- Added deepcopy´s to all from_json() methods to enforce immutability…

… of the input data and implemented first mutability tests (see #2).

- Added Dataset.to_json() test, closes #1

texoopy/model/Annotation.py

@@ -1,3 +1,4 @@
+import copy
 import json
 
 from texoopy.model.Span import Span
@@ -13,6 +14,7 @@ def __init__(self, **kwargs):
 
     @classmethod
     def from_json(cls, json_data: dict):
+        json_data = copy.deepcopy(json_data)
         from texoopy.model.MentionAnnotation import MentionAnnotation
         from texoopy.model.NamedEntityAnnotation import NamedEntityAnnotation
         if json_data['class'] == 'MentionAnnotation':

texoopy/model/Dataset.py

@@ -9,9 +9,11 @@ def __init__(self, **kwargs):
         self.name: str = kwargs.get('name')
         self.language: str = kwargs.get('language')
         self.documents: list = []
+        self.queries: list = []
 
     @classmethod
     def from_json(cls, json_data: dict):
+        json_data = copy.deepcopy(json_data)
         dataset = cls(**json_data)
         for doc_json_data in json_data.get('documents', []):
             dataset.documents.append(Document.from_json(doc_json_data))

texoopy/model/Document.py

@@ -1,3 +1,4 @@
+import copy
 import json
 
 from texoopy.model.Annotation import Annotation
@@ -17,28 +18,29 @@ def __init__(self, **kwargs):
 
     @classmethod
     def from_json(cls, json_data: dict, do_sentence_splitting=False):
-
+        json_data = copy.deepcopy(json_data)
         if json_data.get('class') != 'Document':
-            raise NotATeXooDocumentException('')
+            raise NotATeXooDocumentException('Supplied JSON is not a valid TeXoo document.')
 
         annotations = []
         for json_data_annotation in json_data.get('annotations'):
             annotations.append(Annotation.from_json(json_data_annotation))
         json_data['annotations'] = annotations
 
         if do_sentence_splitting:
-            pass  # TODO add sentence splitting
+            raise NotImplementedError("Sentence splitting is not implemented yet.")  # TODO add sentence splitting
 
         return cls(**json_data)
 
     def to_json(self):
         content = self.to_texoo_dict()
-        content['class'] = 'Document'
         return json.dumps(content, default=lambda o: o.to_texoo_dict())
 
     def to_texoo_dict(self) -> dict:
         content = super().to_texoo_dict()
+        content['class'] = 'Document'
         return content
 
+
 class NotATeXooDocumentException(Exception):
     pass

texoopy/model/MentionAnnotation.py

@@ -1,3 +1,4 @@
+import copy
 import json
 
 from texoopy.model.Annotation import Annotation
@@ -12,6 +13,7 @@ def __init__(self, **kwargs):
 
     @classmethod
     def from_json(cls, json_data: dict):
+        json_data = copy.deepcopy(json_data)
         return cls(**json_data)
 
     def to_json(self):

texoopy/model/NamedEntityAnnotation.py

@@ -1,3 +1,4 @@
+import copy
 import json
 
 from texoopy.model.Annotation import Annotation
@@ -12,6 +13,7 @@ def __init__(self, **kwargs):
 
     @classmethod
     def from_json(cls, json_data: dict):
+        json_data = copy.deepcopy(json_data)
         return cls(**json_data)
 
     def to_json(self):

texoopy/model/Sentence.py

@@ -1,3 +1,4 @@
+import copy
 import json
 
 from texoopy.model.Span import Span
@@ -12,6 +13,7 @@ def __init__(self, **kwargs):
 
     @classmethod
     def from_json(cls, json_data: dict, do_tokenization=False):
+        json_data = copy.deepcopy(json_data)
         if do_tokenization:
             pass  # TODO implement me (later)
         token_json_data = json_data.pop('tokens', [])

texoopy/model/Span.py

@@ -16,6 +16,7 @@ def to_texoo_dict(self) -> dict:
 
     @classmethod
     def from_json(cls, json_data: dict):
+        json_data = copy.deepcopy(json_data)
         return cls(**json_data)
 
     def to_json(self):

texoopy/model/Token.py

@@ -1,3 +1,4 @@
+import copy
 import json
 
 from texoopy.model.Span import Span
@@ -10,6 +11,7 @@ def __init__(self, **kwargs):
 
     @classmethod
     def from_json(cls, json_data: dict):
+        json_data = copy.deepcopy(json_data)
         return cls(**json_data)  # TODO IMPLEMENT ME
 
     def to_json(self):

texoopy/tests/model/tests.py

@@ -14,7 +14,8 @@ class DatasetTest(unittest.TestCase):
 
     def setUp(self) -> None:
         with open('res/texoo_dataset.json', 'r') as test_file:
-            self.dataset = Dataset.from_json(json.load(test_file))
+            self.dataset_json_dict = json.load(test_file)
+            self.dataset = Dataset.from_json(self.dataset_json_dict)
 
     def test_dataset_instance(self):
         self.assertIsInstance(self.dataset, Dataset)
@@ -29,6 +30,9 @@ def test_dataset_document_count(self):
         self.assertEqual(3, len(self.dataset.documents))
 
     def test_dataset_to_json(self):
+        self.assertEqual(self.dataset_json_dict, json.loads(self.dataset.to_json()))
+
+    def test_dataset_toJson_mutability(self):
         pass  # TODO IMPLEMENT ME
 
 
@@ -79,11 +83,17 @@ def test_illegal_teXoo_json_exception(self):
             Document.from_json({'class': 'Span'})
 
     def test_document_to_json(self):
-        orig = self.document_json_dict  # TODO RMD
-        parsed = json.loads(self.document_1.to_json()) # TODO RMD
-
         self.assertEqual(self.document_json_dict, json.loads(self.document_1.to_json()))
 
+    def test_document_toJson_mutability(self):
+        before = copy.deepcopy(self.document_1.__dict__)
+        self.document_1.to_json()
+        after = copy.deepcopy(self.document_1.__dict__)
+        # Annotations should get tested in a separate test so here we only compare the correct count.
+        before['annotations'] = len(before.get('annotations'))
+        after['annotations'] = len(after.get('annotations'))
+        self.assertEqual(before, after)
+
 
 class MentionAnnotationTest(unittest.TestCase):
 
@@ -119,6 +129,9 @@ def test_mention_ann_type(self):
     def test_mention_ann_to_json(self):
         self.assertEqual(self.mention_ann_1_json_dict, json.loads(self.mention_ann_1.to_json()))
 
+    def test_mention_ann_toJson_mutability(self):
+        pass  # TODO IMPLEMENT ME
+
 
 class NamedEntityAnnotationTest(unittest.TestCase):
 
@@ -151,6 +164,9 @@ def test_named_ann_refId(self):
     def test_named_ann_to_json(self):
         self.assertEqual(self.named_entity_ann_json_dict, json.loads(self.named_entity_ann.to_json()))
 
+    def test_named_ann_toJson_mutability(self):
+        pass  # TODO IMPLEMENT ME
+
 
 if __name__ == '__main__':
     unittest.main()