Add files using upload-large-folder tool

.gitattributes

@@ -136,3 +136,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 19.Deeper.Dare[[:space:]]You.101231.mp4 filter=lfs diff=lfs merge=lfs -text
 17.Mofos.Spinner[[:space:]]Sucks[[:space:]]Cock[[:space:]]for[[:space:]]Fame.mp4 filter=lfs diff=lfs merge=lfs -text
 18.Mofos.A[[:space:]]Sneaky[[:space:]]Threesome[[:space:]]Situation.mp4 filter=lfs diff=lfs merge=lfs -text
+17.Clips4Sale.Family[[:space:]]Therapy.Let[[:space:]]Daddy[[:space:]]Help.mp4 filter=lfs diff=lfs merge=lfs -text
+17.CherryPimps.Give[[:space:]]Me[[:space:]]That[[:space:]]Cock[[:space:]]LIVE.mp4 filter=lfs diff=lfs merge=lfs -text
+17.RealityKings.Teaching[[:space:]]Her[[:space:]]To[[:space:]]Suck[[:space:]]Cock.mp4 filter=lfs diff=lfs merge=lfs -text

17.CherryPimps.Give Me That Cock LIVE.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0801b315fee505e3bf8263b012cdbab11bb8eae47f3a921b6695951cbeab4fa7
+size 9009317124

17.Clips4Sale.Family Therapy.Let Daddy Help.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17607983880ebc83ade9468ffa6a25940422058756d187cc60b13908d29cd67a
+size 159946715

17.RealityKings.Teaching Her To Suck Cock.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:752a736937fc5e9bee0a332ed57170cbf4aa42462d587e57d39510388cc98dde
+size 2386235157

venv/lib/python3.12/site-packages/hf_xet-1.1.5.dist-info/licenses/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

venv/lib/python3.12/site-packages/huggingface_hub/utils/_cache_assets.py

@@ -0,0 +1,135 @@
+# coding=utf-8
+# Copyright 2019-present, the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from pathlib import Path
+from typing import Union
+
+from ..constants import HF_ASSETS_CACHE
+
+
+def cached_assets_path(
+    library_name: str,
+    namespace: str = "default",
+    subfolder: str = "default",
+    *,
+    assets_dir: Union[str, Path, None] = None,
+):
+    """Return a folder path to cache arbitrary files.
+
+    `huggingface_hub` provides a canonical folder path to store assets. This is the
+    recommended way to integrate cache in a downstream library as it will benefit from
+    the builtins tools to scan and delete the cache properly.
+
+    The distinction is made between files cached from the Hub and assets. Files from the
+    Hub are cached in a git-aware manner and entirely managed by `huggingface_hub`. See
+    [related documentation](https://huggingface.co/docs/huggingface_hub/how-to-cache).
+    All other files that a downstream library caches are considered to be "assets"
+    (files downloaded from external sources, extracted from a .tar archive, preprocessed
+    for training,...).
+
+    Once the folder path is generated, it is guaranteed to exist and to be a directory.
+    The path is based on 3 levels of depth: the library name, a namespace and a
+    subfolder. Those 3 levels grants flexibility while allowing `huggingface_hub` to
+    expect folders when scanning/deleting parts of the assets cache. Within a library,
+    it is expected that all namespaces share the same subset of subfolder names but this
+    is not a mandatory rule. The downstream library has then full control on which file
+    structure to adopt within its cache. Namespace and subfolder are optional (would
+    default to a `"default/"` subfolder) but library name is mandatory as we want every
+    downstream library to manage its own cache.
+
+    Expected tree:
+    ```text
+        assets/
+        └── datasets/
+        │   ├── SQuAD/
+        │   │   ├── downloaded/
+        │   │   ├── extracted/
+        │   │   └── processed/
+        │   ├── Helsinki-NLP--tatoeba_mt/
+        │       ├── downloaded/
+        │       ├── extracted/
+        │       └── processed/
+        └── transformers/
+            ├── default/
+            │   ├── something/
+            ├── bert-base-cased/
+            │   ├── default/
+            │   └── training/
+        hub/
+        └── models--julien-c--EsperBERTo-small/
+            ├── blobs/
+            │   ├── (...)
+            │   ├── (...)
+            ├── refs/
+            │   └── (...)
+            └── [ 128]  snapshots/
+                ├── 2439f60ef33a0d46d85da5001d52aeda5b00ce9f/
+                │   ├── (...)
+                └── bbc77c8132af1cc5cf678da3f1ddf2de43606d48/
+                    └── (...)
+    ```
+
+
+    Args:
+        library_name (`str`):
+            Name of the library that will manage the cache folder. Example: `"dataset"`.
+        namespace (`str`, *optional*, defaults to "default"):
+            Namespace to which the data belongs. Example: `"SQuAD"`.
+        subfolder (`str`, *optional*, defaults to "default"):
+            Subfolder in which the data will be stored. Example: `extracted`.
+        assets_dir (`str`, `Path`, *optional*):
+            Path to the folder where assets are cached. This must not be the same folder
+            where Hub files are cached. Defaults to `HF_HOME / "assets"` if not provided.
+            Can also be set with `HF_ASSETS_CACHE` environment variable.
+
+    Returns:
+        Path to the cache folder (`Path`).
+
+    Example:
+    ```py
+    >>> from huggingface_hub import cached_assets_path
+
+    >>> cached_assets_path(library_name="datasets", namespace="SQuAD", subfolder="download")
+    PosixPath('/home/wauplin/.cache/huggingface/extra/datasets/SQuAD/download')
+
+    >>> cached_assets_path(library_name="datasets", namespace="SQuAD", subfolder="extracted")
+    PosixPath('/home/wauplin/.cache/huggingface/extra/datasets/SQuAD/extracted')
+
+    >>> cached_assets_path(library_name="datasets", namespace="Helsinki-NLP/tatoeba_mt")
+    PosixPath('/home/wauplin/.cache/huggingface/extra/datasets/Helsinki-NLP--tatoeba_mt/default')
+
+    >>> cached_assets_path(library_name="datasets", assets_dir="/tmp/tmp123456")
+    PosixPath('/tmp/tmp123456/datasets/default/default')
+    ```
+    """
+    # Resolve assets_dir
+    if assets_dir is None:
+        assets_dir = HF_ASSETS_CACHE
+    assets_dir = Path(assets_dir).expanduser().resolve()
+
+    # Avoid names that could create path issues
+    for part in (" ", "/", "\\"):
+        library_name = library_name.replace(part, "--")
+        namespace = namespace.replace(part, "--")
+        subfolder = subfolder.replace(part, "--")
+
+    # Path to subfolder is created
+    path = assets_dir / library_name / namespace / subfolder
+    try:
+        path.mkdir(exist_ok=True, parents=True)
+    except (FileExistsError, NotADirectoryError):
+        raise ValueError(f"Corrupted assets folder: cannot create directory because of an existing file ({path}).")
+
+    # Return
+    return path

venv/lib/python3.12/site-packages/huggingface_hub/utils/_cache_manager.py

@@ -0,0 +1,896 @@
+# coding=utf-8
+# Copyright 2022-present, the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Contains utilities to manage the HF cache directory."""
+
+import os
+import shutil
+import time
+from collections import defaultdict
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Dict, FrozenSet, List, Literal, Optional, Set, Union
+
+from huggingface_hub.errors import CacheNotFound, CorruptedCacheException
+
+from ..commands._cli_utils import tabulate
+from ..constants import HF_HUB_CACHE
+from . import logging
+
+
+logger = logging.get_logger(__name__)
+
+REPO_TYPE_T = Literal["model", "dataset", "space"]
+
+# List of OS-created helper files that need to be ignored
+FILES_TO_IGNORE = [".DS_Store"]
+
+
+@dataclass(frozen=True)
+class CachedFileInfo:
+    """Frozen data structure holding information about a single cached file.
+
+    Args:
+        file_name (`str`):
+            Name of the file. Example: `config.json`.
+        file_path (`Path`):
+            Path of the file in the `snapshots` directory. The file path is a symlink
+            referring to a blob in the `blobs` folder.
+        blob_path (`Path`):
+            Path of the blob file. This is equivalent to `file_path.resolve()`.
+        size_on_disk (`int`):
+            Size of the blob file in bytes.
+        blob_last_accessed (`float`):
+            Timestamp of the last time the blob file has been accessed (from any
+            revision).
+        blob_last_modified (`float`):
+            Timestamp of the last time the blob file has been modified/created.
+
+    <Tip warning={true}>
+
+    `blob_last_accessed` and `blob_last_modified` reliability can depend on the OS you
+    are using. See [python documentation](https://docs.python.org/3/library/os.html#os.stat_result)
+    for more details.
+
+    </Tip>
+    """
+
+    file_name: str
+    file_path: Path
+    blob_path: Path
+    size_on_disk: int
+
+    blob_last_accessed: float
+    blob_last_modified: float
+
+    @property
+    def blob_last_accessed_str(self) -> str:
+        """
+        (property) Timestamp of the last time the blob file has been accessed (from any
+        revision), returned as a human-readable string.
+
+        Example: "2 weeks ago".
+        """
+        return _format_timesince(self.blob_last_accessed)
+
+    @property
+    def blob_last_modified_str(self) -> str:
+        """
+        (property) Timestamp of the last time the blob file has been modified, returned
+        as a human-readable string.
+
+        Example: "2 weeks ago".
+        """
+        return _format_timesince(self.blob_last_modified)
+
+    @property
+    def size_on_disk_str(self) -> str:
+        """
+        (property) Size of the blob file as a human-readable string.
+
+        Example: "42.2K".
+        """
+        return _format_size(self.size_on_disk)
+
+
+@dataclass(frozen=True)
+class CachedRevisionInfo:
+    """Frozen data structure holding information about a revision.
+
+    A revision correspond to a folder in the `snapshots` folder and is populated with
+    the exact tree structure as the repo on the Hub but contains only symlinks. A
+    revision can be either referenced by 1 or more `refs` or be "detached" (no refs).
+
+    Args:
+        commit_hash (`str`):
+            Hash of the revision (unique).
+            Example: `"9338f7b671827df886678df2bdd7cc7b4f36dffd"`.
+        snapshot_path (`Path`):
+            Path to the revision directory in the `snapshots` folder. It contains the
+            exact tree structure as the repo on the Hub.
+        files: (`FrozenSet[CachedFileInfo]`):
+            Set of [`~CachedFileInfo`] describing all files contained in the snapshot.
+        refs (`FrozenSet[str]`):
+            Set of `refs` pointing to this revision. If the revision has no `refs`, it
+            is considered detached.
+            Example: `{"main", "2.4.0"}` or `{"refs/pr/1"}`.
+        size_on_disk (`int`):
+            Sum of the blob file sizes that are symlink-ed by the revision.
+        last_modified (`float`):
+            Timestamp of the last time the revision has been created/modified.
+
+    <Tip warning={true}>
+
+    `last_accessed` cannot be determined correctly on a single revision as blob files
+    are shared across revisions.
+
+    </Tip>
+
+    <Tip warning={true}>
+
+    `size_on_disk` is not necessarily the sum of all file sizes because of possible
+    duplicated files. Besides, only blobs are taken into account, not the (negligible)
+    size of folders and symlinks.
+
+    </Tip>
+    """
+
+    commit_hash: str
+    snapshot_path: Path
+    size_on_disk: int
+    files: FrozenSet[CachedFileInfo]
+    refs: FrozenSet[str]
+
+    last_modified: float
+
+    @property
+    def last_modified_str(self) -> str:
+        """
+        (property) Timestamp of the last time the revision has been modified, returned
+        as a human-readable string.
+
+        Example: "2 weeks ago".
+        """
+        return _format_timesince(self.last_modified)
+
+    @property
+    def size_on_disk_str(self) -> str:
+        """
+        (property) Sum of the blob file sizes as a human-readable string.
+
+        Example: "42.2K".
+        """
+        return _format_size(self.size_on_disk)
+
+    @property
+    def nb_files(self) -> int:
+        """
+        (property) Total number of files in the revision.
+        """
+        return len(self.files)
+
+
+@dataclass(frozen=True)
+class CachedRepoInfo:
+    """Frozen data structure holding information about a cached repository.
+
+    Args:
+        repo_id (`str`):
+            Repo id of the repo on the Hub. Example: `"google/fleurs"`.
+        repo_type (`Literal["dataset", "model", "space"]`):
+            Type of the cached repo.
+        repo_path (`Path`):
+            Local path to the cached repo.
+        size_on_disk (`int`):
+            Sum of the blob file sizes in the cached repo.
+        nb_files (`int`):
+            Total number of blob files in the cached repo.
+        revisions (`FrozenSet[CachedRevisionInfo]`):
+            Set of [`~CachedRevisionInfo`] describing all revisions cached in the repo.
+        last_accessed (`float`):
+            Timestamp of the last time a blob file of the repo has been accessed.
+        last_modified (`float`):
+            Timestamp of the last time a blob file of the repo has been modified/created.
+
+    <Tip warning={true}>
+
+    `size_on_disk` is not necessarily the sum of all revisions sizes because of
+    duplicated files. Besides, only blobs are taken into account, not the (negligible)
+    size of folders and symlinks.
+
+    </Tip>
+
+    <Tip warning={true}>
+
+    `last_accessed` and `last_modified` reliability can depend on the OS you are using.
+    See [python documentation](https://docs.python.org/3/library/os.html#os.stat_result)
+    for more details.
+
+    </Tip>
+    """
+
+    repo_id: str
+    repo_type: REPO_TYPE_T
+    repo_path: Path
+    size_on_disk: int
+    nb_files: int
+    revisions: FrozenSet[CachedRevisionInfo]
+
+    last_accessed: float
+    last_modified: float
+
+    @property
+    def last_accessed_str(self) -> str:
+        """
+        (property) Last time a blob file of the repo has been accessed, returned as a
+        human-readable string.
+
+        Example: "2 weeks ago".
+        """
+        return _format_timesince(self.last_accessed)
+
+    @property
+    def last_modified_str(self) -> str:
+        """
+        (property) Last time a blob file of the repo has been modified, returned as a
+        human-readable string.
+
+        Example: "2 weeks ago".
+        """
+        return _format_timesince(self.last_modified)
+
+    @property
+    def size_on_disk_str(self) -> str:
+        """
+        (property) Sum of the blob file sizes as a human-readable string.
+
+        Example: "42.2K".
+        """
+        return _format_size(self.size_on_disk)
+
+    @property
+    def refs(self) -> Dict[str, CachedRevisionInfo]:
+        """
+        (property) Mapping between `refs` and revision data structures.
+        """
+        return {ref: revision for revision in self.revisions for ref in revision.refs}
+
+
+@dataclass(frozen=True)
+class DeleteCacheStrategy:
+    """Frozen data structure holding the strategy to delete cached revisions.
+
+    This object is not meant to be instantiated programmatically but to be returned by
+    [`~utils.HFCacheInfo.delete_revisions`]. See documentation for usage example.
+
+    Args:
+        expected_freed_size (`float`):
+            Expected freed size once strategy is executed.
+        blobs (`FrozenSet[Path]`):
+            Set of blob file paths to be deleted.
+        refs (`FrozenSet[Path]`):
+            Set of reference file paths to be deleted.
+        repos (`FrozenSet[Path]`):
+            Set of entire repo paths to be deleted.
+        snapshots (`FrozenSet[Path]`):
+            Set of snapshots to be deleted (directory of symlinks).
+    """
+
+    expected_freed_size: int
+    blobs: FrozenSet[Path]
+    refs: FrozenSet[Path]
+    repos: FrozenSet[Path]
+    snapshots: FrozenSet[Path]
+
+    @property
+    def expected_freed_size_str(self) -> str:
+        """
+        (property) Expected size that will be freed as a human-readable string.
+
+        Example: "42.2K".
+        """
+        return _format_size(self.expected_freed_size)
+
+    def execute(self) -> None:
+        """Execute the defined strategy.
+
+        <Tip warning={true}>
+
+        If this method is interrupted, the cache might get corrupted. Deletion order is
+        implemented so that references and symlinks are deleted before the actual blob
+        files.
+
+        </Tip>
+
+        <Tip warning={true}>
+
+        This method is irreversible. If executed, cached files are erased and must be
+        downloaded again.
+
+        </Tip>
+        """
+        # Deletion order matters. Blobs are deleted in last so that the user can't end
+        # up in a state where a `ref`` refers to a missing snapshot or a snapshot
+        # symlink refers to a deleted blob.
+
+        # Delete entire repos
+        for path in self.repos:
+            _try_delete_path(path, path_type="repo")
+
+        # Delete snapshot directories
+        for path in self.snapshots:
+            _try_delete_path(path, path_type="snapshot")
+
+        # Delete refs files
+        for path in self.refs:
+            _try_delete_path(path, path_type="ref")
+
+        # Delete blob files
+        for path in self.blobs:
+            _try_delete_path(path, path_type="blob")
+
+        logger.info(f"Cache deletion done. Saved {self.expected_freed_size_str}.")
+
+
+@dataclass(frozen=True)
+class HFCacheInfo:
+    """Frozen data structure holding information about the entire cache-system.
+
+    This data structure is returned by [`scan_cache_dir`] and is immutable.
+
+    Args:
+        size_on_disk (`int`):
+            Sum of all valid repo sizes in the cache-system.
+        repos (`FrozenSet[CachedRepoInfo]`):
+            Set of [`~CachedRepoInfo`] describing all valid cached repos found on the
+            cache-system while scanning.
+        warnings (`List[CorruptedCacheException]`):
+            List of [`~CorruptedCacheException`] that occurred while scanning the cache.
+            Those exceptions are captured so that the scan can continue. Corrupted repos
+            are skipped from the scan.
+
+    <Tip warning={true}>
+
+    Here `size_on_disk` is equal to the sum of all repo sizes (only blobs). However if
+    some cached repos are corrupted, their sizes are not taken into account.
+
+    </Tip>
+    """
+
+    size_on_disk: int
+    repos: FrozenSet[CachedRepoInfo]
+    warnings: List[CorruptedCacheException]
+
+    @property
+    def size_on_disk_str(self) -> str:
+        """
+        (property) Sum of all valid repo sizes in the cache-system as a human-readable
+        string.
+
+        Example: "42.2K".
+        """
+        return _format_size(self.size_on_disk)
+
+    def delete_revisions(self, *revisions: str) -> DeleteCacheStrategy:
+        """Prepare the strategy to delete one or more revisions cached locally.
+
+        Input revisions can be any revision hash. If a revision hash is not found in the
+        local cache, a warning is thrown but no error is raised. Revisions can be from
+        different cached repos since hashes are unique across repos,
+
+        Examples:
+        ```py
+        >>> from huggingface_hub import scan_cache_dir
+        >>> cache_info = scan_cache_dir()
+        >>> delete_strategy = cache_info.delete_revisions(
+        ...     "81fd1d6e7847c99f5862c9fb81387956d99ec7aa"
+        ... )
+        >>> print(f"Will free {delete_strategy.expected_freed_size_str}.")
+        Will free 7.9K.
+        >>> delete_strategy.execute()
+        Cache deletion done. Saved 7.9K.
+        ```
+
+        ```py
+        >>> from huggingface_hub import scan_cache_dir
+        >>> scan_cache_dir().delete_revisions(
+        ...     "81fd1d6e7847c99f5862c9fb81387956d99ec7aa",
+        ...     "e2983b237dccf3ab4937c97fa717319a9ca1a96d",
+        ...     "6c0e6080953db56375760c0471a8c5f2929baf11",
+        ... ).execute()
+        Cache deletion done. Saved 8.6G.
+        ```
+
+        <Tip warning={true}>
+
+        `delete_revisions` returns a [`~utils.DeleteCacheStrategy`] object that needs to
+        be executed. The [`~utils.DeleteCacheStrategy`] is not meant to be modified but
+        allows having a dry run before actually executing the deletion.
+
+        </Tip>
+        """
+        hashes_to_delete: Set[str] = set(revisions)
+
+        repos_with_revisions: Dict[CachedRepoInfo, Set[CachedRevisionInfo]] = defaultdict(set)
+
+        for repo in self.repos:
+            for revision in repo.revisions:
+                if revision.commit_hash in hashes_to_delete:
+                    repos_with_revisions[repo].add(revision)
+                    hashes_to_delete.remove(revision.commit_hash)
+
+        if len(hashes_to_delete) > 0:
+            logger.warning(f"Revision(s) not found - cannot delete them: {', '.join(hashes_to_delete)}")
+
+        delete_strategy_blobs: Set[Path] = set()
+        delete_strategy_refs: Set[Path] = set()
+        delete_strategy_repos: Set[Path] = set()
+        delete_strategy_snapshots: Set[Path] = set()
+        delete_strategy_expected_freed_size = 0
+
+        for affected_repo, revisions_to_delete in repos_with_revisions.items():
+            other_revisions = affected_repo.revisions - revisions_to_delete
+
+            # If no other revisions, it means all revisions are deleted
+            # -> delete the entire cached repo
+            if len(other_revisions) == 0:
+                delete_strategy_repos.add(affected_repo.repo_path)
+                delete_strategy_expected_freed_size += affected_repo.size_on_disk
+                continue
+
+            # Some revisions of the repo will be deleted but not all. We need to filter
+            # which blob files will not be linked anymore.
+            for revision_to_delete in revisions_to_delete:
+                # Snapshot dir
+                delete_strategy_snapshots.add(revision_to_delete.snapshot_path)
+
+                # Refs dir
+                for ref in revision_to_delete.refs:
+                    delete_strategy_refs.add(affected_repo.repo_path / "refs" / ref)
+
+                # Blobs dir
+                for file in revision_to_delete.files:
+                    if file.blob_path not in delete_strategy_blobs:
+                        is_file_alone = True
+                        for revision in other_revisions:
+                            for rev_file in revision.files:
+                                if file.blob_path == rev_file.blob_path:
+                                    is_file_alone = False
+                                    break
+                            if not is_file_alone:
+                                break
+
+                        # Blob file not referenced by remaining revisions -> delete
+                        if is_file_alone:
+                            delete_strategy_blobs.add(file.blob_path)
+                            delete_strategy_expected_freed_size += file.size_on_disk
+
+        # Return the strategy instead of executing it.
+        return DeleteCacheStrategy(
+            blobs=frozenset(delete_strategy_blobs),
+            refs=frozenset(delete_strategy_refs),
+            repos=frozenset(delete_strategy_repos),
+            snapshots=frozenset(delete_strategy_snapshots),
+            expected_freed_size=delete_strategy_expected_freed_size,
+        )
+
+    def export_as_table(self, *, verbosity: int = 0) -> str:
+        """Generate a table from the [`HFCacheInfo`] object.
+
+        Pass `verbosity=0` to get a table with a single row per repo, with columns
+        "repo_id", "repo_type", "size_on_disk", "nb_files", "last_accessed", "last_modified", "refs", "local_path".
+
+        Pass `verbosity=1` to get a table with a row per repo and revision (thus multiple rows can appear for a single repo), with columns
+        "repo_id", "repo_type", "revision", "size_on_disk", "nb_files", "last_modified", "refs", "local_path".
+
+        Example:
+        ```py
+        >>> from huggingface_hub.utils import scan_cache_dir
+
+        >>> hf_cache_info = scan_cache_dir()
+        HFCacheInfo(...)
+
+        >>> print(hf_cache_info.export_as_table())
+        REPO ID                                             REPO TYPE SIZE ON DISK NB FILES LAST_ACCESSED LAST_MODIFIED REFS LOCAL PATH
+        --------------------------------------------------- --------- ------------ -------- ------------- ------------- ---- --------------------------------------------------------------------------------------------------
+        roberta-base                                        model             2.7M        5 1 day ago     1 week ago    main ~/.cache/huggingface/hub/models--roberta-base
+        suno/bark                                           model             8.8K        1 1 week ago    1 week ago    main ~/.cache/huggingface/hub/models--suno--bark
+        t5-base                                             model           893.8M        4 4 days ago    7 months ago  main ~/.cache/huggingface/hub/models--t5-base
+        t5-large                                            model             3.0G        4 5 weeks ago   5 months ago  main ~/.cache/huggingface/hub/models--t5-large
+
+        >>> print(hf_cache_info.export_as_table(verbosity=1))
+        REPO ID                                             REPO TYPE REVISION                                 SIZE ON DISK NB FILES LAST_MODIFIED REFS LOCAL PATH
+        --------------------------------------------------- --------- ---------------------------------------- ------------ -------- ------------- ---- -----------------------------------------------------------------------------------------------------------------------------------------------------
+        roberta-base                                        model     e2da8e2f811d1448a5b465c236feacd80ffbac7b         2.7M        5 1 week ago    main ~/.cache/huggingface/hub/models--roberta-base/snapshots/e2da8e2f811d1448a5b465c236feacd80ffbac7b
+        suno/bark                                           model     70a8a7d34168586dc5d028fa9666aceade177992         8.8K        1 1 week ago    main ~/.cache/huggingface/hub/models--suno--bark/snapshots/70a8a7d34168586dc5d028fa9666aceade177992
+        t5-base                                             model     a9723ea7f1b39c1eae772870f3b547bf6ef7e6c1       893.8M        4 7 months ago  main ~/.cache/huggingface/hub/models--t5-base/snapshots/a9723ea7f1b39c1eae772870f3b547bf6ef7e6c1
+        t5-large                                            model     150ebc2c4b72291e770f58e6057481c8d2ed331a         3.0G        4 5 months ago  main ~/.cache/huggingface/hub/models--t5-large/snapshots/150ebc2c4b72291e770f58e6057481c8d2ed331a
+        ```
+
+        Args:
+            verbosity (`int`, *optional*):
+                The verbosity level. Defaults to 0.
+
+        Returns:
+            `str`: The table as a string.
+        """
+        if verbosity == 0:
+            return tabulate(
+                rows=[
+                    [
+                        repo.repo_id,
+                        repo.repo_type,
+                        "{:>12}".format(repo.size_on_disk_str),
+                        repo.nb_files,
+                        repo.last_accessed_str,
+                        repo.last_modified_str,
+                        ", ".join(sorted(repo.refs)),
+                        str(repo.repo_path),
+                    ]
+                    for repo in sorted(self.repos, key=lambda repo: repo.repo_path)
+                ],
+                headers=[
+                    "REPO ID",
+                    "REPO TYPE",
+                    "SIZE ON DISK",
+                    "NB FILES",
+                    "LAST_ACCESSED",
+                    "LAST_MODIFIED",
+                    "REFS",
+                    "LOCAL PATH",
+                ],
+            )
+        else:
+            return tabulate(
+                rows=[
+                    [
+                        repo.repo_id,
+                        repo.repo_type,
+                        revision.commit_hash,
+                        "{:>12}".format(revision.size_on_disk_str),
+                        revision.nb_files,
+                        revision.last_modified_str,
+                        ", ".join(sorted(revision.refs)),
+                        str(revision.snapshot_path),
+                    ]
+                    for repo in sorted(self.repos, key=lambda repo: repo.repo_path)
+                    for revision in sorted(repo.revisions, key=lambda revision: revision.commit_hash)
+                ],
+                headers=[
+                    "REPO ID",
+                    "REPO TYPE",
+                    "REVISION",
+                    "SIZE ON DISK",
+                    "NB FILES",
+                    "LAST_MODIFIED",
+                    "REFS",
+                    "LOCAL PATH",
+                ],
+            )
+
+
+def scan_cache_dir(cache_dir: Optional[Union[str, Path]] = None) -> HFCacheInfo:
+    """Scan the entire HF cache-system and return a [`~HFCacheInfo`] structure.
+
+    Use `scan_cache_dir` in order to programmatically scan your cache-system. The cache
+    will be scanned repo by repo. If a repo is corrupted, a [`~CorruptedCacheException`]
+    will be thrown internally but captured and returned in the [`~HFCacheInfo`]
+    structure. Only valid repos get a proper report.
+
+    ```py
+    >>> from huggingface_hub import scan_cache_dir
+
+    >>> hf_cache_info = scan_cache_dir()
+    HFCacheInfo(
+        size_on_disk=3398085269,
+        repos=frozenset({
+            CachedRepoInfo(
+                repo_id='t5-small',
+                repo_type='model',
+                repo_path=PosixPath(...),
+                size_on_disk=970726914,
+                nb_files=11,
+                revisions=frozenset({
+                    CachedRevisionInfo(
+                        commit_hash='d78aea13fa7ecd06c29e3e46195d6341255065d5',
+                        size_on_disk=970726339,
+                        snapshot_path=PosixPath(...),
+                        files=frozenset({
+                            CachedFileInfo(
+                                file_name='config.json',
+                                size_on_disk=1197
+                                file_path=PosixPath(...),
+                                blob_path=PosixPath(...),
+                            ),
+                            CachedFileInfo(...),
+                            ...
+                        }),
+                    ),
+                    CachedRevisionInfo(...),
+                    ...
+                }),
+            ),
+            CachedRepoInfo(...),
+            ...
+        }),
+        warnings=[
+            CorruptedCacheException("Snapshots dir doesn't exist in cached repo: ..."),
+            CorruptedCacheException(...),
+            ...
+        ],
+    )
+    ```
+
+    You can also print a detailed report directly from the `huggingface-cli` using:
+    ```text
+    > huggingface-cli scan-cache
+    REPO ID                     REPO TYPE SIZE ON DISK NB FILES REFS                LOCAL PATH
+    --------------------------- --------- ------------ -------- ------------------- -------------------------------------------------------------------------
+    glue                        dataset         116.3K       15 1.17.0, main, 2.4.0 /Users/lucain/.cache/huggingface/hub/datasets--glue
+    google/fleurs               dataset          64.9M        6 main, refs/pr/1     /Users/lucain/.cache/huggingface/hub/datasets--google--fleurs
+    Jean-Baptiste/camembert-ner model           441.0M        7 main                /Users/lucain/.cache/huggingface/hub/models--Jean-Baptiste--camembert-ner
+    bert-base-cased             model             1.9G       13 main                /Users/lucain/.cache/huggingface/hub/models--bert-base-cased
+    t5-base                     model            10.1K        3 main                /Users/lucain/.cache/huggingface/hub/models--t5-base
+    t5-small                    model           970.7M       11 refs/pr/1, main     /Users/lucain/.cache/huggingface/hub/models--t5-small
+
+    Done in 0.0s. Scanned 6 repo(s) for a total of 3.4G.
+    Got 1 warning(s) while scanning. Use -vvv to print details.
+    ```
+
+    Args:
+        cache_dir (`str` or `Path`, `optional`):
+            Cache directory to cache. Defaults to the default HF cache directory.
+
+    <Tip warning={true}>
+
+    Raises:
+
+        `CacheNotFound`
+          If the cache directory does not exist.
+
+        [`ValueError`](https://docs.python.org/3/library/exceptions.html#ValueError)
+          If the cache directory is a file, instead of a directory.
+
+    </Tip>
+
+    Returns: a [`~HFCacheInfo`] object.
+    """
+    if cache_dir is None:
+        cache_dir = HF_HUB_CACHE
+
+    cache_dir = Path(cache_dir).expanduser().resolve()
+    if not cache_dir.exists():
+        raise CacheNotFound(
+            f"Cache directory not found: {cache_dir}. Please use `cache_dir` argument or set `HF_HUB_CACHE` environment variable.",
+            cache_dir=cache_dir,
+        )
+
+    if cache_dir.is_file():
+        raise ValueError(
+            f"Scan cache expects a directory but found a file: {cache_dir}. Please use `cache_dir` argument or set `HF_HUB_CACHE` environment variable."
+        )
+
+    repos: Set[CachedRepoInfo] = set()
+    warnings: List[CorruptedCacheException] = []
+    for repo_path in cache_dir.iterdir():
+        if repo_path.name == ".locks":  # skip './.locks/' folder
+            continue
+        try:
+            repos.add(_scan_cached_repo(repo_path))
+        except CorruptedCacheException as e:
+            warnings.append(e)
+
+    return HFCacheInfo(
+        repos=frozenset(repos),
+        size_on_disk=sum(repo.size_on_disk for repo in repos),
+        warnings=warnings,
+    )
+
+
+def _scan_cached_repo(repo_path: Path) -> CachedRepoInfo:
+    """Scan a single cache repo and return information about it.
+
+    Any unexpected behavior will raise a [`~CorruptedCacheException`].
+    """
+    if not repo_path.is_dir():
+        raise CorruptedCacheException(f"Repo path is not a directory: {repo_path}")
+
+    if "--" not in repo_path.name:
+        raise CorruptedCacheException(f"Repo path is not a valid HuggingFace cache directory: {repo_path}")
+
+    repo_type, repo_id = repo_path.name.split("--", maxsplit=1)
+    repo_type = repo_type[:-1]  # "models" -> "model"
+    repo_id = repo_id.replace("--", "/")  # google/fleurs -> "google/fleurs"
+
+    if repo_type not in {"dataset", "model", "space"}:
+        raise CorruptedCacheException(
+            f"Repo type must be `dataset`, `model` or `space`, found `{repo_type}` ({repo_path})."
+        )
+
+    blob_stats: Dict[Path, os.stat_result] = {}  # Key is blob_path, value is blob stats
+
+    snapshots_path = repo_path / "snapshots"
+    refs_path = repo_path / "refs"
+
+    if not snapshots_path.exists() or not snapshots_path.is_dir():
+        raise CorruptedCacheException(f"Snapshots dir doesn't exist in cached repo: {snapshots_path}")
+
+    # Scan over `refs` directory
+
+    # key is revision hash, value is set of refs
+    refs_by_hash: Dict[str, Set[str]] = defaultdict(set)
+    if refs_path.exists():
+        # Example of `refs` directory
+        # ── refs
+        #     ├── main
+        #     └── refs
+        #         └── pr
+        #             └── 1
+        if refs_path.is_file():
+            raise CorruptedCacheException(f"Refs directory cannot be a file: {refs_path}")
+
+        for ref_path in refs_path.glob("**/*"):
+            # glob("**/*") iterates over all files and directories -> skip directories
+            if ref_path.is_dir() or ref_path.name in FILES_TO_IGNORE:
+                continue
+
+            ref_name = str(ref_path.relative_to(refs_path))
+            with ref_path.open() as f:
+                commit_hash = f.read()
+
+            refs_by_hash[commit_hash].add(ref_name)
+
+    # Scan snapshots directory
+    cached_revisions: Set[CachedRevisionInfo] = set()
+    for revision_path in snapshots_path.iterdir():
+        # Ignore OS-created helper files
+        if revision_path.name in FILES_TO_IGNORE:
+            continue
+        if revision_path.is_file():
+            raise CorruptedCacheException(f"Snapshots folder corrupted. Found a file: {revision_path}")
+
+        cached_files = set()
+        for file_path in revision_path.glob("**/*"):
+            # glob("**/*") iterates over all files and directories -> skip directories
+            if file_path.is_dir():
+                continue
+
+            blob_path = Path(file_path).resolve()
+            if not blob_path.exists():
+                raise CorruptedCacheException(f"Blob missing (broken symlink): {blob_path}")
+
+            if blob_path not in blob_stats:
+                blob_stats[blob_path] = blob_path.stat()
+
+            cached_files.add(
+                CachedFileInfo(
+                    file_name=file_path.name,
+                    file_path=file_path,
+                    size_on_disk=blob_stats[blob_path].st_size,
+                    blob_path=blob_path,
+                    blob_last_accessed=blob_stats[blob_path].st_atime,
+                    blob_last_modified=blob_stats[blob_path].st_mtime,
+                )
+            )
+
+        # Last modified is either the last modified blob file or the revision folder
+        # itself if it is empty
+        if len(cached_files) > 0:
+            revision_last_modified = max(blob_stats[file.blob_path].st_mtime for file in cached_files)
+        else:
+            revision_last_modified = revision_path.stat().st_mtime
+
+        cached_revisions.add(
+            CachedRevisionInfo(
+                commit_hash=revision_path.name,
+                files=frozenset(cached_files),
+                refs=frozenset(refs_by_hash.pop(revision_path.name, set())),
+                size_on_disk=sum(
+                    blob_stats[blob_path].st_size for blob_path in set(file.blob_path for file in cached_files)
+                ),
+                snapshot_path=revision_path,
+                last_modified=revision_last_modified,
+            )
+        )
+
+    # Check that all refs referred to an existing revision
+    if len(refs_by_hash) > 0:
+        raise CorruptedCacheException(
+            f"Reference(s) refer to missing commit hashes: {dict(refs_by_hash)} ({repo_path})."
+        )
+
+    # Last modified is either the last modified blob file or the repo folder itself if
+    # no blob files has been found. Same for last accessed.
+    if len(blob_stats) > 0:
+        repo_last_accessed = max(stat.st_atime for stat in blob_stats.values())
+        repo_last_modified = max(stat.st_mtime for stat in blob_stats.values())
+    else:
+        repo_stats = repo_path.stat()
+        repo_last_accessed = repo_stats.st_atime
+        repo_last_modified = repo_stats.st_mtime
+
+    # Build and return frozen structure
+    return CachedRepoInfo(
+        nb_files=len(blob_stats),
+        repo_id=repo_id,
+        repo_path=repo_path,
+        repo_type=repo_type,  # type: ignore
+        revisions=frozenset(cached_revisions),
+        size_on_disk=sum(stat.st_size for stat in blob_stats.values()),
+        last_accessed=repo_last_accessed,
+        last_modified=repo_last_modified,
+    )
+
+
+def _format_size(num: int) -> str:
+    """Format size in bytes into a human-readable string.
+
+    Taken from https://stackoverflow.com/a/1094933
+    """
+    num_f = float(num)
+    for unit in ["", "K", "M", "G", "T", "P", "E", "Z"]:
+        if abs(num_f) < 1000.0:
+            return f"{num_f:3.1f}{unit}"
+        num_f /= 1000.0
+    return f"{num_f:.1f}Y"
+
+
+_TIMESINCE_CHUNKS = (
+    # Label, divider, max value
+    ("second", 1, 60),
+    ("minute", 60, 60),
+    ("hour", 60 * 60, 24),
+    ("day", 60 * 60 * 24, 6),
+    ("week", 60 * 60 * 24 * 7, 6),
+    ("month", 60 * 60 * 24 * 30, 11),
+    ("year", 60 * 60 * 24 * 365, None),
+)
+
+
+def _format_timesince(ts: float) -> str:
+    """Format timestamp in seconds into a human-readable string, relative to now.
+
+    Vaguely inspired by Django's `timesince` formatter.
+    """
+    delta = time.time() - ts
+    if delta < 20:
+        return "a few seconds ago"
+    for label, divider, max_value in _TIMESINCE_CHUNKS:  # noqa: B007
+        value = round(delta / divider)
+        if max_value is not None and value <= max_value:
+            break
+    return f"{value} {label}{'s' if value > 1 else ''} ago"
+
+
+def _try_delete_path(path: Path, path_type: str) -> None:
+    """Try to delete a local file or folder.
+
+    If the path does not exists, error is logged as a warning and then ignored.
+
+    Args:
+        path (`Path`)
+            Path to delete. Can be a file or a folder.
+        path_type (`str`)
+            What path are we deleting ? Only for logging purposes. Example: "snapshot".
+    """
+    logger.info(f"Delete {path_type}: {path}")
+    try:
+        if path.is_file():
+            os.remove(path)
+        else:
+            shutil.rmtree(path)
+    except FileNotFoundError:
+        logger.warning(f"Couldn't delete {path_type}: file not found ({path})", exc_info=True)
+    except PermissionError:
+        logger.warning(f"Couldn't delete {path_type}: permission denied ({path})", exc_info=True)

venv/lib/python3.12/site-packages/huggingface_hub/utils/_datetime.py

@@ -0,0 +1,67 @@
+# coding=utf-8
+# Copyright 2022-present, the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Contains utilities to handle datetimes in Huggingface Hub."""
+
+from datetime import datetime, timezone
+
+
+def parse_datetime(date_string: str) -> datetime:
+    """
+    Parses a date_string returned from the server to a datetime object.
+
+    This parser is a weak-parser is the sense that it handles only a single format of
+    date_string. It is expected that the server format will never change. The
+    implementation depends only on the standard lib to avoid an external dependency
+    (python-dateutil). See full discussion about this decision on PR:
+    https://github.com/huggingface/huggingface_hub/pull/999.
+
+    Example:
+        ```py
+        > parse_datetime('2022-08-19T07:19:38.123Z')
+        datetime.datetime(2022, 8, 19, 7, 19, 38, 123000, tzinfo=timezone.utc)
+        ```
+
+    Args:
+        date_string (`str`):
+            A string representing a datetime returned by the Hub server.
+            String is expected to follow '%Y-%m-%dT%H:%M:%S.%fZ' pattern.
+
+    Returns:
+        A python datetime object.
+
+    Raises:
+        :class:`ValueError`:
+            If `date_string` cannot be parsed.
+    """
+    try:
+        # Normalize the string to always have 6 digits of fractional seconds
+        if date_string.endswith("Z"):
+            # Case 1: No decimal point (e.g., "2024-11-16T00:27:02Z")
+            if "." not in date_string:
+                # No fractional seconds - insert .000000
+                date_string = date_string[:-1] + ".000000Z"
+            # Case 2: Has decimal point (e.g., "2022-08-19T07:19:38.123456789Z")
+            else:
+                # Get the fractional and base parts
+                base, fraction = date_string[:-1].split(".")
+                # fraction[:6] takes first 6 digits and :0<6 pads with zeros if less than 6 digits
+                date_string = f"{base}.{fraction[:6]:0<6}Z"
+
+        return datetime.strptime(date_string, "%Y-%m-%dT%H:%M:%S.%fZ").replace(tzinfo=timezone.utc)
+    except ValueError as e:
+        raise ValueError(
+            f"Cannot parse '{date_string}' as a datetime. Date string is expected to"
+            " follow '%Y-%m-%dT%H:%M:%S.%fZ' pattern."
+        ) from e

venv/lib/python3.12/site-packages/huggingface_hub/utils/_deprecation.py

@@ -0,0 +1,136 @@
+import warnings
+from functools import wraps
+from inspect import Parameter, signature
+from typing import Iterable, Optional
+
+
+def _deprecate_positional_args(*, version: str):
+    """Decorator for methods that issues warnings for positional arguments.
+    Using the keyword-only argument syntax in pep 3102, arguments after the
+    * will issue a warning when passed as a positional argument.
+
+    Args:
+        version (`str`):
+            The version when positional arguments will result in error.
+    """
+
+    def _inner_deprecate_positional_args(f):
+        sig = signature(f)
+        kwonly_args = []
+        all_args = []
+        for name, param in sig.parameters.items():
+            if param.kind == Parameter.POSITIONAL_OR_KEYWORD:
+                all_args.append(name)
+            elif param.kind == Parameter.KEYWORD_ONLY:
+                kwonly_args.append(name)
+
+        @wraps(f)
+        def inner_f(*args, **kwargs):
+            extra_args = len(args) - len(all_args)
+            if extra_args <= 0:
+                return f(*args, **kwargs)
+            # extra_args > 0
+            args_msg = [
+                f"{name}='{arg}'" if isinstance(arg, str) else f"{name}={arg}"
+                for name, arg in zip(kwonly_args[:extra_args], args[-extra_args:])
+            ]
+            args_msg = ", ".join(args_msg)
+            warnings.warn(
+                f"Deprecated positional argument(s) used in '{f.__name__}': pass"
+                f" {args_msg} as keyword args. From version {version} passing these"
+                " as positional arguments will result in an error,",
+                FutureWarning,
+            )
+            kwargs.update(zip(sig.parameters, args))
+            return f(**kwargs)
+
+        return inner_f
+
+    return _inner_deprecate_positional_args
+
+
+def _deprecate_arguments(
+    *,
+    version: str,
+    deprecated_args: Iterable[str],
+    custom_message: Optional[str] = None,
+):
+    """Decorator to issue warnings when using deprecated arguments.
+
+    TODO: could be useful to be able to set a custom error message.
+
+    Args:
+        version (`str`):
+            The version when deprecated arguments will result in error.
+        deprecated_args (`List[str]`):
+            List of the arguments to be deprecated.
+        custom_message (`str`, *optional*):
+            Warning message that is raised. If not passed, a default warning message
+            will be created.
+    """
+
+    def _inner_deprecate_positional_args(f):
+        sig = signature(f)
+
+        @wraps(f)
+        def inner_f(*args, **kwargs):
+            # Check for used deprecated arguments
+            used_deprecated_args = []
+            for _, parameter in zip(args, sig.parameters.values()):
+                if parameter.name in deprecated_args:
+                    used_deprecated_args.append(parameter.name)
+            for kwarg_name, kwarg_value in kwargs.items():
+                if (
+                    # If argument is deprecated but still used
+                    kwarg_name in deprecated_args
+                    # And then the value is not the default value
+                    and kwarg_value != sig.parameters[kwarg_name].default
+                ):
+                    used_deprecated_args.append(kwarg_name)
+
+            # Warn and proceed
+            if len(used_deprecated_args) > 0:
+                message = (
+                    f"Deprecated argument(s) used in '{f.__name__}':"
+                    f" {', '.join(used_deprecated_args)}. Will not be supported from"
+                    f" version '{version}'."
+                )
+                if custom_message is not None:
+                    message += "\n\n" + custom_message
+                warnings.warn(message, FutureWarning)
+            return f(*args, **kwargs)
+
+        return inner_f
+
+    return _inner_deprecate_positional_args
+
+
+def _deprecate_method(*, version: str, message: Optional[str] = None):
+    """Decorator to issue warnings when using a deprecated method.
+
+    Args:
+        version (`str`):
+            The version when deprecated arguments will result in error.
+        message (`str`, *optional*):
+            Warning message that is raised. If not passed, a default warning message
+            will be created.
+    """
+
+    def _inner_deprecate_method(f):
+        name = f.__name__
+        if name == "__init__":
+            name = f.__qualname__.split(".")[0]  # class name instead of method name
+
+        @wraps(f)
+        def inner_f(*args, **kwargs):
+            warning_message = (
+                f"'{name}' (from '{f.__module__}') is deprecated and will be removed from version '{version}'."
+            )
+            if message is not None:
+                warning_message += " " + message
+            warnings.warn(warning_message, FutureWarning)
+            return f(*args, **kwargs)
+
+        return inner_f
+
+    return _inner_deprecate_method

venv/lib/python3.12/site-packages/huggingface_hub/utils/_experimental.py

@@ -0,0 +1,68 @@
+# coding=utf-8
+# Copyright 2023-present, the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Contains utilities to flag a feature as "experimental" in Huggingface Hub."""
+
+import warnings
+from functools import wraps
+from typing import Callable
+
+from .. import constants
+
+
+def experimental(fn: Callable) -> Callable:
+    """Decorator to flag a feature as experimental.
+
+    An experimental feature triggers a warning when used as it might be subject to breaking changes without prior notice
+    in the future.
+
+    Warnings can be disabled by setting `HF_HUB_DISABLE_EXPERIMENTAL_WARNING=1` as environment variable.
+
+    Args:
+        fn (`Callable`):
+            The function to flag as experimental.
+
+    Returns:
+        `Callable`: The decorated function.
+
+    Example:
+
+    ```python
+    >>> from huggingface_hub.utils import experimental
+
+    >>> @experimental
+    ... def my_function():
+    ...     print("Hello world!")
+
+    >>> my_function()
+    UserWarning: 'my_function' is experimental and might be subject to breaking changes in the future without prior
+    notice. You can disable this warning by setting `HF_HUB_DISABLE_EXPERIMENTAL_WARNING=1` as environment variable.
+    Hello world!
+    ```
+    """
+    # For classes, put the "experimental" around the "__new__" method => __new__ will be removed in warning message
+    name = fn.__qualname__[: -len(".__new__")] if fn.__qualname__.endswith(".__new__") else fn.__qualname__
+
+    @wraps(fn)
+    def _inner_fn(*args, **kwargs):
+        if not constants.HF_HUB_DISABLE_EXPERIMENTAL_WARNING:
+            warnings.warn(
+                f"'{name}' is experimental and might be subject to breaking changes in the future without prior notice."
+                " You can disable this warning by setting `HF_HUB_DISABLE_EXPERIMENTAL_WARNING=1` as environment"
+                " variable.",
+                UserWarning,
+            )
+        return fn(*args, **kwargs)
+
+    return _inner_fn

venv/lib/python3.12/site-packages/huggingface_hub/utils/_fixes.py

@@ -0,0 +1,133 @@
+# JSONDecodeError was introduced in requests=2.27 released in 2022.
+# This allows us to support older requests for users
+# More information: https://github.com/psf/requests/pull/5856
+try:
+    from requests import JSONDecodeError  # type: ignore  # noqa: F401
+except ImportError:
+    try:
+        from simplejson import JSONDecodeError  # type: ignore # noqa: F401
+    except ImportError:
+        from json import JSONDecodeError  # type: ignore  # noqa: F401
+import contextlib
+import os
+import shutil
+import stat
+import tempfile
+import time
+from functools import partial
+from pathlib import Path
+from typing import Callable, Generator, Optional, Union
+
+import yaml
+from filelock import BaseFileLock, FileLock, SoftFileLock, Timeout
+
+from .. import constants
+from . import logging
+
+
+logger = logging.get_logger(__name__)
+
+# Wrap `yaml.dump` to set `allow_unicode=True` by default.
+#
+# Example:
+# ```py
+# >>> yaml.dump({"emoji": "👀", "some unicode": "日本か"})
+# 'emoji: "\\U0001F440"\nsome unicode: "\\u65E5\\u672C\\u304B"\n'
+#
+# >>> yaml_dump({"emoji": "👀", "some unicode": "日本か"})
+# 'emoji: "👀"\nsome unicode: "日本か"\n'
+# ```
+yaml_dump: Callable[..., str] = partial(yaml.dump, stream=None, allow_unicode=True)  # type: ignore
+
+
+@contextlib.contextmanager
+def SoftTemporaryDirectory(
+    suffix: Optional[str] = None,
+    prefix: Optional[str] = None,
+    dir: Optional[Union[Path, str]] = None,
+    **kwargs,
+) -> Generator[Path, None, None]:
+    """
+    Context manager to create a temporary directory and safely delete it.
+
+    If tmp directory cannot be deleted normally, we set the WRITE permission and retry.
+    If cleanup still fails, we give up but don't raise an exception. This is equivalent
+    to  `tempfile.TemporaryDirectory(..., ignore_cleanup_errors=True)` introduced in
+    Python 3.10.
+
+    See https://www.scivision.dev/python-tempfile-permission-error-windows/.
+    """
+    tmpdir = tempfile.TemporaryDirectory(prefix=prefix, suffix=suffix, dir=dir, **kwargs)
+    yield Path(tmpdir.name).resolve()
+
+    try:
+        # First once with normal cleanup
+        shutil.rmtree(tmpdir.name)
+    except Exception:
+        # If failed, try to set write permission and retry
+        try:
+            shutil.rmtree(tmpdir.name, onerror=_set_write_permission_and_retry)
+        except Exception:
+            pass
+
+    # And finally, cleanup the tmpdir.
+    # If it fails again, give up but do not throw error
+    try:
+        tmpdir.cleanup()
+    except Exception:
+        pass
+
+
+def _set_write_permission_and_retry(func, path, excinfo):
+    os.chmod(path, stat.S_IWRITE)
+    func(path)
+
+
+@contextlib.contextmanager
+def WeakFileLock(
+    lock_file: Union[str, Path], *, timeout: Optional[float] = None
+) -> Generator[BaseFileLock, None, None]:
+    """A filelock with some custom logic.
+
+    This filelock is weaker than the default filelock in that:
+    1. It won't raise an exception if release fails.
+    2. It will default to a SoftFileLock if the filesystem does not support flock.
+
+    An INFO log message is emitted every 10 seconds if the lock is not acquired immediately.
+    If a timeout is provided, a `filelock.Timeout` exception is raised if the lock is not acquired within the timeout.
+    """
+    log_interval = constants.FILELOCK_LOG_EVERY_SECONDS
+    lock = FileLock(lock_file, timeout=log_interval)
+    start_time = time.time()
+
+    while True:
+        elapsed_time = time.time() - start_time
+        if timeout is not None and elapsed_time >= timeout:
+            raise Timeout(str(lock_file))
+
+        try:
+            lock.acquire(timeout=min(log_interval, timeout - elapsed_time) if timeout else log_interval)
+        except Timeout:
+            logger.info(
+                f"Still waiting to acquire lock on {lock_file} (elapsed: {time.time() - start_time:.1f} seconds)"
+            )
+        except NotImplementedError as e:
+            if "use SoftFileLock instead" in str(e):
+                logger.warning(
+                    "FileSystem does not appear to support flock. Falling back to SoftFileLock for %s", lock_file
+                )
+                lock = SoftFileLock(lock_file, timeout=log_interval)
+                continue
+        else:
+            break
+
+    try:
+        yield lock
+    finally:
+        try:
+            lock.release()
+        except OSError:
+            try:
+                Path(lock_file).unlink()
+            except OSError:
+                pass

venv/lib/python3.12/site-packages/huggingface_hub/utils/_http.py

@@ -0,0 +1,637 @@
+# coding=utf-8
+# Copyright 2022-present, the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Contains utilities to handle HTTP requests in Huggingface Hub."""
+
+import io
+import os
+import re
+import threading
+import time
+import uuid
+from functools import lru_cache
+from http import HTTPStatus
+from shlex import quote
+from typing import Any, Callable, List, Optional, Tuple, Type, Union
+
+import requests
+from requests import HTTPError, Response
+from requests.adapters import HTTPAdapter
+from requests.models import PreparedRequest
+
+from huggingface_hub.errors import OfflineModeIsEnabled
+
+from .. import constants
+from ..errors import (
+    BadRequestError,
+    DisabledRepoError,
+    EntryNotFoundError,
+    GatedRepoError,
+    HfHubHTTPError,
+    RepositoryNotFoundError,
+    RevisionNotFoundError,
+)
+from . import logging
+from ._fixes import JSONDecodeError
+from ._lfs import SliceFileObj
+from ._typing import HTTP_METHOD_T
+
+
+logger = logging.get_logger(__name__)
+
+# Both headers are used by the Hub to debug failed requests.
+# `X_AMZN_TRACE_ID` is better as it also works to debug on Cloudfront and ALB.
+# If `X_AMZN_TRACE_ID` is set, the Hub will use it as well.
+X_AMZN_TRACE_ID = "X-Amzn-Trace-Id"
+X_REQUEST_ID = "x-request-id"
+
+REPO_API_REGEX = re.compile(
+    r"""
+        # staging or production endpoint
+        ^https://[^/]+
+        (
+            # on /api/repo_type/repo_id
+            /api/(models|datasets|spaces)/(.+)
+            |
+            # or /repo_id/resolve/revision/...
+            /(.+)/resolve/(.+)
+        )
+    """,
+    flags=re.VERBOSE,
+)
+
+
+class UniqueRequestIdAdapter(HTTPAdapter):
+    X_AMZN_TRACE_ID = "X-Amzn-Trace-Id"
+
+    def add_headers(self, request, **kwargs):
+        super().add_headers(request, **kwargs)
+
+        # Add random request ID => easier for server-side debug
+        if X_AMZN_TRACE_ID not in request.headers:
+            request.headers[X_AMZN_TRACE_ID] = request.headers.get(X_REQUEST_ID) or str(uuid.uuid4())
+
+        # Add debug log
+        has_token = len(str(request.headers.get("authorization", ""))) > 0
+        logger.debug(
+            f"Request {request.headers[X_AMZN_TRACE_ID]}: {request.method} {request.url} (authenticated: {has_token})"
+        )
+
+    def send(self, request: PreparedRequest, *args, **kwargs) -> Response:
+        """Catch any RequestException to append request id to the error message for debugging."""
+        if constants.HF_DEBUG:
+            logger.debug(f"Send: {_curlify(request)}")
+        try:
+            return super().send(request, *args, **kwargs)
+        except requests.RequestException as e:
+            request_id = request.headers.get(X_AMZN_TRACE_ID)
+            if request_id is not None:
+                # Taken from https://stackoverflow.com/a/58270258
+                e.args = (*e.args, f"(Request ID: {request_id})")
+            raise
+
+
+class OfflineAdapter(HTTPAdapter):
+    def send(self, request: PreparedRequest, *args, **kwargs) -> Response:
+        raise OfflineModeIsEnabled(
+            f"Cannot reach {request.url}: offline mode is enabled. To disable it, please unset the `HF_HUB_OFFLINE` environment variable."
+        )
+
+
+def _default_backend_factory() -> requests.Session:
+    session = requests.Session()
+    if constants.HF_HUB_OFFLINE:
+        session.mount("http://", OfflineAdapter())
+        session.mount("https://", OfflineAdapter())
+    else:
+        session.mount("http://", UniqueRequestIdAdapter())
+        session.mount("https://", UniqueRequestIdAdapter())
+    return session
+
+
+BACKEND_FACTORY_T = Callable[[], requests.Session]
+_GLOBAL_BACKEND_FACTORY: BACKEND_FACTORY_T = _default_backend_factory
+
+
+def configure_http_backend(backend_factory: BACKEND_FACTORY_T = _default_backend_factory) -> None:
+    """
+    Configure the HTTP backend by providing a `backend_factory`. Any HTTP calls made by `huggingface_hub` will use a
+    Session object instantiated by this factory. This can be useful if you are running your scripts in a specific
+    environment requiring custom configuration (e.g. custom proxy or certifications).
+
+    Use [`get_session`] to get a configured Session. Since `requests.Session` is not guaranteed to be thread-safe,
+    `huggingface_hub` creates 1 Session instance per thread. They are all instantiated using the same `backend_factory`
+    set in [`configure_http_backend`]. A LRU cache is used to cache the created sessions (and connections) between
+    calls. Max size is 128 to avoid memory leaks if thousands of threads are spawned.
+
+    See [this issue](https://github.com/psf/requests/issues/2766) to know more about thread-safety in `requests`.
+
+    Example:
+    ```py
+    import requests
+    from huggingface_hub import configure_http_backend, get_session
+
+    # Create a factory function that returns a Session with configured proxies
+    def backend_factory() -> requests.Session:
+        session = requests.Session()
+        session.proxies = {"http": "http://10.10.1.10:3128", "https": "https://10.10.1.11:1080"}
+        return session
+
+    # Set it as the default session factory
+    configure_http_backend(backend_factory=backend_factory)
+
+    # In practice, this is mostly done internally in `huggingface_hub`
+    session = get_session()
+    ```
+    """
+    global _GLOBAL_BACKEND_FACTORY
+    _GLOBAL_BACKEND_FACTORY = backend_factory
+    reset_sessions()
+
+
+def get_session() -> requests.Session:
+    """
+    Get a `requests.Session` object, using the session factory from the user.
+
+    Use [`get_session`] to get a configured Session. Since `requests.Session` is not guaranteed to be thread-safe,
+    `huggingface_hub` creates 1 Session instance per thread. They are all instantiated using the same `backend_factory`
+    set in [`configure_http_backend`]. A LRU cache is used to cache the created sessions (and connections) between
+    calls. Max size is 128 to avoid memory leaks if thousands of threads are spawned.
+
+    See [this issue](https://github.com/psf/requests/issues/2766) to know more about thread-safety in `requests`.
+
+    Example:
+    ```py
+    import requests
+    from huggingface_hub import configure_http_backend, get_session
+
+    # Create a factory function that returns a Session with configured proxies
+    def backend_factory() -> requests.Session:
+        session = requests.Session()
+        session.proxies = {"http": "http://10.10.1.10:3128", "https": "https://10.10.1.11:1080"}
+        return session
+
+    # Set it as the default session factory
+    configure_http_backend(backend_factory=backend_factory)
+
+    # In practice, this is mostly done internally in `huggingface_hub`
+    session = get_session()
+    ```
+    """
+    return _get_session_from_cache(process_id=os.getpid(), thread_id=threading.get_ident())
+
+
+def reset_sessions() -> None:
+    """Reset the cache of sessions.
+
+    Mostly used internally when sessions are reconfigured or an SSLError is raised.
+    See [`configure_http_backend`] for more details.
+    """
+    _get_session_from_cache.cache_clear()
+
+
+@lru_cache
+def _get_session_from_cache(process_id: int, thread_id: int) -> requests.Session:
+    """
+    Create a new session per thread using global factory. Using LRU cache (maxsize 128) to avoid memory leaks when
+    using thousands of threads. Cache is cleared when `configure_http_backend` is called.
+    """
+    return _GLOBAL_BACKEND_FACTORY()
+
+
+def http_backoff(
+    method: HTTP_METHOD_T,
+    url: str,
+    *,
+    max_retries: int = 5,
+    base_wait_time: float = 1,
+    max_wait_time: float = 8,
+    retry_on_exceptions: Union[Type[Exception], Tuple[Type[Exception], ...]] = (
+        requests.Timeout,
+        requests.ConnectionError,
+    ),
+    retry_on_status_codes: Union[int, Tuple[int, ...]] = HTTPStatus.SERVICE_UNAVAILABLE,
+    **kwargs,
+) -> Response:
+    """Wrapper around requests to retry calls on an endpoint, with exponential backoff.
+
+    Endpoint call is retried on exceptions (ex: connection timeout, proxy error,...)
+    and/or on specific status codes (ex: service unavailable). If the call failed more
+    than `max_retries`, the exception is thrown or `raise_for_status` is called on the
+    response object.
+
+    Re-implement mechanisms from the `backoff` library to avoid adding an external
+    dependencies to `hugging_face_hub`. See https://github.com/litl/backoff.
+
+    Args:
+        method (`Literal["GET", "OPTIONS", "HEAD", "POST", "PUT", "PATCH", "DELETE"]`):
+            HTTP method to perform.
+        url (`str`):
+            The URL of the resource to fetch.
+        max_retries (`int`, *optional*, defaults to `5`):
+            Maximum number of retries, defaults to 5 (no retries).
+        base_wait_time (`float`, *optional*, defaults to `1`):
+            Duration (in seconds) to wait before retrying the first time.
+            Wait time between retries then grows exponentially, capped by
+            `max_wait_time`.
+        max_wait_time (`float`, *optional*, defaults to `8`):
+            Maximum duration (in seconds) to wait before retrying.
+        retry_on_exceptions (`Type[Exception]` or `Tuple[Type[Exception]]`, *optional*):
+            Define which exceptions must be caught to retry the request. Can be a single type or a tuple of types.
+            By default, retry on `requests.Timeout` and `requests.ConnectionError`.
+        retry_on_status_codes (`int` or `Tuple[int]`, *optional*, defaults to `503`):
+            Define on which status codes the request must be retried. By default, only
+            HTTP 503 Service Unavailable is retried.
+        **kwargs (`dict`, *optional*):
+            kwargs to pass to `requests.request`.
+
+    Example:
+    ```
+    >>> from huggingface_hub.utils import http_backoff
+
+    # Same usage as "requests.request".
+    >>> response = http_backoff("GET", "https://www.google.com")
+    >>> response.raise_for_status()
+
+    # If you expect a Gateway Timeout from time to time
+    >>> http_backoff("PUT", upload_url, data=data, retry_on_status_codes=504)
+    >>> response.raise_for_status()
+    ```
+
+    <Tip warning={true}>
+
+    When using `requests` it is possible to stream data by passing an iterator to the
+    `data` argument. On http backoff this is a problem as the iterator is not reset
+    after a failed call. This issue is mitigated for file objects or any IO streams
+    by saving the initial position of the cursor (with `data.tell()`) and resetting the
+    cursor between each call (with `data.seek()`). For arbitrary iterators, http backoff
+    will fail. If this is a hard constraint for you, please let us know by opening an
+    issue on [Github](https://github.com/huggingface/huggingface_hub).
+
+    </Tip>
+    """
+    if isinstance(retry_on_exceptions, type):  # Tuple from single exception type
+        retry_on_exceptions = (retry_on_exceptions,)
+
+    if isinstance(retry_on_status_codes, int):  # Tuple from single status code
+        retry_on_status_codes = (retry_on_status_codes,)
+
+    nb_tries = 0
+    sleep_time = base_wait_time
+
+    # If `data` is used and is a file object (or any IO), it will be consumed on the
+    # first HTTP request. We need to save the initial position so that the full content
+    # of the file is re-sent on http backoff. See warning tip in docstring.
+    io_obj_initial_pos = None
+    if "data" in kwargs and isinstance(kwargs["data"], (io.IOBase, SliceFileObj)):
+        io_obj_initial_pos = kwargs["data"].tell()
+
+    session = get_session()
+    while True:
+        nb_tries += 1
+        try:
+            # If `data` is used and is a file object (or any IO), set back cursor to
+            # initial position.
+            if io_obj_initial_pos is not None:
+                kwargs["data"].seek(io_obj_initial_pos)
+
+            # Perform request and return if status_code is not in the retry list.
+            response = session.request(method=method, url=url, **kwargs)
+            if response.status_code not in retry_on_status_codes:
+                return response
+
+            # Wrong status code returned (HTTP 503 for instance)
+            logger.warning(f"HTTP Error {response.status_code} thrown while requesting {method} {url}")
+            if nb_tries > max_retries:
+                response.raise_for_status()  # Will raise uncaught exception
+                # We return response to avoid infinite loop in the corner case where the
+                # user ask for retry on a status code that doesn't raise_for_status.
+                return response
+
+        except retry_on_exceptions as err:
+            logger.warning(f"'{err}' thrown while requesting {method} {url}")
+
+            if isinstance(err, requests.ConnectionError):
+                reset_sessions()  # In case of SSLError it's best to reset the shared requests.Session objects
+
+            if nb_tries > max_retries:
+                raise err
+
+        # Sleep for X seconds
+        logger.warning(f"Retrying in {sleep_time}s [Retry {nb_tries}/{max_retries}].")
+        time.sleep(sleep_time)
+
+        # Update sleep time for next retry
+        sleep_time = min(max_wait_time, sleep_time * 2)  # Exponential backoff
+
+
+def fix_hf_endpoint_in_url(url: str, endpoint: Optional[str]) -> str:
+    """Replace the default endpoint in a URL by a custom one.
+
+    This is useful when using a proxy and the Hugging Face Hub returns a URL with the default endpoint.
+    """
+    endpoint = endpoint.rstrip("/") if endpoint else constants.ENDPOINT
+    # check if a proxy has been set => if yes, update the returned URL to use the proxy
+    if endpoint not in (constants._HF_DEFAULT_ENDPOINT, constants._HF_DEFAULT_STAGING_ENDPOINT):
+        url = url.replace(constants._HF_DEFAULT_ENDPOINT, endpoint)
+        url = url.replace(constants._HF_DEFAULT_STAGING_ENDPOINT, endpoint)
+    return url
+
+
+def hf_raise_for_status(response: Response, endpoint_name: Optional[str] = None) -> None:
+    """
+    Internal version of `response.raise_for_status()` that will refine a
+    potential HTTPError. Raised exception will be an instance of `HfHubHTTPError`.
+
+    This helper is meant to be the unique method to raise_for_status when making a call
+    to the Hugging Face Hub.
+
+
+    Example:
+    ```py
+        import requests
+        from huggingface_hub.utils import get_session, hf_raise_for_status, HfHubHTTPError
+
+        response = get_session().post(...)
+        try:
+            hf_raise_for_status(response)
+        except HfHubHTTPError as e:
+            print(str(e)) # formatted message
+            e.request_id, e.server_message # details returned by server
+
+            # Complete the error message with additional information once it's raised
+            e.append_to_message("\n`create_commit` expects the repository to exist.")
+            raise
+    ```
+
+    Args:
+        response (`Response`):
+            Response from the server.
+        endpoint_name (`str`, *optional*):
+            Name of the endpoint that has been called. If provided, the error message
+            will be more complete.
+
+    <Tip warning={true}>
+
+    Raises when the request has failed:
+
+        - [`~utils.RepositoryNotFoundError`]
+            If the repository to download from cannot be found. This may be because it
+            doesn't exist, because `repo_type` is not set correctly, or because the repo
+            is `private` and you do not have access.
+        - [`~utils.GatedRepoError`]
+            If the repository exists but is gated and the user is not on the authorized
+            list.
+        - [`~utils.RevisionNotFoundError`]
+            If the repository exists but the revision couldn't be find.
+        - [`~utils.EntryNotFoundError`]
+            If the repository exists but the entry (e.g. the requested file) couldn't be
+            find.
+        - [`~utils.BadRequestError`]
+            If request failed with a HTTP 400 BadRequest error.
+        - [`~utils.HfHubHTTPError`]
+            If request failed for a reason not listed above.
+
+    </Tip>
+    """
+    try:
+        response.raise_for_status()
+    except HTTPError as e:
+        error_code = response.headers.get("X-Error-Code")
+        error_message = response.headers.get("X-Error-Message")
+
+        if error_code == "RevisionNotFound":
+            message = f"{response.status_code} Client Error." + "\n\n" + f"Revision Not Found for url: {response.url}."
+            raise _format(RevisionNotFoundError, message, response) from e
+
+        elif error_code == "EntryNotFound":
+            message = f"{response.status_code} Client Error." + "\n\n" + f"Entry Not Found for url: {response.url}."
+            raise _format(EntryNotFoundError, message, response) from e
+
+        elif error_code == "GatedRepo":
+            message = (
+                f"{response.status_code} Client Error." + "\n\n" + f"Cannot access gated repo for url {response.url}."
+            )
+            raise _format(GatedRepoError, message, response) from e
+
+        elif error_message == "Access to this resource is disabled.":
+            message = (
+                f"{response.status_code} Client Error."
+                + "\n\n"
+                + f"Cannot access repository for url {response.url}."
+                + "\n"
+                + "Access to this resource is disabled."
+            )
+            raise _format(DisabledRepoError, message, response) from e
+
+        elif error_code == "RepoNotFound" or (
+            response.status_code == 401
+            and error_message != "Invalid credentials in Authorization header"
+            and response.request is not None
+            and response.request.url is not None
+            and REPO_API_REGEX.search(response.request.url) is not None
+        ):
+            # 401 is misleading as it is returned for:
+            #    - private and gated repos if user is not authenticated
+            #    - missing repos
+            # => for now, we process them as `RepoNotFound` anyway.
+            # See https://gist.github.com/Wauplin/46c27ad266b15998ce56a6603796f0b9
+            message = (
+                f"{response.status_code} Client Error."
+                + "\n\n"
+                + f"Repository Not Found for url: {response.url}."
+                + "\nPlease make sure you specified the correct `repo_id` and"
+                " `repo_type`.\nIf you are trying to access a private or gated repo,"
+                " make sure you are authenticated. For more details, see"
+                " https://huggingface.co/docs/huggingface_hub/authentication"
+            )
+            raise _format(RepositoryNotFoundError, message, response) from e
+
+        elif response.status_code == 400:
+            message = (
+                f"\n\nBad request for {endpoint_name} endpoint:" if endpoint_name is not None else "\n\nBad request:"
+            )
+            raise _format(BadRequestError, message, response) from e
+
+        elif response.status_code == 403:
+            message = (
+                f"\n\n{response.status_code} Forbidden: {error_message}."
+                + f"\nCannot access content at: {response.url}."
+                + "\nMake sure your token has the correct permissions."
+            )
+            raise _format(HfHubHTTPError, message, response) from e
+
+        elif response.status_code == 416:
+            range_header = response.request.headers.get("Range")
+            message = f"{e}. Requested range: {range_header}. Content-Range: {response.headers.get('Content-Range')}."
+            raise _format(HfHubHTTPError, message, response) from e
+
+        # Convert `HTTPError` into a `HfHubHTTPError` to display request information
+        # as well (request id and/or server error message)
+        raise _format(HfHubHTTPError, str(e), response) from e
+
+
+def _format(error_type: Type[HfHubHTTPError], custom_message: str, response: Response) -> HfHubHTTPError:
+    server_errors = []
+
+    # Retrieve server error from header
+    from_headers = response.headers.get("X-Error-Message")
+    if from_headers is not None:
+        server_errors.append(from_headers)
+
+    # Retrieve server error from body
+    try:
+        # Case errors are returned in a JSON format
+        data = response.json()
+
+        error = data.get("error")
+        if error is not None:
+            if isinstance(error, list):
+                # Case {'error': ['my error 1', 'my error 2']}
+                server_errors.extend(error)
+            else:
+                # Case {'error': 'my error'}
+                server_errors.append(error)
+
+        errors = data.get("errors")
+        if errors is not None:
+            # Case {'errors': [{'message': 'my error 1'}, {'message': 'my error 2'}]}
+            for error in errors:
+                if "message" in error:
+                    server_errors.append(error["message"])
+
+    except JSONDecodeError:
+        # If content is not JSON and not HTML, append the text
+        content_type = response.headers.get("Content-Type", "")
+        if response.text and "html" not in content_type.lower():
+            server_errors.append(response.text)
+
+    # Strip all server messages
+    server_errors = [str(line).strip() for line in server_errors if str(line).strip()]
+
+    # Deduplicate server messages (keep order)
+    # taken from https://stackoverflow.com/a/17016257
+    server_errors = list(dict.fromkeys(server_errors))
+
+    # Format server error
+    server_message = "\n".join(server_errors)
+
+    # Add server error to custom message
+    final_error_message = custom_message
+    if server_message and server_message.lower() not in custom_message.lower():
+        if "\n\n" in custom_message:
+            final_error_message += "\n" + server_message
+        else:
+            final_error_message += "\n\n" + server_message
+    # Add Request ID
+    request_id = str(response.headers.get(X_REQUEST_ID, ""))
+    if request_id:
+        request_id_message = f" (Request ID: {request_id})"
+    else:
+        # Fallback to X-Amzn-Trace-Id
+        request_id = str(response.headers.get(X_AMZN_TRACE_ID, ""))
+        if request_id:
+            request_id_message = f" (Amzn Trace ID: {request_id})"
+    if request_id and request_id.lower() not in final_error_message.lower():
+        if "\n" in final_error_message:
+            newline_index = final_error_message.index("\n")
+            final_error_message = (
+                final_error_message[:newline_index] + request_id_message + final_error_message[newline_index:]
+            )
+        else:
+            final_error_message += request_id_message
+
+    # Return
+    return error_type(final_error_message.strip(), response=response, server_message=server_message or None)
+
+
+def _curlify(request: requests.PreparedRequest) -> str:
+    """Convert a `requests.PreparedRequest` into a curl command (str).
+
+    Used for debug purposes only.
+
+    Implementation vendored from https://github.com/ofw/curlify/blob/master/curlify.py.
+    MIT License Copyright (c) 2016 Egor.
+    """
+    parts: List[Tuple[Any, Any]] = [
+        ("curl", None),
+        ("-X", request.method),
+    ]
+
+    for k, v in sorted(request.headers.items()):
+        if k.lower() == "authorization":
+            v = "<TOKEN>"  # Hide authorization header, no matter its value (can be Bearer, Key, etc.)
+        parts += [("-H", "{0}: {1}".format(k, v))]
+
+    if request.body:
+        body = request.body
+        if isinstance(body, bytes):
+            body = body.decode("utf-8", errors="ignore")
+        elif hasattr(body, "read"):
+            body = "<file-like object>"  # Don't try to read it to avoid consuming the stream
+        if len(body) > 1000:
+            body = body[:1000] + " ... [truncated]"
+        parts += [("-d", body.replace("\n", ""))]
+
+    parts += [(None, request.url)]
+
+    flat_parts = []
+    for k, v in parts:
+        if k:
+            flat_parts.append(quote(k))
+        if v:
+            flat_parts.append(quote(v))
+
+    return " ".join(flat_parts)
+
+
+# Regex to parse HTTP Range header
+RANGE_REGEX = re.compile(r"^\s*bytes\s*=\s*(\d*)\s*-\s*(\d*)\s*$", re.IGNORECASE)
+
+
+def _adjust_range_header(original_range: Optional[str], resume_size: int) -> Optional[str]:
+    """
+    Adjust HTTP Range header to account for resume position.
+    """
+    if not original_range:
+        return f"bytes={resume_size}-"
+
+    if "," in original_range:
+        raise ValueError(f"Multiple ranges detected - {original_range!r}, not supported yet.")
+
+    match = RANGE_REGEX.match(original_range)
+    if not match:
+        raise RuntimeError(f"Invalid range format - {original_range!r}.")
+    start, end = match.groups()
+
+    if not start:
+        if not end:
+            raise RuntimeError(f"Invalid range format - {original_range!r}.")
+
+        new_suffix = int(end) - resume_size
+        new_range = f"bytes=-{new_suffix}"
+        if new_suffix <= 0:
+            raise RuntimeError(f"Empty new range - {new_range!r}.")
+        return new_range
+
+    start = int(start)
+    new_start = start + resume_size
+    if end:
+        end = int(end)
+        new_range = f"bytes={new_start}-{end}"
+        if new_start > end:
+            raise RuntimeError(f"Empty new range - {new_range!r}.")
+        return new_range
+
+    return f"bytes={new_start}-"

venv/lib/python3.12/site-packages/huggingface_hub/utils/_pagination.py

@@ -0,0 +1,52 @@
+# coding=utf-8
+# Copyright 2022-present, the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Contains utilities to handle pagination on Huggingface Hub."""
+
+from typing import Dict, Iterable, Optional
+
+import requests
+
+from . import get_session, hf_raise_for_status, http_backoff, logging
+
+
+logger = logging.get_logger(__name__)
+
+
+def paginate(path: str, params: Dict, headers: Dict) -> Iterable:
+    """Fetch a list of models/datasets/spaces and paginate through results.
+
+    This is using the same "Link" header format as GitHub.
+    See:
+    - https://requests.readthedocs.io/en/latest/api/#requests.Response.links
+    - https://docs.github.com/en/rest/guides/traversing-with-pagination#link-header
+    """
+    session = get_session()
+    r = session.get(path, params=params, headers=headers)
+    hf_raise_for_status(r)
+    yield from r.json()
+
+    # Follow pages
+    # Next link already contains query params
+    next_page = _get_next_page(r)
+    while next_page is not None:
+        logger.debug(f"Pagination detected. Requesting next page: {next_page}")
+        r = http_backoff("GET", next_page, max_retries=20, retry_on_status_codes=429, headers=headers)
+        hf_raise_for_status(r)
+        yield from r.json()
+        next_page = _get_next_page(r)
+
+
+def _get_next_page(response: requests.Response) -> Optional[str]:
+    return response.links.get("next", {}).get("url")

venv/lib/python3.12/site-packages/huggingface_hub/utils/_paths.py

@@ -0,0 +1,141 @@
+# coding=utf-8
+# Copyright 2022-present, the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Contains utilities to handle paths in Huggingface Hub."""
+
+from fnmatch import fnmatch
+from pathlib import Path
+from typing import Callable, Generator, Iterable, List, Optional, TypeVar, Union
+
+
+T = TypeVar("T")
+
+# Always ignore `.git` and `.cache/huggingface` folders in commits
+DEFAULT_IGNORE_PATTERNS = [
+    ".git",
+    ".git/*",
+    "*/.git",
+    "**/.git/**",
+    ".cache/huggingface",
+    ".cache/huggingface/*",
+    "*/.cache/huggingface",
+    "**/.cache/huggingface/**",
+]
+# Forbidden to commit these folders
+FORBIDDEN_FOLDERS = [".git", ".cache"]
+
+
+def filter_repo_objects(
+    items: Iterable[T],
+    *,
+    allow_patterns: Optional[Union[List[str], str]] = None,
+    ignore_patterns: Optional[Union[List[str], str]] = None,
+    key: Optional[Callable[[T], str]] = None,
+) -> Generator[T, None, None]:
+    """Filter repo objects based on an allowlist and a denylist.
+
+    Input must be a list of paths (`str` or `Path`) or a list of arbitrary objects.
+    In the later case, `key` must be provided and specifies a function of one argument
+    that is used to extract a path from each element in iterable.
+
+    Patterns are Unix shell-style wildcards which are NOT regular expressions. See
+    https://docs.python.org/3/library/fnmatch.html for more details.
+
+    Args:
+        items (`Iterable`):
+            List of items to filter.
+        allow_patterns (`str` or `List[str]`, *optional*):
+            Patterns constituting the allowlist. If provided, item paths must match at
+            least one pattern from the allowlist.
+        ignore_patterns (`str` or `List[str]`, *optional*):
+            Patterns constituting the denylist. If provided, item paths must not match
+            any patterns from the denylist.
+        key (`Callable[[T], str]`, *optional*):
+            Single-argument function to extract a path from each item. If not provided,
+            the `items` must already be `str` or `Path`.
+
+    Returns:
+        Filtered list of objects, as a generator.
+
+    Raises:
+        :class:`ValueError`:
+            If `key` is not provided and items are not `str` or `Path`.
+
+    Example usage with paths:
+    ```python
+    >>> # Filter only PDFs that are not hidden.
+    >>> list(filter_repo_objects(
+    ...     ["aaa.PDF", "bbb.jpg", ".ccc.pdf", ".ddd.png"],
+    ...     allow_patterns=["*.pdf"],
+    ...     ignore_patterns=[".*"],
+    ... ))
+    ["aaa.pdf"]
+    ```
+
+    Example usage with objects:
+    ```python
+    >>> list(filter_repo_objects(
+    ... [
+    ...     CommitOperationAdd(path_or_fileobj="/tmp/aaa.pdf", path_in_repo="aaa.pdf")
+    ...     CommitOperationAdd(path_or_fileobj="/tmp/bbb.jpg", path_in_repo="bbb.jpg")
+    ...     CommitOperationAdd(path_or_fileobj="/tmp/.ccc.pdf", path_in_repo=".ccc.pdf")
+    ...     CommitOperationAdd(path_or_fileobj="/tmp/.ddd.png", path_in_repo=".ddd.png")
+    ... ],
+    ... allow_patterns=["*.pdf"],
+    ... ignore_patterns=[".*"],
+    ... key=lambda x: x.repo_in_path
+    ... ))
+    [CommitOperationAdd(path_or_fileobj="/tmp/aaa.pdf", path_in_repo="aaa.pdf")]
+    ```
+    """
+    if isinstance(allow_patterns, str):
+        allow_patterns = [allow_patterns]
+
+    if isinstance(ignore_patterns, str):
+        ignore_patterns = [ignore_patterns]
+
+    if allow_patterns is not None:
+        allow_patterns = [_add_wildcard_to_directories(p) for p in allow_patterns]
+    if ignore_patterns is not None:
+        ignore_patterns = [_add_wildcard_to_directories(p) for p in ignore_patterns]
+
+    if key is None:
+
+        def _identity(item: T) -> str:
+            if isinstance(item, str):
+                return item
+            if isinstance(item, Path):
+                return str(item)
+            raise ValueError(f"Please provide `key` argument in `filter_repo_objects`: `{item}` is not a string.")
+
+        key = _identity  # Items must be `str` or `Path`, otherwise raise ValueError
+
+    for item in items:
+        path = key(item)
+
+        # Skip if there's an allowlist and path doesn't match any
+        if allow_patterns is not None and not any(fnmatch(path, r) for r in allow_patterns):
+            continue
+
+        # Skip if there's a denylist and path matches any
+        if ignore_patterns is not None and any(fnmatch(path, r) for r in ignore_patterns):
+            continue
+
+        yield item
+
+
+def _add_wildcard_to_directories(pattern: str) -> str:
+    if pattern[-1] == "/":
+        return pattern + "*"
+    return pattern

venv/lib/python3.12/site-packages/huggingface_hub/utils/_safetensors.py

@@ -0,0 +1,111 @@
+import functools
+import operator
+from collections import defaultdict
+from dataclasses import dataclass, field
+from typing import Dict, List, Literal, Optional, Tuple
+
+
+FILENAME_T = str
+TENSOR_NAME_T = str
+DTYPE_T = Literal["F64", "F32", "F16", "BF16", "I64", "I32", "I16", "I8", "U8", "BOOL"]
+
+
+@dataclass
+class TensorInfo:
+    """Information about a tensor.
+
+    For more details regarding the safetensors format, check out https://huggingface.co/docs/safetensors/index#format.
+
+    Attributes:
+        dtype (`str`):
+            The data type of the tensor ("F64", "F32", "F16", "BF16", "I64", "I32", "I16", "I8", "U8", "BOOL").
+        shape (`List[int]`):
+            The shape of the tensor.
+        data_offsets (`Tuple[int, int]`):
+            The offsets of the data in the file as a tuple `[BEGIN, END]`.
+        parameter_count (`int`):
+            The number of parameters in the tensor.
+    """
+
+    dtype: DTYPE_T
+    shape: List[int]
+    data_offsets: Tuple[int, int]
+    parameter_count: int = field(init=False)
+
+    def __post_init__(self) -> None:
+        # Taken from https://stackoverflow.com/a/13840436
+        try:
+            self.parameter_count = functools.reduce(operator.mul, self.shape)
+        except TypeError:
+            self.parameter_count = 1  # scalar value has no shape
+
+
+@dataclass
+class SafetensorsFileMetadata:
+    """Metadata for a Safetensors file hosted on the Hub.
+
+    This class is returned by [`parse_safetensors_file_metadata`].
+
+    For more details regarding the safetensors format, check out https://huggingface.co/docs/safetensors/index#format.
+
+    Attributes:
+        metadata (`Dict`):
+            The metadata contained in the file.
+        tensors (`Dict[str, TensorInfo]`):
+            A map of all tensors. Keys are tensor names and values are information about the corresponding tensor, as a
+            [`TensorInfo`] object.
+        parameter_count (`Dict[str, int]`):
+            A map of the number of parameters per data type. Keys are data types and values are the number of parameters
+            of that data type.
+    """
+
+    metadata: Dict[str, str]
+    tensors: Dict[TENSOR_NAME_T, TensorInfo]
+    parameter_count: Dict[DTYPE_T, int] = field(init=False)
+
+    def __post_init__(self) -> None:
+        parameter_count: Dict[DTYPE_T, int] = defaultdict(int)
+        for tensor in self.tensors.values():
+            parameter_count[tensor.dtype] += tensor.parameter_count
+        self.parameter_count = dict(parameter_count)
+
+
+@dataclass
+class SafetensorsRepoMetadata:
+    """Metadata for a Safetensors repo.
+
+    A repo is considered to be a Safetensors repo if it contains either a 'model.safetensors' weight file (non-shared
+    model) or a 'model.safetensors.index.json' index file (sharded model) at its root.
+
+    This class is returned by [`get_safetensors_metadata`].
+
+    For more details regarding the safetensors format, check out https://huggingface.co/docs/safetensors/index#format.
+
+    Attributes:
+        metadata (`Dict`, *optional*):
+            The metadata contained in the 'model.safetensors.index.json' file, if it exists. Only populated for sharded
+            models.
+        sharded (`bool`):
+            Whether the repo contains a sharded model or not.
+        weight_map (`Dict[str, str]`):
+            A map of all weights. Keys are tensor names and values are filenames of the files containing the tensors.
+        files_metadata (`Dict[str, SafetensorsFileMetadata]`):
+            A map of all files metadata. Keys are filenames and values are the metadata of the corresponding file, as
+            a [`SafetensorsFileMetadata`] object.
+        parameter_count (`Dict[str, int]`):
+            A map of the number of parameters per data type. Keys are data types and values are the number of parameters
+            of that data type.
+    """
+
+    metadata: Optional[Dict]
+    sharded: bool
+    weight_map: Dict[TENSOR_NAME_T, FILENAME_T]  # tensor name -> filename
+    files_metadata: Dict[FILENAME_T, SafetensorsFileMetadata]  # filename -> metadata
+    parameter_count: Dict[DTYPE_T, int] = field(init=False)
+
+    def __post_init__(self) -> None:
+        parameter_count: Dict[DTYPE_T, int] = defaultdict(int)
+        for file_metadata in self.files_metadata.values():
+            for dtype, nb_parameters_ in file_metadata.parameter_count.items():
+                parameter_count[dtype] += nb_parameters_
+        self.parameter_count = dict(parameter_count)

venv/lib/python3.12/site-packages/huggingface_hub/utils/_validators.py

@@ -0,0 +1,226 @@
+# coding=utf-8
+# Copyright 2022-present, the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Contains utilities to validate argument values in `huggingface_hub`."""
+
+import inspect
+import re
+import warnings
+from functools import wraps
+from itertools import chain
+from typing import Any, Dict
+
+from huggingface_hub.errors import HFValidationError
+
+from ._typing import CallableT
+
+
+REPO_ID_REGEX = re.compile(
+    r"""
+    ^
+    (\b[\w\-.]+\b/)? # optional namespace (username or organization)
+    \b               # starts with a word boundary
+    [\w\-.]{1,96}    # repo_name: alphanumeric + . _ -
+    \b               # ends with a word boundary
+    $
+    """,
+    flags=re.VERBOSE,
+)
+
+
+def validate_hf_hub_args(fn: CallableT) -> CallableT:
+    """Validate values received as argument for any public method of `huggingface_hub`.
+
+    The goal of this decorator is to harmonize validation of arguments reused
+    everywhere. By default, all defined validators are tested.
+
+    Validators:
+        - [`~utils.validate_repo_id`]: `repo_id` must be `"repo_name"`
+          or `"namespace/repo_name"`. Namespace is a username or an organization.
+        - [`~utils.smoothly_deprecate_use_auth_token`]: Use `token` instead of
+          `use_auth_token` (only if `use_auth_token` is not expected by the decorated
+          function - in practice, always the case in `huggingface_hub`).
+
+    Example:
+    ```py
+    >>> from huggingface_hub.utils import validate_hf_hub_args
+
+    >>> @validate_hf_hub_args
+    ... def my_cool_method(repo_id: str):
+    ...     print(repo_id)
+
+    >>> my_cool_method(repo_id="valid_repo_id")
+    valid_repo_id
+
+    >>> my_cool_method("other..repo..id")
+    huggingface_hub.utils._validators.HFValidationError: Cannot have -- or .. in repo_id: 'other..repo..id'.
+
+    >>> my_cool_method(repo_id="other..repo..id")
+    huggingface_hub.utils._validators.HFValidationError: Cannot have -- or .. in repo_id: 'other..repo..id'.
+
+    >>> @validate_hf_hub_args
+    ... def my_cool_auth_method(token: str):
+    ...     print(token)
+
+    >>> my_cool_auth_method(token="a token")
+    "a token"
+
+    >>> my_cool_auth_method(use_auth_token="a use_auth_token")
+    "a use_auth_token"
+
+    >>> my_cool_auth_method(token="a token", use_auth_token="a use_auth_token")
+    UserWarning: Both `token` and `use_auth_token` are passed (...)
+    "a token"
+    ```
+
+    Raises:
+        [`~utils.HFValidationError`]:
+            If an input is not valid.
+    """
+    # TODO: add an argument to opt-out validation for specific argument?
+    signature = inspect.signature(fn)
+
+    # Should the validator switch `use_auth_token` values to `token`? In practice, always
+    # True in `huggingface_hub`. Might not be the case in a downstream library.
+    check_use_auth_token = "use_auth_token" not in signature.parameters and "token" in signature.parameters
+
+    @wraps(fn)
+    def _inner_fn(*args, **kwargs):
+        has_token = False
+        for arg_name, arg_value in chain(
+            zip(signature.parameters, args),  # Args values
+            kwargs.items(),  # Kwargs values
+        ):
+            if arg_name in ["repo_id", "from_id", "to_id"]:
+                validate_repo_id(arg_value)
+
+            elif arg_name == "token" and arg_value is not None:
+                has_token = True
+
+        if check_use_auth_token:
+            kwargs = smoothly_deprecate_use_auth_token(fn_name=fn.__name__, has_token=has_token, kwargs=kwargs)
+
+        return fn(*args, **kwargs)
+
+    return _inner_fn  # type: ignore
+
+
+def validate_repo_id(repo_id: str) -> None:
+    """Validate `repo_id` is valid.
+
+    This is not meant to replace the proper validation made on the Hub but rather to
+    avoid local inconsistencies whenever possible (example: passing `repo_type` in the
+    `repo_id` is forbidden).
+
+    Rules:
+    - Between 1 and 96 characters.
+    - Either "repo_name" or "namespace/repo_name"
+    - [a-zA-Z0-9] or "-", "_", "."
+    - "--" and ".." are forbidden
+
+    Valid: `"foo"`, `"foo/bar"`, `"123"`, `"Foo-BAR_foo.bar123"`
+
+    Not valid: `"datasets/foo/bar"`, `".repo_id"`, `"foo--bar"`, `"foo.git"`
+
+    Example:
+    ```py
+    >>> from huggingface_hub.utils import validate_repo_id
+    >>> validate_repo_id(repo_id="valid_repo_id")
+    >>> validate_repo_id(repo_id="other..repo..id")
+    huggingface_hub.utils._validators.HFValidationError: Cannot have -- or .. in repo_id: 'other..repo..id'.
+    ```
+
+    Discussed in https://github.com/huggingface/huggingface_hub/issues/1008.
+    In moon-landing (internal repository):
+    - https://github.com/huggingface/moon-landing/blob/main/server/lib/Names.ts#L27
+    - https://github.com/huggingface/moon-landing/blob/main/server/views/components/NewRepoForm/NewRepoForm.svelte#L138
+    """
+    if not isinstance(repo_id, str):
+        # Typically, a Path is not a repo_id
+        raise HFValidationError(f"Repo id must be a string, not {type(repo_id)}: '{repo_id}'.")
+
+    if repo_id.count("/") > 1:
+        raise HFValidationError(
+            "Repo id must be in the form 'repo_name' or 'namespace/repo_name':"
+            f" '{repo_id}'. Use `repo_type` argument if needed."
+        )
+
+    if not REPO_ID_REGEX.match(repo_id):
+        raise HFValidationError(
+            "Repo id must use alphanumeric chars or '-', '_', '.', '--' and '..' are"
+            " forbidden, '-' and '.' cannot start or end the name, max length is 96:"
+            f" '{repo_id}'."
+        )
+
+    if "--" in repo_id or ".." in repo_id:
+        raise HFValidationError(f"Cannot have -- or .. in repo_id: '{repo_id}'.")
+
+    if repo_id.endswith(".git"):
+        raise HFValidationError(f"Repo_id cannot end by '.git': '{repo_id}'.")
+
+
+def smoothly_deprecate_use_auth_token(fn_name: str, has_token: bool, kwargs: Dict[str, Any]) -> Dict[str, Any]:
+    """Smoothly deprecate `use_auth_token` in the `huggingface_hub` codebase.
+
+    The long-term goal is to remove any mention of `use_auth_token` in the codebase in
+    favor of a unique and less verbose `token` argument. This will be done a few steps:
+
+    0. Step 0: methods that require a read-access to the Hub use the `use_auth_token`
+       argument (`str`, `bool` or `None`). Methods requiring write-access have a `token`
+       argument (`str`, `None`). This implicit rule exists to be able to not send the
+       token when not necessary (`use_auth_token=False`) even if logged in.
+
+    1. Step 1: we want to harmonize everything and use `token` everywhere (supporting
+       `token=False` for read-only methods). In order not to break existing code, if
+       `use_auth_token` is passed to a function, the `use_auth_token` value is passed
+       as `token` instead, without any warning.
+       a. Corner case: if both `use_auth_token` and `token` values are passed, a warning
+          is thrown and the `use_auth_token` value is ignored.
+
+    2. Step 2: Once it is release, we should push downstream libraries to switch from
+       `use_auth_token` to `token` as much as possible, but without throwing a warning
+       (e.g. manually create issues on the corresponding repos).
+
+    3. Step 3: After a transitional period (6 months e.g. until April 2023?), we update
+       `huggingface_hub` to throw a warning on `use_auth_token`. Hopefully, very few
+       users will be impacted as it would have already been fixed.
+       In addition, unit tests in `huggingface_hub` must be adapted to expect warnings
+       to be thrown (but still use `use_auth_token` as before).
+
+    4. Step 4: After a normal deprecation cycle (3 releases ?), remove this validator.
+       `use_auth_token` will definitely not be supported.
+       In addition, we update unit tests in `huggingface_hub` to use `token` everywhere.
+
+    This has been discussed in:
+    - https://github.com/huggingface/huggingface_hub/issues/1094.
+    - https://github.com/huggingface/huggingface_hub/pull/928
+    - (related) https://github.com/huggingface/huggingface_hub/pull/1064
+    """
+    new_kwargs = kwargs.copy()  # do not mutate input !
+
+    use_auth_token = new_kwargs.pop("use_auth_token", None)  # remove from kwargs
+    if use_auth_token is not None:
+        if has_token:
+            warnings.warn(
+                "Both `token` and `use_auth_token` are passed to"
+                f" `{fn_name}` with non-None values. `token` is now the"
+                " preferred argument to pass a User Access Token."
+                " `use_auth_token` value will be ignored."
+            )
+        else:
+            # `token` argument is not passed and a non-None value is passed in
+            # `use_auth_token` => use `use_auth_token` value as `token` kwarg.
+            new_kwargs["token"] = use_auth_token
+
+    return new_kwargs

venv/lib/python3.12/site-packages/huggingface_hub/utils/_xet.py

@@ -0,0 +1,188 @@
+from dataclasses import dataclass
+from enum import Enum
+from typing import Dict, Optional
+
+import requests
+
+from .. import constants
+from . import get_session, hf_raise_for_status, validate_hf_hub_args
+
+
+class XetTokenType(str, Enum):
+    READ = "read"
+    WRITE = "write"
+
+
+@dataclass(frozen=True)
+class XetFileData:
+    file_hash: str
+    refresh_route: str
+
+
+@dataclass(frozen=True)
+class XetConnectionInfo:
+    access_token: str
+    expiration_unix_epoch: int
+    endpoint: str
+
+
+def parse_xet_file_data_from_response(response: requests.Response) -> Optional[XetFileData]:
+    """
+    Parse XET file metadata from an HTTP response.
+
+    This function extracts XET file metadata from the HTTP headers or HTTP links
+    of a given response object. If the required metadata is not found, it returns `None`.
+
+    Args:
+        response (`requests.Response`):
+            The HTTP response object containing headers dict and links dict to extract the XET metadata from.
+    Returns:
+        `Optional[XetFileData]`:
+            An instance of `XetFileData` containing the file hash and refresh route if the metadata
+            is found. Returns `None` if the required metadata is missing.
+    """
+    if response is None:
+        return None
+    try:
+        file_hash = response.headers[constants.HUGGINGFACE_HEADER_X_XET_HASH]
+
+        if constants.HUGGINGFACE_HEADER_LINK_XET_AUTH_KEY in response.links:
+            refresh_route = response.links[constants.HUGGINGFACE_HEADER_LINK_XET_AUTH_KEY]["url"]
+        else:
+            refresh_route = response.headers[constants.HUGGINGFACE_HEADER_X_XET_REFRESH_ROUTE]
+    except KeyError:
+        return None
+
+    return XetFileData(
+        file_hash=file_hash,
+        refresh_route=refresh_route,
+    )
+
+
+def parse_xet_connection_info_from_headers(headers: Dict[str, str]) -> Optional[XetConnectionInfo]:
+    """
+    Parse XET connection info from the HTTP headers or return None if not found.
+    Args:
+        headers (`Dict`):
+           HTTP headers to extract the XET metadata from.
+    Returns:
+        `XetConnectionInfo` or `None`:
+            The information needed to connect to the XET storage service.
+            Returns `None` if the headers do not contain the XET connection info.
+    """
+    try:
+        endpoint = headers[constants.HUGGINGFACE_HEADER_X_XET_ENDPOINT]
+        access_token = headers[constants.HUGGINGFACE_HEADER_X_XET_ACCESS_TOKEN]
+        expiration_unix_epoch = int(headers[constants.HUGGINGFACE_HEADER_X_XET_EXPIRATION])
+    except (KeyError, ValueError, TypeError):
+        return None
+
+    return XetConnectionInfo(
+        endpoint=endpoint,
+        access_token=access_token,
+        expiration_unix_epoch=expiration_unix_epoch,
+    )
+
+
+@validate_hf_hub_args
+def refresh_xet_connection_info(
+    *,
+    file_data: XetFileData,
+    headers: Dict[str, str],
+) -> XetConnectionInfo:
+    """
+    Utilizes the information in the parsed metadata to request the Hub xet connection information.
+    This includes the access token, expiration, and XET service URL.
+    Args:
+        file_data: (`XetFileData`):
+            The file data needed to refresh the xet connection information.
+        headers (`Dict[str, str]`):
+            Headers to use for the request, including authorization headers and user agent.
+    Returns:
+        `XetConnectionInfo`:
+            The connection information needed to make the request to the xet storage service.
+    Raises:
+        [`~utils.HfHubHTTPError`]
+            If the Hub API returned an error.
+        [`ValueError`](https://docs.python.org/3/library/exceptions.html#ValueError)
+            If the Hub API response is improperly formatted.
+    """
+    if file_data.refresh_route is None:
+        raise ValueError("The provided xet metadata does not contain a refresh endpoint.")
+    return _fetch_xet_connection_info_with_url(file_data.refresh_route, headers)
+
+
+@validate_hf_hub_args
+def fetch_xet_connection_info_from_repo_info(
+    *,
+    token_type: XetTokenType,
+    repo_id: str,
+    repo_type: str,
+    revision: Optional[str] = None,
+    headers: Dict[str, str],
+    endpoint: Optional[str] = None,
+    params: Optional[Dict[str, str]] = None,
+) -> XetConnectionInfo:
+    """
+    Uses the repo info to request a xet access token from Hub.
+    Args:
+        token_type (`XetTokenType`):
+            Type of the token to request: `"read"` or `"write"`.
+        repo_id (`str`):
+            A namespace (user or an organization) and a repo name separated by a `/`.
+        repo_type (`str`):
+            Type of the repo to upload to: `"model"`, `"dataset"` or `"space"`.
+        revision (`str`, `optional`):
+            The revision of the repo to get the token for.
+        headers (`Dict[str, str]`):
+            Headers to use for the request, including authorization headers and user agent.
+        endpoint (`str`, `optional`):
+            The endpoint to use for the request. Defaults to the Hub endpoint.
+        params (`Dict[str, str]`, `optional`):
+            Additional parameters to pass with the request.
+    Returns:
+        `XetConnectionInfo`:
+            The connection information needed to make the request to the xet storage service.
+    Raises:
+        [`~utils.HfHubHTTPError`]
+            If the Hub API returned an error.
+        [`ValueError`](https://docs.python.org/3/library/exceptions.html#ValueError)
+            If the Hub API response is improperly formatted.
+    """
+    endpoint = endpoint if endpoint is not None else constants.ENDPOINT
+    url = f"{endpoint}/api/{repo_type}s/{repo_id}/xet-{token_type.value}-token/{revision}"
+    return _fetch_xet_connection_info_with_url(url, headers, params)
+
+
+@validate_hf_hub_args
+def _fetch_xet_connection_info_with_url(
+    url: str,
+    headers: Dict[str, str],
+    params: Optional[Dict[str, str]] = None,
+) -> XetConnectionInfo:
+    """
+    Requests the xet connection info from the supplied URL. This includes the
+    access token, expiration time, and endpoint to use for the xet storage service.
+    Args:
+        url: (`str`):
+            The access token endpoint URL.
+        headers (`Dict[str, str]`):
+            Headers to use for the request, including authorization headers and user agent.
+        params (`Dict[str, str]`, `optional`):
+            Additional parameters to pass with the request.
+    Returns:
+        `XetConnectionInfo`:
+            The connection information needed to make the request to the xet storage service.
+    Raises:
+        [`~utils.HfHubHTTPError`]
+            If the Hub API returned an error.
+        [`ValueError`](https://docs.python.org/3/library/exceptions.html#ValueError)
+            If the Hub API response is improperly formatted.
+    """
+    resp = get_session().get(headers=headers, url=url, params=params)
+    hf_raise_for_status(resp)
+
+    metadata = parse_xet_connection_info_from_headers(resp.headers)  # type: ignore
+    if metadata is None:
+        raise ValueError("Xet headers have not been correctly set by the server.")
+    return metadata

venv/lib/python3.12/site-packages/huggingface_hub/utils/insecure_hashlib.py

@@ -0,0 +1,38 @@
+# Taken from https://github.com/mlflow/mlflow/pull/10119
+#
+# DO NOT use this function for security purposes (e.g., password hashing).
+#
+# In Python >= 3.9, insecure hashing algorithms such as MD5 fail in FIPS-compliant
+# environments unless `usedforsecurity=False` is explicitly passed.
+#
+# References:
+# - https://github.com/mlflow/mlflow/issues/9905
+# - https://github.com/mlflow/mlflow/pull/10119
+# - https://docs.python.org/3/library/hashlib.html
+# - https://github.com/huggingface/transformers/pull/27038
+#
+# Usage:
+#     ```python
+#     # Use
+#     from huggingface_hub.utils.insecure_hashlib import sha256
+#     # instead of
+#     from hashlib import sha256
+#
+#     # Use
+#     from huggingface_hub.utils import insecure_hashlib
+#     # instead of
+#     import hashlib
+#     ```
+import functools
+import hashlib
+import sys
+
+
+if sys.version_info >= (3, 9):
+    md5 = functools.partial(hashlib.md5, usedforsecurity=False)
+    sha1 = functools.partial(hashlib.sha1, usedforsecurity=False)
+    sha256 = functools.partial(hashlib.sha256, usedforsecurity=False)
+else:
+    md5 = hashlib.md5
+    sha1 = hashlib.sha1
+    sha256 = hashlib.sha256

venv/lib/python3.12/site-packages/huggingface_hub/utils/logging.py

@@ -0,0 +1,188 @@
+# coding=utf-8
+# Copyright 2020 Optuna, Hugging Face
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Logging utilities."""
+
+import logging
+import os
+from logging import (
+    CRITICAL,  # NOQA
+    DEBUG,  # NOQA
+    ERROR,  # NOQA
+    FATAL,  # NOQA
+    INFO,  # NOQA
+    NOTSET,  # NOQA
+    WARN,  # NOQA
+    WARNING,  # NOQA
+)
+from typing import Optional
+
+from .. import constants
+
+
+log_levels = {
+    "debug": logging.DEBUG,
+    "info": logging.INFO,
+    "warning": logging.WARNING,
+    "error": logging.ERROR,
+    "critical": logging.CRITICAL,
+}
+
+_default_log_level = logging.WARNING
+
+
+def _get_library_name() -> str:
+    return __name__.split(".")[0]
+
+
+def _get_library_root_logger() -> logging.Logger:
+    return logging.getLogger(_get_library_name())
+
+
+def _get_default_logging_level():
+    """
+    If `HF_HUB_VERBOSITY` env var is set to one of the valid choices return that as the new default level. If it is not
+    - fall back to `_default_log_level`
+    """
+    env_level_str = os.getenv("HF_HUB_VERBOSITY", None)
+    if env_level_str:
+        if env_level_str in log_levels:
+            return log_levels[env_level_str]
+        else:
+            logging.getLogger().warning(
+                f"Unknown option HF_HUB_VERBOSITY={env_level_str}, has to be one of: {', '.join(log_levels.keys())}"
+            )
+    return _default_log_level
+
+
+def _configure_library_root_logger() -> None:
+    library_root_logger = _get_library_root_logger()
+    library_root_logger.addHandler(logging.StreamHandler())
+    library_root_logger.setLevel(_get_default_logging_level())
+
+
+def _reset_library_root_logger() -> None:
+    library_root_logger = _get_library_root_logger()
+    library_root_logger.setLevel(logging.NOTSET)
+
+
+def get_logger(name: Optional[str] = None) -> logging.Logger:
+    """
+        Returns a logger with the specified name. This function is not supposed
+        to be directly accessed by library users.
+
+        Args:
+            name (`str`, *optional*):
+                The name of the logger to get, usually the filename
+
+        Example:
+
+    ```python
+    >>> from huggingface_hub import get_logger
+
+    >>> logger = get_logger(__file__)
+    >>> logger.set_verbosity_info()
+    ```
+    """
+
+    if name is None:
+        name = _get_library_name()
+
+    return logging.getLogger(name)
+
+
+def get_verbosity() -> int:
+    """Return the current level for the HuggingFace Hub's root logger.
+
+    Returns:
+        Logging level, e.g., `huggingface_hub.logging.DEBUG` and
+        `huggingface_hub.logging.INFO`.
+
+    <Tip>
+
+    HuggingFace Hub has following logging levels:
+
+    - `huggingface_hub.logging.CRITICAL`, `huggingface_hub.logging.FATAL`
+    - `huggingface_hub.logging.ERROR`
+    - `huggingface_hub.logging.WARNING`, `huggingface_hub.logging.WARN`
+    - `huggingface_hub.logging.INFO`
+    - `huggingface_hub.logging.DEBUG`
+
+    </Tip>
+    """
+    return _get_library_root_logger().getEffectiveLevel()
+
+
+def set_verbosity(verbosity: int) -> None:
+    """
+    Sets the level for the HuggingFace Hub's root logger.
+
+    Args:
+        verbosity (`int`):
+            Logging level, e.g., `huggingface_hub.logging.DEBUG` and
+            `huggingface_hub.logging.INFO`.
+    """
+    _get_library_root_logger().setLevel(verbosity)
+
+
+def set_verbosity_info():
+    """
+    Sets the verbosity to `logging.INFO`.
+    """
+    return set_verbosity(INFO)
+
+
+def set_verbosity_warning():
+    """
+    Sets the verbosity to `logging.WARNING`.
+    """
+    return set_verbosity(WARNING)
+
+
+def set_verbosity_debug():
+    """
+    Sets the verbosity to `logging.DEBUG`.
+    """
+    return set_verbosity(DEBUG)
+
+
+def set_verbosity_error():
+    """
+    Sets the verbosity to `logging.ERROR`.
+    """
+    return set_verbosity(ERROR)
+
+
+def disable_propagation() -> None:
+    """
+    Disable propagation of the library log outputs. Note that log propagation is
+    disabled by default.
+    """
+    _get_library_root_logger().propagate = False
+
+
+def enable_propagation() -> None:
+    """
+    Enable propagation of the library log outputs. Please disable the
+    HuggingFace Hub's default handler to prevent double logging if the root
+    logger has been configured.
+    """
+    _get_library_root_logger().propagate = True
+
+
+_configure_library_root_logger()
+
+if constants.HF_DEBUG:
+    # If `HF_DEBUG` environment variable is set, set the verbosity of `huggingface_hub` logger to `DEBUG`.
+    set_verbosity_debug()

venv/lib/python3.12/site-packages/huggingface_hub/utils/sha.py

@@ -0,0 +1,64 @@
+"""Utilities to efficiently compute the SHA 256 hash of a bunch of bytes."""
+
+from typing import BinaryIO, Optional
+
+from .insecure_hashlib import sha1, sha256
+
+
+def sha_fileobj(fileobj: BinaryIO, chunk_size: Optional[int] = None) -> bytes:
+    """
+    Computes the sha256 hash of the given file object, by chunks of size `chunk_size`.
+
+    Args:
+        fileobj (file-like object):
+            The File object to compute sha256 for, typically obtained with `open(path, "rb")`
+        chunk_size (`int`, *optional*):
+            The number of bytes to read from `fileobj` at once, defaults to 1MB.
+
+    Returns:
+        `bytes`: `fileobj`'s sha256 hash as bytes
+    """
+    chunk_size = chunk_size if chunk_size is not None else 1024 * 1024
+
+    sha = sha256()
+    while True:
+        chunk = fileobj.read(chunk_size)
+        sha.update(chunk)
+        if not chunk:
+            break
+    return sha.digest()
+
+
+def git_hash(data: bytes) -> str:
+    """
+    Computes the git-sha1 hash of the given bytes, using the same algorithm as git.
+
+    This is equivalent to running `git hash-object`. See https://git-scm.com/docs/git-hash-object
+    for more details.
+
+    Note: this method is valid for regular files. For LFS files, the proper git hash is supposed to be computed on the
+          pointer file content, not the actual file content. However, for simplicity, we directly compare the sha256 of
+          the LFS file content when we want to compare LFS files.
+
+    Args:
+        data (`bytes`):
+            The data to compute the git-hash for.
+
+    Returns:
+        `str`: the git-hash of `data` as an hexadecimal string.
+
+    Example:
+    ```python
+    >>> from huggingface_hub.utils.sha import git_hash
+    >>> git_hash(b"Hello, World!")
+    'b45ef6fec89518d314f546fd6c3025367b721684'
+    ```
+    """
+    # Taken from https://gist.github.com/msabramo/763200
+    # Note: no need to optimize by reading the file in chunks as we're not supposed to hash huge files (5MB maximum).
+    sha = sha1()
+    sha.update(b"blob ")
+    sha.update(str(len(data)).encode())
+    sha.update(b"\0")
+    sha.update(data)
+    return sha.hexdigest()

venv/lib/python3.12/site-packages/packaging-25.0.dist-info/licenses/LICENSE

@@ -0,0 +1,3 @@
+This software is made available under the terms of *either* of the licenses
+found in LICENSE.APACHE or LICENSE.BSD. Contributions to this software is made
+under the terms of *both* these licenses.

venv/lib/python3.12/site-packages/packaging-25.0.dist-info/licenses/LICENSE.APACHE

@@ -0,0 +1,177 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS

venv/lib/python3.12/site-packages/packaging-25.0.dist-info/licenses/LICENSE.BSD

@@ -0,0 +1,23 @@
+Copyright (c) Donald Stufft and individual contributors.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+    1. Redistributions of source code must retain the above copyright notice,
+       this list of conditions and the following disclaimer.
+
+    2. Redistributions in binary form must reproduce the above copyright
+       notice, this list of conditions and the following disclaimer in the
+       documentation and/or other materials provided with the distribution.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

venv/lib/python3.12/site-packages/pip/_internal/__init__.py

@@ -0,0 +1,18 @@
+from typing import List, Optional
+
+from pip._internal.utils import _log
+
+# init_logging() must be called before any call to logging.getLogger()
+# which happens at import of most modules.
+_log.init_logging()
+
+
+def main(args: (Optional[List[str]]) = None) -> int:
+    """This is preserved for old console scripts that may still be referencing
+    it.
+
+    For additional details, see https://github.com/pypa/pip/issues/7498.
+    """
+    from pip._internal.utils.entrypoints import _wrapper
+
+    return _wrapper(args)

venv/lib/python3.12/site-packages/pip/_internal/build_env.py

@@ -0,0 +1,311 @@
+"""Build Environment used for isolation during sdist building
+"""
+
+import logging
+import os
+import pathlib
+import site
+import sys
+import textwrap
+from collections import OrderedDict
+from types import TracebackType
+from typing import TYPE_CHECKING, Iterable, List, Optional, Set, Tuple, Type, Union
+
+from pip._vendor.certifi import where
+from pip._vendor.packaging.requirements import Requirement
+from pip._vendor.packaging.version import Version
+
+from pip import __file__ as pip_location
+from pip._internal.cli.spinners import open_spinner
+from pip._internal.locations import get_platlib, get_purelib, get_scheme
+from pip._internal.metadata import get_default_environment, get_environment
+from pip._internal.utils.subprocess import call_subprocess
+from pip._internal.utils.temp_dir import TempDirectory, tempdir_kinds
+
+if TYPE_CHECKING:
+    from pip._internal.index.package_finder import PackageFinder
+
+logger = logging.getLogger(__name__)
+
+
+def _dedup(a: str, b: str) -> Union[Tuple[str], Tuple[str, str]]:
+    return (a, b) if a != b else (a,)
+
+
+class _Prefix:
+    def __init__(self, path: str) -> None:
+        self.path = path
+        self.setup = False
+        scheme = get_scheme("", prefix=path)
+        self.bin_dir = scheme.scripts
+        self.lib_dirs = _dedup(scheme.purelib, scheme.platlib)
+
+
+def get_runnable_pip() -> str:
+    """Get a file to pass to a Python executable, to run the currently-running pip.
+
+    This is used to run a pip subprocess, for installing requirements into the build
+    environment.
+    """
+    source = pathlib.Path(pip_location).resolve().parent
+
+    if not source.is_dir():
+        # This would happen if someone is using pip from inside a zip file. In that
+        # case, we can use that directly.
+        return str(source)
+
+    return os.fsdecode(source / "__pip-runner__.py")
+
+
+def _get_system_sitepackages() -> Set[str]:
+    """Get system site packages
+
+    Usually from site.getsitepackages,
+    but fallback on `get_purelib()/get_platlib()` if unavailable
+    (e.g. in a virtualenv created by virtualenv<20)
+
+    Returns normalized set of strings.
+    """
+    if hasattr(site, "getsitepackages"):
+        system_sites = site.getsitepackages()
+    else:
+        # virtualenv < 20 overwrites site.py without getsitepackages
+        # fallback on get_purelib/get_platlib.
+        # this is known to miss things, but shouldn't in the cases
+        # where getsitepackages() has been removed (inside a virtualenv)
+        system_sites = [get_purelib(), get_platlib()]
+    return {os.path.normcase(path) for path in system_sites}
+
+
+class BuildEnvironment:
+    """Creates and manages an isolated environment to install build deps"""
+
+    def __init__(self) -> None:
+        temp_dir = TempDirectory(kind=tempdir_kinds.BUILD_ENV, globally_managed=True)
+
+        self._prefixes = OrderedDict(
+            (name, _Prefix(os.path.join(temp_dir.path, name)))
+            for name in ("normal", "overlay")
+        )
+
+        self._bin_dirs: List[str] = []
+        self._lib_dirs: List[str] = []
+        for prefix in reversed(list(self._prefixes.values())):
+            self._bin_dirs.append(prefix.bin_dir)
+            self._lib_dirs.extend(prefix.lib_dirs)
+
+        # Customize site to:
+        # - ensure .pth files are honored
+        # - prevent access to system site packages
+        system_sites = _get_system_sitepackages()
+
+        self._site_dir = os.path.join(temp_dir.path, "site")
+        if not os.path.exists(self._site_dir):
+            os.mkdir(self._site_dir)
+        with open(
+            os.path.join(self._site_dir, "sitecustomize.py"), "w", encoding="utf-8"
+        ) as fp:
+            fp.write(
+                textwrap.dedent(
+                    """
+                import os, site, sys
+
+                # First, drop system-sites related paths.
+                original_sys_path = sys.path[:]
+                known_paths = set()
+                for path in {system_sites!r}:
+                    site.addsitedir(path, known_paths=known_paths)
+                system_paths = set(
+                    os.path.normcase(path)
+                    for path in sys.path[len(original_sys_path):]
+                )
+                original_sys_path = [
+                    path for path in original_sys_path
+                    if os.path.normcase(path) not in system_paths
+                ]
+                sys.path = original_sys_path
+
+                # Second, add lib directories.
+                # ensuring .pth file are processed.
+                for path in {lib_dirs!r}:
+                    assert not path in sys.path
+                    site.addsitedir(path)
+                """
+                ).format(system_sites=system_sites, lib_dirs=self._lib_dirs)
+            )
+
+    def __enter__(self) -> None:
+        self._save_env = {
+            name: os.environ.get(name, None)
+            for name in ("PATH", "PYTHONNOUSERSITE", "PYTHONPATH")
+        }
+
+        path = self._bin_dirs[:]
+        old_path = self._save_env["PATH"]
+        if old_path:
+            path.extend(old_path.split(os.pathsep))
+
+        pythonpath = [self._site_dir]
+
+        os.environ.update(
+            {
+                "PATH": os.pathsep.join(path),
+                "PYTHONNOUSERSITE": "1",
+                "PYTHONPATH": os.pathsep.join(pythonpath),
+            }
+        )
+
+    def __exit__(
+        self,
+        exc_type: Optional[Type[BaseException]],
+        exc_val: Optional[BaseException],
+        exc_tb: Optional[TracebackType],
+    ) -> None:
+        for varname, old_value in self._save_env.items():
+            if old_value is None:
+                os.environ.pop(varname, None)
+            else:
+                os.environ[varname] = old_value
+
+    def check_requirements(
+        self, reqs: Iterable[str]
+    ) -> Tuple[Set[Tuple[str, str]], Set[str]]:
+        """Return 2 sets:
+        - conflicting requirements: set of (installed, wanted) reqs tuples
+        - missing requirements: set of reqs
+        """
+        missing = set()
+        conflicting = set()
+        if reqs:
+            env = (
+                get_environment(self._lib_dirs)
+                if hasattr(self, "_lib_dirs")
+                else get_default_environment()
+            )
+            for req_str in reqs:
+                req = Requirement(req_str)
+                # We're explicitly evaluating with an empty extra value, since build
+                # environments are not provided any mechanism to select specific extras.
+                if req.marker is not None and not req.marker.evaluate({"extra": ""}):
+                    continue
+                dist = env.get_distribution(req.name)
+                if not dist:
+                    missing.add(req_str)
+                    continue
+                if isinstance(dist.version, Version):
+                    installed_req_str = f"{req.name}=={dist.version}"
+                else:
+                    installed_req_str = f"{req.name}==={dist.version}"
+                if not req.specifier.contains(dist.version, prereleases=True):
+                    conflicting.add((installed_req_str, req_str))
+                # FIXME: Consider direct URL?
+        return conflicting, missing
+
+    def install_requirements(
+        self,
+        finder: "PackageFinder",
+        requirements: Iterable[str],
+        prefix_as_string: str,
+        *,
+        kind: str,
+    ) -> None:
+        prefix = self._prefixes[prefix_as_string]
+        assert not prefix.setup
+        prefix.setup = True
+        if not requirements:
+            return
+        self._install_requirements(
+            get_runnable_pip(),
+            finder,
+            requirements,
+            prefix,
+            kind=kind,
+        )
+
+    @staticmethod
+    def _install_requirements(
+        pip_runnable: str,
+        finder: "PackageFinder",
+        requirements: Iterable[str],
+        prefix: _Prefix,
+        *,
+        kind: str,
+    ) -> None:
+        args: List[str] = [
+            sys.executable,
+            pip_runnable,
+            "install",
+            "--ignore-installed",
+            "--no-user",
+            "--prefix",
+            prefix.path,
+            "--no-warn-script-location",
+        ]
+        if logger.getEffectiveLevel() <= logging.DEBUG:
+            args.append("-v")
+        for format_control in ("no_binary", "only_binary"):
+            formats = getattr(finder.format_control, format_control)
+            args.extend(
+                (
+                    "--" + format_control.replace("_", "-"),
+                    ",".join(sorted(formats or {":none:"})),
+                )
+            )
+
+        index_urls = finder.index_urls
+        if index_urls:
+            args.extend(["-i", index_urls[0]])
+            for extra_index in index_urls[1:]:
+                args.extend(["--extra-index-url", extra_index])
+        else:
+            args.append("--no-index")
+        for link in finder.find_links:
+            args.extend(["--find-links", link])
+
+        for host in finder.trusted_hosts:
+            args.extend(["--trusted-host", host])
+        if finder.allow_all_prereleases:
+            args.append("--pre")
+        if finder.prefer_binary:
+            args.append("--prefer-binary")
+        args.append("--")
+        args.extend(requirements)
+        extra_environ = {"_PIP_STANDALONE_CERT": where()}
+        with open_spinner(f"Installing {kind}") as spinner:
+            call_subprocess(
+                args,
+                command_desc=f"pip subprocess to install {kind}",
+                spinner=spinner,
+                extra_environ=extra_environ,
+            )
+
+
+class NoOpBuildEnvironment(BuildEnvironment):
+    """A no-op drop-in replacement for BuildEnvironment"""
+
+    def __init__(self) -> None:
+        pass
+
+    def __enter__(self) -> None:
+        pass
+
+    def __exit__(
+        self,
+        exc_type: Optional[Type[BaseException]],
+        exc_val: Optional[BaseException],
+        exc_tb: Optional[TracebackType],
+    ) -> None:
+        pass
+
+    def cleanup(self) -> None:
+        pass
+
+    def install_requirements(
+        self,
+        finder: "PackageFinder",
+        requirements: Iterable[str],
+        prefix_as_string: str,
+        *,
+        kind: str,
+    ) -> None:
+        raise NotImplementedError()