Adding initial tests

README.md

@@ -1,2 +1,5 @@
+[![Pytest](https://github.com/Thomas-George-T/Ecommerce-Data-MLOps/actions/workflows/pytest.yml/badge.svg)](https://github.com/Thomas-George-T/Ecommerce-Data-MLOps/actions/workflows/pytest.yml)
+
+
 # MLOps-Ecomm
 Work in Progress

requirements.txt

@@ -3,4 +3,6 @@ numpy
 pandas
 pytest
 ipykernel
-mlflow
+mlflow
+requests
+zipfile

src/datapipeline.py

@@ -0,0 +1,34 @@
+import requests
+import zipfile
+
+def ingest_data():
+
+  file_url = "https://archive.ics.uci.edu/static/public/352/online+retail.zip"
+
+  # Send an HTTP GET request to the URL
+  response = requests.get(file_url)
+
+  # Check if the request was successful (status code 200)
+  if response.status_code == 200:
+      # Save file to data
+      with open("data/data.zip", "wb") as file:
+          file.write(response.content)
+      print("File downloaded successfully.")
+  else:
+      print(f"Failed to download the file. Status code: {response.status_code}")
+
+
+def unzip_file():
+    zip_filename ='data/data.zip'
+    extract_to = 'data/'
+    try:
+        with zipfile.ZipFile(zip_filename, 'r') as zip_ref:
+            zip_ref.extractall(extract_to)
+        print(f"File {zip_filename} successfully unzipped to {extract_to}")
+    except Exception as e:
+        print(f"Failed to unzip {zip_filename}: {e}")
+
+
+if __name__ == "__main__":
+    ingest_data()
+    unzip_file()

test/test_datapipeline.py

@@ -0,0 +1,32 @@
+import requests
+import pytest
+from src import datapipeline
+
+def test_ingest_data(mocker):
+    # arrange:
+    # mocked dependencies
+    mock_print = mocker.MagicMock(name='print')
+    mocker.patch('src.datapipeline.print', new=mock_print) 
+
+    # act: invoking the tested code
+    datapipeline.ingest_data()
+
+    # assert:
+    assert 1 == mock_print.call_count
+
+
+def test_unzip_file(mocker):
+    # arrange:
+    # mocked dependencies
+    mock_ZipFile = mocker.MagicMock(name='ZipFile')
+    mocker.patch('src.datapipeline.zipfile.ZipFile', new=mock_ZipFile)
+    mock_print = mocker.MagicMock(name='print')
+    mocker.patch('src.datapipeline.print', new=mock_print)
+    mock_Exception = mocker.MagicMock(name='Exception')
+    mocker.patch('src.datapipeline.Exception', new=mock_Exception)
+
+    # act: invoking the tested code
+    datapipeline.unzip_file()
+
+    # assert:
+    mock_Exception.assert_not_called()