test: update python backend test cases, update logs

my_ghost_writer/app.py

@@ -109,7 +109,7 @@ def health_mongo() -> str:
 def get_words_frequency(body: RequestTextFrequencyBody | str) -> JSONResponse:
     t0 = datetime.now()
     app_logger.info(f"body type: {type(body)}.")
-    app_logger.info(f"body: {body}.")
+    app_logger.debug(f"body: {body}.")
     body_validated = RequestTextFrequencyBody.model_validate_json(body)
     text = body_validated.text
     app_logger.info(f"LOG_LEVEL: '{LOG_LEVEL}', length of text: {len(text)}, type of 'text':'{type(text)}'.")
@@ -130,7 +130,7 @@ def get_words_frequency(body: RequestTextFrequencyBody | str) -> JSONResponse:
 def get_sentence_sliced_by_word_and_positions(body: RequestSplitText | str) -> JSONResponse:
     t0 = datetime.now()
     app_logger.info(f"body type: {type(body)}.")
-    app_logger.info(f"body: {body}.")
+    app_logger.debug(f"body: {body}.")
     try:
         try:
             body_validated = RequestSplitText.model_validate_json(body)
@@ -148,8 +148,8 @@ def get_sentence_sliced_by_word_and_positions(body: RequestSplitText | str) -> J
             sentence, start_in_sentence, end_in_sentence = text_parsers.get_sentence_by_word(text, word, start, end)
         except Exception as e0:
             app_logger.error(f"end:'{end}', start:'{start}', word:'{word}'.")
-            app_logger.info("text:")
-            app_logger.info(text)
+            app_logger.error("text:")
+            app_logger.error(text)
             app_logger.error("## error:")
             app_logger.error(e0)
             raise e0
@@ -300,9 +300,15 @@ async def get_synonyms(body: RequestQueryThesaurusInflatedBody):
             end,
             word
         )
+        t1 = datetime.now()
+        duration = (t1 - t0).total_seconds()
+        app_logger.info(f"got extract_contextual_info_by_indices() result in: {duration:.3f}s. ...")
 
         # Process synonym groups
         processed_synonyms = process_synonym_groups(body.word, context_info)
+        t2 = datetime.now()
+        duration = (t2 - t1).total_seconds()
+        app_logger.info(f"got process_synonym_groups() result in: {duration:.3f}s. ...")
 
         if not processed_synonyms:
             return JSONResponse(
@@ -369,7 +375,7 @@ async def get_synonyms_for_phrase(body: RequestQueryThesaurusInflatedBody):
     synonym groups for each.
     """
     app_logger.info(f"body tye:{type(body)}!")
-    app_logger.info(f"body:{body}!")
+    app_logger.debug(f"body:{body}!")
     t0 = datetime.now()
     try:
         body_validated = RequestQueryThesaurusInflatedBody.model_validate_json(body)
@@ -398,7 +404,7 @@ async def get_synonyms_for_phrase(body: RequestQueryThesaurusInflatedBody):
         t1 = datetime.now()
         duration = (t1 - t0).total_seconds()
         app_logger.info(f"got find_synonyms_for_phrase() result in: {duration:.3f}s. ...")
-        app_logger.info(results)
+        app_logger.debug(results)
 
         message = f"Got {len(results)} synonym groups." if results else "No words with synonyms found in the selected phrase."
 

my_ghost_writer/text_parsers2.py

@@ -132,7 +132,7 @@ def extract_contextual_info_by_indices(text: str, start_idx: int, end_idx: int,
         # If the primary loop didn't find a token, it's an unexpected state,
         # but the original code to handle this was unreachable.
         # The most likely failure is now a word/index mismatch, handled above.
-        if target_token is None:
+        if target_token is None or str(target_token) != target_word:
             raise HTTPException(
                 status_code=400,
                 detail=f"Could not find token for word '{target_word}' at indices {start_idx}-{end_idx}"

tests/test_app.py

@@ -315,7 +315,3 @@ class TestAppEndpoints(unittest.TestCase):
             self.assertEqual(response.status_code, 503)
             # Verify the CORS header is set by our custom handler
             self.assertEqual(response.headers["access-control-allow-origin"], allowed_origin)
-
-
-if __name__ == "__main__":
-    unittest.main()

tests/test_text_parsers2.py

@@ -53,14 +53,15 @@ class TestTextParsers2(unittest.TestCase):
         self.assertEqual(context.exception.detail, "spaCy model not available")
 
     def test_extract_contextual_info_word_mismatch(self):
-        """Tests that a 400 HTTPException is raised for a word/index mismatch."""
+        """Tests that a 400->500 HTTPException is raised for a word/index mismatch."""
         text = "The quick brown fox"
+        start_idx, end_idx, target_word = 4, 9, "brown"
         # Indices point to "quick", but target_word is "brown"
         with self.assertRaises(HTTPException) as context:
-            extract_contextual_info_by_indices(text, 4, 9, "brown")
+            extract_contextual_info_by_indices(text, start_idx, end_idx, target_word)
 
-        self.assertEqual(context.exception.status_code, 400)
-        self.assertIn("Word mismatch", context.exception.detail)
+        self.assertEqual(context.exception.status_code, 500)
+        self.assertIn(f"Error analyzing context: 400: Could not find token for word '{target_word}' at indices {start_idx}-{end_idx}", context.exception.detail)
 
     @patch("my_ghost_writer.text_parsers2.nlp")
     def test_extract_contextual_info_word_none(self, nlp_mock):
@@ -259,7 +260,7 @@ class TestTextParsers2(unittest.TestCase):
     @patch("my_ghost_writer.text_parsers2.wn.synsets")
     def test_process_synonym_groups_not_synonyms_by_sense(self, mock_synsets):
         mock_synsets.return_value = []
-        context_info = {'pos': 'VERB'}
+        context_info = {'pos': 'VERB', 'lemma': 'look'}
         result = process_synonym_groups("look", context_info)
         self.assertListEqual(result, [])