pycldf 2.1.0__tar.gz → 2.2.0__tar.gz

{pycldf-2.1.0 → pycldf-2.2.0}/CHANGELOG.md

@@ -3,6 +3,22 @@
 The `pycldf` package adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.html).
 
 
+## [2.2.0] - 2026-11-11
+
+- Fixed bug where missing words in glossed examples would lead to downstream
+  issues (thanks @antonsoo).
+- Fixed bug where explicit foreign key declaration was not honored when 
+  resolving list-valued foreign keys (thanks @antonsoo).
+- Fixed documentation issue with confusing error message.
+- Enhanced support for copying datasets (thanks @antonsoo).
+- Dropped py3.9 support.
+
+
+## [2.1.1] - 2026-07-17 
+
+Fixed incomplete requirements.
+
+
 ## [2.1.0] - 2026-07-17 
 
 - Added support for reading media file data from locally downloaded zip archives. This helps with the

{pycldf-2.1.0 → pycldf-2.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pycldf
-Version: 2.1.0
+Version: 2.2.0
 Summary: A python library to read and write CLDF datasets
 Home-page: https://github.com/cldf/pycldf
 Author: Robert Forkel
@@ -15,19 +15,19 @@ Classifier: Intended Audience :: Science/Research
 Classifier: Natural Language :: English
 Classifier: Operating System :: OS Independent
 Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
 Classifier: Programming Language :: Python :: 3.14
+Classifier: Programming Language :: Python :: 3.15
 Classifier: Programming Language :: Python :: Implementation :: CPython
 Classifier: Programming Language :: Python :: Implementation :: PyPy
 Classifier: License :: OSI Approved :: Apache Software License
-Requires-Python: >=3.9
+Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: python-frontmatter<1.2; python_version < "3.10"
+Requires-Dist: python-frontmatter
 Requires-Dist: csvw>=4.0
 Requires-Dist: clldutils>=4.0
 Requires-Dist: uritemplate>=3.0

{pycldf-2.1.0 → pycldf-2.2.0}/setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name = pycldf
-version = 2.1.0
+version = 2.2.0
 author = Robert Forkel
 author_email = robert_forkel@eva.mpg.de
 description = A python library to read and write CLDF datasets
@@ -20,12 +20,12 @@ classifiers =
 	Natural Language :: English
 	Operating System :: OS Independent
 	Programming Language :: Python :: 3
-	Programming Language :: Python :: 3.9
 	Programming Language :: Python :: 3.10
 	Programming Language :: Python :: 3.11
 	Programming Language :: Python :: 3.12
 	Programming Language :: Python :: 3.13
 	Programming Language :: Python :: 3.14
+	Programming Language :: Python :: 3.15
 	Programming Language :: Python :: Implementation :: CPython
 	Programming Language :: Python :: Implementation :: PyPy
 	License :: OSI Approved :: Apache Software License
@@ -35,9 +35,9 @@ zip_safe = False
 packages = find:
 package_dir = 
 	= src
-python_requires = >=3.9
+python_requires = >=3.10
 install_requires = 
-	python-frontmatter<1.2;python_version<"3.10"
+	python-frontmatter
 	csvw>=4.0
 	clldutils>=4.0
 	uritemplate>=3.0
@@ -116,7 +116,7 @@ show_missing = true
 skip_covered = true
 
 [tox:tox]
-envlist = py39, py310, py311, py312, py313, py314
+envlist = py310, py311, py312, py313, py314, py315
 isolated_build = true
 skip_missing_interpreter = true
 

{pycldf-2.1.0 → pycldf-2.2.0}/src/pycldf/__init__.py

@@ -9,4 +9,4 @@ from pycldf.sources import *
 from pycldf.terms import *
 
 # flake8: noqa
-__version__ = "2.1.0"
+__version__ = "2.2.0"

{pycldf-2.1.0 → pycldf-2.2.0}/src/pycldf/dataset.py

@@ -3,7 +3,7 @@ An implementation of a CLDF dataset object.
 """
 import re
 import types
-from typing import Union, Optional, Type, Any
+from typing import Optional, Type, Any
 import logging
 import pathlib
 import functools
@@ -39,8 +39,8 @@ __all__ = [
     'TextCorpus', 'iter_datasets', 'sniff', 'SchemaError']
 
 ORM_CLASSES = {cls.component_name(): cls for cls in orm.Object.__subclasses__()}
-TableSpecType = Union[str, Link, Table]
-SchemaObjectType = Union[TableSpecType, tuple[TableSpecType, ColType]]
+TableSpecType = str | Link | Table
+SchemaObjectType = TableSpecType | tuple[TableSpecType, ColType]
 ODict = collections.OrderedDict
 RowType = ODict[str, Any]
 
@@ -285,7 +285,7 @@ class Dataset:  # pylint: disable=too-many-public-methods
                 pass
         return None
 
-    def __getitem__(self, item: SchemaObjectType) -> Union[csvw.Table, csvw.Column]:
+    def __getitem__(self, item: SchemaObjectType) -> csvw.Table | csvw.Column:
         """
         Access to tables and columns.
 
@@ -365,7 +365,7 @@ class Dataset:  # pylint: disable=too-many-public-methods
         """
         return bool(self.get(item))
 
-    def get(self, item: SchemaObjectType, default=None) -> Union[csvw.Table, csvw.Column, None]:
+    def get(self, item: SchemaObjectType, default=None) -> csvw.Table | csvw.Column | None:
         """
         Acts like `dict.get`.
 
@@ -485,7 +485,7 @@ class Dataset:  # pylint: disable=too-many-public-methods
         # Now remove the table:
         self.tablegroup.tables = [t for t in self.tablegroup.tables if t.url != table.url]
 
-    def add_component(self, component: Union[str, dict], *cols: ColSpecType, **kw) -> csvw.Table:
+    def add_component(self, component: str | dict, *cols: ColSpecType, **kw) -> csvw.Table:
         """
         Add a CLDF component to a dataset.
 
@@ -632,7 +632,7 @@ class Dataset:  # pylint: disable=too-many-public-methods
         return add_auto_constraints(self, component)
 
     # Add data -------------------------------------------------------------------------------------
-    def add_sources(self, *sources: Union[str, Source], **kw) -> None:
+    def add_sources(self, *sources: str | Source, **kw) -> None:
         """
         Add sources to the dataset.
 
@@ -676,7 +676,7 @@ class Dataset:  # pylint: disable=too-many-public-methods
                 return row
         raise ValueError(id_)  # pragma: no cover
 
-    def get_row_url(self, table: TableType, row: Union[RowType, str]) -> Optional[str]:
+    def get_row_url(self, table: TableType, row: RowType | str) -> Optional[str]:
         """
         Get a URL associated with a row. Tables can specify associated row URLs by
 

{pycldf-2.1.0 → pycldf-2.2.0}/src/pycldf/ext/discovery.py

@@ -16,7 +16,7 @@ Additional resolvers:
   resolver for DOI URLs pointing to the Zenodo archive.
 """
 import re
-from typing import Optional, Union
+from typing import Optional
 import pathlib
 import zipfile
 import warnings
@@ -29,7 +29,6 @@ from csvw.utils import is_url
 
 from pycldf import Dataset, iter_datasets, sniff
 from pycldf.urlutil import url_without_fragment
-from pycldf._compat import entry_points_select
 
 __all__ = ['get_dataset', 'DatasetResolver']
 EP = 'pycldf_dataset_resolver'
@@ -47,11 +46,7 @@ class DatasetResolver:  # pylint: disable=R0903
     """
     priority = 5
 
-    def __call__(
-            self,
-            loc: str,
-            download_dir: pathlib.Path,
-    ) -> Union[None, Dataset, pathlib.Path]:
+    def __call__(self, loc: str, download_dir: pathlib.Path) -> None | Dataset | pathlib.Path:
         """
         :param loc: URL pointing to a place where datasets are archived.
         :param download_dir: A directory to which resolvers can download data.
@@ -145,7 +140,7 @@ class DatasetLocator(str):
 def get_resolvers() -> list[type]:
     """Register resolvers defined via entry points."""
     if not _resolvers:
-        for ep in set(entry_points_select(entry_points(), EP)):
+        for ep in set(entry_points().select(group=EP)):
             try:
                 _resolvers.append(ep.load()())
             except ImportError:  # pragma: no cover
@@ -156,9 +151,9 @@ def get_resolvers() -> list[type]:
 
 def _get_dataset(
         locator: DatasetLocator,
-        location: Union[None, Dataset, pathlib.Path],
+        location: None | Dataset | pathlib.Path,
         get_all: bool = False,
-) -> Optional[Union[Dataset, list[Dataset]]]:
+) -> Optional[Dataset | list[Dataset]]:
     """Determine whether locator matches location and if so, resolve to a Dataset instance."""
     if isinstance(location, Dataset):
         if locator.match(location):
@@ -180,7 +175,7 @@ def _get_dataset(
 def get_dataset(locator: str,
                 download_dir: pathlib.Path,
                 base: Optional[pathlib.Path] = None,
-                get_all: bool = False) -> Union[Dataset, list[Dataset]]:
+                get_all: bool = False) -> Dataset | list[Dataset]:
     """
     :param locator: Dataset locator as specified in "Dataset discovery".
     :param download_dir: Directory to which to download remote data if necessary.

{pycldf-2.1.0 → pycldf-2.2.0}/src/pycldf/ext/markdown.py

@@ -4,7 +4,7 @@ This module provides tools to build a CLDF Markdown renderer.
 For an example, see :class:`FilenameToComponent`.
 """
 import re
-from typing import Optional, Union, Any
+from typing import Optional, Any
 import pathlib
 import warnings
 import collections
@@ -65,7 +65,7 @@ class DatasetMapping(Mapping):
             if not isinstance(self.m[k], Dataset):
                 self.m[k] = get_dataset(self.m[k], download_dir, doc_path)
 
-    def __getitem__(self, prefix: Union[str, None]) -> Dataset:
+    def __getitem__(self, prefix: str | None) -> Dataset:
         """
         Get a `Dataset` mapped to a prefix.
         """
@@ -130,9 +130,7 @@ class CLDFMarkdownLink(MarkdownLink):
         return None  # pragma: no cover
 
     def component(
-            self,
-            cldf: Optional[Union[Dataset, dict[str, Dataset], DatasetMapping]] = None,
-    ) -> Union[str, None]:
+            self, cldf: None | Dataset | dict[str, Dataset] | DatasetMapping = None) -> str | None:
         """
         :param cldf: `pycldf.Dataset` instance to which the link refers.
         :return: Name of the CLDF component the link pertains to or `None`.
@@ -173,7 +171,7 @@ class CLDFMarkdownLink(MarkdownLink):
         """
         return self.objid == '__all__'
 
-    def get_row(self, cldf: Union[Dataset, DatasetMapping]) -> dict:
+    def get_row(self, cldf: Dataset | DatasetMapping) -> dict:
         """
         Resolve the reference in a CLDF Markdown link to a row (`dict`) in the CLDF `Dataset`.
         """
@@ -181,7 +179,7 @@ class CLDFMarkdownLink(MarkdownLink):
         ds = DatasetMapping(cldf)[self.prefix]
         return ds.get_row(self.component(cldf=ds), self.objid)
 
-    def get_object(self, cldf: Union[Dataset, DatasetMapping]) -> orm.Object:
+    def get_object(self, cldf: Dataset | DatasetMapping) -> orm.Object:
         """
         Resolve the reference in a CLDF Markdown link to an ORM object in the CLDF `Dataset`.
         """
@@ -219,8 +217,8 @@ class CLDFMarkdownText:
     :cvar metadata_component: Name of the special "Metadata" component.
     """
     def __init__(self,
-                 text: Union[pathlib.Path, str],
-                 dataset_mapping: Optional[Union[str, Dataset, dict]] = None,
+                 text: pathlib.Path | str,
+                 dataset_mapping: Optional[str | Dataset | dict] = None,
                  download_dir: Optional[pathlib.Path] = None):
         """
         :param text: CLDF Markdown text either to be read from a path or specified as `str`.
@@ -230,7 +228,7 @@ class CLDFMarkdownText:
         """
         p = frontmatter.loads(text) if isinstance(text, str) else frontmatter.load(str(text))
         self.metadata: dict[str, Any] = p.metadata
-        self.dataset_mapping: Mapping[Union[str, None], Dataset] = DatasetMapping(
+        self.dataset_mapping: Mapping[str | None, Dataset] = DatasetMapping(
             p.get(DATASETS_MAPPING),
             dataset_mapping,
             text.parent if isinstance(text, pathlib.Path) else None,
@@ -249,7 +247,7 @@ class CLDFMarkdownText:
         """
         return f'---\n{yaml.dump(self.metadata)}---'
 
-    def get_object(self, ml: CLDFMarkdownLink) -> Union[list, orm.Object, Source, dict]:
+    def get_object(self, ml: CLDFMarkdownLink) -> list | orm.Object | Source | dict:
         """
         Resolve the reference in a CLDF Markdown link to the matching object from a mapped dataset.
 
@@ -279,13 +277,13 @@ class CLDFMarkdownText:
         return list(self._datadict[ml.prefix][key].values()) if ml.all \
             else self._datadict[ml.prefix][key][ml.objid]
 
-    def _render_link(self, link: CLDFMarkdownLink) -> Union[str, CLDFMarkdownLink]:
+    def _render_link(self, link: CLDFMarkdownLink) -> str | CLDFMarkdownLink:
         """Dispatches to custom rendering in case of CLDF links."""
         if link.is_cldf_link:
             return self.render_link(link)
         return link
 
-    def render_link(self, cldf_link: CLDFMarkdownLink) -> Union[str, CLDFMarkdownLink]:
+    def render_link(self, cldf_link: CLDFMarkdownLink) -> str | CLDFMarkdownLink:
         """
         CLDF Markdown renderers must implement this method.
         """

{pycldf-2.1.0 → pycldf-2.2.0}/src/pycldf/fileutil.py

@@ -4,12 +4,12 @@ Functionality to access and manipulate files.
 import re
 import math
 import string
-from typing import Union, Optional
+from typing import Optional
 import pathlib
 import itertools
 
 
-PathType = Union[str, pathlib.Path]
+PathType = str | pathlib.Path
 
 
 def splitfile(p: PathType, chunksize: int, total: Optional[int] = None) -> list[pathlib.Path]:

{pycldf-2.1.0 → pycldf-2.2.0}/src/pycldf/media.py

@@ -24,7 +24,7 @@ or instantiate a :class:`.File` from a :class:`pycldf.orm.Object`:
 import io
 import json
 import base64
-from typing import Union, TYPE_CHECKING, Optional, Callable
+from typing import TYPE_CHECKING, Optional, Callable
 import pathlib
 import zipfile
 import functools
@@ -47,7 +47,7 @@ if TYPE_CHECKING:
 
 __all__ = ['Mimetype', 'MediaTable', 'File']
 
-StrOrBytes = Union[str, bytes]
+StrOrBytes = str | bytes
 
 
 class File:  # pylint: disable=too-many-instance-attributes
@@ -91,7 +91,7 @@ class File:  # pylint: disable=too-many-instance-attributes
 
     @classmethod
     def from_dataset(
-            cls, ds: 'Dataset', row_or_object: Union[dict, orm.Media]) -> 'File':
+            cls, ds: 'Dataset', row_or_object: dict | orm.Media) -> 'File':
         """
         Factory method to instantiate a `File` bypassing the `Media` wrapper.
         """
@@ -350,7 +350,7 @@ class Mimetype:
         return self.type == 'text'
 
     @property
-    def extension(self) -> Union[None, str]:
+    def extension(self) -> None | str:
         """Return a suitable filename extension for the mimetype."""
         return mimetypes.guess_extension(f'{self.type}/{self.subtype}')
 
@@ -360,7 +360,7 @@ class Mimetype:
             return data.decode(self.encoding)
         return data
 
-    def write(self, data: StrOrBytes, p: Optional[pathlib.Path] = None) -> Union[int, StrOrBytes]:
+    def write(self, data: StrOrBytes, p: Optional[pathlib.Path] = None) -> int | StrOrBytes:
         """The mimetype dictates how/if to encode data."""
         res = data.encode(self.encoding) if self.is_text else data
         return p.write_bytes(res) if p else res

{pycldf-2.1.0 → pycldf-2.2.0}/src/pycldf/module.py

@@ -2,7 +2,7 @@
 Functionality to manage modules, i.e. `Dataset` subclasses implementing particular CLDF modules.
 """
 import dataclasses
-from typing import Union, Optional, Type
+from typing import Optional, Type
 
 from csvw.metadata import TableGroup
 
@@ -33,7 +33,7 @@ class Module:
         """
         return self.uri.split('#')[1]
 
-    def match(self, thing: Union[TableGroup, str]) -> bool:
+    def match(self, thing: TableGroup | str) -> bool:
         """Check if the module described here matches thing."""
         if isinstance(thing, TableGroup):
             return thing.common_props.get('dc:conformsTo') == term_uri(self.id)
@@ -45,7 +45,7 @@ class Module:
 _modules = []
 
 
-def get_module_impl(base_class, spec: Union[TableGroup, str]) -> Optional[Type]:
+def get_module_impl(base_class, spec: TableGroup | str) -> Optional[Type]:
     """
     Returns an implementation (aka Dataset subclass) for a particular CLDF module.
     """

{pycldf-2.1.0 → pycldf-2.2.0}/src/pycldf/orm.py

@@ -46,7 +46,7 @@ Limitations:
   * ~35secs iterating over ``pycldf.Dataset.objects('ValueTable')``
 """
 import types
-from typing import TYPE_CHECKING, Union, Optional, Any
+from typing import TYPE_CHECKING, Optional, Any
 import decimal
 import functools
 import collections
@@ -63,7 +63,7 @@ if TYPE_CHECKING:
     from pycldf.media import File  # pragma: no cover
 
 
-def to_json(s: Any) -> Union[str, float, None, list, dict]:
+def to_json(s: Any) -> str | float | int | None | list | dict:
     """Converts `s` to an object that can be serialized as JSON."""
     if isinstance(s, (list, tuple)):
         return [to_json(ss) for ss in s]
@@ -162,19 +162,19 @@ class Object:  # pylint: disable=too-many-instance-attributes
             return getattr(col, attr).expand(**variables)
         return None  # pragma: no cover
 
-    def aboutUrl(self, col: str = 'id') -> Union[str, None]:  # pylint: disable=invalid-name
+    def aboutUrl(self, col: str = 'id') -> str | None:  # pylint: disable=invalid-name
         """
         The table's `aboutUrl` property, expanded with the object's row as context.
         """
         return self._expand_uritemplate('aboutUrl', col)
 
-    def valueUrl(self, col: str = 'id') -> Union[str, None]:  # pylint: disable=invalid-name
+    def valueUrl(self, col: str = 'id') -> str | None:  # pylint: disable=invalid-name
         """
         The table's `valueUrl` property, expanded with the object's row as context.
         """
         return self._expand_uritemplate('valueUrl', col)
 
-    def propertyUrl(self, col: str = 'id') -> Union[str, None]:  # pylint: disable=invalid-name
+    def propertyUrl(self, col: str = 'id') -> str | None:  # pylint: disable=invalid-name
         """
         The table's `propertyUrl` property, expanded with the object's row as context.
         """
@@ -218,7 +218,7 @@ class Object:  # pylint: disable=too-many-instance-attributes
                                           'referencing columns other than CLDF id or primary key.')
         return None  # pragma: no cover
 
-    def all_related(self, relation: str) -> Union[DictTuple, list]:
+    def all_related(self, relation: str) -> DictTuple | list:
         """
         CLDF reference properties can be list-valued. This method returns all related objects for
         such a property.
@@ -227,7 +227,16 @@ class Object:  # pylint: disable=too-many-instance-attributes
         if fks and not isinstance(fks, list):
             fks = [fks]
         if fks:
-            return DictTuple(self.dataset.get_object(TERMS[relation].references, fk) for fk in fks)
+            pk = False
+            ref = self.dataset.get_foreign_key_reference(self.component_name(), relation)
+            if ref and str(ref[1].propertyUrl) != term_uri('id'):
+                if [ref[1].name] != self.dataset[TERMS[relation].references].tableSchema.primaryKey:
+                    raise NotImplementedError(
+                        'pycldf does not support foreign key constraints '
+                        'referencing columns other than CLDF id or primary key.')
+                pk = True
+            return DictTuple(
+                self.dataset.get_object(TERMS[relation].references, fk, pk=pk) for fk in fks)
         return []
 
 
@@ -237,7 +246,7 @@ class _WithLanguageMixin:
         return self.related('languageReference')
 
     @property
-    def languages(self) -> Union[DictTuple, list]:  # pylint: disable=C0116
+    def languages(self) -> DictTuple | list:  # pylint: disable=C0116
         return self.all_related('languageReference')
 
 
@@ -247,7 +256,7 @@ class _WithParameterMixin:
         return self.related('parameterReference')
 
     @property
-    def parameters(self) -> Union[DictTuple, list]:  # pylint: disable=C0116
+    def parameters(self) -> DictTuple | list:  # pylint: disable=C0116
         return self.all_related('parameterReference')
 
 
@@ -379,7 +388,7 @@ class Language(Object):
         return None  # pragma: no cover
 
     @property
-    def as_geojson_feature(self) -> Union[None, dict[str, Any]]:
+    def as_geojson_feature(self) -> None | dict[str, Any]:
         """
         `dict` suitable for serialization as GeoJSON Feature object, with the point coordinate as
         geographic data.

{pycldf-2.1.0 → pycldf-2.2.0}/src/pycldf/schemautil.py

@@ -1,18 +1,16 @@
 """
 Functionality to create schema objects.
 """
-from typing import Union
-
 from csvw.metadata import Column, Table
 from clldutils import jsonlib
 
 from pycldf.terms import TERMS
 from pycldf.util import MD_SUFFIX, pkg_path
 
-ColSpecType = Union[str, dict, Column]
-TableSpecType = Union[str, dict, Table]
-TableType = Union[str, Table]
-ColType = Union[str, Column]
+ColSpecType = str | dict | Column
+TableSpecType = str | dict | Table
+TableType = str | Table
+ColType = str | Column
 
 
 def make_column(spec: ColSpecType) -> Column:

{pycldf-2.1.0 → pycldf-2.2.0}/src/pycldf/sliceutil.py

@@ -5,14 +5,14 @@ In addition to Python's way of specifying slices as triples of integers (start,
 allow specification as strings like '1' or '2:5', where the numbers are interpreted as **1-based**
 indices, specifying **inclusive** boundaries. I.e. '2:5' is equivalent to `slice(1:5).`
 """
-from typing import Union, TypeVar
+from typing import TypeVar
 import itertools
 from collections.abc import Sequence, Iterable
 
 __all__ = ['multislice', 'multislice_with_split']
 
 T = TypeVar('T')
-SliceType = Union[str, tuple[int], tuple[int, int], tuple[int, int, int], slice]
+SliceType = str | tuple[int] | tuple[int, int] | tuple[int, int, int] | slice
 
 
 def multislice(sliceable: Sequence[T], *slices: SliceType) -> Sequence[T]:

{pycldf-2.1.0 → pycldf-2.2.0}/src/pycldf/sources.py

@@ -3,7 +3,7 @@ Functionality to handle BibTeX source data of Datasets.
 """
 import re
 import types
-from typing import Optional, Union, Literal
+from typing import Optional, Literal
 import pathlib
 import zipfile
 import tempfile
@@ -91,7 +91,7 @@ class Source(BaseSource):
                 else:
                     yield database.Person(name)
 
-    def refkey(self, year_brackets: Union[None, Literal["round", "square", "curly"]] = 'round'):
+    def refkey(self, year_brackets: None | Literal["round", "square", "curly"] = 'round'):
         """Compute an author-year type reference key for the item."""
         brackets = {
             None: ('', ''),
@@ -205,7 +205,7 @@ class Sources:
             pages = pages[:-1].strip()
         return sid, pages
 
-    def validate(self, refs: Union[str, list[str]]) -> None:
+    def validate(self, refs: str | list[str]) -> None:
         """Make sure refs are valid. If not, raises Exceptions."""
         if not isinstance(refs, str) and any(r is None for r in refs):
             raise ValueError('empty reference in ref list (possibly caused by trailing separator)')
@@ -232,14 +232,14 @@ class Sources:
                 self._add_entries(Source('misc', sid, glottolog_id=sid), **kw)
             yield Reference(self[sid], pages)
 
-    def _add_entries(self, data: Union[Source, database.BibliographyData], **kw) -> None:
+    def _add_entries(self, data: Source | database.BibliographyData, **kw) -> None:
         if isinstance(data, Source):
             entries = [(data.id, data.entry)]
         elif hasattr(data, 'entries'):
             entries = data.entries.items()
         else:
             msg = (
-                'expected `clldutils.source.Source`,'
+                'expected `pycldf.sources.Source`,'
                 ' `pybtex.database.BibliographyData`,'
                 ' or `simplepybtex.database.BibliographyData`;'
                 f' got {type(data)}')
@@ -302,7 +302,7 @@ class Sources:
             return fname
         return None
 
-    def add(self, *entries: Union[str, Source], **kw) -> None:
+    def add(self, *entries: str | Source, **kw) -> None:
         """
         Add a source, either specified as BibTeX string or as :class:`Source`.
         """

{pycldf-2.1.0 → pycldf-2.2.0}/src/pycldf/terms.py

@@ -8,7 +8,7 @@ import pathlib
 import warnings
 import dataclasses
 import urllib.parse
-from typing import Optional, Union, Callable, Any, TYPE_CHECKING, Literal, get_args
+from typing import Optional, Callable, Any, TYPE_CHECKING, Literal, get_args
 from collections.abc import Container
 from xml.etree import ElementTree
 
@@ -197,7 +197,7 @@ class Terms(dict):
         return {k: v for k, v in self.items() if v.subtype == 'component'}
 
 
-def term_uri(name: Union[Term, str], terms: Container[str] = None, ns: str = URL) -> Optional[str]:
+def term_uri(name: Term | str, terms: Container[str] = None, ns: str = URL) -> Optional[str]:
     """
     Returns a full term URI associated with `name`.
 

{pycldf-2.1.0 → pycldf-2.2.0}/src/pycldf/trees.py

@@ -43,6 +43,11 @@ class Tree:
     """
     Represents a tree object as specified in a row of `TreeTable`.
     """
+    description: Optional[str]
+    tree_type: Optional[str]
+    tree_is_rooted: Optional[bool]
+    tree_branch_length_unit: Optional[str]
+
     def __init__(self, trees: 'TreeTable', row: 'RowType', file: File):
         self.row: 'RowType' = row
         self.id: str = row[trees.cols['id'].name]

{pycldf-2.1.0 → pycldf-2.2.0}/src/pycldf/urlutil.py

@@ -1,16 +1,16 @@
 """
 Functionality to manipulate URLs.
 """
-from typing import Callable, Union
+from typing import Callable
 import urllib.parse
 
 __all__ = ['update_url', 'sanitize_url', 'url_without_fragment']
 
 
 def update_url(
-        url: Union[str, urllib.parse.ParseResult],
+        url: str | urllib.parse.ParseResult,
         updater: Callable[[urllib.parse.ParseResult], tuple[str, str, str, str, str]],
-) -> Union[str, None]:
+) -> str | None:
     """Generic update function for URLs."""
     if url is None:
         return None
@@ -32,6 +32,6 @@ def sanitize_url(url: str) -> str:
     return update_url(url, fix)
 
 
-def url_without_fragment(url: Union[str, urllib.parse.ParseResult]) -> str:
+def url_without_fragment(url: str | urllib.parse.ParseResult) -> str:
     """Removes fragment from URL."""
     return update_url(url, lambda u: (u.scheme, u.hostname, u.path, u.query, ''))