datamaestro 1.5.1__py3-none-any.whl → 1.6.0__py3-none-any.whl

datamaestro/__init__.py

@@ -7,7 +7,6 @@ from .context import (
     prepare_dataset,
 )
 
-from pkg_resources import get_distribution, DistributionNotFound
 from .definitions import dataset, metadata
 from .data import Base
-from .version import version, version_tuple
+from .version import __version__

datamaestro/__main__.py

@@ -1,6 +1,7 @@
 #!/usr/bin/env python3
 # flake8: noqa: T201
 
+from importlib.metadata import entry_points
 import sys
 import logging
 from functools import update_wrapper
@@ -38,7 +39,7 @@ def pass_cfg(f):
 # Get all the available repositories
 
 REPOSITORIES = {}
-for entry_point in pkg_resources.iter_entry_points("datamaestro.repositories"):
+for entry_point in entry_points(group="datamaestro.repositories"):
     REPOSITORIES[entry_point.name] = entry_point
 
 

datamaestro/context.py

@@ -1,5 +1,5 @@
 from pathlib import Path
-from typing import Iterable, Iterator, Dict, Union
+from typing import Iterable, Iterator, Dict, Optional, Union
 import importlib
 import os
 import hashlib
@@ -8,8 +8,7 @@ import inspect
 import json
 from abc import ABC, abstractmethod
 from experimaestro import Config
-import pkg_resources
-from experimaestro.compat import cached_property
+from functools import cached_property
 from experimaestro.mkdocs.metaloader import Module
 from .utils import CachedFile, downloadURL
 from .settings import UserSettings, Settings
@@ -18,6 +17,22 @@ from typing import TYPE_CHECKING
 if TYPE_CHECKING:
     from datamaestro.definitions import AbstractDataset, DatasetWrapper
 
+from importlib.metadata import (
+    entry_points as _entry_points,
+    version as _version,
+    PackageNotFoundError as _PackageNotFoundError,
+)
+
+
+def iter_entry_points(group, name=None):
+    """Yield entry points for a given group (and optional name) using importlib.metadata."""
+    eps = _entry_points()
+    selected = eps.select(group=group)
+    if name:
+        selected = [ep for ep in selected if ep.name == name]
+    for ep in selected:
+        yield ep
+
 
 class Compression:
     @staticmethod
@@ -106,7 +121,7 @@ class Context:
 
     def repositories(self) -> Iterable["Repository"]:
         """Returns an iterator over repositories"""
-        for entry_point in pkg_resources.iter_entry_points("datamaestro.repositories"):
+        for entry_point in iter_entry_points("datamaestro.repositories"):
             yield entry_point.load().instance()
 
     def repository(self, repositoryid):
@@ -114,10 +129,7 @@ class Context:
             return None
 
         entry_points = [
-            x
-            for x in pkg_resources.iter_entry_points(
-                "datamaestro.repositories", repositoryid
-            )
+            x for x in iter_entry_points("datamaestro.repositories", repositoryid)
         ]
         if not entry_points:
             raise Exception("No datasets repository named %s", repositoryid)
@@ -299,8 +311,7 @@ class BaseRepository(ABC):
         self.basedir = Path(p).parent
 
     @abstractmethod
-    def __iter__(self) -> Iterator["AbstractDataset"]:
-        ...
+    def __iter__(self) -> Iterator["AbstractDataset"]: ...
 
     def search(self, name: str):
         """Search for a dataset in the definitions"""
@@ -353,11 +364,9 @@ class Repository(BaseRepository):
 
     @classmethod
     def version(cls):
-        from pkg_resources import get_distribution, DistributionNotFound
-
         try:
-            return get_distribution(cls.__module__).version
-        except DistributionNotFound:
+            return _version(cls.__module__)
+        except _PackageNotFoundError:
             return None
 
     def __repr__(self):
@@ -423,16 +432,23 @@ def find_dataset(dataset_id: str):
     return AbstractDataset.find(dataset_id)
 
 
-def prepare_dataset(dataset_id: Union[str, "DatasetWrapper", Config]):
+def prepare_dataset(
+    dataset_id: Union[str, "DatasetWrapper", Config],
+    context: Optional[Union[Context, Path]] = None,
+):
     """Find a dataset given its id and download the resources"""
     from .definitions import AbstractDataset, DatasetWrapper
 
+    match context:
+        case Path() | str():
+            context = Context(Path(context))
+
     if isinstance(dataset_id, DatasetWrapper):
         ds = dataset_id
     elif isinstance(dataset_id, Config):
         ds = dataset_id.__datamaestro_dataset__
     else:
-        ds = AbstractDataset.find(dataset_id)
+        ds = AbstractDataset.find(dataset_id, context=context)
 
     return ds.prepare(download=True)
 

datamaestro/definitions.py

@@ -236,10 +236,12 @@ class AbstractDataset(AbstractData):
         return success
 
     @staticmethod
-    def find(name: str) -> "DataDefinition":
+    def find(name: str, context: Optional["Context"] = None) -> "DataDefinition":
         """Find a dataset given its name"""
         from datamaestro.context import Context  # noqa: F811
 
+        context = Context.instance() if context is None else context
+
         logging.debug("Searching dataset %s", name)
         for repository in Context.instance().repositories():
             logging.debug("Searching dataset %s in %s", name, repository)

datamaestro/version.py

@@ -1,21 +1,4 @@
-# file generated by setuptools-scm
-# don't change, don't track in version control
-
-__all__ = ["__version__", "__version_tuple__", "version", "version_tuple"]
-
-TYPE_CHECKING = False
-if TYPE_CHECKING:
-    from typing import Tuple
-    from typing import Union
-
-    VERSION_TUPLE = Tuple[Union[int, str], ...]
-else:
-    VERSION_TUPLE = object
-
-version: str
-__version__: str
-__version_tuple__: VERSION_TUPLE
-version_tuple: VERSION_TUPLE
-
-__version__ = version = '1.5.1'
-__version_tuple__ = version_tuple = (1, 5, 1)
+# This file is auto-generated by Hatchling. As such, do not:
+#   - modify
+#   - track in version control e.g. be sure to add to .gitignore
+__version__ = VERSION = '1.6.0'

{datamaestro-1.5.1.dist-info → datamaestro-1.6.0.dist-info}/METADATA

@@ -1,42 +1,31 @@
 Metadata-Version: 2.4
 Name: datamaestro
-Version: 1.5.1
-Summary: "Dataset management command line and API"
-Home-page: https://github.com/experimaestro/datamaestro
-Author: Benjamin Piwowarski
-Author-email: benjamin@piwowarski.fr
-License: GPL-3
-Keywords: dataset manager
-Platform: any
+Version: 1.6.0
+Summary: Add your description here
+Author-email: Benjamin Piwowarski <benjamin@piwowarski.fr>
+License-File: LICENSE
 Classifier: Development Status :: 4 - Beta
 Classifier: Intended Audience :: Science/Research
 Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3)
 Classifier: Operating System :: OS Independent
 Classifier: Programming Language :: Python
-Classifier: Programming Language :: Python :: 3.9
-Classifier: Programming Language :: Python :: 3.10
-Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
-Requires-Python: >=3.8
-Description-Content-Type: text/markdown
-License-File: LICENSE
-Requires-Dist: click
-Requires-Dist: tqdm
-Requires-Dist: urllib3
-Requires-Dist: marshmallow
-Requires-Dist: cached_property
-Requires-Dist: requests
-Requires-Dist: bitmath
+Requires-Python: >=3.10
+Requires-Dist: bitmath>=1.3.3.1
+Requires-Dist: cached-property>=2.0.1
+Requires-Dist: click>=8.2.1
+Requires-Dist: docstring-parser>=0.16
 Requires-Dist: experimaestro>=1.8.9
-Requires-Dist: mkdocs
-Requires-Dist: pymdown-extensions
-Requires-Dist: mkdocs-material
-Requires-Dist: docstring_parser
+Requires-Dist: marshmallow>=3.26.1
+Requires-Dist: mkdocs-material>=9.6.15
+Requires-Dist: mkdocs>=1.6.1
 Requires-Dist: numpy
-Provides-Extra: test
-Requires-Dist: tox; extra == "test"
-Dynamic: license-file
-Dynamic: requires-dist
+Requires-Dist: pymdown-extensions>=10.16
+Requires-Dist: requests>=2.32.4
+Requires-Dist: tqdm>=4.67.1
+Requires-Dist: urllib3>=2.5.0
+Description-Content-Type: text/markdown
 
 [![PyPI version](https://badge.fury.io/py/datamaestro.svg)](https://badge.fury.io/py/datamaestro) [![pre-commit](https://img.shields.io/badge/pre--commit-enabled-brightgreen?logo=pre-commit&logoColor=white)](https://github.com/pre-commit/pre-commit) [![DOI](https://zenodo.org/badge/4573876.svg)](https://zenodo.org/badge/latestdoi/4573876)
 
@@ -192,22 +181,3 @@ This will allow to
 
 1. Document the dataset
 2. Allow to use the command line interface to manipulate it (download resources, etc.)
-
-# 0.8.0
-
-- Integration with other repositories: abstracting away the notion of dataset
-- Repository prefix
-- Set sub-datasets IDs automatically
-
-# 0.7.3
-
-- Updates for new experimaestro (0.8.5)
-- Search types with "type:..."
-
-# 0.6.17
-
-- Allow remote access through rpyc
-
-# 0.6.9
-
-`version` command

{datamaestro-1.5.1.dist-info → datamaestro-1.6.0.dist-info}/RECORD

@@ -1,14 +1,14 @@
-datamaestro/__init__.py,sha256=LR8nx7H3Fo97O0gJXV2PxQezsmSTDLAg_nQEXB5QAjc,322
-datamaestro/__main__.py,sha256=2p36ZcJcZAL9NZBUkMaYRUhKyqhheVPXMGw6K1KNwhk,9196
-datamaestro/context.py,sha256=KsXYNTt4xX4zEVrnd2hciP7PVCh1StRzjU1Ih6VeCtU,13532
-datamaestro/definitions.py,sha256=FiM_WZhF91f1H2c8ZardEV9jw5R-4zlkFWmzuTGQxos,19705
+datamaestro/__init__.py,sha256=oh9M4VODuvTc9EFHKirtDxpCJkLUANzpzBOIwzHc_mw,246
+datamaestro/__main__.py,sha256=esbe4LcRA9SrRdBzMnJ02R7LvY3KeK_UqlkXbboHhxc,9227
+datamaestro/context.py,sha256=AL2BTi6dLA8rDGBE0PFyfV9ua29JHvBgx6_w6hDj9Dg,13977
+datamaestro/definitions.py,sha256=-9rKgFxqn103HslqWLovq4EdoSkJmx9FNTKneLGBfTo,19812
 datamaestro/record.py,sha256=IxxcrSIf99iluohtpnuMBTFkqeHRe5S-T_hWEqBgeME,5812
 datamaestro/registry.py,sha256=M7QJkcWJP_cxAoqIioLQ01ou2Zg9RqGQvW0XGVspYFE,1421
 datamaestro/search.py,sha256=bRT-91-2VJJ2JSfNaS1mzaVfqq_HMVBVs-RBj0w-ypM,2906
 datamaestro/settings.py,sha256=HYSElTUYZ6DZocBb9o3ifm6WW9knRO64XJUwxGIpvwQ,1304
 datamaestro/sphinx.py,sha256=bp7x_2BFoTSwTqcVZDM8R8cWa7G2pz0Zb8GS054lLYM,6996
 datamaestro/utils.py,sha256=9m-AVVww6InAZfGFiGy6XJzfExpYNqH1fhWQEezjafA,6536
-datamaestro/version.py,sha256=FasBapRYrTzegTWgTLGzFg84_c9eAAt8ZkOEeQxlLC4,511
+datamaestro/version.py,sha256=lfHpvMHdtMD7zPueNTzYopOoKdIsSEJoCUS3amAUa5g,171
 datamaestro/annotations/__init__.py,sha256=jLprrxSBa5QIqc--vqycEcxU4CR9WjVNRaqR5lH0EuE,39
 datamaestro/annotations/agreement.py,sha256=xEH0ddZxdJ_oG_150PoOa-WjY_OaeQja3FzMzY5IB6k,955
 datamaestro/commands/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -40,9 +40,8 @@ datamaestro/test/conftest.py,sha256=it4S5Qq1CA_U8qM0pr4m7v-1dhLj5Y49WjVg5Ee3mpM,
 datamaestro/test/test_annotations.py,sha256=XUjDWb3FJimSD91wcItJ0lLwTBmvN4wVu_EgTKSvV2c,278
 datamaestro/test/test_download_handlers.py,sha256=-Gofr89zqIyeI8C4rZqfYR3JfiZVImdcSz9s6q361zQ,641
 datamaestro/test/test_record.py,sha256=hNZ3uo2i5FZ0VsOHRwvLO1Z6Zce92PdipAF65UptPB8,1156
-datamaestro-1.5.1.dist-info/licenses/LICENSE,sha256=WJ7YI-moTFb-uVrFjnzzhGJrnL9P2iqQe8NuED3hutI,35141
-datamaestro-1.5.1.dist-info/METADATA,sha256=r0FpdXB_X6gXcy3hdBDq06N1lP-6DN-RQ_65tYjRixc,8191
-datamaestro-1.5.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-datamaestro-1.5.1.dist-info/entry_points.txt,sha256=8qMhwSRvFG2iBqtJYVD22Zd4s4c3YkODtcp0Ajw1knw,133
-datamaestro-1.5.1.dist-info/top_level.txt,sha256=XSznaMNAA8jELV7-TOqaAgDsjLzUf9G9MxL7C4helT0,12
-datamaestro-1.5.1.dist-info/RECORD,,
+datamaestro-1.6.0.dist-info/METADATA,sha256=RmJeWy1yUhVfI5_S6Z85Z5rrMcdM6oZMjybvyalE284,7635
+datamaestro-1.6.0.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
+datamaestro-1.6.0.dist-info/entry_points.txt,sha256=8qMhwSRvFG2iBqtJYVD22Zd4s4c3YkODtcp0Ajw1knw,133
+datamaestro-1.6.0.dist-info/licenses/LICENSE,sha256=WJ7YI-moTFb-uVrFjnzzhGJrnL9P2iqQe8NuED3hutI,35141
+datamaestro-1.6.0.dist-info/RECORD,,

{datamaestro-1.5.1.dist-info → datamaestro-1.6.0.dist-info}/WHEEL

@@ -1,5 +1,4 @@
 Wheel-Version: 1.0
-Generator: setuptools (80.9.0)
+Generator: hatchling 1.28.0
 Root-Is-Purelib: true
 Tag: py3-none-any
-

datamaestro-1.5.1.dist-info/top_level.txt

@@ -1 +0,0 @@
-datamaestro