mmif-python 1.0.0__tar.gz → 1.0.2__tar.gz

{mmif-python-1.0.0/mmif_python.egg-info → mmif-python-1.0.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mmif-python
-Version: 1.0.0
+Version: 1.0.2
 Summary: Python implementation of MultiMedia Interchange Format specification. (https://mmif.clams.ai)
 Home-page: https://mmif.clams.ai
 Author: Brandeis Lab for Linguistics and Computation

mmif-python-1.0.2/VERSION

@@ -0,0 +1 @@
+1.0.2

mmif-python-1.0.2/mmif/__init__.py

@@ -0,0 +1,36 @@
+import importlib
+import pkgutil
+import re
+from collections import defaultdict
+from typing import Dict, Type, Callable, Set
+
+import pkg_resources
+
+# DO NOT CHANGE THIS ORDER, important to prevent circular imports
+from mmif.ver import __version__
+from mmif.ver import __specver__
+from mmif.vocabulary import *
+from mmif.serialize import *
+
+_res_pkg = 'res'
+_ver_pkg = 'ver'
+_vocabulary_pkg = 'vocabulary'
+_schema_res_name = 'mmif.json'
+
+
+def get_mmif_json_schema():
+    res = pkg_resources.resource_stream(f'{__name__}.{_res_pkg}', _schema_res_name)
+    res_str = res.read().decode('utf-8')
+    res.close()
+    return res_str
+
+
+patches: Dict[Type, Set[Callable]] = defaultdict(set)
+for _, name, ispkg in pkgutil.iter_modules():
+    if ispkg and re.match(r'mmif[-_]utils[-_]', name):
+        mod = importlib.import_module(name)
+        for c, ms in mod.patches.items():
+            for m in ms:
+                if m in patches[c]:
+                    raise ValueError(f'Patch for {c}::{m.__name__} already exists.')
+                patches[c].add(m)

{mmif-python-1.0.0 → mmif-python-1.0.2}/mmif/serialize/annotation.py

@@ -6,21 +6,28 @@ In MMIF, annotations are created by apps in a pipeline as a part
 of a view. For documentation on how views are represented, see
 :mod:`mmif.serialize.view`.
 """
+import importlib
 import itertools
 import pathlib
-from typing import Union, Dict, List, Type, Optional, Iterator, MutableMapping, TypeVar
+import pkgutil
+import re
+import typing
+import warnings
+from typing import Union, Dict, List, Optional, Iterator, MutableMapping, TypeVar
 from urllib.parse import urlparse
 
 from mmif.vocabulary import ThingTypesBase, DocumentTypesBase
-from .model import MmifObject
+from .model import MmifObject, JSON_PRMTV_TYPES
 
 __all__ = ['Annotation', 'AnnotationProperties', 'Document', 'DocumentProperties', 'Text']
 
 T = TypeVar('T')
 
-from .. import DocumentTypes
+from .. import DocumentTypes, AnnotationTypes
 
-JSON_COMPATIBLE_PRIMITIVES: Type = Union[str, int, float, bool, None]
+discovered_docloc_plugins = {
+    name[len('mmif_docloc_'):]: importlib.import_module(name) for _, name, _ in pkgutil.iter_modules() if re.match(r'mmif[-_]docloc[-_]', name)
+}
 
 
 class Annotation(MmifObject):
@@ -30,6 +37,9 @@ class Annotation(MmifObject):
 
     def __init__(self, anno_obj: Optional[Union[bytes, str, dict]] = None) -> None:
         self._type: ThingTypesBase = ThingTypesBase('')
+        # to store the parent view ID
+        self._parent_view_id = ''
+        self.reserved_names.add('_parent_view_id')
         if not hasattr(self, 'properties'):  # don't overwrite DocumentProperties on super() call
             self.properties: AnnotationProperties = AnnotationProperties()
             self._attribute_classes = {'properties': AnnotationProperties}
@@ -62,6 +72,16 @@ class Annotation(MmifObject):
         else:
             self._type = at_type
 
+    @property
+    def parent(self) -> str:
+        return self._parent_view_id
+
+    @parent.setter
+    def parent(self, parent_view_id: str) -> None:
+        # I want to make this to accept `View` object as an input too,
+        # but import `View` will break the code due to circular imports
+        self._parent_view_id = parent_view_id
+
     @property
     def id(self) -> str:
         return self.properties.id
@@ -69,32 +89,52 @@ class Annotation(MmifObject):
     @id.setter
     def id(self, aid: str) -> None:
         self.properties.id = aid
+        
+    @staticmethod
+    def check_prop_value_is_simple_enough(
+            value: Union[JSON_PRMTV_TYPES, List[JSON_PRMTV_TYPES], List[List[JSON_PRMTV_TYPES]]]):
+        def json_primitives(x): 
+            return isinstance(x, typing.get_args(JSON_PRMTV_TYPES))
+        return json_primitives(value) \
+            or (isinstance(value, list) and all(map(json_primitives, value))) \
+            or (all(map(lambda elem: isinstance(elem, list), value)) and map(json_primitives, [subelem for elem in value for subelem in elem]))
 
     def add_property(self, name: str,
-                     value: Union[JSON_COMPATIBLE_PRIMITIVES,
-                                  List[JSON_COMPATIBLE_PRIMITIVES],
-                                  List[List[JSON_COMPATIBLE_PRIMITIVES]]
-                    ]) -> None:
+                     value: Union[JSON_PRMTV_TYPES, List[JSON_PRMTV_TYPES], List[List[JSON_PRMTV_TYPES]]]
+                     ) -> None:
         """
         Adds a property to the annotation's properties.
         :param name: the name of the property
         :param value: the property's desired value
         :return: None
         """
-        json_primitives = lambda x:isinstance(x, JSON_COMPATIBLE_PRIMITIVES.__args__)
-        if json_primitives(value) or (
-                isinstance(value,list)
-                and all(map(json_primitives, value)) or (
-                        all(map(lambda elem: isinstance(elem, list), value))
-                        and map(json_primitives, [subelem for elem in value for subelem in elem])
-                )
-        ):
+        if self.check_prop_value_is_simple_enough(value):
             self.properties[name] = value
         else:
             raise ValueError("Property values cannot be a complex object. It must be "
                              "either string, number, boolean, None, or a list of them."
                              f"(\"{name}\": \"{str(value)}\"")
 
+    def get(self, prop_name: str) -> Union['AnnotationProperties', JSON_PRMTV_TYPES, List[JSON_PRMTV_TYPES], List[List[JSON_PRMTV_TYPES]]]:
+        """
+        A special getter for Annotation properties. This is to allow for
+        directly accessing properties without having to go through the
+        properties object.
+        """
+        if prop_name in {'at_type', '@type'}:
+            return str(self._type)
+        elif prop_name == 'properties':
+            return self.properties
+        elif prop_name in self.properties:
+            return self.properties[prop_name]
+        else:
+            raise KeyError(f"Property {prop_name} does not exist in this annotation.")
+
+    get_property = get
+
+    def __getitem__(self, prop_name: str):
+        return self.get(prop_name)
+    
     def is_document(self):
         return isinstance(self.at_type, DocumentTypesBase)
 
@@ -113,36 +153,81 @@ class Document(Annotation):
     :param document_obj: the JSON data that defines the document
     """
     def __init__(self, doc_obj: Optional[Union[bytes, str, dict]] = None) -> None:
-        # to store the parent view ID
-        self._parent_view_id = ''
-        self.reserved_names.add('_parent_view_id')
+        # see https://github.com/clamsproject/mmif-python/issues/226 for discussion
+        # around the use of these three dictionaries
+        self._props_original: DocumentProperties = DocumentProperties()
+        self._props_existing: AnnotationProperties = AnnotationProperties()
+        self._props_temporary: AnnotationProperties = AnnotationProperties()
+        self.reserved_names.update(('_props_original', '_props_existing', '_props_temporary'))
         
         self._type: Union[ThingTypesBase, DocumentTypesBase] = ThingTypesBase('')
-        self.properties: DocumentProperties = DocumentProperties()
+        self.properties = self._props_original
         self.disallow_additional_properties()
         self._attribute_classes = {'properties': DocumentProperties}
         super().__init__(doc_obj)
-
-    @property
-    def parent(self) -> str:
-        return self._parent_view_id
-
-    @parent.setter
-    def parent(self, parent_view_id: str) -> None:
-        # I want to make this to accept `View` object as an input too,
-        # but import `View` will break the code due to circular imports
-        self._parent_view_id = parent_view_id
+    
+    def _add_property_from_annotation(self, annotation: Annotation):
+        if annotation.at_type != AnnotationTypes.Annotation:
+            raise ValueError("Only `Annotation` type can be added as a property to a `Document` object.")
+        for prop_name, prop_value in annotation.properties.items():
+            self._props_existing[prop_name] = prop_value
 
     def add_property(self, name: str,
-                     value: Union[JSON_COMPATIBLE_PRIMITIVES,
-                                  List[JSON_COMPATIBLE_PRIMITIVES]]) -> None:
+                     value: Union[JSON_PRMTV_TYPES, List[JSON_PRMTV_TYPES]]
+                     ) -> None:
+        """
+        Adds a property to the document's properties.
+        
+        Unlike the parent :class:`Annotation` class, added properties of a 
+        ``Document`` object can be lost during serialization unless it belongs 
+        to somewhere in a ``Mmif`` object. This is because we want to keep 
+        ``Document`` object as "read-only" as possible. Thus, if you want to add 
+        a property to a ``Document`` object, 
+        
+        * add the document to a ``Mmif`` object (either in the documents list or 
+          in a view from the views list), or
+        * directly write to ``Document.properties`` instead of using this method
+          (which is not recommended). 
+        
+        With the former method, the SDK will record the added property as a 
+        `Annotation` annotation object, separate from the original `Document` 
+        object. See :meth:`.Mmif.generate_capital_annotations()` for more.
+        
+        A few notes to keep in mind:
+        
+        #. You can't overwrite an existing property of a ``Document`` object. 
+        #. A MMIF can have multiple ``Annotation`` objects with the same 
+           property name but different values. When this happens, the SDK will
+           only keep the latest value (in order of appearances in views list) of 
+           the property, effectively overwriting the previous values.
+        """
         if name == "text":
             self.properties.text = Text(value)
+        elif name == "mime":
+            self.properties.mime = str(value)
         elif name == "location":
             self.location = value
+        elif name not in self._props_original:
+            if self.check_prop_value_is_simple_enough(value):
+                self._props_temporary[name] = value
+            else:
+                super().add_property(name, value)
+
+    def get(self, prop_name):
+        """
+        A special getter for Document properties. This is to allow for reading 
+        the three properties in a specific order so that the latest value is 
+        returned, in case there are multiple values for the same key.
+        """
+        if prop_name in self._props_temporary:
+            return self._props_temporary[prop_name]
+        elif prop_name in self._props_existing:
+            return self._props_existing[prop_name]
         else:
-            super().add_property(name, value)
+            return super().get(prop_name)
 
+    get_property = get
+    
     @property
     def text_language(self) -> str:
         if self.at_type == DocumentTypes.TextDocument:
@@ -210,11 +295,12 @@ class Document(Annotation):
 
     def location_path(self) -> Optional[str]:
         """
-        Retrieves only path name of the document location (hostname is ignored). 
-        Useful to get a path of a local file.
+        Retrieves a path that's resolved to a pathname in the local file system.
+        To obtain the original value of the "path" part in the location string
+        (before resolving), use ``properties.location_path_literal`` method.
         Returns None when no location is set.
         """
-        return self.properties.location_path()
+        return self.properties.location_path_resolved()
 
 
 class AnnotationProperties(MmifObject, MutableMapping[str, T]):
@@ -241,7 +327,7 @@ class AnnotationProperties(MmifObject, MutableMapping[str, T]):
         empty props are ignored (note that emtpy but required props are serialized 
         with the *emtpy* value). 
         Hence, this ``__iter__`` method should also work in the same way and 
-        ignored empty but optional props. 
+        ignore empty optional props. 
         """
         for key in itertools.chain(self._named_attributes(), self._unnamed_attributes):
             if key in self._required_attributes:
@@ -349,9 +435,32 @@ class DocumentProperties(AnnotationProperties):
             return "".join((parsed_location.netloc, parsed_location.path))
 
     def location_path(self) -> Optional[str]:
+        warnings.warn('location_path() is deprecated. Use location_path_resolved() instead.', DeprecationWarning)
+        return self.location_path_resolved()
+    
+    def location_path_resolved(self) -> Optional[str]:
+        """
+        Retrieves only path name of the document location (hostname is ignored), 
+        and then try to resolve the path name in the local file system.
+        This method should be used when the document scheme is ``file`` or empty.
+        For other schemes, users should install ``mmif-locdoc-<scheme>`` plugin.
+        
+        Returns None when no location is set.
+        Raise ValueError when no code found to resolve the given location scheme.
+        """
+        if self.location is None:
+            return None
+        scheme = self.location_scheme()
+        if scheme in ('', 'file'):
+            return urlparse(self.location).path
+        elif scheme in discovered_docloc_plugins:
+            return discovered_docloc_plugins[scheme].resolve(self.location)
+        else:
+            raise ValueError(f'Cannot resolve location of scheme "{scheme}". Interested in developing mmif-locdoc-{scheme} plugin? See https://clams.ai/mmif-python/plugins')
+
+    def location_path_literal(self) -> Optional[str]:
         """
         Retrieves only path name of the document location (hostname is ignored). 
-        Useful to get a path of a local file.
         Returns None when no location is set.
         """
         if self.location is None:

{mmif-python-1.0.0 → mmif-python-1.0.2}/mmif/serialize/mmif.py

@@ -6,6 +6,8 @@ See the specification docs and the JSON Schema file for more information.
 """
 
 import json
+import warnings
+from collections import defaultdict
 from datetime import datetime
 from typing import List, Union, Optional, Dict, ClassVar, cast
 
@@ -68,11 +70,113 @@ class Mmif(MmifObject):
             json_str = json.loads(json_str)
         jsonschema.validators.validate(json_str, schema)
 
-    def serialize(self, pretty: bool = False, sanitize: bool = False) -> str:
+    def serialize(self, pretty: bool = False, sanitize: bool = False, autogenerate_capital_annotations=True) -> str:
+        """
+        Serializes the MMIF object to a JSON string.
+
+        :param sanitize: If True, performs some sanitization of before returning 
+            the JSON string. See :meth:`sanitize` for details.
+        :param autogenerate_capital_annotations: If True, automatically convert 
+            any "pending" temporary properties from `Document` objects to 
+            `Annotation` objects. See :meth:`generate_capital_annotations` for 
+            details.
+        :param pretty: If True, returns string representation with indentation.
+        :return: JSON string of the MMIF object.
+        """
+        if autogenerate_capital_annotations:
+            self.generate_capital_annotations()
+        # sanitization should be done after `Annotation` annotations are generated
         if sanitize:
             self.sanitize()
         return super().serialize(pretty)
 
+    def _deserialize(self, input_dict: dict) -> None:
+        """
+        Deserializes the MMIF JSON string into a Mmif object.
+        This will read in existing ``Annotation`` typed annotations and 
+        attach the document-level properties to the ``Document`` objects, 
+        using a volatile property dict. This will allow apps to access the
+        document-level properties without having too much hassle to iterate
+        views and manually collect the properties.
+        """
+        super()._deserialize(input_dict)
+        for view in self.views:
+            doc_id = None
+            if AnnotationTypes.Annotation in view.metadata.contains:
+                if 'document' in view.metadata.contains[AnnotationTypes.Annotation]:
+                    doc_id = view.metadata.contains[AnnotationTypes.Annotation]['document']
+                # in a view, it is guaranteed that all Annotation objects are not duplicates
+                for ann in view.get_annotations(AnnotationTypes.Annotation):
+                    if doc_id is None:
+                        doc_id = ann.get_property('document')
+                    try:
+                        self.get_document_by_id(doc_id)._add_property_from_annotation(ann)
+                    except KeyError:
+                        warnings.warn(f"Annotation {ann.id} has a document ID {doc_id} that does not exist in the MMIF object. Skipping.", RuntimeWarning)
+
+    def generate_capital_annotations(self):
+        """
+        Automatically convert any "pending" temporary properties from 
+        `Document` objects to `Annotation` objects . The generated `Annotation` 
+        objects are then added to the last `View` in the views lists. 
+        
+        See https://github.com/clamsproject/mmif-python/issues/226 for rationale
+        behind this behavior and discussion.
+        """
+        # this view will be the default kitchen sink for all generated annotations
+        last_view = self.views.get_last()
+        # proceed only when there's at least one view
+        if last_view:
+            # to avoid duplicate property recording, this will be populated with
+            # existing Annotation objects from all existing views
+            existing_anns = defaultdict(lambda: defaultdict(dict))
+            
+            # new properties to record in the current serialization call
+            anns_to_write = defaultdict(dict)
+            for view in self.views:
+                doc_id = None
+                if AnnotationTypes.Annotation in view.metadata.contains:
+                    if 'document' in view.metadata.contains[AnnotationTypes.Annotation]:
+                        doc_id = view.metadata.contains[AnnotationTypes.Annotation]['document']
+                    for ann in view.get_annotations(AnnotationTypes.Annotation):
+                        if doc_id is None:
+                            doc_id = ann.get_property('document')
+                        existing_anns[doc_id].update(ann.properties)
+                for doc in view.get_documents():
+                    anns_to_write[doc.id].update(doc._props_temporary)
+            for doc in self.documents:
+                anns_to_write[doc.id].update(doc._props_temporary)
+            # additional iteration of views, to find a proper view to add the 
+            # generated annotations. If none found, use the last view as the kitchen sink
+            last_view_for_docs = defaultdict(lambda: last_view)
+            doc_ids = set(anns_to_write.keys())
+            for doc_id in doc_ids:
+                for view in reversed(self.views):
+                    # first try to find out if this view "contains" any annotation to the doc
+                    # then, check for individual annotations
+                    if [cont for cont in view.metadata.contains.values() if cont.get('document', None) == doc_id] \
+                            or list(view.get_annotations(document=doc_id)):
+                        last_view_for_docs[doc_id] = view
+                        break
+            for doc_id, found_props in anns_to_write.items():
+                # ignore the "empty" id property from temporary dict 
+                # `id` is "required" attribute for `AnnotationProperty` class 
+                # thus will always be present in `props` dict as a key with emtpy value
+                # also ignore duplicate k-v pairs
+                props = {}
+                for k, v in found_props.items():
+                    if k != 'id' and existing_anns[doc_id][k] != v:
+                        props[k] = v
+                if props:
+                    if len(anns_to_write) == 1:
+                        # if there's only one document, we can record the doc_id in the contains metadata
+                        last_view_for_docs[doc_id].metadata.new_contain(AnnotationTypes.Annotation, document=doc_id)
+                        props.pop('document', None)
+                    else:
+                        # otherwise, doc_id needs to be recorded in the annotation property
+                        props['document'] = doc_id
+                    last_view_for_docs[doc_id].new_annotation(AnnotationTypes.Annotation, **props)
+
     def sanitize(self):
         """
         Sanitizes a Mmif object by running some safeguards.
@@ -195,9 +299,9 @@ class Mmif(MmifObject):
         docs = []
         for view in self.views:
             for doc in view.get_documents():
-                if prop_key in doc and doc.properties[prop_key] == prop_value:
+                if prop_key in doc and doc.get(prop_key) == prop_value:
                     docs.append(doc)
-        docs.extend([document for document in self.documents if document.properties[prop_key] == prop_value])
+        docs.extend([document for document in self.documents if document[prop_key] == prop_value])
         return docs
 
     def get_documents_locations(self, m_type: Union[DocumentTypes, str], path_only=False) -> List[Union[str, None]]:
@@ -231,7 +335,7 @@ class Mmif(MmifObject):
 
         :param doc_id: the ID to search for
         :return: a reference to the corresponding document, if it exists
-        :raises Exception: if there is no corresponding document
+        :raises KeyError: if there is no corresponding document
         """
         if Mmif.id_delimiter in doc_id:
             vid, did = doc_id.split(Mmif.id_delimiter)
@@ -279,7 +383,7 @@ class Mmif(MmifObject):
             else:
                 for alignment in alignment_view.get_annotations(AnnotationTypes.Alignment):
                     aligned_types = set()
-                    for ann_id in [alignment.properties['target'], alignment.properties['source']]:
+                    for ann_id in [alignment['target'], alignment['source']]:
                         ann_id = cast(str, ann_id)
                         if Mmif.id_delimiter in ann_id:
                             view_id, ann_id = ann_id.split(Mmif.id_delimiter)
@@ -294,10 +398,10 @@ class Mmif(MmifObject):
                 v_and_a[alignment_view.id] = alignments
         return v_and_a
 
-    def get_views_for_document(self, doc_id: str):
+    def get_views_for_document(self, doc_id: str) -> List[View]:
         """
         Returns the list of all views that have annotations anchored on a particular document.
-        Note that when the document is insids a view (generated during the pipeline's running),
+        Note that when the document is inside a view (generated during the pipeline's running),
         doc_id must be prefixed with the view_id.
         """
         views = []
@@ -447,6 +551,9 @@ class ViewsList(DataList[View]):
     for :class:`mmif.serialize.view.View`.
     """
     _items: Dict[str, View]
+    
+    def __init__(self, mmif_obj: Optional[Union[bytes, str, list]] = None):
+        super().__init__(mmif_obj)
 
     def _deserialize(self, input_list: list) -> None:  # pytype: disable=signature-mismatch
         """
@@ -456,7 +563,8 @@ class ViewsList(DataList[View]):
         :param input_list: the JSON data that defines the list of views
         :return: None
         """
-        self._items = {item['id']: View(item) for item in input_list}
+        if input_list:
+            self._items = {item['id']: View(item) for item in input_list}
 
     def append(self, value: View, overwrite=False) -> None:
         """
@@ -475,3 +583,11 @@ class ViewsList(DataList[View]):
         :return: None
         """
         super()._append_with_key(value.id, value, overwrite)
+
+    def get_last(self) -> Optional[View]:
+        """
+        Returns the last view appended to the list.
+        """
+        for view in reversed(self._items.values()):
+            if 'error' not in view.metadata and 'warning' not in view.metadata:
+                return view

{mmif-python-1.0.0 → mmif-python-1.0.2}/mmif/serialize/model.py

@@ -11,21 +11,25 @@ core functionality for deserializing MMIF JSON data into live objects
 and serializing live objects into MMIF JSON data. Specialized behavior
 for the different components of MMIF is added in the subclasses.
 """
-
 import json
 from datetime import datetime
+from types import MethodType
 from typing import Union, Any, Dict, Optional, TypeVar, Generic, Generator, Iterator, Type, Set
 
 from deepdiff import DeepDiff
 
+import mmif
+
 T = TypeVar('T')
 S = TypeVar('S')
+JSON_PRMTV_TYPES: Type = Union[str, int, float, bool, None]
 
 __all__ = [
     'MmifObject',
     'MmifObjectEncoder',
     'DataList',
-    'DataDict'
+    'DataDict',
+    'JSON_PRMTV_TYPES'
 ]
 
 
@@ -77,6 +81,8 @@ class MmifObject(object):
      an ID value automatically generated, based on its parent object.
     """
     
+    # these are the reserved names that cannot be used as attribute names, and 
+    # they won't be serialized
     reserved_names: Set[str] = {
         'reserved_names',
         '_unnamed_attributes',
@@ -99,6 +105,8 @@ class MmifObject(object):
             self._unnamed_attributes = {}
         if mmif_obj is not None:
             self.deserialize(mmif_obj)
+        for method in mmif.patches[self.__class__]:
+            setattr(self, method.__name__, MethodType(method, self))
 
     def disallow_additional_properties(self) -> None:
         """
@@ -419,7 +427,7 @@ class DataList(MmifObject, Generic[T]):
         return self._items.__len__()
 
     def __reversed__(self) -> Iterator[T]:
-        return reversed(list(self._items.values()))
+        return reversed(self._items.values())
 
     def __contains__(self, item) -> bool:
         return item in self._items
@@ -440,9 +448,6 @@ class DataDict(MmifObject, Generic[T, S]):
     def _serialize(self, *args, **kwargs) -> dict:
         return super()._serialize(self._items)
 
-    def _deserialize(self, input_dict: dict) -> None:
-        raise NotImplementedError()
-
     def get(self, key: T, default=None) -> Optional[S]:
         return self._items.get(key, default)
 

{mmif-python-1.0.0 → mmif-python-1.0.2}/mmif/serialize/view.py

@@ -47,8 +47,7 @@ class View(MmifObject):
         self._required_attributes = ["id", "metadata", "annotations"]
         super().__init__(view_obj)
         for item in self.annotations:
-            if isinstance(item, Document):
-                item.parent = self.id
+            item.parent = self.id
 
     def new_contain(self, at_type: Union[str, ThingTypesBase], **contains_metadata) -> Optional['Contain']:
         """
@@ -116,6 +115,7 @@ class View(MmifObject):
                           in the view
         :return: the same Annotation object passed in as ``annotation``
         """
+        annotation.parent = self.id
         self.annotations.append(annotation, overwrite)
         self.new_contain(annotation.at_type)
         return annotation
@@ -162,7 +162,6 @@ class View(MmifObject):
                           an existing view with the same ID
         :return: None
         """
-        document.parent = self.id
         return self.add_annotation(document, overwrite)
 
     def get_annotations(self, at_type: Optional[Union[str, ThingTypesBase]] = None, 
@@ -278,8 +277,11 @@ class ViewMetadata(MmifObject):
             
         if at_type not in self.contains:
             new_contain = Contain(contains_metadata)
-            self.contains[at_type] = new_contain
+            self.add_contain(new_contain, at_type)
             return new_contain
+    
+    def add_contain(self, contain: 'Contain', at_type: Union[str, ThingTypesBase]):
+        self.contains[at_type] = contain
 
     def add_parameters(self, **runtime_params):
         self.parameters.update(dict(runtime_params))
@@ -315,7 +317,7 @@ class ErrorDict(MmifObject):
         super().__init__(error_obj)
         
 
-class Contain(MmifObject):
+class Contain(DataDict[str, str]):
     """
     Contain object that represents the metadata of a single
     annotation type in the ``contains`` metadata of a MMIF view.

mmif-python-1.0.2/mmif/ver/__init__.py

@@ -0,0 +1,2 @@
+__version__ = "1.0.2"
+__specver__ = "1.0.0"

{mmif-python-1.0.0 → mmif-python-1.0.2}/mmif/vocabulary/annotation_types.py

@@ -1,4 +1,4 @@
-# Spec version 0.5.0
+# Spec version 1.0.0
 # This file is auto-generated by setup.py
 
 from .base_types import AnnotationTypesBase
@@ -7,7 +7,7 @@ from .base_types import AnnotationTypesBase
 class AnnotationTypes(AnnotationTypesBase):
     """
     This class contains the CLAMS annotation types 
-    defined in the spec version 0.5.0 as class variables. 
+    defined in the spec version 1.0.0 as class variables. 
     """
     Annotation = AnnotationTypesBase('http://mmif.clams.ai/vocabulary/Annotation/v2')
     Region = AnnotationTypesBase('http://mmif.clams.ai/vocabulary/Region/v1')

{mmif-python-1.0.0 → mmif-python-1.0.2}/mmif/vocabulary/base_types.py

@@ -196,7 +196,7 @@ class DocumentTypesBase(ClamsTypesBase):
 class ThingType(ThingTypesBase):
     """
     This class contains the topmost CLAMS thing type 
-    defined in the spec version 0.5.0 as a class variable. 
+    defined in the spec version 1.0.0 as a class variable. 
     """
     Thing = ClamsTypesBase('http://mmif.clams.ai/vocabulary/Thing/v1')
     typevers = {'Thing': 'v1'}

{mmif-python-1.0.0 → mmif-python-1.0.2}/mmif/vocabulary/document_types.py

@@ -1,4 +1,4 @@
-# Spec version 0.5.0
+# Spec version 1.0.0
 # This file is auto-generated by setup.py
 
 from .base_types import DocumentTypesBase
@@ -7,7 +7,7 @@ from .base_types import DocumentTypesBase
 class DocumentTypes(DocumentTypesBase):
     """
     This class contains the CLAMS document types 
-    defined in the spec version 0.5.0 as class variables. 
+    defined in the spec version 1.0.0 as class variables. 
     """
     Document = DocumentTypesBase('http://mmif.clams.ai/vocabulary/Document/v1')
     VideoDocument = DocumentTypesBase('http://mmif.clams.ai/vocabulary/VideoDocument/v1')

{mmif-python-1.0.0 → mmif-python-1.0.2/mmif_python.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mmif-python
-Version: 1.0.0
+Version: 1.0.2
 Summary: Python implementation of MultiMedia Interchange Format specification. (https://mmif.clams.ai)
 Home-page: https://mmif.clams.ai
 Author: Brandeis Lab for Linguistics and Computation

{mmif-python-1.0.0 → mmif-python-1.0.2}/setup.py

@@ -162,11 +162,17 @@ def generate_vocabulary(spec_version, clams_types_vers):
 
 
 def get_latest_mmif_gittag():
-    # vmaj, vmin, vpat = version.split('.')[0:3]
-    res = request.urlopen('https://api.github.com/repos/clamsproject/mmif/git/refs/tags')
-    body = json.loads(res.read())
-    tags = [os.path.basename(tag['ref']) for tag in body]
+    cur_p = 1
+    body = [None]
+    tags = []
+    while len(body) > 0:
+        # for when we have more than 30 (default pagination size) tags
+        res = request.urlopen(f'https://api.github.com/repos/clamsproject/mmif/tags?per_page=100&page={cur_p}')
+        body = json.loads(res.read())
+        tags.extend([tag['name'] for tag in body])
+        cur_p += 1
     # sort and return highest version
+    print(tags)
     mmif_ver_format = lambda x: re.match(r'\d+\.\d+\.\d$', x)
     return sorted([tag for tag in tags if mmif_ver_format(tag)])[-1]
 

mmif-python-1.0.0/VERSION

@@ -1 +0,0 @@
-1.0.0

mmif-python-1.0.0/mmif/__init__.py

@@ -1,20 +0,0 @@
-import contextlib
-
-import pkg_resources
-
-from mmif.ver import __version__
-from mmif.ver import __specver__
-from mmif.vocabulary import *
-from mmif.serialize import *
-
-_res_pkg = 'res'
-_ver_pkg = 'ver'
-_vocabulary_pkg = 'vocabulary'
-_schema_res_name = 'mmif.json'
-
-
-def get_mmif_json_schema():
-    res = pkg_resources.resource_stream(f'{__name__}.{_res_pkg}', _schema_res_name)
-    res_str = res.read().decode('utf-8')
-    res.close()
-    return res_str

mmif-python-1.0.0/mmif/ver/__init__.py

@@ -1,2 +0,0 @@
-__version__ = "1.0.0"
-__specver__ = "0.5.0"