def test_clean_lazy():
G = LineFile("tests/smallcorpus-malformed.txt.bz2", header="foo bar baz qux".split(),
path="tests/tmp/testcorpus")
len_G = len(G)
G.clean(columns=4, lower=False, alphanumeric=False, count_columns=True,
nounderscores=False, echo_toss=True, lazy=True)
assert_equal(len(G), len_G - 2)
G.delete()
G = LineFile("tests/smallcorpus-malformed.txt.bz2", header="foo bar baz qux".split(),
path="tests/tmp/testcorpus")
G.clean(lower=True, alphanumeric=True, count_columns=False, echo_toss=True, lazy=True)
assert_equal(len(G), 8562)
G.delete()
G = LineFile("tests/smallcorpus-malformed.txt.bz2", header="foo bar baz qux".split(),
path="tests/tmp/testcorpus")
G.clean(lower=True, alphanumeric=True, count_columns=False, echo_toss=True,
filter_fn=lambda x: False, lazy=True)
assert_equal(len(G), 0)
G.delete()
G = LineFile("tests/smallcorpus-malformed.txt.bz2", header="foo bar baz qux".split(),
path="tests/tmp/testcorpus")
G.clean(lower=True, alphanumeric=False, count_columns=False, echo_toss=True,
modifier_fn=lambda x: "hello", lazy=True)
for line in G.lines(parts=False):
assert_equal(line, "hello")
G.delete()
def test_unicode():
""" test unicode
replace every word in the test corpus with random unicode
and see if we get the same surprisal scores.
"""
def generate_random_unicode():
for _ in xrange(5):
yield unichr(random.choice((0x300, 0x9999)) + random.randint(0, 0xff))
scramblemap = {}
G = LineFile("tests/smallcorpus.txt.bz2", header="foo bar baz qux".split(),
path="tests/tmp/testcorpus")
G.clean(lower=True, alphanumeric=False, count_columns=False, echo_toss=True, lazy=True)
G.make_marginal_column("quux", "foo bar".split(), "qux", lazy=True)
G.sort("baz")
len_G = len(G)
sum_counts = G.sum_column("quux")
sum_surprisal = math.fsum(line[2] for line in G.average_surprisal("baz", "qux", "quux", assert_sorted=True))
G.delete()
G = LineFile("tests/smallcorpus.txt.bz2", header="foo bar baz qux".split(),
path="tests/tmp/testcorpus")
def scramble(line):
words = line.split()[:3]
count = line.split()[-1]
for i, word in enumerate(words):
if word in scramblemap:
words[i] = scramblemap[word]
else:
garbage = u"".join(generate_random_unicode())
words[i] = garbage
scramblemap[word] = garbage
return "\t".join(words + [count])
G.clean(lower=True, alphanumeric=False, count_columns=False, echo_toss=True,
modifier_fn=scramble)
G.make_marginal_column("quux", "foo bar".split(), "qux")
G.sort("baz")
sum_counts_scrambled = G.sum_column("quux")
assert_equal(sum_counts, sum_counts_scrambled)
assert_equal(len_G, len(G))
sum_surprisal_scrambled = math.fsum(line[2] for line in G.average_surprisal("baz", "qux", "quux", assert_sorted=True))
G.delete()
assert_equal(sum_surprisal, sum_surprisal_scrambled)
def test_clean_lazy():
G = LineFile("tests/smallcorpus-malformed.txt.bz2",
header="foo bar baz qux".split(),
path="tests/tmp/testcorpus")
len_G = len(G)
G.clean(columns=4,
lower=False,
alphanumeric=False,
count_columns=True,
nounderscores=False,
echo_toss=True,
lazy=True)
assert_equal(len(G), len_G - 2)
G.delete()
G = LineFile("tests/smallcorpus-malformed.txt.bz2",
header="foo bar baz qux".split(),
path="tests/tmp/testcorpus")
G.clean(lower=True,
alphanumeric=True,
count_columns=False,
echo_toss=True,
lazy=True)
assert_equal(len(G), 8562)
G.delete()
G = LineFile("tests/smallcorpus-malformed.txt.bz2",
header="foo bar baz qux".split(),
path="tests/tmp/testcorpus")
G.clean(lower=True,
alphanumeric=True,
count_columns=False,
echo_toss=True,
filter_fn=lambda x: False,
lazy=True)
assert_equal(len(G), 0)
G.delete()
G = LineFile("tests/smallcorpus-malformed.txt.bz2",
header="foo bar baz qux".split(),
path="tests/tmp/testcorpus")
G.clean(lower=True,
alphanumeric=False,
count_columns=False,
echo_toss=True,
modifier_fn=lambda x: "hello",
lazy=True)
for line in G.lines(parts=False):
assert_equal(line, "hello")
G.delete()
def test_unicode():
""" test unicode
replace every word in the test corpus with random unicode
and see if we get the same surprisal scores.
"""
def generate_random_unicode():
for _ in xrange(5):
yield unichr(
random.choice((0x300, 0x9999)) + random.randint(0, 0xff))
scramblemap = {}
G = LineFile("tests/smallcorpus.txt.bz2",
header="foo bar baz qux".split(),
path="tests/tmp/testcorpus")
G.clean(lower=True,
alphanumeric=False,
count_columns=False,
echo_toss=True,
lazy=True)
G.make_marginal_column("quux", "foo bar".split(), "qux", lazy=True)
G.sort("baz")
len_G = len(G)
sum_counts = G.sum_column("quux")
sum_surprisal = math.fsum(line[2] for line in G.average_surprisal(
"baz", "qux", "quux", assert_sorted=True))
G.delete()
G = LineFile("tests/smallcorpus.txt.bz2",
header="foo bar baz qux".split(),
path="tests/tmp/testcorpus")
def scramble(line):
words = line.split()[:3]
count = line.split()[-1]
for i, word in enumerate(words):
if word in scramblemap:
words[i] = scramblemap[word]
else:
garbage = u"".join(generate_random_unicode())
words[i] = garbage
scramblemap[word] = garbage
return "\t".join(words + [count])
G.clean(lower=True,
alphanumeric=False,
count_columns=False,
echo_toss=True,
modifier_fn=scramble)
G.make_marginal_column("quux", "foo bar".split(), "qux")
G.sort("baz")
sum_counts_scrambled = G.sum_column("quux")
assert_equal(sum_counts, sum_counts_scrambled)
assert_equal(len_G, len(G))
sum_surprisal_scrambled = math.fsum(
line[2] for line in G.average_surprisal(
"baz", "qux", "quux", assert_sorted=True))
G.delete()
assert_equal(sum_surprisal, sum_surprisal_scrambled)
