def test_two_level_grouping_with_collect(self):
data = [
datum(name='Jill', place='Kettering', age=[34, 36], count=2),
datum(name='Jack', place='Kennington', age=[23], count=1),
datum(name='James', place='Keswick', age=[10, 21, 32], count=3),
datum(name='James', place='Kettering', age=[43, 87], count=2),
datum(name='Jill', place='Keswick', age=[76, 32], count=2),
]
results = nested_merge([['name'], ['place']], [('age', 'mean')], data)
assert_that(
results,
contains(
has_entries({
'name':
'Jack',
'age:mean':
23,
'_subgroup':
contains(
has_entries({
'place': 'Kennington',
'age:mean': 23
}))
}),
has_entries({
'name':
'James',
'age:mean':
38.6,
'_subgroup':
contains(
has_entries({
'place': 'Keswick',
'age:mean': 21.0
}),
has_entries({
'place': 'Kettering',
'age:mean': 65.0
}))
}),
has_entries({
'name':
'Jill',
'age:mean':
44.5,
'_subgroup':
contains(
has_entries({
'place': 'Keswick',
'age:mean': 54.0
}),
has_entries({
'place': 'Kettering',
'age:mean': 35.0
}))
}),
))
def test_two_level_grouping_with_collect(self):
data = [
datum(name='Jill', place='Kettering', age=[34, 36], count=2),
datum(name='Jack', place='Kennington', age=[23], count=1),
datum(name='James', place='Keswick', age=[10, 21, 32], count=3),
datum(name='James', place='Kettering', age=[43, 87], count=2),
datum(name='Jill', place='Keswick', age=[76, 32], count=2),
]
results = nested_merge([['name'], ['place']], [('age', 'mean')], data)
assert_that(results,
contains(
has_entries({
'name': 'Jack',
'age:mean': 23,
'_subgroup': contains(
has_entries({
'place': 'Kennington',
'age:mean': 23
})
)}),
has_entries({
'name': 'James',
'age:mean': 38.6,
'_subgroup': contains(
has_entries({
'place': 'Keswick',
'age:mean': 21.0
}),
has_entries({
'place': 'Kettering',
'age:mean': 65.0
})
)
}),
has_entries({
'name': 'Jill',
'age:mean': 44.5,
'_subgroup': contains(
has_entries({
'place': 'Keswick',
'age:mean': 54.0
}),
has_entries({
'place': 'Kettering',
'age:mean': 35.0
})
)
}),
))
def test_one_level_grouping_with_collect(self):
data = [
datum(name='Jill', age=[12, 45]),
datum(name='Jack', age=[34, 34]),
datum(name='John', age=[56, 65])
]
results = nested_merge([['name']], [('age', 'mean')], data)
assert_that(results,
contains(
has_entries({'name': 'Jack', 'age:mean': 34}),
has_entries({'name': 'Jill', 'age:mean': 28.5}),
has_entries({'name': 'John', 'age:mean': 60.5}),
))
def _group(self, keys, query, sort=None, limit=None, collect=None):
collect_fields = unique_collect_fields(collect)
results = self._mongo_driver.group(keys, query, list(collect_fields))
results = nested_merge(keys, collect, results)
if sort:
sorters = {
"ascending": lambda a, b: cmp(a, b),
"descending": lambda a, b: cmp(b, a)
}
sorter = sorters[sort[1]]
try:
results.sort(cmp=sorter, key=lambda a: a[sort[0]])
except KeyError:
raise InvalidSortError('Invalid sort key {0}'.format(sort[0]))
if limit:
results = results[:limit]
return results
def test_one_level_grouping_with_collect(self):
data = [
datum(name='Jill', age=[12, 45]),
datum(name='Jack', age=[34, 34]),
datum(name='John', age=[56, 65])
]
results = nested_merge([['name']], [('age', 'mean')], data)
assert_that(
results,
contains(
has_entries({
'name': 'Jack',
'age:mean': 34
}),
has_entries({
'name': 'Jill',
'age:mean': 28.5
}),
has_entries({
'name': 'John',
'age:mean': 60.5
}),
))
def test_two_level_grouping_combination_of_keys(self):
data = [
datum(name='IE',
version='6',
place='England',
age=[13, 12],
count=2),
datum(name='IE', version='6', place='Wales', age=[13, 14],
count=2),
datum(name='IE', version='7', place='England', age=[8, 7],
count=2),
datum(name='IE', version='7', place='Wales', age=[8, 9], count=2),
datum(name='IE', version='8', place='England', age=[5, 4],
count=2),
datum(name='IE', version='8', place='Wales', age=[5, 6], count=2),
datum(name='Chrome',
version='20',
place='England',
age=[2, 1],
count=2),
datum(name='Chrome',
version='20',
place='Wales',
age=[2, 3],
count=2),
]
results = nested_merge([['name', 'version'], ['place']],
[('age', 'mean')], data)
assert_that(
results,
contains(
has_entries({
'name':
'Chrome',
'version':
'20',
'age:mean':
2,
'_subgroup':
contains(
has_entries({
'place': 'England',
'age:mean': 1.5
}), has_entries({
'place': 'Wales',
'age:mean': 2.5
}))
}),
has_entries({
'name':
'IE',
'version':
'6',
'age:mean':
13,
'_subgroup':
contains(
has_entries({
'place': 'England',
'age:mean': 12.5
}), has_entries({
'place': 'Wales',
'age:mean': 13.5
}))
}),
has_entries({
'name':
'IE',
'version':
'7',
'age:mean':
8,
'_subgroup':
contains(
has_entries({
'place': 'England',
'age:mean': 7.5
}), has_entries({
'place': 'Wales',
'age:mean': 8.5
}))
}),
has_entries({
'name':
'IE',
'version':
'8',
'age:mean':
5,
'_subgroup':
contains(
has_entries({
'place': 'England',
'age:mean': 4.5
}), has_entries({
'place': 'Wales',
'age:mean': 5.5
}))
}),
))
def test_two_level_grouping_combination_of_keys(self):
data = [
datum(name='IE', version='6', place='England', age=[13, 12], count=2),
datum(name='IE', version='6', place='Wales', age=[13, 14], count=2),
datum(name='IE', version='7', place='England', age=[8, 7], count=2),
datum(name='IE', version='7', place='Wales', age=[8, 9], count=2),
datum(name='IE', version='8', place='England', age=[5, 4], count=2),
datum(name='IE', version='8', place='Wales', age=[5, 6], count=2),
datum(name='Chrome', version='20', place='England', age=[2, 1], count=2),
datum(name='Chrome', version='20', place='Wales', age=[2, 3], count=2),
]
results = nested_merge([['name', 'version'], ['place']], [('age', 'mean')], data)
assert_that(results,
contains(
has_entries({
'name': 'Chrome',
'version': '20',
'age:mean': 2,
'_subgroup': contains(
has_entries({
'place': 'England',
'age:mean': 1.5
}),
has_entries({
'place': 'Wales',
'age:mean': 2.5
})
)
}),
has_entries({
'name': 'IE',
'version': '6',
'age:mean': 13,
'_subgroup': contains(
has_entries({
'place': 'England',
'age:mean': 12.5
}),
has_entries({
'place': 'Wales',
'age:mean': 13.5
})
)
}),
has_entries({
'name': 'IE',
'version': '7',
'age:mean': 8,
'_subgroup': contains(
has_entries({
'place': 'England',
'age:mean': 7.5
}),
has_entries({
'place': 'Wales',
'age:mean': 8.5
})
)
}),
has_entries({
'name': 'IE',
'version': '8',
'age:mean': 5,
'_subgroup': contains(
has_entries({
'place': 'England',
'age:mean': 4.5
}),
has_entries({
'place': 'Wales',
'age:mean': 5.5
})
)
}),
))
