Fix devergence multitime frame check

Author: MDUYN

pyindicators/__init__.py

@@ -10,7 +10,20 @@
 from .exceptions import PyIndicatorException
 from .date_range import DateRange
 
+# Version information
+__version__ = "0.15.0"
+__version_info__ = tuple(int(x) for x in __version__.split("."))
+
+
+def get_version():
+    """Return the version string of pyindicators."""
+    return __version__
+
+
 __all__ = [
+    '__version__',
+    '__version_info__',
+    'get_version',
     'sma',
     'wma',
     'is_crossover',

pyindicators/indicators/divergence.py

@@ -215,6 +215,17 @@ def check_divergence_pattern(series_a, series_b, target_a=-1, target_b=1):
     Returns:
         bool: True if pattern is found, False otherwise
     """
+    # Convert to flat numpy arrays for consistent integer indexing
+    # This handles pandas Series with DatetimeIndex correctly
+    if hasattr(series_a, 'values'):
+        series_a = series_a.values.flatten()
+    elif hasattr(series_a, 'flatten'):
+        series_a = series_a.flatten()
+    if hasattr(series_b, 'values'):
+        series_b = series_b.values.flatten()
+    elif hasattr(series_b, 'flatten'):
+        series_b = series_b.flatten()
+
     try:
         # Find the first index of `target_a` (e.g., -1 in the indicator)
         a_index = next(i for i, val in enumerate(series_a) if val == target_a)
@@ -223,9 +234,13 @@ def check_divergence_pattern(series_a, series_b, target_a=-1, target_b=1):
 
     # From that point forward, check if series_b has a target_b
     for j in range(a_index, len(series_b)):
-        if series_b[j] == -1:
+        val = series_b[j]
+        # Handle numpy scalar comparison explicitly
+        if hasattr(val, 'item'):
+            val = val.item()
+        if val == -1:
             return False  # Higher low before lower low — invalid
-        if series_b[j] == target_b:
+        if val == target_b:
             return True  # Valid divergence pattern
     return False
 

tests/indicators/test_divergence.py

@@ -830,3 +830,134 @@ def test_standard_situations(self):
             window_size=7
         )
         self.assertTrue(any(df["bearish_divergence"]))
+
+
+class TestCheckDivergencePatternWithDatetimeIndex(TestCase):
+    """Tests to verify divergence functions work correctly with DatetimeIndex."""
+
+    def test_check_divergence_pattern_with_datetime_index(self):
+        """Test that check_divergence_pattern works with DatetimeIndex Series."""
+        from pyindicators.indicators.divergence import check_divergence_pattern
+
+        series_a = pd.Series(
+            [np.nan, -1, np.nan],
+            index=pd.date_range('2024-01-01', periods=3, freq='1h')
+        )
+        series_b = pd.Series(
+            [np.nan, np.nan, 1],
+            index=pd.date_range('2024-01-01', periods=3, freq='1h')
+        )
+
+        result = check_divergence_pattern(series_a, series_b, target_a=-1, target_b=1)
+        self.assertTrue(result)
+
+    def test_check_divergence_pattern_with_integer_index(self):
+        """Test that check_divergence_pattern still works with integer index."""
+        from pyindicators.indicators.divergence import check_divergence_pattern
+
+        series_a = pd.Series([np.nan, -1, np.nan])
+        series_b = pd.Series([np.nan, np.nan, 1])
+
+        result = check_divergence_pattern(series_a, series_b, target_a=-1, target_b=1)
+        self.assertTrue(result)
+
+    def test_check_divergence_pattern_with_numpy_arrays(self):
+        """Test that check_divergence_pattern works with numpy arrays."""
+        from pyindicators.indicators.divergence import check_divergence_pattern
+
+        series_a = np.array([np.nan, -1, np.nan])
+        series_b = np.array([np.nan, np.nan, 1])
+
+        result = check_divergence_pattern(series_a, series_b, target_a=-1, target_b=1)
+        self.assertTrue(result)
+
+    def test_bearish_divergence_multi_dataframe_with_datetime_index(self):
+        """Test bearish_divergence_multi_dataframe with DatetimeIndex DataFrames."""
+        from pyindicators import macd, detect_peaks
+
+        # Data with DatetimeIndex (common when loading from CSV or APIs)
+        df = pd.DataFrame({
+            'Close': [100, 101, 102, 101, 100, 99, 100, 101, 102, 103, 102, 101]
+        }, index=pd.date_range('2024-01-01', periods=12, freq='1h'))
+
+        # Apply indicators
+        macd_data = macd(
+            df.copy(),
+            source_column="Close",
+            short_period=3,
+            long_period=6,
+            signal_period=3
+        )
+        close_data = detect_peaks(
+            df.copy(),
+            source_column="Close",
+            number_of_neighbors_to_compare=2,
+            min_consecutive=1
+        )
+        macd_data = detect_peaks(
+            macd_data,
+            source_column="macd_histogram",
+            number_of_neighbors_to_compare=2,
+            min_consecutive=1
+        )
+
+        # This should not raise ValueError with DatetimeIndex
+        divergence_data = bearish_divergence_multi_dataframe(
+            first_df=macd_data,
+            second_df=close_data,
+            result_df=close_data.copy(),
+            first_column="macd_histogram",
+            second_column="Close",
+            window_size=5,
+            result_column="bearish_divergence"
+        )
+
+        # Verify the result has the expected column and index
+        self.assertIn("bearish_divergence", divergence_data.columns)
+        self.assertIsInstance(divergence_data.index, pd.DatetimeIndex)
+
+    def test_bullish_divergence_multi_dataframe_with_datetime_index(self):
+        """Test bullish_divergence_multi_dataframe with DatetimeIndex DataFrames."""
+        from pyindicators import macd, detect_peaks
+
+        # Data with DatetimeIndex
+        df = pd.DataFrame({
+            'Close': [103, 102, 101, 100, 99, 100, 101, 102, 101, 100, 99, 98]
+        }, index=pd.date_range('2024-01-01', periods=12, freq='1h'))
+
+        # Apply indicators
+        macd_data = macd(
+            df.copy(),
+            source_column="Close",
+            short_period=3,
+            long_period=6,
+            signal_period=3
+        )
+        close_data = detect_peaks(
+            df.copy(),
+            source_column="Close",
+            number_of_neighbors_to_compare=2,
+            min_consecutive=1
+        )
+        macd_data = detect_peaks(
+            macd_data,
+            source_column="macd_histogram",
+            number_of_neighbors_to_compare=2,
+            min_consecutive=1
+        )
+
+        # This should not raise ValueError with DatetimeIndex
+        divergence_data = bullish_divergence_multi_dataframe(
+            first_df=macd_data,
+            second_df=close_data,
+            result_df=close_data.copy(),
+            first_column="macd_histogram",
+            second_column="Close",
+            window_size=5,
+            result_column="bullish_divergence"
+        )
+
+        # Verify the result has the expected column and index
+        self.assertIn("bullish_divergence", divergence_data.columns)
+        self.assertIsInstance(divergence_data.index, pd.DatetimeIndex)
+