#170 - Handling of the "uima.noNamespace" prefix (#178)

- Move no namespace handling to xmi - Add better tests for this
dkpro · Aug 14, 2021 · f5d396c · f5d396c
1 parent 9140380
commit f5d396c
Show file tree

Hide file tree

Showing 6 changed files with 42 additions and 46 deletions.
diff --git a/cassis/typesystem.py b/cassis/typesystem.py
@@ -782,12 +782,6 @@ def deserialize(self, source: Union[IO, str]) -> TypeSystem:
             description = self._get_elem_as_str(elem.find("{*}description"))
             supertypeName = self._get_elem_as_str(elem.find("{*}supertypeName"))
 
-            if "." not in type_name:
-                type_name = "uima.noNamespace." + type_name
-
-            if "." not in supertypeName:
-                supertypeName = "uima.noNamespace." + supertypeName
-
             types[type_name] = Type(name=type_name, supertypeName=supertypeName, description=description)
             type_dependencies[type_name].add(supertypeName)
 
@@ -923,20 +917,13 @@ def _serialize_type(self, xf: IO, type_: Type):
         typeDescription = etree.Element("typeDescription")
 
         name = etree.SubElement(typeDescription, "name")
-        type_name = type_.name
-        if type_name.startswith("uima.noNamespace."):
-            type_name = type_name.replace("uima.noNamespace.", "")
-
-        name.text = type_name
+        name.text = type_.name
 
         description = etree.SubElement(typeDescription, "description")
         description.text = type_.description
 
         supertype_name_node = etree.SubElement(typeDescription, "supertypeName")
-        supertype_name = type_.supertypeName
-        if supertype_name.startswith("uima.noNamespace."):
-            supertype_name = supertype_name.replace("uima.noNamespace.", "")
-        supertype_name_node.text = supertype_name
+        supertype_name_node.text = type_.supertypeName
 
         # Only create the `feature` element if there is at least one feature
         feature_list = list(type_.features)

diff --git a/cassis/xmi.py b/cassis/xmi.py
@@ -262,9 +262,12 @@ def _parse_view(self, elem) -> ProtoView:
     def _parse_feature_structure(self, typesystem: TypeSystem, elem, children: Dict[str, List[str]]):
         # Strip the http prefix, replace / with ., remove the ecore part
         # TODO: Error checking
-        typename = elem.tag[9:].replace("/", ".").replace("ecore}", "").strip()
+        type_name: str = elem.tag[9:].replace("/", ".").replace("ecore}", "").strip()
 
-        AnnotationType = typesystem.get_type(typename)
+        if type_name.startswith("uima.noNamespace."):
+            type_name = type_name[17:]
+
+        AnnotationType = typesystem.get_type(type_name)
         attributes = dict(elem.attrib)
         attributes.update(children)
 
@@ -363,8 +366,12 @@ def _serialize_cas_null(self, root: etree.Element):
     def _serialize_feature_structure(self, cas: Cas, root: etree.Element, fs: FeatureStructure):
         ts = cas.typesystem
 
+        type_name = fs.type
+        if "." not in type_name:
+            type_name = f"uima.noNamespace.{type_name}"
+
         # The type name is a Java package, e.g. `org.myproj.Foo`.
-        parts = fs.type.split(".")
+        parts = type_name.split(".")
 
         # The CAS type namespace is converted to an XML namespace URI by the following rule:
         # replace all dots with slashes, prepend http:///, and append .ecore.

diff --git a/tests/fixtures.py b/tests/fixtures.py
@@ -160,6 +160,20 @@ def cas_with_leniency_xmi(cas_with_leniency_path):
         return f.read()
 
 
+# CAS and with feature structures whose types have no namespace
+
+
+@pytest.fixture
+def cas_has_fs_with_no_namespace_path():
+    return os.path.join(FIXTURE_DIR, "xmi", "cas_has_fs_with_no_namespace.xmi")
+
+
+@pytest.fixture
+def cas_has_fs_with_no_namespace_xmi(cas_has_fs_with_no_namespace_path):
+    with open(cas_has_fs_with_no_namespace_path, "r") as f:
+        return f.read()
+
+
 # Small type system
 
 

diff --git a/tests/test_files/typesystems/typesystem_has_types_with_no_namespace.xml b/tests/test_files/typesystems/typesystem_has_types_with_no_namespace.xml
@@ -2,41 +2,18 @@
 <typeSystemDescription xmlns="http://uima.apache.org/resourceSpecifier">
     <types>
         <typeDescription>
-            <name>ArtifactID</name>
+            <name>TypeWithNoNameSpace</name>
             <description>A unique artifact identifier.</description>
-            <supertypeName>uima.cas.TOP</supertypeName>
+            <supertypeName>uima.tcas.Annotation</supertypeName>
             <features>
                 <featureDescription>
-                    <name>artifactID</name>
-                    <description>A unique identification string for the artifact. This should be the file name for files,
-            or the unique identifier used in a database if the document source is a database
-            collection reader.</description>
+                    <name>testFeature</name>
+                    <description>I am the testFeatureDescription</description>
                     <rangeTypeName>uima.cas.String</rangeTypeName>
                 </featureDescription>
             </features>
         </typeDescription>
-        <typeDescription>
-            <name>ArtifactMetadata</name>
-            <description>A piece of metadata about the artifact in the form of a key value pair.</description>
-            <supertypeName>uima.cas.TOP</supertypeName>
-            <features>
-                <featureDescription>
-                    <name>key</name>
-                    <description/>
-                    <rangeTypeName>uima.cas.String</rangeTypeName>
-                </featureDescription>
-                <featureDescription>
-                    <name>value</name>
-                    <description/>
-                    <rangeTypeName>uima.cas.String</rangeTypeName>
-                </featureDescription>
-            </features>
-        </typeDescription>
-        <typeDescription>
-            <name>ChildOfArtifactID</name>
-            <supertypeName>ArtifactID</supertypeName>
-            <description/>
 
-        </typeDescription>
+
     </types>
 </typeSystemDescription>
diff --git a/tests/test_files/xmi/cas_has_fs_with_no_namespace.xmi b/tests/test_files/xmi/cas_has_fs_with_no_namespace.xmi
@@ -0,0 +1,7 @@
+<?xml version="1.0" encoding="UTF-8"?><xmi:XMI xmlns:noNamespace="http:///uima/noNamespace.ecore" xmlns:tcas="http:///uima/tcas.ecore" xmlns:xmi="http://www.omg.org/XMI" xmlns:cas="http:///uima/cas.ecore" xmi:version="2.0">
+    <cas:NULL xmi:id="0"/>
+    <noNamespace:TypeWithNoNameSpace xmi:id="2" sofa="1" begin="0" end="9"/>
+    <tcas:DocumentAnnotation xmi:id="3" sofa="1" begin="0" end="9" language="x-unspecified"/>
+    <cas:Sofa xmi:id="1" sofaNum="1" sofaID="_InitialView" mimeType="text" sofaString="I am test"/>
+    <cas:View sofa="1" members="2 3"/>
+</xmi:XMI>
diff --git a/tests/test_xmi.py b/tests/test_xmi.py
@@ -19,6 +19,10 @@
     (pytest.lazy_fixture("cas_with_reserved_names_xmi"), pytest.lazy_fixture("typesystem_with_reserved_names_xml")),
     (pytest.lazy_fixture("cas_with_two_sofas_xmi"), pytest.lazy_fixture("small_typesystem_xml")),
     (pytest.lazy_fixture("cas_with_smileys_xmi"), pytest.lazy_fixture("dkpro_typesystem_xml")),
+    (
+        pytest.lazy_fixture("cas_has_fs_with_no_namespace_xmi"),
+        pytest.lazy_fixture("typesystem_has_types_with_no_namespace_xml"),
+    ),
 ]