pycldf 2.0.2__tar.gz → 2.1.0__tar.gz

{pycldf-2.0.2 → pycldf-2.1.0}/CHANGELOG.md

@@ -3,6 +3,13 @@
 The `pycldf` package adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.html).
 
 
+## [2.1.0] - 2026-07-17 
+
+- Added support for reading media file data from locally downloaded zip archives. This helps with the
+  use case of datasets linking to media in separate Zenodo deposits.
+- Added support for discovering multiple datasets from the same locator.
+
+
 ## [2.0.2] - 2026-05-14
 
 - Fix bug whereby pipes in regular expression patterns would screw up table

{pycldf-2.0.2 → pycldf-2.1.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pycldf
-Version: 2.0.2
+Version: 2.1.0
 Summary: A python library to read and write CLDF datasets
 Home-page: https://github.com/cldf/pycldf
 Author: Robert Forkel
@@ -27,6 +27,7 @@ Classifier: License :: OSI Approved :: Apache Software License
 Requires-Python: >=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE
+Requires-Dist: python-frontmatter<1.2; python_version < "3.10"
 Requires-Dist: csvw>=4.0
 Requires-Dist: clldutils>=4.0
 Requires-Dist: uritemplate>=3.0
@@ -34,7 +35,6 @@ Requires-Dist: python-dateutil
 Requires-Dist: simplepybtex
 Requires-Dist: newick
 Requires-Dist: commonnexus>=1.2.0
-Requires-Dist: python-frontmatter
 Requires-Dist: termcolor
 Requires-Dist: PyYAML
 Requires-Dist: jmespath

{pycldf-2.0.2 → pycldf-2.1.0}/RELEASING.md

@@ -12,7 +12,7 @@ Releasing pycldf
   ```shell
   flake8 src
   ```
-- Make sure pylint passes with a score of 10:
+- Make sure pylint passes with a score >= 9.99:
   ```shell
   pylint src
   ```

{pycldf-2.0.2 → pycldf-2.1.0}/setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name = pycldf
-version = 2.0.2
+version = 2.1.0
 author = Robert Forkel
 author_email = robert_forkel@eva.mpg.de
 description = A python library to read and write CLDF datasets
@@ -37,6 +37,7 @@ package_dir =
 	= src
 python_requires = >=3.9
 install_requires = 
+	python-frontmatter<1.2;python_version<"3.10"
 	csvw>=4.0
 	clldutils>=4.0
 	uritemplate>=3.0
@@ -44,7 +45,6 @@ install_requires =
 	simplepybtex
 	newick
 	commonnexus>=1.2.0
-	python-frontmatter
 	termcolor
 	PyYAML
 	jmespath

{pycldf-2.0.2 → pycldf-2.1.0}/src/pycldf/__init__.py

@@ -9,4 +9,4 @@ from pycldf.sources import *
 from pycldf.terms import *
 
 # flake8: noqa
-__version__ = "2.0.2"
+__version__ = "2.1.0"

{pycldf-2.0.2 → pycldf-2.1.0}/src/pycldf/ext/discovery.py

@@ -157,13 +157,16 @@ def get_resolvers() -> list[type]:
 def _get_dataset(
         locator: DatasetLocator,
         location: Union[None, Dataset, pathlib.Path],
-) -> Optional[Dataset]:
+        get_all: bool = False,
+) -> Optional[Union[Dataset, list[Dataset]]]:
     """Determine whether locator matches location and if so, resolve to a Dataset instance."""
     if isinstance(location, Dataset):
         if locator.match(location):
             return location
         return None
     if location.is_dir():
+        if get_all:
+            return [ds for ds in iter_datasets(location) if locator.match(ds)]
         for ds in iter_datasets(location):
             if locator.match(ds):
                 return ds
@@ -176,7 +179,8 @@ def _get_dataset(
 
 def get_dataset(locator: str,
                 download_dir: pathlib.Path,
-                base: Optional[pathlib.Path] = None) -> Dataset:
+                base: Optional[pathlib.Path] = None,
+                get_all: bool = False) -> Union[Dataset, list[Dataset]]:
     """
     :param locator: Dataset locator as specified in "Dataset discovery".
     :param download_dir: Directory to which to download remote data if necessary.
@@ -191,7 +195,7 @@ def get_dataset(locator: str,
         else:
             res = resolver(locator.url_without_fragment, download_dir)
         if res:
-            res = _get_dataset(locator, res)
+            res = _get_dataset(locator, res, get_all=get_all)
             if res:
                 return res
     raise ValueError(f'Could not resolve dataset locator {locator}')

{pycldf-2.0.2 → pycldf-2.1.0}/src/pycldf/media.py

@@ -139,12 +139,19 @@ class File:  # pylint: disable=too-many-instance-attributes
 
     def local_path(self, d: pathlib.Path = None) -> Optional[pathlib.Path]:
         """
-        :return: The expected path of the file in the directory `d`.
+        :return: The expected path of the file in the directory `d` in the case of files \
+        downloaded using the `downloadmedia` command. If `d` is a file it is accepted as full \
+        local path if it has the same name as the filename in the URL.
         """
         if d is None:
             if self.scheme == 'file':
                 return self._dsdir / urllib.parse.unquote(self.relpath)
             return None
+        if d.is_file() and self.parsed_url:
+            if d.name == pathlib.Path(self.parsed_url.path).name:
+                # Support the use case of files in zip archives, where the archives are available
+                # locally, e.g. as download of a separate media file deposit from Zenodo.
+                return d
         zip_ext = '.zip' if self.path_in_zip else (self.mimetype.extension or '')
         return d.joinpath(f'{self.id}{zip_ext}')
 
@@ -155,14 +162,14 @@ class File:  # pylint: disable=too-many-instance-attributes
 
     def read(self, d: Optional[pathlib.Path] = None) -> Optional[StrOrBytes]:
         """
-        :param d: A local directory where the file has been saved before. If `None`, the content \
-        will be read from the file's URL.
+        :param d: A local path where the file has been saved before - as expected by `local_path`. \
+        If `None`, the content will be read from the file's URL.
         """
         if self.path_in_zip:
             zipcontent = None
             if d:
                 zipcontent = self.local_path(d).read_bytes()
-            if self.url:
+            if zipcontent is None and self.url:
                 zipcontent = self.url_reader[self.scheme](
                     self.parsed_url, Mimetype('application/zip'))
             if zipcontent:

{pycldf-2.0.2 → pycldf-2.1.0}/src/pycldf.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pycldf
-Version: 2.0.2
+Version: 2.1.0
 Summary: A python library to read and write CLDF datasets
 Home-page: https://github.com/cldf/pycldf
 Author: Robert Forkel
@@ -27,6 +27,7 @@ Classifier: License :: OSI Approved :: Apache Software License
 Requires-Python: >=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE
+Requires-Dist: python-frontmatter<1.2; python_version < "3.10"
 Requires-Dist: csvw>=4.0
 Requires-Dist: clldutils>=4.0
 Requires-Dist: uritemplate>=3.0
@@ -34,7 +35,6 @@ Requires-Dist: python-dateutil
 Requires-Dist: simplepybtex
 Requires-Dist: newick
 Requires-Dist: commonnexus>=1.2.0
-Requires-Dist: python-frontmatter
 Requires-Dist: termcolor
 Requires-Dist: PyYAML
 Requires-Dist: jmespath

{pycldf-2.0.2 → pycldf-2.1.0}/src/pycldf.egg-info/requires.txt

@@ -5,11 +5,13 @@ python-dateutil
 simplepybtex
 newick
 commonnexus>=1.2.0
-python-frontmatter
 termcolor
 PyYAML
 jmespath
 
+[:python_version < "3.10"]
+python-frontmatter<1.2
+
 [catalogs]
 cldfcatalog
 pyglottolog>=4.0

{pycldf-2.0.2 → pycldf-2.1.0}/tests/test_ext_discovery.py

@@ -10,6 +10,8 @@ from pycldf.ext.discovery import *
 
 def test_get_dataset_local(data, tmp_path):
     assert get_dataset('structuredataset_with_examples', tmp_path, base=data)
+    res = get_dataset('structuredataset_with_examples', tmp_path, base=data, get_all=True)
+    assert isinstance(res, list) and isinstance(res[0], Dataset)
 
 
 def test_get_dataset_github(data, tmp_path, mocker):

{pycldf-2.0.2 → pycldf-2.1.0}/tests/test_media.py

@@ -194,6 +194,14 @@ def test_Media(tmp_path, ds_factory):
     assert list(MediaTable(ds))[0].read() == 'äöü'
     assert list(MediaTable(ds))[0].read(d=tmp_path) == 'äöü'
 
+    ds = ds_factory(dict(
+        ID='x',
+        Download_URL="http://example.com/path/123.zip",
+        Media_Type='text/plain',
+        Path_In_Zip='arc/name',
+    ))
+    assert list(MediaTable(ds))[0].read(d=tmp_path / '123.zip') == 'äöü'
+
 
 def test_save_read_zipped_media(dataset_with_trees, tmp_path):
     zipped = None