Merge pull request #2 from AndreaSottana/improve_test_coverage

improve test coverage

tests/test_process_user_queries.py

@@ -1,6 +1,9 @@
 import os
 import pytest
+import logging
 import numpy as np
+import pandas as pd
+import pandas.util.testing as pdt
 from pdf2emb_nlp.process_user_queries import query_embeddings
 
 
@@ -43,6 +46,38 @@ def test_query_embeddings_with_word2vec_tfidf_weighted_with_exact_query():
     np.testing.assert_array_equal(embedding, trained_df['Word2Vec_with_TfIdf_weights'][0])
 
 
+def test_query_embeddings_with_word2vec_raises_logger_error_when_all_words_out_of_vocabulary(caplog):
+    with caplog.at_level(logging.ERROR):
+        embedding, trained_df = query_embeddings(
+            "Hello there how are you?",
+            os.path.join(os.getenv('FIXTURES_DIR'), 'full_df_with_embeddings.parquet.gzip'),
+            'Word2Vec',
+            'Word2Vec',
+            os.path.join(os.getenv('FIXTURES_DIR'), 'word2vec.pickle')
+        )
+    expected_log_message = \
+        'None of the words inputted are in the Word2Vec vocabulary. Please change your input or try a different ' \
+        'model, such as ELMo or BERT. Returning empty array and DataFrame.'
+    assert expected_log_message in caplog.text
+    np.testing.assert_array_equal(embedding, np.array([]))
+    pdt.assert_frame_equal(trained_df, pd.DataFrame())
+
+
+def test_query_embeddings_with_word2vec_raises_logger_warning_when_some_words_out_of_vocabulary(caplog):
+    with caplog.at_level(logging.WARNING):
+        query_embeddings(
+            "Hello Michael, this is a trial sentence!",
+            os.path.join(os.getenv('FIXTURES_DIR'), 'full_df_with_embeddings.parquet.gzip'),
+            'Word2Vec',
+            'Word2Vec',
+            os.path.join(os.getenv('FIXTURES_DIR'), 'word2vec.pickle')
+        )
+    expected_log_message = \
+        "The following words are not in the trained vocabulary and were therefore excluded from the search: " \
+        "['Hello', 'trial', '!']"
+    assert expected_log_message in caplog.text
+
+
 def test_query_embeddings_with_elmo_with_exact_query():
     embedding, trained_df = query_embeddings(
         "Michael went to the store to buy some eggs .",

tests/test_scraper.py

@@ -1,4 +1,5 @@
 import os
+import logging
 import numpy as np
 import pandas as pd
 from pdf2emb_nlp.scraper import DocumentScraper
@@ -12,6 +13,15 @@ def test_class_instantiation(self, text_cleaning_json):
         assert scraper.pdf_folder == os.getenv("FIXTURES_DIR")
         assert scraper.open_json == text_cleaning_json
 
+    def test_class_instantiation_when_no_text_cleaning_json_provided(self, caplog):
+        with caplog.at_level(logging.WARNING):
+            scraper = DocumentScraper(os.getenv("FIXTURES_DIR"))
+        assert scraper.pdf_folder == os.getenv("FIXTURES_DIR")
+        expected_log_message = \
+            'No .json file for text cleaning was provided. Ad-hoc text cleaning will not be performed.'
+        assert expected_log_message in caplog.text
+        assert scraper.open_json == dict()
+
     def test_document_corpus_to_pandas_df(self):
         expected_scraped_df = pd.DataFrame(
             {'test_pdf_1': [