From 8b8c2e397c830024b497eeec8af556e33a82cb31 Mon Sep 17 00:00:00 2001
From: R-Palazzo <romainpalazzo@gmail.com>
Date: Fri, 17 Nov 2023 07:34:46 -0600
Subject: [PATCH] tests

---
 .../single_table/test_diagnostic_report.py    | 37 +++++++++++++++++++
 .../single_table/test_diagnostic_report.py    | 36 ++++++++++++++++++
 tests/unit/reports/test_base_report.py        | 30 +++++++++++++++
 3 files changed, 103 insertions(+)

diff --git a/tests/integration/reports/single_table/test_diagnostic_report.py b/tests/integration/reports/single_table/test_diagnostic_report.py
index 3e33e99f..a64774a4 100644
--- a/tests/integration/reports/single_table/test_diagnostic_report.py
+++ b/tests/integration/reports/single_table/test_diagnostic_report.py
@@ -204,3 +204,40 @@ def test_get_details_with_errors(self):
             report.get_details('Data Validity'),
             expected_details
         )
+
+    def test_report_runs_with_mismatch_data_metadata(self):
+        """Test that the report runs with mismatched data and metadata."""
+        # Setup
+        data = pd.DataFrame({
+            'id': [0, 1, 2],
+            'val1': ['a', 'a', 'b'],
+            'val2': [0.1, 2.4, 5.7]
+        })
+        synthetic_data = pd.DataFrame({
+            'id': [1, 2, 3],
+            'extra_col': ['x', 'y', 'z'],
+            'val1': ['c', 'd', 'd']
+        })
+
+        metadata = {
+            'columns': {
+                'id': {'sdtype': 'id'},
+                'val1': {'sdtype': 'categorical'},
+                'val2': {'sdtype': 'numerical'}
+            },
+            'primary_key': 'id'
+        }
+        report = DiagnosticReport()
+
+        # Run
+        report.generate(data, synthetic_data, metadata)
+
+        # Assert
+        expected_properties = pd.DataFrame({
+            'Property': ['Data Validity', 'Data Structure'],
+            'Score': [0.5, 0.4444444]
+        })
+        assert report.get_score() == 0.47222222222222227
+        pd.testing.assert_frame_equal(
+            report.get_properties(), expected_properties
+        )
diff --git a/tests/unit/reports/single_table/test_diagnostic_report.py b/tests/unit/reports/single_table/test_diagnostic_report.py
index b744ce98..fb1af0be 100644
--- a/tests/unit/reports/single_table/test_diagnostic_report.py
+++ b/tests/unit/reports/single_table/test_diagnostic_report.py
@@ -1,3 +1,7 @@
+from unittest.mock import Mock
+
+import pandas as pd
+
 from sdmetrics.reports.single_table import DiagnosticReport
 from sdmetrics.reports.single_table._properties import DataValidity, Structure
 
@@ -14,3 +18,35 @@ def test___init__(self):
         assert report.is_generated is False
         assert isinstance(report._properties['Data Validity'], DataValidity)
         assert isinstance(report._properties['Data Structure'], Structure)
+
+    def test__validate_with_data_metadata_mismatch(self):
+        """Test the ``_validate`` method doesn't raise an error."""
+        # Setup
+        base_report = DiagnosticReport()
+        mock__validate_metadata_matches_data = Mock(
+            side_effect=ValueError('error message')
+        )
+        base_report._validate_metadata_matches_data = mock__validate_metadata_matches_data
+
+        real_data = pd.DataFrame({
+            'column1': [1, 2, 3],
+            'column2': ['a', 'b', 'c'],
+            'column3': [4, 5, 6]
+        })
+        synthetic_data = pd.DataFrame({
+            'column1': [1, 2, 3],
+            'column2': ['a', 'b', 'c'],
+            'column4': [4, 5, 6]
+        })
+        metadata = {
+            'columns': {
+                'column1': {'sdtype': 'numerical'},
+                'column2': {'sdtype': 'categorical'},
+            }
+        }
+
+        # Run
+        result = base_report._validate(real_data, synthetic_data, metadata)
+
+        # Assert
+        assert result is None
diff --git a/tests/unit/reports/test_base_report.py b/tests/unit/reports/test_base_report.py
index 8e56d474..09b16ea8 100644
--- a/tests/unit/reports/test_base_report.py
+++ b/tests/unit/reports/test_base_report.py
@@ -137,6 +137,36 @@ def test__validate(self):
             real_data, synthetic_data, metadata
         )
 
+    def test__validate_with_value_error(self):
+        """Test the ``_validate`` method with a ValueError."""
+        # Setup
+        base_report = BaseReport()
+        mock__validate_metadata_matches_data = Mock(
+            side_effect=ValueError('error message')
+        )
+        base_report._validate_metadata_matches_data = mock__validate_metadata_matches_data
+
+        real_data = pd.DataFrame({
+            'column1': [1, 2, 3],
+            'column2': ['a', 'b', 'c'],
+            'column3': [4, 5, 6]
+        })
+        synthetic_data = pd.DataFrame({
+            'column1': [1, 2, 3],
+            'column2': ['a', 'b', 'c'],
+            'column4': [4, 5, 6]
+        })
+        metadata = {
+            'columns': {
+                'column1': {'sdtype': 'numerical'},
+                'column2': {'sdtype': 'categorical'},
+            }
+        }
+
+        # Run and Assert
+        with pytest.raises(ValueError, match='error message'):
+            base_report._validate(real_data, synthetic_data, metadata)
+
     def test_convert_datetimes(self):
         """Test that ``_convert_datetimes`` tries to convert datetime columns."""
         # Setup