cpython/Lib/test/test_collections.py

506 lines
22 KiB
Python
Raw Normal View History

2008-01-10 20:23:13 -04:00
import unittest, doctest
2007-03-01 02:16:43 -04:00
from test import test_support
from collections import namedtuple, Counter, Mapping
import pickle, cPickle, copy
from random import randrange
import operator
from collections import Hashable, Iterable, Iterator
from collections import Sized, Container, Callable
from collections import Set, MutableSet
from collections import Mapping, MutableMapping
from collections import Sequence, MutableSequence
TestNT = namedtuple('TestNT', 'x y z') # type used for pickle tests
2007-03-01 02:16:43 -04:00
class TestNamedTuple(unittest.TestCase):
def test_factory(self):
2007-10-23 17:37:41 -03:00
Point = namedtuple('Point', 'x y')
2007-03-01 02:16:43 -04:00
self.assertEqual(Point.__name__, 'Point')
self.assertEqual(Point.__doc__, 'Point(x, y)')
self.assertEqual(Point.__slots__, ())
self.assertEqual(Point.__module__, __name__)
self.assertEqual(Point.__getitem__, tuple.__getitem__)
self.assertEqual(Point._fields, ('x', 'y'))
2007-10-16 18:28:32 -03:00
2007-10-23 17:37:41 -03:00
self.assertRaises(ValueError, namedtuple, 'abc%', 'efg ghi') # type has non-alpha char
self.assertRaises(ValueError, namedtuple, 'class', 'efg ghi') # type has keyword
self.assertRaises(ValueError, namedtuple, '9abc', 'efg ghi') # type starts with digit
2007-10-16 18:28:32 -03:00
2007-10-23 17:37:41 -03:00
self.assertRaises(ValueError, namedtuple, 'abc', 'efg g%hi') # field with non-alpha char
self.assertRaises(ValueError, namedtuple, 'abc', 'abc class') # field has keyword
self.assertRaises(ValueError, namedtuple, 'abc', '8efg 9ghi') # field starts with digit
2007-12-13 22:49:47 -04:00
self.assertRaises(ValueError, namedtuple, 'abc', '_efg ghi') # field with leading underscore
2007-10-23 17:37:41 -03:00
self.assertRaises(ValueError, namedtuple, 'abc', 'efg efg ghi') # duplicate field
2007-10-16 18:28:32 -03:00
2007-10-23 17:37:41 -03:00
namedtuple('Point0', 'x1 y2') # Verify that numbers are allowed in names
2007-12-13 22:49:47 -04:00
namedtuple('_', 'a b c') # Test leading underscores in a typename
2007-03-01 02:16:43 -04:00
nt = namedtuple('nt', u'the quick brown fox') # check unicode input
self.assert_("u'" not in repr(nt._fields))
nt = namedtuple('nt', (u'the', u'quick')) # check unicode input
self.assert_("u'" not in repr(nt._fields))
self.assertRaises(TypeError, Point._make, [11]) # catch too few args
self.assertRaises(TypeError, Point._make, [11, 22, 33]) # catch too many args
2007-03-01 02:16:43 -04:00
def test_instance(self):
2007-10-23 17:37:41 -03:00
Point = namedtuple('Point', 'x y')
2007-03-01 02:16:43 -04:00
p = Point(11, 22)
self.assertEqual(p, Point(x=11, y=22))
self.assertEqual(p, Point(11, y=22))
self.assertEqual(p, Point(y=22, x=11))
self.assertEqual(p, Point(*(11, 22)))
self.assertEqual(p, Point(**dict(x=11, y=22)))
self.assertRaises(TypeError, Point, 1) # too few args
self.assertRaises(TypeError, Point, 1, 2, 3) # too many args
self.assertRaises(TypeError, eval, 'Point(XXX=1, y=2)', locals()) # wrong keyword argument
self.assertRaises(TypeError, eval, 'Point(x=1)', locals()) # missing keyword argument
self.assertEqual(repr(p), 'Point(x=11, y=22)')
self.assert_('__dict__' not in dir(p)) # verify instance has no dict
self.assert_('__weakref__' not in dir(p))
self.assertEqual(p, Point._make([11, 22])) # test _make classmethod
2007-12-13 22:49:47 -04:00
self.assertEqual(p._fields, ('x', 'y')) # test _fields attribute
self.assertEqual(p._replace(x=1), (1, 22)) # test _replace method
self.assertEqual(p._asdict(), dict(x=11, y=22)) # test _asdict method
try:
p._replace(x=1, error=2)
except ValueError:
pass
else:
self._fail('Did not detect an incorrect fieldname')
# verify that field string can have commas
2007-10-23 17:37:41 -03:00
Point = namedtuple('Point', 'x, y')
p = Point(x=11, y=22)
self.assertEqual(repr(p), 'Point(x=11, y=22)')
2007-03-01 02:16:43 -04:00
# verify that fieldspec can be a non-string sequence
2007-10-23 17:37:41 -03:00
Point = namedtuple('Point', ('x', 'y'))
p = Point(x=11, y=22)
self.assertEqual(repr(p), 'Point(x=11, y=22)')
2007-03-01 02:16:43 -04:00
def test_tupleness(self):
2007-10-23 17:37:41 -03:00
Point = namedtuple('Point', 'x y')
2007-03-01 02:16:43 -04:00
p = Point(11, 22)
self.assert_(isinstance(p, tuple))
self.assertEqual(p, (11, 22)) # matches a real tuple
self.assertEqual(tuple(p), (11, 22)) # coercable to a real tuple
self.assertEqual(list(p), [11, 22]) # coercable to a list
self.assertEqual(max(p), 22) # iterable
self.assertEqual(max(*p), 22) # star-able
x, y = p
self.assertEqual(p, (x, y)) # unpacks like a tuple
self.assertEqual((p[0], p[1]), (11, 22)) # indexable like a tuple
self.assertRaises(IndexError, p.__getitem__, 3)
self.assertEqual(p.x, x)
self.assertEqual(p.y, y)
self.assertRaises(AttributeError, eval, 'p.z', locals())
def test_odd_sizes(self):
2007-10-23 17:37:41 -03:00
Zero = namedtuple('Zero', '')
self.assertEqual(Zero(), ())
self.assertEqual(Zero._make([]), ())
self.assertEqual(repr(Zero()), 'Zero()')
self.assertEqual(Zero()._asdict(), {})
self.assertEqual(Zero()._fields, ())
2007-10-23 17:37:41 -03:00
Dot = namedtuple('Dot', 'd')
self.assertEqual(Dot(1), (1,))
self.assertEqual(Dot._make([1]), (1,))
self.assertEqual(Dot(1).d, 1)
self.assertEqual(repr(Dot(1)), 'Dot(d=1)')
self.assertEqual(Dot(1)._asdict(), {'d':1})
self.assertEqual(Dot(1)._replace(d=999), (999,))
self.assertEqual(Dot(1)._fields, ('d',))
n = 5000
import string, random
names = list(set(''.join([random.choice(string.ascii_letters)
for j in range(10)]) for i in range(n)))
n = len(names)
Big = namedtuple('Big', names)
b = Big(*range(n))
self.assertEqual(b, tuple(range(n)))
self.assertEqual(Big._make(range(n)), tuple(range(n)))
for pos, name in enumerate(names):
self.assertEqual(getattr(b, name), pos)
repr(b) # make sure repr() doesn't blow-up
d = b._asdict()
d_expected = dict(zip(names, range(n)))
self.assertEqual(d, d_expected)
b2 = b._replace(**dict([(names[1], 999),(names[-5], 42)]))
b2_expected = range(n)
b2_expected[1] = 999
b2_expected[-5] = 42
self.assertEqual(b2, tuple(b2_expected))
self.assertEqual(b._fields, tuple(names))
def test_pickle(self):
p = TestNT(x=10, y=20, z=30)
for module in pickle, cPickle:
loads = getattr(module, 'loads')
dumps = getattr(module, 'dumps')
for protocol in -1, 0, 1, 2:
q = loads(dumps(p, protocol))
self.assertEqual(p, q)
self.assertEqual(p._fields, q._fields)
def test_copy(self):
p = TestNT(x=10, y=20, z=30)
for copier in copy.copy, copy.deepcopy:
q = copier(p)
self.assertEqual(p, q)
self.assertEqual(p._fields, q._fields)
class TestOneTrickPonyABCs(unittest.TestCase):
def test_Hashable(self):
# Check some non-hashables
non_samples = [list(), set(), dict()]
for x in non_samples:
self.failIf(isinstance(x, Hashable), repr(x))
self.failIf(issubclass(type(x), Hashable), repr(type(x)))
# Check some hashables
samples = [None,
int(), float(), complex(),
str(),
tuple(), frozenset(),
int, list, object, type,
]
for x in samples:
self.failUnless(isinstance(x, Hashable), repr(x))
self.failUnless(issubclass(type(x), Hashable), repr(type(x)))
self.assertRaises(TypeError, Hashable)
# Check direct subclassing
class H(Hashable):
def __hash__(self):
return super(H, self).__hash__()
__eq__ = Hashable.__eq__ # Silence Py3k warning
self.assertEqual(hash(H()), 0)
self.failIf(issubclass(int, H))
def test_Iterable(self):
# Check some non-iterables
non_samples = [None, 42, 3.14, 1j]
for x in non_samples:
self.failIf(isinstance(x, Iterable), repr(x))
self.failIf(issubclass(type(x), Iterable), repr(type(x)))
# Check some iterables
samples = [str(),
tuple(), list(), set(), frozenset(), dict(),
dict().keys(), dict().items(), dict().values(),
(lambda: (yield))(),
(x for x in []),
]
for x in samples:
self.failUnless(isinstance(x, Iterable), repr(x))
self.failUnless(issubclass(type(x), Iterable), repr(type(x)))
# Check direct subclassing
class I(Iterable):
def __iter__(self):
return super(I, self).__iter__()
self.assertEqual(list(I()), [])
self.failIf(issubclass(str, I))
def test_Iterator(self):
non_samples = [None, 42, 3.14, 1j, "".encode('ascii'), "", (), [],
{}, set()]
for x in non_samples:
self.failIf(isinstance(x, Iterator), repr(x))
self.failIf(issubclass(type(x), Iterator), repr(type(x)))
samples = [iter(str()),
iter(tuple()), iter(list()), iter(dict()),
iter(set()), iter(frozenset()),
iter(dict().keys()), iter(dict().items()),
iter(dict().values()),
(lambda: (yield))(),
(x for x in []),
]
for x in samples:
self.failUnless(isinstance(x, Iterator), repr(x))
self.failUnless(issubclass(type(x), Iterator), repr(type(x)))
def test_Sized(self):
non_samples = [None, 42, 3.14, 1j,
(lambda: (yield))(),
(x for x in []),
]
for x in non_samples:
self.failIf(isinstance(x, Sized), repr(x))
self.failIf(issubclass(type(x), Sized), repr(type(x)))
samples = [str(),
tuple(), list(), set(), frozenset(), dict(),
dict().keys(), dict().items(), dict().values(),
]
for x in samples:
self.failUnless(isinstance(x, Sized), repr(x))
self.failUnless(issubclass(type(x), Sized), repr(type(x)))
def test_Container(self):
non_samples = [None, 42, 3.14, 1j,
(lambda: (yield))(),
(x for x in []),
]
for x in non_samples:
self.failIf(isinstance(x, Container), repr(x))
self.failIf(issubclass(type(x), Container), repr(type(x)))
samples = [str(),
tuple(), list(), set(), frozenset(), dict(),
dict().keys(), dict().items(),
]
for x in samples:
self.failUnless(isinstance(x, Container), repr(x))
self.failUnless(issubclass(type(x), Container), repr(type(x)))
def test_Callable(self):
non_samples = [None, 42, 3.14, 1j,
"", "".encode('ascii'), (), [], {}, set(),
(lambda: (yield))(),
(x for x in []),
]
for x in non_samples:
self.failIf(isinstance(x, Callable), repr(x))
self.failIf(issubclass(type(x), Callable), repr(type(x)))
samples = [lambda: None,
type, int, object,
len,
list.append, [].append,
]
for x in samples:
self.failUnless(isinstance(x, Callable), repr(x))
self.failUnless(issubclass(type(x), Callable), repr(type(x)))
def test_direct_subclassing(self):
for B in Hashable, Iterable, Iterator, Sized, Container, Callable:
class C(B):
pass
self.failUnless(issubclass(C, B))
self.failIf(issubclass(int, C))
def test_registration(self):
for B in Hashable, Iterable, Iterator, Sized, Container, Callable:
class C:
__metaclass__ = type
__hash__ = None # Make sure it isn't hashable by default
self.failIf(issubclass(C, B), B.__name__)
B.register(C)
self.failUnless(issubclass(C, B))
class TestCollectionABCs(unittest.TestCase):
# XXX For now, we only test some virtual inheritance properties.
# We should also test the proper behavior of the collection ABCs
# as real base classes or mix-in classes.
def test_Set(self):
for sample in [set, frozenset]:
self.failUnless(isinstance(sample(), Set))
self.failUnless(issubclass(sample, Set))
2008-06-23 00:29:28 -03:00
def test_hash_Set(self):
class OneTwoThreeSet(Set):
def __init__(self):
self.contents = [1, 2, 3]
def __contains__(self, x):
return x in self.contents
def __len__(self):
return len(self.contents)
def __iter__(self):
return iter(self.contents)
def __hash__(self):
return self._hash()
a, b = OneTwoThreeSet(), OneTwoThreeSet()
self.failUnless(hash(a) == hash(b))
def test_MutableSet(self):
self.failUnless(isinstance(set(), MutableSet))
self.failUnless(issubclass(set, MutableSet))
self.failIf(isinstance(frozenset(), MutableSet))
self.failIf(issubclass(frozenset, MutableSet))
def test_Mapping(self):
for sample in [dict]:
self.failUnless(isinstance(sample(), Mapping))
self.failUnless(issubclass(sample, Mapping))
def test_MutableMapping(self):
for sample in [dict]:
self.failUnless(isinstance(sample(), MutableMapping))
self.failUnless(issubclass(sample, MutableMapping))
def test_Sequence(self):
for sample in [tuple, list, str]:
self.failUnless(isinstance(sample(), Sequence))
self.failUnless(issubclass(sample, Sequence))
self.failUnless(issubclass(basestring, Sequence))
def test_MutableSequence(self):
for sample in [tuple, str]:
self.failIf(isinstance(sample(), MutableSequence))
self.failIf(issubclass(sample, MutableSequence))
for sample in [list]:
self.failUnless(isinstance(sample(), MutableSequence))
self.failUnless(issubclass(sample, MutableSequence))
self.failIf(issubclass(basestring, MutableSequence))
class TestCounter(unittest.TestCase):
def test_basics(self):
c = Counter('abcaba')
self.assertEqual(c, Counter({'a':3 , 'b': 2, 'c': 1}))
self.assertEqual(c, Counter(a=3, b=2, c=1))
self.assert_(isinstance(c, dict))
self.assert_(isinstance(c, Mapping))
self.assert_(issubclass(Counter, dict))
self.assert_(issubclass(Counter, Mapping))
self.assertEqual(len(c), 3)
self.assertEqual(sum(c.values()), 6)
self.assertEqual(sorted(c.values()), [1, 2, 3])
self.assertEqual(sorted(c.keys()), ['a', 'b', 'c'])
self.assertEqual(sorted(c), ['a', 'b', 'c'])
self.assertEqual(sorted(c.items()),
[('a', 3), ('b', 2), ('c', 1)])
self.assertEqual(c['b'], 2)
self.assertEqual(c['z'], 0)
self.assertEqual(c.has_key('c'), True)
self.assertEqual(c.has_key('z'), False)
self.assertEqual(c.__contains__('c'), True)
self.assertEqual(c.__contains__('z'), False)
self.assertEqual(c.get('b', 10), 2)
self.assertEqual(c.get('z', 10), 10)
self.assertEqual(c, dict(a=3, b=2, c=1))
self.assertEqual(repr(c), "Counter({'a': 3, 'b': 2, 'c': 1})")
self.assertEqual(c.most_common(), [('a', 3), ('b', 2), ('c', 1)])
for i in range(5):
self.assertEqual(c.most_common(i),
[('a', 3), ('b', 2), ('c', 1)][:i])
self.assertEqual(''.join(sorted(c.elements())), 'aaabbc')
c['a'] += 1 # increment an existing value
c['b'] -= 2 # sub existing value to zero
del c['c'] # remove an entry
del c['c'] # make sure that del doesn't raise KeyError
c['d'] -= 2 # sub from a missing value
c['e'] = -5 # directly assign a missing value
c['f'] += 4 # add to a missing value
self.assertEqual(c, dict(a=4, b=0, d=-2, e=-5, f=4))
self.assertEqual(''.join(sorted(c.elements())), 'aaaaffff')
self.assertEqual(c.pop('f'), 4)
self.assertEqual('f' in c, False)
for i in range(3):
elem, cnt = c.popitem()
self.assertEqual(elem in c, False)
c.clear()
self.assertEqual(c, {})
self.assertEqual(repr(c), 'Counter()')
self.assertRaises(NotImplementedError, Counter.fromkeys, 'abc')
self.assertRaises(TypeError, hash, c)
c.update(dict(a=5, b=3))
c.update(c=1)
c.update(Counter('a' * 50 + 'b' * 30))
c.update() # test case with no args
c.__init__('a' * 500 + 'b' * 300)
c.__init__('cdc')
c.__init__()
self.assertEqual(c, dict(a=555, b=333, c=3, d=1))
self.assertEqual(c.setdefault('d', 5), 1)
self.assertEqual(c['d'], 1)
self.assertEqual(c.setdefault('e', 5), 5)
self.assertEqual(c['e'], 5)
def test_copying(self):
# Check that counters are copyable, deepcopyable, picklable, and
#have a repr/eval round-trip
words = Counter('which witch had which witches wrist watch'.split())
update_test = Counter()
update_test.update(words)
for i, dup in enumerate([
words.copy(),
copy.copy(words),
copy.deepcopy(words),
pickle.loads(pickle.dumps(words, 0)),
pickle.loads(pickle.dumps(words, 1)),
pickle.loads(pickle.dumps(words, 2)),
pickle.loads(pickle.dumps(words, -1)),
cPickle.loads(cPickle.dumps(words, 0)),
cPickle.loads(cPickle.dumps(words, 1)),
cPickle.loads(cPickle.dumps(words, 2)),
cPickle.loads(cPickle.dumps(words, -1)),
eval(repr(words)),
update_test,
Counter(words),
]):
msg = (i, dup, words)
self.assert_(dup is not words)
self.assertEquals(dup, words)
self.assertEquals(len(dup), len(words))
self.assertEquals(type(dup), type(words))
def test_conversions(self):
# Convert to: set, list, dict
s = 'she sells sea shells by the sea shore'
self.assertEqual(sorted(Counter(s).elements()), sorted(s))
self.assertEqual(sorted(Counter(s)), sorted(set(s)))
self.assertEqual(dict(Counter(s)), dict(Counter(s).items()))
self.assertEqual(set(Counter(s)), set(s))
2009-01-21 19:12:51 -04:00
def test_invariant_for_the_in_operator(self):
c = Counter(a=10, b=-2, c=0)
for elem in c:
self.assert_(elem in c)
def test_multiset_operations(self):
# Verify that adding a zero counter will strip zeros and negatives
c = Counter(a=10, b=-2, c=0) + Counter()
self.assertEqual(dict(c), dict(a=10))
elements = 'abcd'
for i in range(1000):
# test random pairs of multisets
p = Counter(dict((elem, randrange(-2,4)) for elem in elements))
p.update(e=1, f=-1, g=0)
q = Counter(dict((elem, randrange(-2,4)) for elem in elements))
q.update(h=1, i=-1, j=0)
for counterop, numberop in [
(Counter.__add__, lambda x, y: max(0, x+y)),
(Counter.__sub__, lambda x, y: max(0, x-y)),
(Counter.__or__, lambda x, y: max(0,x,y)),
(Counter.__and__, lambda x, y: max(0, min(x,y))),
]:
result = counterop(p, q)
for x in elements:
self.assertEqual(numberop(p[x], q[x]), result[x],
(counterop, x, p, q))
# verify that results exclude non-positive counts
self.assert_(x>0 for x in result.values())
elements = 'abcdef'
for i in range(100):
# verify that random multisets with no repeats are exactly like sets
p = Counter(dict((elem, randrange(0, 2)) for elem in elements))
q = Counter(dict((elem, randrange(0, 2)) for elem in elements))
for counterop, setop in [
(Counter.__sub__, set.__sub__),
(Counter.__or__, set.__or__),
(Counter.__and__, set.__and__),
]:
counter_result = counterop(p, q)
set_result = setop(set(p.elements()), set(q.elements()))
self.assertEqual(counter_result, dict.fromkeys(set_result, 1))
2008-01-10 20:23:13 -04:00
import doctest, collections
2007-03-01 02:16:43 -04:00
def test_main(verbose=None):
NamedTupleDocs = doctest.DocTestSuite(module=collections)
test_classes = [TestNamedTuple, NamedTupleDocs, TestOneTrickPonyABCs,
TestCollectionABCs, TestCounter]
2007-03-01 02:16:43 -04:00
test_support.run_unittest(*test_classes)
test_support.run_doctest(collections, verbose)
2007-03-01 02:16:43 -04:00
if __name__ == "__main__":
test_main(verbose=True)