def test_if_first_column_is_non_numeric_then_raise_error(self):
test_df = pd.DataFrame(
{"non_text_column": [8000, 200], "description": ["spam", "ham"]}
)
with pytest.raises(TypeError):
ngram_analysis.clean_input_data(test_df)
def test_for_special_keywords(self):
test_df = pd.DataFrame(
{
"description": [
"Stress Free Planning For Your Big Day! Customized & All Inclusive Packages. | no_description2",
"{=VenuePrice.Venue} Venue Packages Starting At {=VenuePrice.Start_price} Per Person.\nLearn More!",
"{KeyWord:Exceptional Services}. $90,000 | Views of Vegas Strip Skyline & Sunrise Mountain Range!",
"Tie The Knot At One Of The Best Weddings'",
]
}
)
result_df = ngram_analysis.clean_input_data(test_df)
result_series = result_df["cleaned_text"]
assert_series = pd.Series(
[
"stress free planning for your big day customized & all inclusive packages | no_description2",
"{=venuepricevenue} venue packages starting at {=venuepricestart_price} per person learn more ",
"{keywordexceptionalservices} $90000 | views of vegas strip skyline & sunrise mountain range ",
"tie the knot at one of the best weddings'",
],
name="cleaned_text",
)
pandas.util.testing.assert_series_equal(result_series, assert_series)
def test_digits_after_cleaning_being_togheter(self):
test_df = pd.DataFrame({"description": ["Number is 1 800 800", "$200,000.45"]})
result_df = ngram_analysis.clean_input_data(test_df)
result_series = result_df["cleaned_text"]
assert_series = pd.Series(
["number is 1800800", "$20000045"], name="cleaned_text"
)
pandas.util.testing.assert_series_equal(result_series, assert_series)
def test_for_no_multiple_spaces_present(self):
test_df = pd.DataFrame(
{"description": ["Num.ber ... is 1 800 800", "$200,000...45"]}
)
result_df = ngram_analysis.clean_input_data(test_df)
test_series = result_df["cleaned_text"]
assert_series = pd.Series(
["num ber is 1800800", "$20000045"], name="cleaned_text"
)
pandas.util.testing.assert_series_equal(test_series, assert_series)
def test_lemmatization_support(self):
test_df = pd.DataFrame(
{
"description": [
"Who's getting popcorn?",
"Lost your card? Freeze it in seconds to keep it safe ❄️\nDefrost it if you find it again 🔥",
"stress free planning for your big day customized & all inclusive packages"
]
}
)
result_df = ngram_analysis.clean_input_data(test_df, lemmatize=True)
result_series = result_df["cleaned_text"]
assert_series = pd.Series(
[
"who's get popcorn",
"lose -PRON- card freeze -PRON- in second to keep -PRON- safe ❄️ defrost -PRON- if -PRON- find -PRON- again 🔥",
"stress free planning for -PRON- big day customize & all inclusive package",
],
name="cleaned_text",
)
pandas.util.testing.assert_series_equal(result_series, assert_series)
def test_for_emoji_support(self):
test_df = pd.DataFrame(
{
"description": [
"Top marks for customer communication and our test_company app 🤓",
"Get great exchange rates abroad, and no fees on card payments! 🙌",
"Freeze it in seconds to keep it safe ❄️\nDefrost it if you find it again 🔥",
]
}
)
result_df = ngram_analysis.clean_input_data(test_df)
result_series = result_df["cleaned_text"]
assert_series = pd.Series(
[
"top marks for customer communication and our test_company app 🤓",
"get great exchange rates abroad and no fees on card payments 🙌",
"freeze it in seconds to keep it safe ❄️ defrost it if you find it again 🔥",
],
name="cleaned_text",
)
pandas.util.testing.assert_series_equal(result_series, assert_series)