def test_filter_and(self):
coll = self.coll
df = self.df
y = int(df.y.unique()[0])
result = Filter(coll, x=0, y=y).value
testdf = df[(df.x == 0) & (df.y == y)]
self.assertTrue(result.equals(testdf))
def test_filter_near(self):
om = self.om
# create a dataframe with geo locations
geodf = pd.DataFrame(locations)
geodf['location'] = geodf.location.apply(lambda v: GeoJSON(v))
om.datasets.put(geodf, 'geosample', append=False, index='@location')
coll = om.datasets.collection('geosample')
# closest place
result = Filter(coll,
location__near=dict(location=(8.541694, 47.3768866),
maxd=1))
places = result.value.place.unique()
self.assertEqual(places, ['Zurich'])
# ordered by distance
result = Filter(coll,
location__near=dict(location=(8.541694, 47.3768866)))
places = list(result.value.place.unique())
self.assertListEqual(places, 'Zurich,Bern,Geneva,New York'.split(','))
# use tuple (lon, lat, maxd)
result = Filter(coll, location__near=(8.541694, 47.3768866, 1))
places = list(result.value.place.unique())
self.assertListEqual(places, 'Zurich'.split(','))
# use tuple (lon, lat, mind, maxd)
result = Filter(coll, location__near=(8.541694, 47.3768866, 0, 100))
places = list(result.value.place.unique())
self.assertListEqual(places, 'Zurich'.split(','))
def __iter__(self):
""" for each group returns the key and a Filter object"""
groups = self._count()
for group in groups:
keys = group.get('_id')
data = Filter(self.collection, **keys)
yield keys, data
def _get_filter_criteria(self, *args, **kwargs):
"""
return mongo query from filter specs
this uses a Filter to produce the query from the kwargs.
:param args: a Q object or logical combination of Q objects
(optional)
:param kwargs: all AND filter criteria
"""
if len(args) > 0:
q = args[0]
if isinstance(q, MongoQ):
filter_criteria = Filter(self.collection, q).query
elif isinstance(q, Filter):
filter_criteria = Filter(self.collection, q.q).query
else:
filter_criteria = Filter(self.collection, **kwargs).query
return filter_criteria
def __init__(self, collection, query=None, projection=None, **kwargs):
if isinstance(collection, (Collection, PickableCollection)):
database = collection.database
name = collection.name
else:
raise ValueError('collection should be a pymongo.Collection')
query = query or {}
super(FilteredCollection, self).__init__(
database, name, create=False, **kwargs)
self.query = Filter(self, **query).query
self.projection = projection
def query(self):
return Filter(self.collection, **self._fixed_query).query
def test_filter_or(self):
coll = self.coll
df = self.df
result = Filter(coll, x=0, y__gt=5).value
testdf = df[(df.x == 0) & (df.y > 5)]
self.assertTrue(result.equals(testdf))
def test_filter_in(self):
coll = self.coll
df = self.df
result = Filter(coll, x__in=[1, 2, 3]).value
testdf = df[df.x.isin([1, 2, 3])]
self.assertTrue(result.equals(testdf))
def test_filter(self):
coll = self.coll
df = self.df
result = Filter(coll, x=0).value
testdf = df[df.x == 0]
self.assertTrue(result.equals(testdf))
def test_filter_subdoc(self):
coll = self.coll
coll.update_many(qops.IS(x=qops.LT(5)), qops.SET('subdoc.a', 99))
coll.update_many(qops.IS(x=qops.GTE(5)), qops.SET('subdoc.a', 0))
result = Filter(coll, subdoc__a__lt=10).value
self.assertEqual(set(result.x.unique()), set(range(5, 10)))