ENH: support reading numeric subtypes (bool, int16, float32) (#83)

pyogrio/_io.pyx

@@ -47,6 +47,13 @@ FIELD_TYPES = [
     None            # OFTInteger64List, List of 64bit integers, not supported
 ]
 
+FIELD_SUBTYPES = {
+    OFSTNone: None,           # No subtype
+    OFSTBoolean: "bool",      # Boolean integer
+    OFSTInt16: "int16",       # Signed 16-bit integer
+    OFSTFloat32: "float32",   # Single precision (32 bit) floating point
+}
+
 # Mapping of numpy ndarray dtypes to (field type, subtype)
 DTYPE_OGR_FIELD_TYPES = {
     'int8': (OFTInteger, OFSTInt16),
@@ -310,6 +317,7 @@ cdef get_fields(OGRLayerH ogr_layer, str encoding):
     cdef int field_count
     cdef OGRFeatureDefnH ogr_featuredef = NULL
     cdef OGRFieldDefnH ogr_fielddef = NULL
+    cdef int field_subtype
     cdef const char *key_c
 
     try:
@@ -345,6 +353,12 @@ cdef get_fields(OGRLayerH ogr_layer, str encoding):
                 f"Skipping field {field_name}: unsupported OGR type: {field_type}")
             continue
 
+        field_subtype = OGR_Fld_GetSubType(ogr_fielddef)
+        subtype = FIELD_SUBTYPES.get(field_subtype)
+        if subtype is not None:
+            # bool, int16, float32 dtypes
+            np_type = subtype
+
         fields_view[i,0] = i
         fields_view[i,1] = field_type
         fields_view[i,2] = field_name
@@ -1250,10 +1264,10 @@ def ogr_write(str path, str layer, str driver, geometry, field_data, fields,
             ogr_fielddef = exc_wrap_pointer(OGR_Fld_Create(name_b, field_type))
 
             # subtypes, see: https://gdal.org/development/rfc/rfc50_ogr_field_subtype.html
-            if field_type != OFSTNone:
+            if field_subtype != OFSTNone:
                 OGR_Fld_SetSubType(ogr_fielddef, field_subtype)
 
-            if field_type:
+            if width:
                 OGR_Fld_SetWidth(ogr_fielddef, width)
 
             # TODO: set precision

pyogrio/tests/test_raw_io.py

@@ -363,3 +363,43 @@ def test_read_from_file_like(tmpdir, naturalearth_lowres, driver, ext):
         result2 = read(f)
 
     assert_equal_result((meta, index, geometry, field_data), result2)
+
+
+@pytest.mark.parametrize("ext", ["gpkg", "fgb"])
+def test_read_write_data_types_numeric(tmp_path, ext):
+    # Point(0, 0)
+    geometry = np.array(
+        [bytes.fromhex("010100000000000000000000000000000000000000")] * 3, dtype=object
+    )
+    field_data = [
+        np.array([True, False, True], dtype="bool"),
+        np.array([1, 2, 3], dtype="int16"),
+        np.array([1, 2, 3], dtype="int32"),
+        np.array([1, 2, 3], dtype="int64"),
+        np.array([1, 2, 3], dtype="float32"),
+        np.array([1, 2, 3], dtype="float64"),
+    ]
+    fields = ["bool", "int16", "int32", "int64", "float32", "float64"]
+    meta = dict(geometry_type="Point", crs="EPSG:4326", spatial_index=False)
+
+    filename = tmp_path / f"test.{ext}"
+    write(filename, geometry, field_data, fields, **meta)
+    result = read(filename)[3]
+    assert all([np.array_equal(f1, f2) for f1, f2 in zip(result, field_data)])
+    assert all([f1.dtype == f2.dtype for f1, f2 in zip(result, field_data)])
+
+    # other integer data types that don't roundtrip exactly
+    # these are generally promoted to a larger integer type except for uint64
+    for i, (dtype, result_dtype) in enumerate([
+        ("int8", "int16"),
+        ("uint8", "int16"),
+        ("uint16", "int32"),
+        ("uint32", "int64"),
+        ("uint64", "int64")
+    ]):
+        field_data = [np.array([1, 2, 3], dtype=dtype)]
+        filename = tmp_path / f"test{i}.{ext}"
+        write(filename, geometry, field_data, ["col"], **meta)
+        result = read(filename)[3][0]
+        assert np.array_equal(result, np.array([1, 2, 3]))
+        assert result.dtype == result_dtype