Gridded data null

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "hec-dss-python"
-version = "0.1.31"
+version = "0.1.32"
 description = "Python wrapper for the HEC-DSS file database C library."
 authors = ["Hydrologic Engineering Center"]
 license = "MIT"

setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name = hecdss
-version = 0.1.31
+version = 0.1.32
 author = Hydrologic Engineering Center
 author_email =hec.dss@usace.army.mil
 description = Python wrapper for the HEC-DSS file database C library.

src/hecdss/gridded_data.py

@@ -85,7 +85,7 @@ def __init__(self):
         self.data = np.zeros(0)
         self.location_info = None
 
-    def range_limit_table(self, minval, maxval, range_, bins, datasize, data):
+    def range_limit_table(self, minval, maxval, range_, bins, data):
         """
         Calculate the range limit table and the number of values equal or exceeding each range limit.
 
@@ -94,7 +94,6 @@ def range_limit_table(self, minval, maxval, range_, bins, datasize, data):
             maxval (float): Maximum value.
             range_ (float): Range of values.
             bins (int): Number of bins.
-            datasize (int): Size of the data.
             data (numpy.ndarray): Data array.
         """
         max_bins = 15
@@ -103,7 +102,7 @@ def range_limit_table(self, minval, maxval, range_, bins, datasize, data):
             bins = max_bins
 
         self.rangeLimitTable = np.empty(bins, dtype=float)
-        self.rangeLimitTable[0] = NULL_INT
+        self.rangeLimitTable[0] = self.nullValue
         self.rangeLimitTable[1] = minval
 
         step = range_ / bins
@@ -118,6 +117,7 @@ def update_grid_info(self):
         """
         Update grid information based on the data array.
         """
+        self.data[self.data == self.nullValue] = np.nan
         self.numberOfCellsX = len(self.data[0])
         self.numberOfCellsY = len(self.data)
         n = np.size(self.data)
@@ -126,10 +126,14 @@ def update_grid_info(self):
         bin_range = int(math.ceil(self.maxDataValue) - math.floor(self.minDataValue))
         self.meanDataValue = np.nanmean(self.data)
 
-        self.data = np.nan_to_num(self.data, nan=NULL_INT)
-        self.numberOfRanges = math.floor(2 + 3.322 * math.log10(n))
+        self.data = np.nan_to_num(self.data, nan=self.nullValue)
+        if bin_range == 0:
+            self.numberOfRanges = 2
+        else:
+            self.numberOfRanges = math.floor(2 + 3.322 * math.log10(n))
         flat_data = self.data.flatten()
-        self.range_limit_table(self.minDataValue, self.maxDataValue, bin_range, self.numberOfRanges, n, flat_data)
+        self.range_limit_table(self.minDataValue, self.maxDataValue,
+                               bin_range, self.numberOfRanges, flat_data)
 
     @staticmethod
     def create(path=None,
@@ -223,7 +227,7 @@ def create(path=None,
         gd.meanDataValue = meanDataValue
         gd.rangeLimitTable = rangeLimitTable
         gd.numberEqualOrExceedingRangeLimit = numberEqualOrExceedingRangeLimit
-        gd.data = np.array(data)
+        gd.data = np.array(data).astype(float)
         gd.location_info = location_info
 
         gd.update_grid_info()

tests/test_gridded_data.py

@@ -9,6 +9,8 @@
 from hecdss.gridded_data import GriddedData
 
 
+NULL_INT = -3.4028234663852886e+38
+
 class TestGriddedData(unittest.TestCase):
 
     def setUp(self) -> None:
@@ -125,6 +127,40 @@ def test_gridded_data_write_precompressed(self):
             assert gd_original.numberOfCellsY == gd_readback.numberOfCellsY, "numberOfCellsY mismatch"
             assert gd_original.dataUnits == gd_readback.dataUnits, "dataUnits mismatch"
 
+    def test_null_consistency(self):
+        """
+        This test serves to ensure that the range limit table remains consistent with different cases of missing values.
+        """
+        gd_nan = self._create_half_nul_gd(np.nan)
+        gd_negative = self._create_half_nul_gd(-9999)
+        gd_null_int = self._create_half_nul_gd(NULL_INT)
+        gd_zero = self._create_half_nul_gd(0)
+
+        assert (gd_nan.numberOfRanges == 2)
+        assert (gd_nan.numberOfRanges == gd_negative.numberOfRanges)
+        assert (gd_nan.numberOfRanges == gd_null_int.numberOfRanges)
+        assert (gd_nan.numberOfRanges == gd_zero.numberOfRanges)
+
+        assert (gd_nan.maxDataValue == gd_negative.maxDataValue)
+        assert (gd_nan.maxDataValue == gd_null_int.maxDataValue)
+        assert (gd_nan.maxDataValue == gd_zero.maxDataValue)
+
+        assert (gd_nan.minDataValue == gd_negative.minDataValue)
+        assert (gd_nan.minDataValue == gd_null_int.minDataValue)
+        assert (gd_nan.minDataValue == gd_zero.minDataValue)
+
+
+    def _create_half_nul_gd(self, default_value):
+        gd_data = [[1 for _ in range(100)] for _ in range(50)]
+        gd_data.extend([[default_value for _ in range(100)] for _ in range(50)])
+
+        gd_test = GriddedData.create(
+            data=gd_data,
+            nullValue=default_value
+        )
+
+        return gd_test
+
 
 if __name__ == "__main__":
     unittest.main()