Enh/update xsd

CHANGELOG.md

@@ -2,6 +2,10 @@
 ==================
 
 - Add support for <EnumTypeDef>. By @bzah
+- Update XSD schema and dataclasses to latest version from netcdf-java to add support
+  for unsigned types. By @bzah
+- Add support for scalar variables. By @Bzah
+- [fix] empty attributes now are parsed into an empty string instead of crashing the parser.  By @Bzah
 
 0.3.1 (2023-11-10)
 ==================

schemas/ncml-2.2.xsd

@@ -114,7 +114,7 @@
       <xsd:attribute name="value" type="xsd:string"/>
       <xsd:attribute name="separator" type="xsd:string"/>
       <xsd:attribute name="orgName" type="xsd:string"/>
-      <xsd:attribute name="isUnsigned" type="xsd:boolean"/>
+      <xsd:attribute name="isUnsigned" type="xsd:boolean"/> <!-- deprecated - use unsigned type -->
     </xsd:complexType>
   </xsd:element>
 
@@ -136,12 +136,16 @@
       <xsd:enumeration value="enum1"/>
       <xsd:enumeration value="enum2"/>
       <xsd:enumeration value="enum4"/>
+      <xsd:enumeration value="ubyte"/>
+      <xsd:enumeration value="ushort"/>
+      <xsd:enumeration value="uint"/>
+      <xsd:enumeration value="ulong"/>
     </xsd:restriction>
   </xsd:simpleType>
 
   <!-- XML encoding of enum typedef object -->
  <xsd:element name="enumTypedef">
-    <xsd:complexType>
+   <xsd:complexType mixed="true">
       <xsd:sequence>
         <xsd:element name="enum" minOccurs="1" maxOccurs="unbounded">
           <xsd:complexType mixed="true">

tests/data/testEmptyAttr.xml

@@ -0,0 +1,4 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<netcdf xmlns="http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2">
+    <attribute name="comment" value="" />
+</netcdf>

tests/data/testEmptyScalar.xml

@@ -0,0 +1,5 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<netcdf xmlns="http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2">
+    <variable name="empty_scalar_var" type="double" shape="">
+    </variable>
+</netcdf>

tests/data/testEmptyScalar_withMultipleValues.xml

@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<netcdf xmlns="http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2">
+    <variable name="empty_scalar_var" type="double" shape="">
+        <values>1 2 3 42</values>
+    </variable>
+</netcdf>

tests/data/testEmptyScalar_withValuesTag.xml

@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<netcdf xmlns="http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2">
+    <variable name="empty_scalar_var" type="double" shape="">
+        <values></values>
+    </variable>
+</netcdf>

tests/data/testEnums.xml

@@ -5,5 +5,6 @@
     <enum key="1">true</enum>
   </enumTypedef>
   <variable name="be_or_not_to_be" shape="" type="enum1" typedef="boolean">
+    <values>0</values>
   </variable>
 </netcdf>

tests/data/testUnsignedType.xml

@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<netcdf xmlns="http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2">
+  <variable name="be_or_not_to_be" shape="" type="uint">
+    <values>42</values>
+  </variable>
+</netcdf>

tests/test_parser.py

@@ -303,12 +303,46 @@ def test_read_meta_data():
     assert ds.variables['T'].attrs['units'] == 'degC'
 
 
+def test_unsigned_type():
+    ds = xncml.open_ncml(data / 'testUnsignedType.xml')
+    assert ds['be_or_not_to_be'].dtype == np.uintc
+
+
+def test_empty_scalar__no_values_tag():
+    """
+    Scalar without values loose their type because we can't create a typed numpy
+    scalar which is empty
+    """
+    ds = xncml.open_ncml(data / 'testEmptyScalar.xml')
+    assert ds['empty_scalar_var'].dtype == np.dtype('O')
+    assert ds['empty_scalar_var'].item() is None
+
+
+def test_empty_scalar__with_empty_values_tag():
+    """A scalar variable with an empty <values> tag is invalid."""
+    with pytest.raises(ValueError, match='No values found for variable .*'):
+        xncml.open_ncml(data / 'testEmptyScalar_withValuesTag.xml')
+
+
+def test_multiple_values_for_scalar():
+    """Scalar with an multiple values in <values> tag is invalid."""
+    with pytest.raises(ValueError, match='The expected size for variable .* was 1, .*'):
+        xncml.open_ncml(data / 'testEmptyScalar_withMultipleValues.xml')
+
+
 def test_read_enum():
+    """A enum should be turned into CF flag_values and flag_meanings attributes."""
     ds = xncml.open_ncml(data / 'testEnums.xml')
     assert ds['be_or_not_to_be'].attrs['flag_values'] == [0, 1]
     assert ds['be_or_not_to_be'].attrs['flag_meanings'] == ['false', 'true']
 
 
+def test_empty_attr():
+    """A empty attribute is valid."""
+    ds = xncml.open_ncml(data / 'testEmptyAttr.xml')
+    assert ds.attrs['comment'] == ''
+
+
 # --- #
 def check_dimension(ds):
     assert len(ds['lat']) == 3

xncml/generated/ncml_2_2.py

@@ -32,6 +32,10 @@ class DataType(Enum):
     ENUM1 = 'enum1'
     ENUM2 = 'enum2'
     ENUM4 = 'enum4'
+    UBYTE = 'ubyte'
+    USHORT = 'ushort'
+    UINT = 'uint'
+    ULONG = 'ulong'
 
 
 class ObjectType(Enum):
@@ -289,13 +293,6 @@ class Meta:
         name = 'enumTypedef'
         namespace = 'http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2'
 
-    enum: List['EnumTypedef.EnumType'] = field(
-        default_factory=list,
-        metadata={
-            'type': 'Element',
-            'min_occurs': 1,
-        },
-    )
     name: Optional[str] = field(
         default=None,
         metadata={
@@ -309,6 +306,20 @@ class Meta:
             'type': 'Attribute',
         },
     )
+    content: List[object] = field(
+        default_factory=list,
+        metadata={
+            'type': 'Wildcard',
+            'namespace': '##any',
+            'mixed': True,
+            'choices': (
+                {
+                    'name': 'enum',
+                    'type': Type['EnumTypedef.EnumType'],
+                },
+            ),
+        },
+    )
 
     @dataclass
     class EnumType: