Adding the ability to configure the timeout of get request

docs/source/en/package_reference/environment_variables.md

@@ -74,6 +74,14 @@ small files will be duplicated to ease user experience while bigger files are sy
 
 For more details, see the [download guide](../guides/download#download-files-to-local-folder).
 
+### HF_HUB_ETAG_TIMEOUT
+
+Integer value to define etag timeout when fetching ETag on the system level.
+
+### HF_HUB_DOWNLOAD_TIMEOUT
+
+Integer value to define get request timeout when fetching a file on the system level.
+
 ## Boolean values
 
 The following environment variables expect a boolean value. The variable will be considered

src/huggingface_hub/_snapshot_download.py

@@ -8,8 +8,10 @@
 from .constants import (
     DEFAULT_REVISION,
     HF_HUB_ENABLE_HF_TRANSFER,
+    HF_HUB_ETAG_TIMEOUT,
     HUGGINGFACE_HUB_CACHE,
     REPO_TYPES,
+    DEFAULT_ETAG_TIMEOUT,
 )
 from .file_download import REGEX_COMMIT_HASH, hf_hub_download, repo_folder_name
 from .hf_api import HfApi
@@ -34,7 +36,7 @@ def snapshot_download(
     library_version: Optional[str] = None,
     user_agent: Optional[Union[Dict, str]] = None,
     proxies: Optional[Dict] = None,
-    etag_timeout: float = 10,
+    etag_timeout: float = DEFAULT_ETAG_TIMEOUT,
     resume_download: bool = False,
     force_download: bool = False,
     token: Optional[Union[bool, str]] = None,
@@ -145,6 +147,9 @@ def snapshot_download(
 
     </Tip>
     """
+    if HF_HUB_ETAG_TIMEOUT != DEFAULT_ETAG_TIMEOUT:
+        etag_timeout = HF_HUB_ETAG_TIMEOUT
+
     if cache_dir is None:
         cache_dir = HUGGINGFACE_HUB_CACHE
     if revision is None:

src/huggingface_hub/constants.py

@@ -29,6 +29,9 @@ def _as_int(value: Optional[str]) -> Optional[int]:
 FLAX_WEIGHTS_NAME = "flax_model.msgpack"
 CONFIG_NAME = "config.json"
 REPOCARD_NAME = "README.md"
+DEFAULT_ETAG_TIMEOUT = 10
+DEFAULT_DOWNLOAD_TIMEOUT = 10
+DEFAULT_REQUEST_TIMEOUT = 10
 
 # Git-related constants
 
@@ -130,6 +133,12 @@ def _as_int(value: Optional[str]) -> Optional[int]:
     _as_int(os.environ.get("HF_HUB_LOCAL_DIR_AUTO_SYMLINK_THRESHOLD")) or 5 * 1024 * 1024
 )
 
+# Used to override the etag timeout on a system level
+HF_HUB_ETAG_TIMEOUT: int = _as_int(os.environ.get("HF_HUB_ETAG_TIMEOUT")) or DEFAULT_ETAG_TIMEOUT
+
+# Used to override the get request timeout on a system level
+HF_HUB_DOWNLOAD_TIMEOUT: int = _as_int(os.environ.get("HF_HUB_DOWNLOAD_TIMEOUT")) or DEFAULT_DOWNLOAD_TIMEOUT
+
 # List frameworks that are handled by the InferenceAPI service. Useful to scan endpoints and check which models are
 # deployed and running. Since 95% of the models are using the top 4 frameworks listed below, we scan only those by
 # default. We still keep the full list of supported frameworks in case we want to scan all of them.

src/huggingface_hub/file_download.py

@@ -33,10 +33,15 @@
     HUGGINGFACE_HEADER_X_LINKED_ETAG,
     HUGGINGFACE_HEADER_X_LINKED_SIZE,
     HUGGINGFACE_HEADER_X_REPO_COMMIT,
+    HF_HUB_ETAG_TIMEOUT,
+    HF_HUB_DOWNLOAD_TIMEOUT,
     HUGGINGFACE_HUB_CACHE,
     REPO_ID_SEPARATOR,
     REPO_TYPES,
     REPO_TYPES_URL_PREFIXES,
+    DEFAULT_DOWNLOAD_TIMEOUT,
+    DEFAULT_ETAG_TIMEOUT,
+    DEFAULT_REQUEST_TIMEOUT,
 )
 from .utils import (
     EntryNotFoundError,
@@ -373,7 +378,7 @@ def _request_wrapper(
     max_retries: int = 0,
     base_wait_time: float = 0.5,
     max_wait_time: float = 2,
-    timeout: Optional[float] = 10.0,
+    timeout: Optional[float] = DEFAULT_REQUEST_TIMEOUT,
     follow_relative_redirects: bool = False,
     **params,
 ) -> requests.Response:
@@ -478,13 +483,16 @@ def http_get(
     proxies=None,
     resume_size: float = 0,
     headers: Optional[Dict[str, str]] = None,
-    timeout: Optional[float] = 10.0,
+    timeout: Optional[float] = DEFAULT_DOWNLOAD_TIMEOUT,
     max_retries: int = 0,
     expected_size: Optional[int] = None,
 ):
     """
     Download a remote file. Do not gobble up errors, and will return errors tailored to the Hugging Face Hub.
     """
+    if HF_HUB_DOWNLOAD_TIMEOUT != DEFAULT_DOWNLOAD_TIMEOUT:
+        timeout = HF_HUB_DOWNLOAD_TIMEOUT
+
     if not resume_size:
         if HF_HUB_ENABLE_HF_TRANSFER:
             try:
@@ -576,7 +584,7 @@ def cached_download(
     force_download: bool = False,
     force_filename: Optional[str] = None,
     proxies: Optional[Dict] = None,
-    etag_timeout: float = 10,
+    etag_timeout: float = DEFAULT_ETAG_TIMEOUT,
     resume_download: bool = False,
     token: Union[bool, str, None] = None,
     local_files_only: bool = False,
@@ -656,6 +664,9 @@ def cached_download(
 
     </Tip>
     """
+    if HF_HUB_ETAG_TIMEOUT != DEFAULT_ETAG_TIMEOUT:
+        etag_timeout = HF_HUB_ETAG_TIMEOUT
+
     if not legacy_cache_layout:
         warnings.warn(
             "'cached_download' is the legacy way to download files from the HF hub, please consider upgrading to"
@@ -1005,7 +1016,7 @@ def hf_hub_download(
     force_download: bool = False,
     force_filename: Optional[str] = None,
     proxies: Optional[Dict] = None,
-    etag_timeout: float = 10,
+    etag_timeout: float = DEFAULT_ETAG_TIMEOUT,
     resume_download: bool = False,
     token: Union[bool, str, None] = None,
     local_files_only: bool = False,
@@ -1138,6 +1149,9 @@ def hf_hub_download(
 
     </Tip>
     """
+    if HF_HUB_ETAG_TIMEOUT != DEFAULT_ETAG_TIMEOUT:
+        etag_timeout = HF_HUB_ETAG_TIMEOUT
+
     if force_filename is not None:
         warnings.warn(
             "The `force_filename` parameter is deprecated as a new caching system, "

src/huggingface_hub/hf_api.py

@@ -93,6 +93,8 @@
 )
 from .constants import (
     DEFAULT_REVISION,
+    HF_HUB_ETAG_TIMEOUT,
+    DEFAULT_ETAG_TIMEOUT,
     ENDPOINT,
     REGEX_COMMIT_OID,
     REPO_TYPE_MODEL,
@@ -4041,7 +4043,7 @@ def hf_hub_download(
         force_download: bool = False,
         force_filename: Optional[str] = None,
         proxies: Optional[Dict] = None,
-        etag_timeout: float = 10,
+        etag_timeout: float = DEFAULT_ETAG_TIMEOUT,
         resume_download: bool = False,
         local_files_only: bool = False,
         legacy_cache_layout: bool = False,
@@ -4164,6 +4166,9 @@ def hf_hub_download(
         """
         from .file_download import hf_hub_download
 
+        if HF_HUB_ETAG_TIMEOUT != DEFAULT_ETAG_TIMEOUT:
+            etag_timeout = HF_HUB_ETAG_TIMEOUT
+
         return hf_hub_download(
             repo_id=repo_id,
             filename=filename,
@@ -4198,7 +4203,7 @@ def snapshot_download(
         local_dir: Union[str, Path, None] = None,
         local_dir_use_symlinks: Union[bool, Literal["auto"]] = "auto",
         proxies: Optional[Dict] = None,
-        etag_timeout: float = 10,
+        etag_timeout: float = DEFAULT_ETAG_TIMEOUT,
         resume_download: bool = False,
         force_download: bool = False,
         local_files_only: bool = False,
@@ -4296,6 +4301,9 @@ def snapshot_download(
         """
         from ._snapshot_download import snapshot_download
 
+        if HF_HUB_ETAG_TIMEOUT != DEFAULT_ETAG_TIMEOUT:
+            etag_timeout = HF_HUB_ETAG_TIMEOUT
+
         return snapshot_download(
             repo_id=repo_id,
             repo_type=repo_type,

src/huggingface_hub/utils/_runtime.py

@@ -315,6 +315,8 @@ def dump_environment_info() -> Dict[str, Any]:
     info["HF_HUB_DISABLE_EXPERIMENTAL_WARNING"] = constants.HF_HUB_DISABLE_EXPERIMENTAL_WARNING
     info["HF_HUB_DISABLE_IMPLICIT_TOKEN"] = constants.HF_HUB_DISABLE_IMPLICIT_TOKEN
     info["HF_HUB_ENABLE_HF_TRANSFER"] = constants.HF_HUB_ENABLE_HF_TRANSFER
+    info["HF_HUB_ETAG_TIMEOUT"] = constants.HF_HUB_ETAG_TIMEOUT
+    info["HF_HUB_DOWNLOAD_TIMEOUT"] = constants.HF_HUB_DOWNLOAD_TIMEOUT
 
     print("\nCopy-and-paste the text below in your GitHub issue.\n")
     print("\n".join([f"- {prop}: {val}" for prop, val in info.items()]) + "\n")