tfds-nightly 4.9.9.dev202508210044__py3-none-any.whl → 4.9.9.dev202508230044__py3-none-any.whl

tensorflow_datasets/core/dataset_builders/croissant_builder_test.py

@@ -15,6 +15,7 @@
 
 """Tests for croissant_builder."""
 
+from typing import Any, Dict, List, Type
 import numpy as np
 import pytest
 from tensorflow_datasets import testing
@@ -146,7 +147,10 @@ def _create_mlc_field(
     ],
 )
 def test_simple_datatype_converter(
-    mlc_field, expected_feature, int_dtype, float_dtype
+    mlc_field: mlc.Field,
+    expected_feature: type[Any],
+    int_dtype: np.dtype | None,
+    float_dtype: np.dtype | None,
 ):
   actual_feature = croissant_builder.datatype_converter(
       mlc_field,
@@ -252,7 +256,11 @@ def test_datatype_converter_bbox_with_invalid_format():
         ),
     ],
 )
-def test_datatype_converter_complex(mlc_field, feature_type, subfield_types):
+def test_datatype_converter_complex(
+    mlc_field: mlc.Field,
+    feature_type: Type[Any],
+    subfield_types: Dict[str, Type[Any]] | None,
+):
   actual_feature = croissant_builder.datatype_converter(mlc_field)
   assert actual_feature.doc.desc == mlc_field.description
   assert isinstance(actual_feature, feature_type)
@@ -411,7 +419,9 @@ def test_version_converter(tmp_path):
 
 
 @pytest.fixture(name="crs_builder")
-def mock_croissant_dataset_builder(tmp_path, request):
+def mock_croissant_dataset_builder(
+    tmp_path, request
+) -> croissant_builder.CroissantBuilder:
   dataset_name = request.param["dataset_name"]
   with testing.dummy_croissant_file(
       dataset_name=dataset_name,
@@ -477,7 +487,11 @@ def test_croissant_builder(crs_builder):
     indirect=["crs_builder"],
 )
 @pytest.mark.parametrize("split_name", ["train", "test"])
-def test_download_and_prepare(crs_builder, expected_entries, split_name):
+def test_download_and_prepare(
+    crs_builder: croissant_builder.CroissantBuilder,
+    expected_entries: List[Dict[str, Any]],
+    split_name: str,
+):
   crs_builder.download_and_prepare()
   data_source = crs_builder.as_data_source(split=split_name)
   expected_entries = [

tensorflow_datasets/core/read_only_builder.py

@@ -314,10 +314,11 @@ def builder_from_files(
     DatasetNotFoundError: If the dataset cannot be loaded.
   """
   # Find and load dataset builder.
-  builder_dir = _find_builder_dir(name, **builder_kwargs)
+  copy_builder_kwargs = dict(builder_kwargs)
+  builder_dir = _find_builder_dir(name, **copy_builder_kwargs)
   if builder_dir is None:
     data_dirs = file_utils.list_data_dirs(
-        given_data_dir=builder_kwargs.get('data_dir')
+        given_data_dir=copy_builder_kwargs.get('data_dir')
     )
     raise registered.DatasetNotFoundError(
         f'Could not find dataset files for: {name}. Make sure you have the'
@@ -325,7 +326,7 @@ def builder_from_files(
         f'and that it has been generated in: {data_dirs}. If the dataset has'
         ' configs, you might have to specify the config name.'
     )
-  file_format = builder_kwargs.pop('file_format', None)
+  file_format = copy_builder_kwargs.pop('file_format', None)
   return builder_from_directory(builder_dir, file_format=file_format)
 
 

tensorflow_datasets/datasets/multi_news/multi_news_dataset_builder.py

@@ -34,9 +34,16 @@ class Builder(tfds.core.GeneratorBasedBuilder):
   VERSION = tfds.core.Version("2.1.0")
   RELEASE_NOTES = {
       "1.0.0": "Initial release.",
-      "2.0.0": "Update the dataset with valid URLs.",
-      "2.1.0": "Update the dataset with cleaned URLs.",
+      "2.0.0": "[Do not use] Update the dataset with valid URLs.",
+      "2.1.0": (
+          "Update the dataset with the correct URLs. The URLs in this version"
+          " come from HuggingFace's dataset repo, which is curated by the same"
+          " author: https://huggingface.co/datasets/alexfabbri/multi_news."
+      ),
   }
+  BLOCKED_VERSIONS = tfds.core.utils.BlockedVersions(
+      versions={"2.0.0": "The URLs of this version are invalid."}
+  )
 
   def _info(self) -> tfds.core.DatasetInfo:
     """Returns the dataset metadata."""
@@ -77,9 +84,10 @@ class Builder(tfds.core.GeneratorBasedBuilder):
     ).open() as tgt_f:
       for i, (src_line, tgt_line) in enumerate(zip(src_f, tgt_f)):
         yield i, {
-            # In original file, each line has one example and natural newline
-            # tokens "\n" are being replaced with "NEWLINE_CHAR". Here restore
-            # the natural newline token to avoid special vocab "NEWLINE_CHAR".
+            # In the original file, each line has one example and natural
+            # newline tokens "\n" are being replaced with "NEWLINE_CHAR"
+            # Here, we restore the natural newline token to avoid the special
+            # vocab token "NEWLINE_CHAR".
             _DOCUMENT: src_line.strip().replace("NEWLINE_CHAR", "\n"),
             _SUMMARY: tgt_line.strip().lstrip(),
         }

tensorflow_datasets/scripts/documentation/build_community_catalog.py

@@ -168,7 +168,7 @@ class DatasetDocumentation:
     )
 
   def to_details_markdown(self) -> str:
-    """ "Markdown to be shown on the details page for the namespace."""
+    """Markdown to be shown on the details page for the namespace."""
     extra_links = self.format_extra_links(prefix='*   ', infix='\n')
     details = self.templates.dataset_details_template.format(
         name=self.name,
@@ -194,9 +194,6 @@ class DatasetDocumentation:
 
   def documentation(self, keep_short: bool = False) -> str:
     """Returns detailed documentation for all configs of this dataset."""
-    # TODO(weide): if e.g. the description contains markdown chars, then it
-    # messes up the page. Try escaping backticks or using code blocks.
-    # TODO(weide): how to format citation?
     header_template = '## {config_name}'
     template = textwrap.dedent("""
       Use the following command to load this dataset in TFDS:
@@ -207,9 +204,7 @@ class DatasetDocumentation:
 
       *   **Description**:
 
-      ```
       {description}
-      ```
 
       *   **License**: {license}
       *   **Version**: {version}
@@ -364,6 +359,7 @@ class HuggingfaceDatasetDocumentation(GithubDatasetDocumentation):
           )
       )
 
+    version = None
     if isinstance(config['version'], dict):
       version = config['version']['version_str']
     elif isinstance(config['version'], str):

tensorflow_datasets/scripts/documentation/dataset_markdown_builder.py

@@ -222,9 +222,9 @@ class VersionSection(Section):
     all_versions = set(tfds.core.Version(v) for v in all_versions)
     for v in sorted(all_versions):  # List all available versions
       if v == builder.version:  # Highlight the default version
-        version_name = '**`{}`** (default)'.format(str(v))
+        version_name = f'**`{v}`** (default)'
       else:
-        version_name = '`{}`'.format(str(v))
+        version_name = f'`{v}`'
       if (
           v in curr_versions  # Filter versions only present in RELEASE_NOTES
           and self._nightly_doc_util
@@ -322,14 +322,14 @@ class AutocacheSection(Section):
       autocached_info_parts = []
       if always_cached:
         split_names_str = ', '.join(always_cached)
-        autocached_info_parts.append('Yes ({})'.format(split_names_str))
+        autocached_info_parts.append(f'Yes ({split_names_str})')
       if never_cached:
         split_names_str = ', '.join(never_cached)
-        autocached_info_parts.append('No ({})'.format(split_names_str))
+        autocached_info_parts.append(f'No ({split_names_str})')
       if unshuffle_cached:
         split_names_str = ', '.join(unshuffle_cached)
         autocached_info_parts.append(
-            'Only when `shuffle_files=False` ({})'.format(split_names_str)
+            f'Only when `shuffle_files=False` ({split_names_str})'
         )
       autocached_info = ', '.join(autocached_info_parts)
     return autocached_info
@@ -346,7 +346,7 @@ class SplitInfoSection(Section):
 
   def _get_num_examples(self, split_info):
     if split_info.num_examples:
-      return '{:,}'.format(split_info.num_examples)
+      return f'{split_info.num_examples:,}'
     return 'Not computed'
 
   def get_key(self, builder: tfds.core.DatasetBuilder):

{tfds_nightly-4.9.9.dev202508210044.dist-info → tfds_nightly-4.9.9.dev202508230044.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tfds-nightly
-Version: 4.9.9.dev202508210044
+Version: 4.9.9.dev202508230044
 Summary: tensorflow/datasets is a library of datasets ready to use with TensorFlow.
 Home-page: https://github.com/tensorflow/datasets
 Download-URL: https://github.com/tensorflow/datasets/tags

{tfds_nightly-4.9.9.dev202508210044.dist-info → tfds_nightly-4.9.9.dev202508230044.dist-info}/RECORD

@@ -90,7 +90,7 @@ tensorflow_datasets/core/load.py,sha256=1FQVnKwn8OVS_IgDbs9XN7aIVxQnyfrS0pI2X9dh
 tensorflow_datasets/core/load_test.py,sha256=EEa8GuSIrEbn0RcGrWS3hmmatKBqBA3QOQWpQ1WjVgA,6490
 tensorflow_datasets/core/naming.py,sha256=B_P77QDA4lkG2FUl4PrzZR0U6qqae_fLxruGBw3ZSVc,25614
 tensorflow_datasets/core/naming_test.py,sha256=SwydgLjf2Mouow1yVZlc73sb8rp4522NhkTSEmg31vo,30112
-tensorflow_datasets/core/read_only_builder.py,sha256=R0QIqckUjl74G7oBj1uCRm_g9e0omstDMTbbwC25B88,22146
+tensorflow_datasets/core/read_only_builder.py,sha256=08BmsgEBXhX0ydGo9-9qHLTjBE6pIvAC6VMmc3b9S8U,22206
 tensorflow_datasets/core/read_only_builder_test.py,sha256=Nw2KQCHBdTW7210Um2K3SzfqAOJB1v1r2yJkzdFehWA,24174
 tensorflow_datasets/core/reader.py,sha256=s65FNOUDyAhd4OgHOSvE5lr4rnlUnOILjlVcRS6Qbhw,17345
 tensorflow_datasets/core/reader_test.py,sha256=VcbUIDtvwjTRZs-0beQIiz26TALqLM5FgBsB-Gtw4kw,17882
@@ -142,7 +142,7 @@ tensorflow_datasets/core/dataset_builders/__init__.py,sha256=StTA3euephqDZdpTzJQ
 tensorflow_datasets/core/dataset_builders/adhoc_builder.py,sha256=1a-5hVjf9t24SD9fWzDDuKoOrA-Vmydf5QxvU7ap-sI,9263
 tensorflow_datasets/core/dataset_builders/adhoc_builder_test.py,sha256=yhRwrznK78MvHeWGRggnMTiyx_SlR1z30iD5VU3Gweo,13096
 tensorflow_datasets/core/dataset_builders/croissant_builder.py,sha256=XmnbIKiEN9OnY_RC8P7-83hbUfvtuJhbm24HfNFpiQs,17088
-tensorflow_datasets/core/dataset_builders/croissant_builder_test.py,sha256=42HpBr3pANVKiok4lcx6xqwf0fY7kma6WIGA8WehNSs,15072
+tensorflow_datasets/core/dataset_builders/croissant_builder_test.py,sha256=_8JVvhkv_QRUhN4GEw6V1PEryJXp8-DLzuVKzjkozgo,15370
 tensorflow_datasets/core/dataset_builders/huggingface_dataset_builder.py,sha256=Loq3qeGk1Ias-d2oT_dK47BRNgTA4LKJchNGh7aA4a0,18313
 tensorflow_datasets/core/dataset_builders/huggingface_dataset_builder_test.py,sha256=6N3DLsry9LhDqhpleaoXrrhaGiLJMBgUlwDnAji-1fI,4389
 tensorflow_datasets/core/dataset_builders/view_builder.py,sha256=eaCtjN5Vg4rK8JD3auA4PhF9mjH5HvQ9dslDX8LbwyM,11907
@@ -935,7 +935,7 @@ tensorflow_datasets/datasets/multi_news/README.md,sha256=s0XL9ddJL7oNJ9r7mSG8_Hd
 tensorflow_datasets/datasets/multi_news/TAGS.txt,sha256=OPDe1XqRiLYpvmXuPX2_aMaOKIXYsl562usmTEEqkwg,449
 tensorflow_datasets/datasets/multi_news/__init__.py,sha256=eFqnTjU7s5iubj6XcKoU8lZUSHecOdnebZFm1vTkjbA,612
 tensorflow_datasets/datasets/multi_news/checksums.tsv,sha256=S-8k82snl0zj1rjjO5LW7svXRNnDuWRc72qpIcBu6WA,1031
-tensorflow_datasets/datasets/multi_news/multi_news_dataset_builder.py,sha256=-ZOuQ7BfPN6_DkrUddcE2qm76eUpmlo8nHizd8DsstQ,3046
+tensorflow_datasets/datasets/multi_news/multi_news_dataset_builder.py,sha256=6ZeVdbrtXKIu1sjxURsoGdPLRW0SXmK0BhOnmyrwpk4,3419
 tensorflow_datasets/datasets/multi_news/multi_news_dataset_builder_test.py,sha256=5amBMQ7PKbPLeZ2kiT18tEb_Z-CMS0DasTRT6goTjXQ,1259
 tensorflow_datasets/datasets/natural_instructions/CITATIONS.bib,sha256=tcQG5eEGL_wr_5MEnZ6Q_ce2oZm6InbbRKiFqee9g7I,412
 tensorflow_datasets/datasets/natural_instructions/README.md,sha256=mceGvviI62PO5mh59sYPP_9vuuwKo0g-m7LQilP1mBI,370
@@ -2007,11 +2007,11 @@ tensorflow_datasets/scripts/deployment/export_community_datasets_test.py,sha256=
 tensorflow_datasets/scripts/documentation/__init__.py,sha256=Z8UWkv0wbzS4AzaLgSpYVGApYv5j57RWY0vN5Z553BQ,613
 tensorflow_datasets/scripts/documentation/build_catalog.py,sha256=SYJoNW-VxvL8xx85uYlFBwbr1k64HcmRBfxsj9-sdYA,8680
 tensorflow_datasets/scripts/documentation/build_catalog_test.py,sha256=qjnqK6lhBh-uNrjLQkEs3AbKFBo5uz_sxhhdT4ibOyA,2532
-tensorflow_datasets/scripts/documentation/build_community_catalog.py,sha256=gh84xnKbL_ndR4GGbgBNLJ0nxjFwiAPLuhUvzeKPZAo,19902
+tensorflow_datasets/scripts/documentation/build_community_catalog.py,sha256=58CT0UaHxw0-mZX1a1aoW96NkszgsntDXuxS_OOZtc8,19709
 tensorflow_datasets/scripts/documentation/build_community_catalog_test.py,sha256=KvCmBzIePyztWPSrCqTJ_j_3puNWXxgSWSfvcMgQPgk,6352
 tensorflow_datasets/scripts/documentation/collection_markdown_builder.py,sha256=4Oofl2dQjlvHTir46x2K6Vpa3amwPaB-3dm43f_GcS8,7287
 tensorflow_datasets/scripts/documentation/collection_markdown_builder_test.py,sha256=t8KEbotAk6zH09HFvJhsrHW23uE8H3_UGLOHOFcvFeQ,3909
-tensorflow_datasets/scripts/documentation/dataset_markdown_builder.py,sha256=DWILh5kphLecCt77GSHXHDXrvFZOwi5AuiT3qnUQ7Ng,25556
+tensorflow_datasets/scripts/documentation/dataset_markdown_builder.py,sha256=4W-L77aM2B4xGNQC7i1p5yHvOlIJfC0dONTM-86yeoY,25498
 tensorflow_datasets/scripts/documentation/dataset_markdown_builder_test.py,sha256=WsDbmAO6TYGFpn4VxF49FYvRy3ujNiysk38cWKlFC10,4219
 tensorflow_datasets/scripts/documentation/doc_utils.py,sha256=DKHGhF7I4ZkKcDAJWYC8mxWBBtbWD211Yv6CTqOBSTw,10325
 tensorflow_datasets/scripts/documentation/doc_utils_test.py,sha256=FSncjt0UCgvdN9WcvqzswwEuf7ZGmdUIRL480PzHxNw,5805
@@ -2399,7 +2399,6 @@ tensorflow_datasets/url_checksums/movie_lens.txt,sha256=DmPIlh1aM7PxNzI5sVmOGwC4
 tensorflow_datasets/url_checksums/movie_rationales.txt,sha256=1GweBeFRzD61ISAkTR5MNiWuujW6PQymgp7ISGBgsAU,139
 tensorflow_datasets/url_checksums/movielens.txt,sha256=i6St5kA_ZV6y8_mk_b47eE9RIf9Pc1VH6asv58kNPlo,731
 tensorflow_datasets/url_checksums/moving_mnist.txt,sha256=OtC5WoEUStRKL2I7jAwIEFF6WvZ-z_1vDGPzxpnGxXA,166
-tensorflow_datasets/url_checksums/multi_news.txt,sha256=noajcrnQ_UK7sh-uRR9CJYaeBFenCmj_ZXr_5ih3Gu0,201
 tensorflow_datasets/url_checksums/multi_nli.txt,sha256=LXDz04hlq0b9au9DDHaX_P-KGVi4ZHWV5wEGJcvD8bA,148
 tensorflow_datasets/url_checksums/multi_nli_mismatch.txt,sha256=LXDz04hlq0b9au9DDHaX_P-KGVi4ZHWV5wEGJcvD8bA,148
 tensorflow_datasets/url_checksums/omniglot.txt,sha256=4KFU4nJ5H772d1JmRBmQ2bzoL0rJqCzMdf6XCx1Xa_c,728
@@ -2472,10 +2471,10 @@ tensorflow_datasets/vision_language/wit/wit_test.py,sha256=PXS8DMNW-MDrT2p5oy4Ic
 tensorflow_datasets/vision_language/wit_kaggle/__init__.py,sha256=vGwSGeM8WE4Q-l0-eEE1sBojmk6YT0l1OO60AWa4Q40,719
 tensorflow_datasets/vision_language/wit_kaggle/wit_kaggle.py,sha256=q-vX_FBzIwsFxL4sY9vuyQ3UQD2PLM4yhUR4U6l-qao,16903
 tensorflow_datasets/vision_language/wit_kaggle/wit_kaggle_test.py,sha256=ZymHT1NkmD-pUnh3BmM3_g30c5afsWYnmqDD9dVyDSA,1778
-tfds_nightly-4.9.9.dev202508210044.dist-info/licenses/AUTHORS,sha256=nvBG4WwfgjuOu1oZkuQKw9kg7X6rve679ObS-YDDmXg,309
-tfds_nightly-4.9.9.dev202508210044.dist-info/licenses/LICENSE,sha256=z8d0m5b2O9McPEK1xHG_dWgUBT6EfBDz6wA0F7xSPTA,11358
-tfds_nightly-4.9.9.dev202508210044.dist-info/METADATA,sha256=wYNNz571WHEAExSi_IJGNjB8LPNS6tiseK1QVGIcc3s,11291
-tfds_nightly-4.9.9.dev202508210044.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-tfds_nightly-4.9.9.dev202508210044.dist-info/entry_points.txt,sha256=eHEL7nF5y1uCY2FgkuYIdE062epJXlAQTSdq89px4p4,73
-tfds_nightly-4.9.9.dev202508210044.dist-info/top_level.txt,sha256=bAevmk9209s_oxVZVlN6hSDIVS423qrMQvmcWSvW4do,20
-tfds_nightly-4.9.9.dev202508210044.dist-info/RECORD,,
+tfds_nightly-4.9.9.dev202508230044.dist-info/licenses/AUTHORS,sha256=nvBG4WwfgjuOu1oZkuQKw9kg7X6rve679ObS-YDDmXg,309
+tfds_nightly-4.9.9.dev202508230044.dist-info/licenses/LICENSE,sha256=z8d0m5b2O9McPEK1xHG_dWgUBT6EfBDz6wA0F7xSPTA,11358
+tfds_nightly-4.9.9.dev202508230044.dist-info/METADATA,sha256=5HX3mZXPihfkdO-s-nCCoBocQswqBhL_-dCkKIMRnlU,11291
+tfds_nightly-4.9.9.dev202508230044.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+tfds_nightly-4.9.9.dev202508230044.dist-info/entry_points.txt,sha256=eHEL7nF5y1uCY2FgkuYIdE062epJXlAQTSdq89px4p4,73
+tfds_nightly-4.9.9.dev202508230044.dist-info/top_level.txt,sha256=bAevmk9209s_oxVZVlN6hSDIVS423qrMQvmcWSvW4do,20
+tfds_nightly-4.9.9.dev202508230044.dist-info/RECORD,,

tensorflow_datasets/url_checksums/multi_news.txt

@@ -1 +0,0 @@
-https://drive.google.com/uc?export=download&id=1vRY2wM6rlOZrf9exGTm5pXj5ExlVwJ0C	256966232	64ae4d2483b248c9664b50bacfab6821f8a3e93f382c7587686fa4a127f77626	multi-news-original-20190725T164630Z-001.zip