def test_case_insensitivity(self):
print("test_case_insensitivity")
sample_data = [Row("A TEST OF SPLITTING A DATASET OF WORDS BY SPACES")]
sample_data_df = self.spark.createDataFrame(sample_data)
sample_data_df_split = w.splitWords(self.spark, sample_data_df)
word_count_df = w.countByWord(self.spark, sample_data_df_split)
actual_word_counts = dataframe_converter(word_count_df)
expected_word_counts = [('a', 2), ('by', 1), ('dataset', 1), ('of', 2), ('spaces', 1), ('splitting', 1), ('test', 1), ('words', 1)]
self.assertEqual(actual_word_counts,expected_word_counts)
def test_should_not_aggregate_dissimilar_words(self):
print("test_should_not_aggregate_dissimilar_words")
sample_data = [Row("a test of splitting a dataset of words by spaces")]
sample_data_df = self.spark.createDataFrame(sample_data)
sample_data_df_split = w.splitWords(self.spark, sample_data_df)
word_count_df = w.countByWord(self.spark, sample_data_df_split)
actual_word_counts = dataframe_converter(word_count_df)
expected_word_counts = [('a', 2), ('by', 1), ('dataset', 1), ('of', 2), ('spaces', 1), ('splitting', 1), ('test', 1), ('words', 1)]
self.assertEqual(actual_word_counts,expected_word_counts)
def test_count_words_basic(self):
print("test_count_words_basic")
sample_data = [Row("a test of splitting a dataset of words by spaces")]
sample_data_df = self.spark.createDataFrame(sample_data)
sample_data_df_split = w.splitWords(self.spark, sample_data_df)
word_count_df = w.countByWord(self.spark, sample_data_df_split)
actual_word_counts = dataframe_converter(word_count_df)
counts = [word_count[1] for word_count in actual_word_counts]
total_count = sum(counts)
self.assertEqual(total_count,10)
def test_ordering_words(self):
print("test_ordering_words")
sample_data = [Row("a test of splitting a dataset of words by spaces")]
sample_data_df = self.spark.createDataFrame(sample_data)
sample_data_df_split = w.splitWords(self.spark, sample_data_df)
word_count_df = w.countByWord(self.spark, sample_data_df_split)
actual_word_counts = dataframe_converter(word_count_df)
actual_word_order = [word_count[0] for word_count in actual_word_counts]
expected_word_order = ['a', 'by', 'dataset', 'of', 'spaces', 'splitting', 'test', 'words']
self.assertEqual(actual_word_order, expected_word_order)
def run(spark, inputPath, outputPath):
logging.info("Reading text file from: " + inputPath)
input_df = spark.read.text(inputPath)
split_df = wordcount_utils.splitWords(spark, input_df)
count_df = wordcount_utils.countByWord(spark, split_df)
logging.info("Writing csv to directory: " + outputPath)
count_df.show
count_df.write.csv(outputPath, mode='append')