def test_verify_bool():
ex = "Wherefore art thou Romeo! Wherefore art thou Romeo. Wherefore art thou Romeo?"
with pytest.raises(TypeError) as e:
text_grams(ex, stop_remove="text")
assert str(
e.value
) == "stop_remove, remove_punctuation, remove_number and case_sensitive must be boolean"
def test_verify_k():
ex = "Wherefore art thou Romeo! Wherefore art thou Romeo. Wherefore art thou Romeo?"
ex_output1 = text_grams(ex, k=0)
assert ex_output1.shape[0] == 0 # If k is 0, dataframe should be empty
with pytest.raises(ValueError) as e:
text_grams(ex, k=-1)
assert str(e.value) == "k must be 0 or greater"
def test_verify_n():
ex = "Wherefore art thou Romeo! Wherefore art thou Romeo. Wherefore art thou Romeo?"
with pytest.raises(ValueError) as e:
text_grams(ex, n=[])
assert str(e.value) == "n must have at least one positive value"
with pytest.raises(ValueError) as e:
text_grams(ex, n=[-1])
assert str(e.value) == "Values of n must be greater than 0"
def test_verify_input2():
"""
Test if input is not empty
"""
text = " "
with pytest.raises(ValueError) as e:
text_grams(text)
assert str(e.value) == "Input text is empty."
def test_verify_input1():
"""
Test if input string is valid (not numeric)
"""
text = 100
with pytest.raises(ValueError) as e:
text_grams(text)
assert str(e.value) == "Input must be a string"
def test_normal_function():
ex = "Today is a sunny day. We should go to a beach on this sunny day"
k = 1
n = [2]
ex_output = text_grams(ex, k, n)
assert ex_output["2gram"][0] == sample_out['2gram'][0]
assert len(ex_output['2gram'][0]) == n[0]
assert ex_output.shape[0] == sample_out.shape[0]
n = 2
ex_output = text_grams(ex, k, n)
ex_output
assert ex_output["2gram"][0] == ("sunny", "day")
assert ex_output["Number of Instances"][0] == 2
def test_verify_output():
ex = "Today is a sunny day. We should go to a beach on this sunny day"
k = 1
n = [2]
ex_output = text_grams(ex, k, n)
assert type(ex_output) == type(sample_out)
assert type(ex_output['2gram'][0]) == tuple
assert type(ex_output['Number of Instances'][0]) == numpy.int64
def test_verify_n():
ex = "Wherefore art thou Romeo! Wherefore art thou Romeo. Wherefore art thou Romeo?"
with pytest.raises(TypeError) as e:
text_grams(ex, n="text")
assert str(e.value) == "n must be an integer list"
with pytest.raises(ValueError) as e:
text_grams(ex, n=[-1, -2])
assert str(e.value) == "Values of n must be greater than 0"
with pytest.raises(ValueError) as e:
text_grams(ex, n=-1)
assert str(e.value) == "n must be 0 or greater"
def test_verify_case_sensitive():
ex = "Hey Guys. Hey Guys. hey guys. hey guys."
ex_output = text_grams(ex, case_sensitive=True)
assert ex_output.shape[
0] > 1 # If case_sensitive is true, only (hey, guys) will be considered
def test_verify_remove_number():
ex = "This is 123 a sentence. This is 123 also a sentence. This is 123 also a sentence."
ex_output = text_grams(ex, remove_number=False)
assert ex_output.shape[
0] > 1 # If numbers is removed, only (also, sentence) will be considered
def test_verify_remove_punctuation():
ex = "This is / a sentence. This is / also a sentence. This is / also a sentence."
ex_output = text_grams(ex, remove_punctuation=False)
assert ex_output.shape[
0] > 1 # If punctuation is removed, only (also, sentence) will be considered
def test_verify_stop_remove():
ex = "This is a sentence. This is also a sentence. This is also a sentence."
ex_output = text_grams(ex, stop_remove=False)
assert ex_output.shape[
0] > 1 # If stopwords are removed, only (also, sentence) will be considered
def test_verify_sentence_endings():
ex = "Wherefore art thou Romeo! Wherefore art thou Romeo. Wherefore art thou Romeo?"
ex_output = text_grams(ex)
assert ex_output.shape[
0] != 4 # If it split sentences properly, (Romeo, Wherefore) will not be a gram combination
def test_verify_n_larger_than_sentence():
ex = "short sentence. short sentence."
ex_output = text_grams(ex, n=[3])
assert ex_output.shape[
0] == 0 # If n is larger than sentence length, dataframe should be empty