chore: rename create_udf to create_function (#914)

👋 Thanks for submitting a Pull Request to EvaDB! 🙌 We want to make contributing to EvaDB as easy and transparent as possible. Here are a few tips to get you started: - 🔍 Search existing EvaDB [PRs](https://github.com/georgia-tech-db/eva/pulls) to see if a similar PR already exists. - 🔗 Link this PR to a EvaDB [issue](https://github.com/georgia-tech-db/eva/issues) to help us understand what bug fix or feature is being implemented. - 📈 Provide before and after profiling results to help us quantify the improvement your PR provides (if applicable). 👉 Please see our ✅ [Contributing Guide](https://evadb.readthedocs.io/en/stable/source/contribute/index.html) for more details.
georgia-tech-db · Jun 30, 2023 · 777d391 · 777d391
1 parent f00a07a
commit 777d391
Show file tree

Hide file tree

Showing 12 changed files with 54 additions and 18 deletions.
diff --git a/apps/youtube_channel_qa/youtube_channel_qa.py b/apps/youtube_channel_qa/youtube_channel_qa.py
@@ -145,7 +145,7 @@ def generate_online_video_transcript(cursor) -> str:
 
     # bootstrap speech analyzer udf and chatgpt udf for analysis
     args = {"task": "automatic-speech-recognition", "model": "openai/whisper-base"}
-    speech_analyzer_udf_rel = cursor.create_udf(
+    speech_analyzer_udf_rel = cursor.create_function(
         "SpeechRecognizer", type="HuggingFace", **args
     )
     speech_analyzer_udf_rel.execute()
@@ -340,8 +340,8 @@ def cleanup():
 
         print("Creating embeddings and Vector Index")
 
-        cursor.drop_udf("embedding", if_exists=True).execute()
-        cursor.create_udf(
+        cursor.drop_function("embedding", if_exists=True).execute()
+        cursor.create_function(
             "embedding",
             if_not_exists=True,
             impl_path=SENTENCE_FEATURE_EXTRACTOR_UDF_PATH,

diff --git a/evadb/interfaces/relational/db.py b/evadb/interfaces/relational/db.py
@@ -41,7 +41,7 @@
     parse_table_clause,
 )
 from evadb.udfs.udf_bootstrap_queries import init_builtin_udfs
-from evadb.utils.generic_utils import is_ray_enabled_and_installed
+from evadb.utils.generic_utils import find_nearest_word, is_ray_enabled_and_installed
 from evadb.utils.logging_manager import logger
 
 
@@ -137,6 +137,30 @@ def __getattr__(self, name):
         Auto generate sync function calls from async
         Sync function calls should not be used in an async environment.
         """
+        function_name_list = [
+            "table",
+            "load",
+            "execute",
+            "query",
+            "create_function",
+            "create_table",
+            "create_vector_index",
+            "drop_table",
+            "drop_function",
+            "drop_index",
+            "df",
+            "show",
+            "insert" "explain",
+            "rename",
+            "fetch_one",
+        ]
+
+        if name not in function_name_list:
+            nearest_function = find_nearest_word(name, function_name_list)
+            raise AttributeError(
+                f"EvaDBCursor does not contain a function named: '{name}'. Did you mean to run: '{nearest_function}()'?"
+            )
+
         try:
             func = object.__getattribute__(self, "%s_async" % name)
         except Exception as e:
@@ -342,7 +366,7 @@ def create_function(
         if_not_exists: bool = True,
         impl_path: str = None,
         type: str = None,
-        **kwargs
+        **kwargs,
     ) -> "EvaDBQuery":
         """
         Create a udf in the database.

diff --git a/evadb/udfs/chatgpt.py b/evadb/udfs/chatgpt.py
@@ -118,13 +118,7 @@ def forward(self, text_df):
 
         @retry(tries=6, delay=20)
         def completion_with_backoff(**kwargs):
-            try:
-                response = openai.ChatCompletion.create(**kwargs)
-                answer = response.choices[0].message.content
-            # ignore API rate limit error etc.
-            except Exception as e:
-                answer = f"{e}"
-            return answer
+            return openai.ChatCompletion.create(**kwargs)
 
         # Register API key, try configuration manager first
         openai.api_key = ConfigurationManager().get_value("third_party", "OPENAI_KEY")
@@ -177,7 +171,8 @@ def completion_with_backoff(**kwargs):
                 ],
             )
 
-            answer = completion_with_backoff(**params)
+            response = completion_with_backoff(**params)
+            answer = response.choices[0].message.content
             results.append(answer)
 
         df = pd.DataFrame({"response": results})

diff --git a/evadb/utils/generic_utils.py b/evadb/utils/generic_utils.py
@@ -220,6 +220,15 @@ def remove_directory_contents(dir_path):
                 logger.warning(f"Failed to delete {file_path}. Reason: {str(e)}")
 
 
+def find_nearest_word(word, word_list):
+    from thefuzz import process
+
+    nearest_word_and_score = process.extractOne(word, word_list)
+    nearest_word = nearest_word_and_score[0]
+
+    return nearest_word
+
+
 ##############################
 ## TRY TO IMPORT PACKAGES
 ##############################

diff --git a/script/test/test.sh b/script/test/test.sh
@@ -105,7 +105,7 @@ fi
 
 if [[ ( "$OSTYPE" != "msys" ) && ( "$MODE" = "NOTEBOOK" || "$MODE" = "ALL" ) ]];
 then 
-    PYTHONPATH=./ python -m pytest --durations=5 --nbmake --overwrite "./tutorials" --capture=sys --tb=short -v --log-level=WARNING --nbmake-timeout=3000
+    PYTHONPATH=./ python -m pytest --durations=5 --nbmake --overwrite "./tutorials" --capture=sys --tb=short -v --log-level=WARNING --nbmake-timeout=3000 --ignore="tutorials/11-similarity-search-for-motif-mining.ipynb" --ignore="tutorials/08-chatgpt.ipynb" 
     notebook_test_code=$?
     if [ "$notebook_test_code" != "0" ];
     then

diff --git a/setup.py b/setup.py
@@ -53,6 +53,7 @@ def read(path, encoding="utf-8"):
     "diskcache>=5.4.0",
     "retry>=0.9.2",
     "psutil",
+    "thefuzz"
 ]
 
 vision_libs = [
@@ -84,7 +85,6 @@ def read(path, encoding="utf-8"):
 
 udf_libs = [
     "facenet-pytorch>=2.5.2",  # FACE DETECTION
-    "thefuzz",  # FUZZY STRING MATCHING
     "pytube",  # YOUTUBE QA APP
     "youtube-transcript-api",  # YOUTUBE QA APP
     "boto3",  # AWS

diff --git a/test/app_tests/test_pandas_qa.py b/test/app_tests/test_pandas_qa.py
@@ -16,6 +16,7 @@
 import subprocess
 import unittest
 from pathlib import Path
+from test.markers import chatgpt_skip_marker
 from test.util import get_evadb_for_testing, shutdown_ray
 
 
@@ -36,6 +37,7 @@ def setUp(self):
     def tearDown(self) -> None:
         shutdown_ray()
 
+    @chatgpt_skip_marker
     def test_should_run_pandas_qa_app(self):
         app_path = Path("apps", "pandas_qa", "pandas_qa.py")
         input1 = "\n"  # use default csv

diff --git a/test/app_tests/test_youtube_qa.py b/test/app_tests/test_youtube_qa.py
@@ -16,6 +16,7 @@
 import subprocess
 import unittest
 from pathlib import Path
+from test.markers import chatgpt_skip_marker
 from test.util import get_evadb_for_testing, shutdown_ray
 
 
@@ -36,6 +37,7 @@ def setUp(self):
     def tearDown(self) -> None:
         shutdown_ray()
 
+    @chatgpt_skip_marker
     def test_should_run_youtube_qa_app(self):
         app_path = Path("apps", "youtube_qa", "youtube_qa.py")
         input1 = "yes\n\n"  # Go with online video and default URL

diff --git a/test/integration_tests/test_text_filtering.py b/test/integration_tests/test_text_filtering.py
@@ -38,7 +38,7 @@ def test_text_filter(self):
         cursor = self.conn.cursor()
         cursor.load(pdf_path, "MyPDFs", "pdf").df()
         load_pdf_data = cursor.table("MyPDFs").df()
-        cursor.create_udf(
+        cursor.create_function(
             "TextFilterKeyword",
             True,
             f"{EvaDB_ROOT_DIR}/evadb/udfs/text_filter_keyword.py",

diff --git a/test/markers.py b/test/markers.py
@@ -61,3 +61,7 @@
 gpu_skip_marker = pytest.mark.skipif(
     is_gpu_available() is False, reason="Run only if gpu is available"
 )
+
+chatgpt_skip_marker = pytest.mark.skip(
+    reason="requires chatgpt",
+)
diff --git a/test/server/test_db_api.py b/test/server/test_db_api.py
@@ -32,14 +32,12 @@ def __init__(self, *args, **kwargs):
             super().__init__(*args, **kwargs)
 
         def setUp(self) -> None:
-            print("setUp")
             f = open(suffix_pytest_xdist_worker_id_to_dir("upload.txt"), "w")
             f.write("dummy data")
             f.close()
             return super().setUp()
 
         def tearDown(self) -> None:
-            print("tearDown")
             os.remove(suffix_pytest_xdist_worker_id_to_dir("upload.txt"))
             return super().tearDown()
 

diff --git a/test/udfs/test_chatgpt.py b/test/udfs/test_chatgpt.py
@@ -14,6 +14,7 @@
 # limitations under the License.
 
 import unittest
+from test.markers import chatgpt_skip_marker
 from test.util import get_evadb_for_testing
 
 import pandas as pd
@@ -56,6 +57,7 @@ def setUp(self) -> None:
     def tearDown(self) -> None:
         execute_query_fetch_all(self.evadb, "DROP TABLE IF EXISTS MyTextCSV;")
 
+    @chatgpt_skip_marker
     def test_openai_chat_completion_udf(self):
         udf_name = "OpenAIChatCompletion"
         execute_query_fetch_all(self.evadb, f"DROP UDF IF EXISTS {udf_name};")