🧪4️⃣ test: level 3 fix no input

.github/workflows/ci.yml

@@ -28,7 +28,7 @@ jobs:
         id: test
         run: |
           pytest -vs test/test_generator.py::test_generation_level_${{ matrix.group }}
-        timeout-minutes: 10
+        timeout-minutes: 15
         env:
           OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
           SCENEX_API_KEY: ${{ secrets.SCENEX_API_KEY }}

src/options/generate/templates_user.py

@@ -118,6 +118,10 @@ The function must full-fill: '{microservice_description}'.
 It will be tested with the following scenario: '{test_description}'.
 For the implementation use the following package(s): '{packages}'.
 
+The code must start with the following import:
+```
+from .apis import GPT_3_5_Turbo_API
+```
 Obey the following rules:
 ''' + not_allowed_function_string + '''
 
@@ -138,11 +142,10 @@ template_generate_test = PromptTemplate.from_template(
 Write a single pytest case that tests the following scenario: '{test_description}'. In case the test scenario is not precise enough, test a general case without any assumptions.
 Start the test with an extensive comment about the test case. If gpt_3_5_turbo_api is used in the executor, then the test must not check the exact output of the executor as it is not deterministic. 
 
-You must use the following import to import the function:
+The test must start with the following import:
 ```
 from .implementation import func
 ```
-
 ''' + not_allowed_function_string + '''
 The test must not open local files.
 The test must not mock a function of the executor.

test/test_generator.py

@@ -68,13 +68,43 @@ def test_generation_level_2(tmpdir):
     )
     assert generator.generate() == 0
 
-
 def test_generation_level_3(tmpdir):
     """
     Requirements:
     coding challenge: ❌
-    pip packages: ✅ (text to speech)
+    pip packages: ✅ (csv parser)
     environment: ❌
+    GPT-3.5-turbo: ✅ (for processing the text)
+    APIs: ✅ (financial data API)
+    Databases: ❌
+    """
+    os.environ['VERBOSE'] = 'true'
+    generator = Generator(
+        f'''The input is a stock symbol (e.g., AAPL for Apple Inc.). 
+1. Fetch stock data (open, high, low, close, volume) for the past 30 days using a financial data API (e.g., Alpha Vantage, Yahoo Finance, or any other API). 
+2. Calculate the average closing price over the 30 days.
+3. Read a CSV file containing a list of stock symbols and their company names.
+4. Find the input stock symbol in the CSV file and get the corresponding company name.
+5. Generate a brief summary of the company's stock performance over the past 30 days, including the average closing price and the company name.
+6. Return the summary as a string.
+Example input: 'AAPL'
+Example CSV file format:
+symbol,company_name
+AAPL,Apple Inc.
+GOOGL,Alphabet Inc.
+AMZN,Amazon.com, Inc.
+''',
+        str(tmpdir) + 'microservice',
+        'gpt-3.5-turbo'
+    )
+    assert generator.generate() == 0
+
+def test_generation_level_4(tmpdir):
+    """
+    Requirements:
+    coding challenge: ❌
+    pip packages: ✅ (text to speech)
+    environment: ✅ (tts library)
     GPT-3.5-turbo: ✅ (summarizing the text)
     APIs: ✅ (whisper for speech to text)
     Databases: ❌
@@ -101,17 +131,17 @@ print('This is the text from the audio file:', response.json()['text'])
 Example input file: https://www.signalogic.com/melp/EngSamples/Orig/ENG_M.wav
 ''',
         str(tmpdir) + 'microservice',
-        'gpt-3.5-turbo'
+        'gpt-4'
     )
     assert generator.generate() == 0
 
 
-def test_generation_level_4(tmpdir):
+def test_generation_level_5(tmpdir):
     """
     Requirements:
     coding challenge: ✅ (putting text on the image)
     pip packages: ✅ (Pillow for image processing)
-    environment: ❌
+    environment: ✅ (image library)
     GPT-3.5-turbo: ✅ (for writing the joke)
     APIs: ✅ (scenex for image description)
     Databases: ❌