tests/test_distance.py Sat Jun 24 14:48:31 2017 -0700 @@ -10,6 +10,7 @@ from globalneighbors import distance from globalneighbors.constants import Rearth from globalneighbors.locations import locations +from globalneighbors.read import read_cities from globalneighbors.read import read_city_list from globalneighbors.schema import primary_key @@ -17,6 +18,7 @@ data = os.path.join(here, 'data') full_tsv_lines = 149092 + class DistanceTests(unittest.TestCase): # created with @@ -28,6 +30,9 @@ full_tsv = os.path.join(data, 'cities1000.txt') full_tsv_lines = 149092 + # here's a smaller one + moderate_tsv = os.path.join(data, '10000cities.tsv') + def test_haversine(self): # a simple canned case @@ -102,6 +107,26 @@ for i in range(1, len(distances)): assert distances[i] >= distances[i-1] + def test_10000cities(self): + """a moderate size test""" + + assert os.path.exists(self.moderate_tsv) + with open(self.moderate_tsv) as f: + cities = locations(read_cities(f)) + + # test over different values of # of neighbors + for k in (10, 100, 1000): + neighbors = distance.calculate_neighbors(cities, + k=k) + + # ensure you have no more neighbors than you ask for + assert max([len(value) for value in neighbors.values()]) <= k + + # assert distances increase + for value in neighbors.values(): + distances = [i[-1] for i in value] + assert distances == sorted(distances) + if __name__ == '__main__': unittest.main()
--- a/tests/test_grid.py Sat Jun 24 14:04:00 2017 -0700 +++ b/tests/test_grid.py Sat Jun 24 14:48:31 2017 -0700 @@ -49,9 +49,19 @@ def grid_locations(self, locations): """grid locations + test created grid""" + # create a grid grid = LatLonGrid(8, 8) + + # add the items to it for geoid, (lat, lon) in locations.items(): grid.add(geoid, lat, lon) + # iterate over the grid + n_locations = 0 + for i in range(grid.n[0]): + for j in range(grid.n[1]): + n_locations += len(grid[(i,j)]) + assert n_locations == len(locations) + if __name__ == '__main__': unittest.main()
--- a/tests/test_read.py Sat Jun 24 14:04:00 2017 -0700 +++ b/tests/test_read.py Sat Jun 24 14:48:31 2017 -0700 @@ -7,7 +7,9 @@ import os import unittest from globalneighbors import schema +from globalneighbors.locations import locations from globalneighbors.read import read_tsv +from globalneighbors.read import read_cities from globalneighbors.read import read_city_list @@ -52,6 +54,19 @@ """ensure we can read the cities as unicode""" cities = read_city_list(self.full_tsv) + for city in cities: + for field in schema.unicode_fields: + assert isinstance(city[field], unicode) + + def test_iterative_locations(self): + """assert we can read into locations as a generator""" + + with open(self.test_tsv) as f: + cities = locations(read_cities(f)) + for geonameid, (lat, lon) in cities.items(): + assert -90. <= lat <= 90. + assert -180. <= lon <= 180. + assert type(geonameid) == schema.types['geonameid'] if __name__ == '__main__':