huggingface-hub 0.35.0rc0__py3-none-any.whl → 0.35.1__py3-none-any.whl

huggingface_hub/__init__.py

@@ -46,7 +46,7 @@ import sys
 from typing import TYPE_CHECKING
 
 
-__version__ = "0.35.0.rc0"
+__version__ = "0.35.1"
 
 # Alphabetical order of definitions is ensured in tests
 # WARNING: any comment added in this dictionary definition will be lost when
@@ -182,6 +182,8 @@ _SUBMOD_ATTRS = {
         "create_inference_endpoint_from_catalog",
         "create_pull_request",
         "create_repo",
+        "create_scheduled_job",
+        "create_scheduled_uv_job",
         "create_tag",
         "create_webhook",
         "dataset_info",
@@ -192,6 +194,7 @@ _SUBMOD_ATTRS = {
         "delete_folder",
         "delete_inference_endpoint",
         "delete_repo",
+        "delete_scheduled_job",
         "delete_space_secret",
         "delete_space_storage",
         "delete_space_variable",
@@ -219,6 +222,7 @@ _SUBMOD_ATTRS = {
         "get_webhook",
         "grant_access",
         "inspect_job",
+        "inspect_scheduled_job",
         "list_accepted_access_requests",
         "list_collections",
         "list_datasets",
@@ -259,6 +263,7 @@ _SUBMOD_ATTRS = {
         "request_space_storage",
         "restart_space",
         "resume_inference_endpoint",
+        "resume_scheduled_job",
         "revision_exists",
         "run_as_future",
         "run_job",
@@ -267,6 +272,7 @@ _SUBMOD_ATTRS = {
         "set_space_sleep_time",
         "space_info",
         "super_squash_history",
+        "suspend_scheduled_job",
         "unlike",
         "update_collection_item",
         "update_collection_metadata",
@@ -828,6 +834,8 @@ __all__ = [
     "create_inference_endpoint_from_catalog",
     "create_pull_request",
     "create_repo",
+    "create_scheduled_job",
+    "create_scheduled_uv_job",
     "create_tag",
     "create_webhook",
     "dataset_info",
@@ -838,6 +846,7 @@ __all__ = [
     "delete_folder",
     "delete_inference_endpoint",
     "delete_repo",
+    "delete_scheduled_job",
     "delete_space_secret",
     "delete_space_storage",
     "delete_space_variable",
@@ -878,6 +887,7 @@ __all__ = [
     "hf_hub_download",
     "hf_hub_url",
     "inspect_job",
+    "inspect_scheduled_job",
     "interpreter_login",
     "list_accepted_access_requests",
     "list_collections",
@@ -933,6 +943,7 @@ __all__ = [
     "request_space_storage",
     "restart_space",
     "resume_inference_endpoint",
+    "resume_scheduled_job",
     "revision_exists",
     "run_as_future",
     "run_job",
@@ -949,6 +960,7 @@ __all__ = [
     "split_tf_state_dict_into_shards",
     "split_torch_state_dict_into_shards",
     "super_squash_history",
+    "suspend_scheduled_job",
     "try_to_load_from_cache",
     "unlike",
     "update_collection_item",
@@ -1190,6 +1202,8 @@ if TYPE_CHECKING:  # pragma: no cover
         create_inference_endpoint_from_catalog,  # noqa: F401
         create_pull_request,  # noqa: F401
         create_repo,  # noqa: F401
+        create_scheduled_job,  # noqa: F401
+        create_scheduled_uv_job,  # noqa: F401
         create_tag,  # noqa: F401
         create_webhook,  # noqa: F401
         dataset_info,  # noqa: F401
@@ -1200,6 +1214,7 @@ if TYPE_CHECKING:  # pragma: no cover
         delete_folder,  # noqa: F401
         delete_inference_endpoint,  # noqa: F401
         delete_repo,  # noqa: F401
+        delete_scheduled_job,  # noqa: F401
         delete_space_secret,  # noqa: F401
         delete_space_storage,  # noqa: F401
         delete_space_variable,  # noqa: F401
@@ -1227,6 +1242,7 @@ if TYPE_CHECKING:  # pragma: no cover
         get_webhook,  # noqa: F401
         grant_access,  # noqa: F401
         inspect_job,  # noqa: F401
+        inspect_scheduled_job,  # noqa: F401
         list_accepted_access_requests,  # noqa: F401
         list_collections,  # noqa: F401
         list_datasets,  # noqa: F401
@@ -1267,6 +1283,7 @@ if TYPE_CHECKING:  # pragma: no cover
         request_space_storage,  # noqa: F401
         restart_space,  # noqa: F401
         resume_inference_endpoint,  # noqa: F401
+        resume_scheduled_job,  # noqa: F401
         revision_exists,  # noqa: F401
         run_as_future,  # noqa: F401
         run_job,  # noqa: F401
@@ -1275,6 +1292,7 @@ if TYPE_CHECKING:  # pragma: no cover
         set_space_sleep_time,  # noqa: F401
         space_info,  # noqa: F401
         super_squash_history,  # noqa: F401
+        suspend_scheduled_job,  # noqa: F401
         unlike,  # noqa: F401
         update_collection_item,  # noqa: F401
         update_collection_metadata,  # noqa: F401

huggingface_hub/_jobs_api.py

@@ -15,7 +15,7 @@
 from dataclasses import dataclass
 from datetime import datetime
 from enum import Enum
-from typing import Any, Dict, List, Optional
+from typing import Any, Dict, List, Optional, Union
 
 from huggingface_hub import constants
 from huggingface_hub._space_api import SpaceHardware
@@ -47,15 +47,12 @@ class JobStatus:
     stage: JobStage
     message: Optional[str]
 
-    def __init__(self, **kwargs) -> None:
-        self.stage = kwargs["stage"]
-        self.message = kwargs.get("message")
-
 
 @dataclass
 class JobOwner:
     id: str
     name: str
+    type: str
 
 
 @dataclass
@@ -88,8 +85,8 @@ class JobInfo:
         status: (`JobStatus` or `None`):
             Status of the Job, e.g. `JobStatus(stage="RUNNING", message=None)`
             See [`JobStage`] for possible stage values.
-        status: (`JobOwner` or `None`):
-            Owner of the Job, e.g. `JobOwner(id="5e9ecfc04957053f60648a3e", name="lhoestq")`
+        owner: (`JobOwner` or `None`):
+            Owner of the Job, e.g. `JobOwner(id="5e9ecfc04957053f60648a3e", name="lhoestq", type="user")`
 
     Example:
 
@@ -100,7 +97,7 @@ class JobInfo:
     ...     command=["python", "-c", "print('Hello from the cloud!')"]
     ... )
     >>> job
-    JobInfo(id='687fb701029421ae5549d998', created_at=datetime.datetime(2025, 7, 22, 16, 6, 25, 79000, tzinfo=datetime.timezone.utc), docker_image='python:3.12', space_id=None, command=['python', '-c', "print('Hello from the cloud!')"], arguments=[], environment={}, secrets={}, flavor='cpu-basic', status=JobStatus(stage='RUNNING', message=None), owner=JobOwner(id='5e9ecfc04957053f60648a3e', name='lhoestq'), endpoint='https://huggingface.co', url='https://huggingface.co/jobs/lhoestq/687fb701029421ae5549d998')
+    JobInfo(id='687fb701029421ae5549d998', created_at=datetime.datetime(2025, 7, 22, 16, 6, 25, 79000, tzinfo=datetime.timezone.utc), docker_image='python:3.12', space_id=None, command=['python', '-c', "print('Hello from the cloud!')"], arguments=[], environment={}, secrets={}, flavor='cpu-basic', status=JobStatus(stage='RUNNING', message=None), owner=JobOwner(id='5e9ecfc04957053f60648a3e', name='lhoestq', type='user'), endpoint='https://huggingface.co', url='https://huggingface.co/jobs/lhoestq/687fb701029421ae5549d998')
     >>> job.id
     '687fb701029421ae5549d998'
     >>> job.url
@@ -119,8 +116,8 @@ class JobInfo:
     environment: Optional[Dict[str, Any]]
     secrets: Optional[Dict[str, Any]]
     flavor: Optional[SpaceHardware]
-    status: Optional[JobStatus]
-    owner: Optional[JobOwner]
+    status: JobStatus
+    owner: JobOwner
 
     # Inferred fields
     endpoint: str
@@ -132,14 +129,173 @@ class JobInfo:
         self.created_at = parse_datetime(created_at) if created_at else None
         self.docker_image = kwargs.get("dockerImage") or kwargs.get("docker_image")
         self.space_id = kwargs.get("spaceId") or kwargs.get("space_id")
-        self.owner = JobOwner(**(kwargs["owner"] if isinstance(kwargs.get("owner"), dict) else {}))
+        owner = kwargs.get("owner", {})
+        self.owner = JobOwner(id=owner["id"], name=owner["name"], type=owner["type"])
         self.command = kwargs.get("command")
         self.arguments = kwargs.get("arguments")
         self.environment = kwargs.get("environment")
         self.secrets = kwargs.get("secrets")
         self.flavor = kwargs.get("flavor")
-        self.status = JobStatus(**(kwargs["status"] if isinstance(kwargs.get("status"), dict) else {}))
+        status = kwargs.get("status", {})
+        self.status = JobStatus(stage=status["stage"], message=status.get("message"))
 
         # Inferred fields
         self.endpoint = kwargs.get("endpoint", constants.ENDPOINT)
         self.url = f"{self.endpoint}/jobs/{self.owner.name}/{self.id}"
+
+
+@dataclass
+class JobSpec:
+    docker_image: Optional[str]
+    space_id: Optional[str]
+    command: Optional[List[str]]
+    arguments: Optional[List[str]]
+    environment: Optional[Dict[str, Any]]
+    secrets: Optional[Dict[str, Any]]
+    flavor: Optional[SpaceHardware]
+    timeout: Optional[int]
+    tags: Optional[List[str]]
+    arch: Optional[str]
+
+    def __init__(self, **kwargs) -> None:
+        self.docker_image = kwargs.get("dockerImage") or kwargs.get("docker_image")
+        self.space_id = kwargs.get("spaceId") or kwargs.get("space_id")
+        self.command = kwargs.get("command")
+        self.arguments = kwargs.get("arguments")
+        self.environment = kwargs.get("environment")
+        self.secrets = kwargs.get("secrets")
+        self.flavor = kwargs.get("flavor")
+        self.timeout = kwargs.get("timeout")
+        self.tags = kwargs.get("tags")
+        self.arch = kwargs.get("arch")
+
+
+@dataclass
+class LastJobInfo:
+    id: str
+    at: datetime
+
+    def __init__(self, **kwargs) -> None:
+        self.id = kwargs["id"]
+        self.at = parse_datetime(kwargs["at"])
+
+
+@dataclass
+class ScheduledJobStatus:
+    last_job: Optional[LastJobInfo]
+    next_job_run_at: Optional[datetime]
+
+    def __init__(self, **kwargs) -> None:
+        last_job = kwargs.get("lastJob") or kwargs.get("last_job")
+        self.last_job = LastJobInfo(**last_job) if last_job else None
+        next_job_run_at = kwargs.get("nextJobRunAt") or kwargs.get("next_job_run_at")
+        self.next_job_run_at = parse_datetime(str(next_job_run_at)) if next_job_run_at else None
+
+
+@dataclass
+class ScheduledJobInfo:
+    """
+    Contains information about a Job.
+
+    Args:
+        id (`str`):
+            Scheduled Job ID.
+        created_at (`datetime` or `None`):
+            When the scheduled Job was created.
+        tags (`List[str]` or `None`):
+            The tags of the scheduled Job.
+        schedule (`str` or `None`):
+            One of "@annually", "@yearly", "@monthly", "@weekly", "@daily", "@hourly", or a
+            CRON schedule expression (e.g., '0 9 * * 1' for 9 AM every Monday).
+        suspend (`bool` or `None`):
+            Whether the scheduled job is suspended (paused).
+        concurrency (`bool` or `None`):
+            Whether multiple instances of this Job can run concurrently.
+        status (`ScheduledJobStatus` or `None`):
+            Status of the scheduled Job.
+        owner: (`JobOwner` or `None`):
+            Owner of the scheduled Job, e.g. `JobOwner(id="5e9ecfc04957053f60648a3e", name="lhoestq", type="user")`
+        job_spec: (`JobSpec` or `None`):
+            Specifications of the Job.
+
+    Example:
+
+    ```python
+    >>> from huggingface_hub import run_job
+    >>> scheduled_job = create_scheduled_job(
+    ...     image="python:3.12",
+    ...     command=["python", "-c", "print('Hello from the cloud!')"],
+    ...     schedule="@hourly",
+    ... )
+    >>> scheduled_job.id
+    '687fb701029421ae5549d999'
+    >>> scheduled_job.status.next_job_run_at
+    datetime.datetime(2025, 7, 22, 17, 6, 25, 79000, tzinfo=datetime.timezone.utc)
+    ```
+    """
+
+    id: str
+    created_at: Optional[datetime]
+    job_spec: JobSpec
+    schedule: Optional[str]
+    suspend: Optional[bool]
+    concurrency: Optional[bool]
+    status: ScheduledJobStatus
+    owner: JobOwner
+
+    def __init__(self, **kwargs) -> None:
+        self.id = kwargs["id"]
+        created_at = kwargs.get("createdAt") or kwargs.get("created_at")
+        self.created_at = parse_datetime(created_at) if created_at else None
+        self.job_spec = JobSpec(**(kwargs.get("job_spec") or kwargs.get("jobSpec", {})))
+        self.schedule = kwargs.get("schedule")
+        self.suspend = kwargs.get("suspend")
+        self.concurrency = kwargs.get("concurrency")
+        status = kwargs.get("status", {})
+        self.status = ScheduledJobStatus(
+            last_job=status.get("last_job") or status.get("lastJob"),
+            next_job_run_at=status.get("next_job_run_at") or status.get("nextJobRunAt"),
+        )
+        owner = kwargs.get("owner", {})
+        self.owner = JobOwner(id=owner["id"], name=owner["name"], type=owner["type"])
+
+
+def _create_job_spec(
+    *,
+    image: str,
+    command: List[str],
+    env: Optional[Dict[str, Any]],
+    secrets: Optional[Dict[str, Any]],
+    flavor: Optional[SpaceHardware],
+    timeout: Optional[Union[int, float, str]],
+) -> Dict[str, Any]:
+    # prepare job spec to send to HF Jobs API
+    job_spec: Dict[str, Any] = {
+        "command": command,
+        "arguments": [],
+        "environment": env or {},
+        "flavor": flavor or SpaceHardware.CPU_BASIC,
+    }
+    # secrets are optional
+    if secrets:
+        job_spec["secrets"] = secrets
+    # timeout is optional
+    if timeout:
+        time_units_factors = {"s": 1, "m": 60, "h": 3600, "d": 3600 * 24}
+        if isinstance(timeout, str) and timeout[-1] in time_units_factors:
+            job_spec["timeoutSeconds"] = int(float(timeout[:-1]) * time_units_factors[timeout[-1]])
+        else:
+            job_spec["timeoutSeconds"] = int(timeout)
+    # input is either from docker hub or from HF spaces
+    for prefix in (
+        "https://huggingface.co/spaces/",
+        "https://hf.co/spaces/",
+        "huggingface.co/spaces/",
+        "hf.co/spaces/",
+    ):
+        if image.startswith(prefix):
+            job_spec["spaceId"] = image[len(prefix) :]
+            break
+    else:
+        job_spec["dockerImage"] = image
+    return job_spec

huggingface_hub/_local_folder.py

@@ -90,7 +90,7 @@ class LocalDownloadFilePaths:
         resolved_path = str(path.resolve())
         # Some Windows versions do not allow for paths longer than 255 characters.
         # In this case, we must specify it as an extended path by using the "\\?\" prefix.
-        if len(resolved_path) > 255 and not resolved_path.startswith("\\\\?\\"):
+        if os.name == "nt" and len(resolved_path) > 255 and not resolved_path.startswith("\\\\?\\"):
             path = Path("\\\\?\\" + resolved_path)
         return path
 

huggingface_hub/_oauth.py

@@ -6,7 +6,7 @@ import time
 import urllib.parse
 import warnings
 from dataclasses import dataclass
-from typing import TYPE_CHECKING, Dict, List, Optional, Tuple, Union
+from typing import TYPE_CHECKING, Dict, List, Literal, Optional, Tuple, Union
 
 from . import constants
 from .hf_api import whoami
@@ -39,10 +39,8 @@ class OAuthOrgInfo:
             Whether the org has a payment method set up. Hugging Face field.
         role_in_org (`Optional[str]`, *optional*):
             The user's role in the org. Hugging Face field.
-        pending_sso (`Optional[bool]`, *optional*):
-            Indicates if the user granted the OAuth app access to the org but didn't complete SSO. Hugging Face field.
-        missing_mfa (`Optional[bool]`, *optional*):
-            Indicates if the user granted the OAuth app access to the org but didn't complete MFA. Hugging Face field.
+        security_restrictions (`Optional[List[Literal["ip", "token-policy", "mfa", "sso"]]]`, *optional*):
+            Array of security restrictions that the user hasn't completed for this org. Possible values: "ip", "token-policy", "mfa", "sso". Hugging Face field.
     """
 
     sub: str
@@ -52,8 +50,7 @@ class OAuthOrgInfo:
     is_enterprise: bool
     can_pay: Optional[bool] = None
     role_in_org: Optional[str] = None
-    pending_sso: Optional[bool] = None
-    missing_mfa: Optional[bool] = None
+    security_restrictions: Optional[List[Literal["ip", "token-policy", "mfa", "sso"]]] = None
 
 
 @dataclass
@@ -221,8 +218,7 @@ def parse_huggingface_oauth(request: "fastapi.Request") -> Optional[OAuthInfo]:
                 is_enterprise=org.get("isEnterprise"),
                 can_pay=org.get("canPay"),
                 role_in_org=org.get("roleInOrg"),
-                pending_sso=org.get("pendingSSO"),
-                missing_mfa=org.get("missingMFA"),
+                security_restrictions=org.get("securityRestrictions"),
             )
             for org in orgs_data
         ]

huggingface_hub/_tensorboard_logger.py

@@ -14,7 +14,7 @@
 """Contains a logger to push training logs to the Hub, using Tensorboard."""
 
 from pathlib import Path
-from typing import TYPE_CHECKING, List, Optional, Union
+from typing import List, Optional, Union
 
 from ._commit_scheduler import CommitScheduler
 from .errors import EntryNotFoundError
@@ -26,25 +26,24 @@ from .utils import experimental
 # or from 'torch.utils.tensorboard'. Both are compatible so let's try to load
 # from either of them.
 try:
-    from tensorboardX import SummaryWriter
+    from tensorboardX import SummaryWriter as _RuntimeSummaryWriter
 
     is_summary_writer_available = True
-
 except ImportError:
     try:
-        from torch.utils.tensorboard import SummaryWriter
+        from torch.utils.tensorboard import SummaryWriter as _RuntimeSummaryWriter
 
-        is_summary_writer_available = False
+        is_summary_writer_available = True
     except ImportError:
         # Dummy class to avoid failing at import. Will raise on instance creation.
-        SummaryWriter = object
-        is_summary_writer_available = False
+        class _DummySummaryWriter:
+            pass
 
-if TYPE_CHECKING:
-    from tensorboardX import SummaryWriter
+        _RuntimeSummaryWriter = _DummySummaryWriter  # type: ignore[assignment]
+        is_summary_writer_available = False
 
 
-class HFSummaryWriter(SummaryWriter):
+class HFSummaryWriter(_RuntimeSummaryWriter):
     """
     Wrapper around the tensorboard's `SummaryWriter` to push training logs to the Hub.
 

huggingface_hub/_upload_large_folder.py

@@ -24,7 +24,7 @@ import traceback
 from datetime import datetime
 from pathlib import Path
 from threading import Lock
-from typing import TYPE_CHECKING, List, Optional, Tuple, Union
+from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple, Union
 from urllib.parse import quote
 
 from . import constants
@@ -49,6 +49,108 @@ COMMIT_SIZE_SCALE: List[int] = [20, 50, 75, 100, 125, 200, 250, 400, 600, 1000]
 UPLOAD_BATCH_SIZE_XET = 256  # Max 256 files per upload batch for XET-enabled repos
 UPLOAD_BATCH_SIZE_LFS = 1  # Otherwise, batches of 1 for regular LFS upload
 
+# Repository limits (from https://huggingface.co/docs/hub/repositories-recommendations)
+MAX_FILES_PER_REPO = 100_000  # Recommended maximum number of files per repository
+MAX_FILES_PER_FOLDER = 10_000  # Recommended maximum number of files per folder
+MAX_FILE_SIZE_GB = 50  # Hard limit for individual file size
+RECOMMENDED_FILE_SIZE_GB = 20  # Recommended maximum for individual file size
+
+
+def _validate_upload_limits(paths_list: List[LocalUploadFilePaths]) -> None:
+    """
+    Validate upload against repository limits and warn about potential issues.
+
+    Args:
+        paths_list: List of file paths to be uploaded
+
+    Warns about:
+        - Too many files in the repository (>100k)
+        - Too many entries (files or subdirectories) in a single folder (>10k)
+        - Files exceeding size limits (>20GB recommended, >50GB hard limit)
+    """
+    logger.info("Running validation checks on files to upload...")
+
+    # Check 1: Total file count
+    if len(paths_list) > MAX_FILES_PER_REPO:
+        logger.warning(
+            f"You are about to upload {len(paths_list):,} files. "
+            f"This exceeds the recommended limit of {MAX_FILES_PER_REPO:,} files per repository.\n"
+            f"Consider:\n"
+            f"  - Splitting your data into multiple repositories\n"
+            f"  - Using fewer, larger files (e.g., parquet files)\n"
+            f"  - See: https://huggingface.co/docs/hub/repositories-recommendations"
+        )
+
+    # Check 2: Files and subdirectories per folder
+    # Track immediate children (files and subdirs) for each folder
+    from collections import defaultdict
+
+    entries_per_folder: Dict[str, Any] = defaultdict(lambda: {"files": 0, "subdirs": set()})
+
+    for paths in paths_list:
+        path = Path(paths.path_in_repo)
+        parts = path.parts
+
+        # Count this file in its immediate parent directory
+        parent = str(path.parent) if str(path.parent) != "." else "."
+        entries_per_folder[parent]["files"] += 1
+
+        # Track immediate subdirectories for each parent folder
+        # Walk through the path components to track parent-child relationships
+        for i, child in enumerate(parts[:-1]):
+            parent = "." if i == 0 else "/".join(parts[:i])
+            entries_per_folder[parent]["subdirs"].add(child)
+
+    # Check limits for each folder
+    for folder, data in entries_per_folder.items():
+        file_count = data["files"]
+        subdir_count = len(data["subdirs"])
+        total_entries = file_count + subdir_count
+
+        if total_entries > MAX_FILES_PER_FOLDER:
+            folder_display = "root" if folder == "." else folder
+            logger.warning(
+                f"Folder '{folder_display}' contains {total_entries:,} entries "
+                f"({file_count:,} files and {subdir_count:,} subdirectories). "
+                f"This exceeds the recommended {MAX_FILES_PER_FOLDER:,} entries per folder.\n"
+                "Consider reorganising into sub-folders."
+            )
+
+    # Check 3: File sizes
+    large_files = []
+    very_large_files = []
+
+    for paths in paths_list:
+        size = paths.file_path.stat().st_size
+        size_gb = size / 1_000_000_000  # Use decimal GB as per Hub limits
+
+        if size_gb > MAX_FILE_SIZE_GB:
+            very_large_files.append((paths.path_in_repo, size_gb))
+        elif size_gb > RECOMMENDED_FILE_SIZE_GB:
+            large_files.append((paths.path_in_repo, size_gb))
+
+    # Warn about very large files (>50GB)
+    if very_large_files:
+        files_str = "\n  - ".join(f"{path}: {size:.1f}GB" for path, size in very_large_files[:5])
+        more_str = f"\n  ... and {len(very_large_files) - 5} more files" if len(very_large_files) > 5 else ""
+        logger.warning(
+            f"Found {len(very_large_files)} files exceeding the {MAX_FILE_SIZE_GB}GB hard limit:\n"
+            f"  - {files_str}{more_str}\n"
+            f"These files may fail to upload. Consider splitting them into smaller chunks."
+        )
+
+    # Warn about large files (>20GB)
+    if large_files:
+        files_str = "\n  - ".join(f"{path}: {size:.1f}GB" for path, size in large_files[:5])
+        more_str = f"\n  ... and {len(large_files) - 5} more files" if len(large_files) > 5 else ""
+        logger.warning(
+            f"Found {len(large_files)} files larger than {RECOMMENDED_FILE_SIZE_GB}GB (recommended limit):\n"
+            f"  - {files_str}{more_str}\n"
+            f"Large files may slow down loading and processing."
+        )
+
+    logger.info("Validation checks complete.")
+
 
 def upload_large_folder_internal(
     api: "HfApi",
@@ -118,6 +220,11 @@ def upload_large_folder_internal(
     paths_list = [get_local_upload_paths(folder_path, relpath) for relpath in filtered_paths_list]
     logger.info(f"Found {len(paths_list)} candidate files to upload")
 
+    # Validate upload against repository limits
+    _validate_upload_limits(paths_list)
+
+    logger.info("Starting upload...")
+
     # Read metadata for each file
     items = [
         (paths, read_upload_metadata(folder_path, paths.path_in_repo))

huggingface_hub/cli/auth.py

@@ -62,6 +62,9 @@ class AuthCommands(BaseHuggingfaceCLICommand):
         auth_parser = parser.add_parser("auth", help="Manage authentication (login, logout, etc.).")
         auth_subparsers = auth_parser.add_subparsers(help="Authentication subcommands")
 
+        # Show help if no subcommand is provided
+        auth_parser.set_defaults(func=lambda args: auth_parser.print_help())
+
         # Add 'login' as a subcommand of 'auth'
         login_parser = auth_subparsers.add_parser(
             "login", help="Log in using a token from huggingface.co/settings/tokens"
@@ -197,7 +200,7 @@ class AuthWhoami(BaseAuthCommand):
             exit()
         try:
             info = self._api.whoami(token)
-            print(info["name"])
+            print(ANSI.bold("user: "), info["name"])
             orgs = [org["name"] for org in info["orgs"]]
             if orgs:
                 print(ANSI.bold("orgs: "), ",".join(orgs))

huggingface_hub/cli/cache.py

@@ -21,13 +21,7 @@ from functools import wraps
 from tempfile import mkstemp
 from typing import Any, Callable, Iterable, List, Literal, Optional, Union
 
-from ..utils import (
-    CachedRepoInfo,
-    CachedRevisionInfo,
-    CacheNotFound,
-    HFCacheInfo,
-    scan_cache_dir,
-)
+from ..utils import CachedRepoInfo, CachedRevisionInfo, CacheNotFound, HFCacheInfo, scan_cache_dir
 from . import BaseHuggingfaceCLICommand
 from ._cli_utils import ANSI, tabulate
 
@@ -52,7 +46,7 @@ def require_inquirer_py(fn: Callable) -> Callable:
         if not _inquirer_py_available:
             raise ImportError(
                 "The 'cache delete' command requires extra dependencies for the TUI.\n"
-                "Please run 'pip install huggingface_hub[cli]' to install them.\n"
+                "Please run 'pip install \"huggingface_hub[cli]\"' to install them.\n"
                 "Otherwise, disable TUI using the '--disable-tui' flag."
             )
         return fn(*args, **kwargs)
@@ -65,6 +59,10 @@ class CacheCommand(BaseHuggingfaceCLICommand):
     def register_subcommand(parser: _SubParsersAction):
         cache_parser = parser.add_parser("cache", help="Manage local cache directory.")
         cache_subparsers = cache_parser.add_subparsers(dest="cache_command", help="Cache subcommands")
+
+        # Show help if no subcommand is provided
+        cache_parser.set_defaults(func=lambda args: cache_parser.print_help())
+
         # Scan subcommand
         scan_parser = cache_subparsers.add_parser("scan", help="Scan cache directory.")
         scan_parser.add_argument(
@@ -145,7 +143,7 @@ class CacheCommand(BaseHuggingfaceCLICommand):
             if self.verbosity >= 3:
                 print(ANSI.gray(message))
                 for warning in hf_cache_info.warnings:
-                    print(ANSI.gray(warning))
+                    print(ANSI.gray(str(warning)))
             else:
                 print(ANSI.gray(message + " Use -vvv to print details."))
 

huggingface_hub/cli/hf.py

@@ -47,10 +47,6 @@ def main():
     # LFS commands (hidden in --help)
     LfsCommands.register_subcommand(commands_parser)
 
-    # Legacy commands
-
-    # Experimental
-
     # Let's go
     args = parser.parse_args()
     if not hasattr(args, "func"):
@@ -59,7 +55,8 @@ def main():
 
     # Run
     service = args.func(args)
-    service.run()
+    if service is not None:
+        service.run()
 
 
 if __name__ == "__main__":