use huggingface_hub offline mode

.github/conda/meta.yaml

@@ -24,7 +24,7 @@ requirements:
     - dataclasses
     - multiprocess
     - fsspec
-    - huggingface_hub >=0.22.0,<1.0.0
+    - huggingface_hub >=0.23.0,<1.0.0
     - packaging
     - aiohttp
   run:
@@ -41,7 +41,7 @@ requirements:
     - dataclasses
     - multiprocess
     - fsspec
-    - huggingface_hub >=0.22.0,<1.0.0
+    - huggingface_hub >=0.23.0,<1.0.0
     - packaging
     - aiohttp
 

.github/workflows/ci.yml

@@ -62,7 +62,7 @@ jobs:
         run: uv pip install --system --upgrade pyarrow huggingface-hub "dill<0.3.9"
       - name: Install dependencies (minimum versions)
         if: ${{ matrix.deps_versions != 'deps-latest' }}
-        run: uv pip install --system pyarrow==15.0.0 huggingface-hub==0.22.0 transformers dill==0.3.1.1
+        run: uv pip install --system pyarrow==15.0.0 huggingface-hub==0.23.5 transformers dill==0.3.1.1
       - name: Test with pytest
         run: |
           python -m pytest -rfExX -m ${{ matrix.test }} -n 2 --dist loadfile -sv ./tests/

setup.py

@@ -133,7 +133,7 @@
     # for data streaming via http
     "aiohttp",
     # To get datasets from the Datasets Hub on huggingface.co
-    "huggingface-hub>=0.22.0",
+    "huggingface-hub>=0.23.0",
     # Utilities from PyPA to e.g., compare versions
     "packaging",
     # To parse YAML metadata from dataset cards

src/datasets/load.py

@@ -40,6 +40,8 @@
 from huggingface_hub.utils import (
     EntryNotFoundError,
     GatedRepoError,
+    LocalEntryNotFoundError,
+    OfflineModeIsEnabled,
     RepositoryNotFoundError,
     RevisionNotFoundError,
     get_session,
@@ -79,7 +81,6 @@
 from .splits import Split
 from .utils import _dataset_viewer
 from .utils.file_utils import (
-    OfflineModeIsEnabled,
     _raise_if_offline_mode_is_enabled,
     cached_path,
     get_datasets_user_agent,
@@ -1603,9 +1604,19 @@ def dataset_module_factory(
                     proxies=download_config.proxies,
                 )
                 commit_hash = os.path.basename(os.path.dirname(dataset_readme_path))
-            except EntryNotFoundError as e:
-                if "internet connection" in str(e).lower():
+            except LocalEntryNotFoundError as e:
+                if isinstance(
+                    e.__cause__,
+                    (
+                        OfflineModeIsEnabled,
+                        requests.exceptions.ConnectTimeout,
+                        requests.exceptions.ConnectionError,
+                    ),
+                ):
                     raise ConnectionError(f"Couldn't reach '{path}' on the Hub ({e.__class__.__name__})") from e
+                else:
+                    raise
+            except EntryNotFoundError:
                 commit_hash = api.dataset_info(
                     path,
                     revision=revision,

src/datasets/utils/file_utils.py

@@ -282,14 +282,10 @@ def get_authentication_headers_for_url(url: str, token: Optional[Union[str, bool
         return {}
 
 
-class OfflineModeIsEnabled(ConnectionError):
-    pass
-
-
 def _raise_if_offline_mode_is_enabled(msg: Optional[str] = None):
     """Raise an OfflineModeIsEnabled error (subclass of ConnectionError) if HF_HUB_OFFLINE is True."""
     if config.HF_HUB_OFFLINE:
-        raise OfflineModeIsEnabled(
+        raise huggingface_hub.errors.OfflineModeIsEnabled(
             "Offline mode is enabled." if msg is None else "Offline mode is enabled. " + str(msg)
         )
 

tests/test_file_utils.py

@@ -7,10 +7,10 @@
 import zstandard as zstd
 from fsspec.registry import _registry as _fsspec_registry
 from fsspec.spec import AbstractBufferedFile, AbstractFileSystem
+from huggingface_hub.errors import OfflineModeIsEnabled
 
 from datasets.download.download_config import DownloadConfig
 from datasets.utils.file_utils import (
-    OfflineModeIsEnabled,
     _get_extraction_protocol,
     _prepare_single_hop_path_and_storage_options,
     cached_path,