huggingface-hub 1.0.0rc2__py3-none-any.whl → 1.0.0rc3__py3-none-any.whl

huggingface_hub/file_download.py

@@ -9,7 +9,7 @@ import uuid
 import warnings
 from dataclasses import dataclass
 from pathlib import Path
-from typing import Any, BinaryIO, NoReturn, Optional, Union
+from typing import Any, BinaryIO, Literal, NoReturn, Optional, Union, overload
 from urllib.parse import quote, urlparse
 
 import httpx
@@ -149,6 +149,34 @@ class HfFileMetadata:
     xet_file_data: Optional[XetFileData]
 
 
+@dataclass
+class DryRunFileInfo:
+    """Information returned when performing a dry run of a file download.
+
+    Returned by [`hf_hub_download`] when `dry_run=True`.
+
+    Args:
+        commit_hash (`str`):
+            The commit_hash related to the file.
+        file_size (`int`):
+            Size of the file. In case of an LFS file, contains the size of the actual LFS file, not the pointer.
+        filename (`str`):
+            Name of the file in the repo.
+        is_cached (`bool`):
+            Whether the file is already cached locally.
+        will_download (`bool`):
+            Whether the file will be downloaded if `hf_hub_download` is called with `dry_run=False`.
+            In practice, will_download is `True` if the file is not cached or if `force_download=True`.
+    """
+
+    commit_hash: str
+    file_size: int
+    filename: str
+    local_path: str
+    is_cached: bool
+    will_download: bool
+
+
 @validate_hf_hub_args
 def hf_hub_url(
     repo_id: str,
@@ -763,6 +791,75 @@ def _check_disk_space(expected_size: int, target_dir: Union[str, Path]) -> None:
             pass
 
 
+@overload
+def hf_hub_download(
+    repo_id: str,
+    filename: str,
+    *,
+    subfolder: Optional[str] = None,
+    repo_type: Optional[str] = None,
+    revision: Optional[str] = None,
+    library_name: Optional[str] = None,
+    library_version: Optional[str] = None,
+    cache_dir: Union[str, Path, None] = None,
+    local_dir: Union[str, Path, None] = None,
+    user_agent: Union[dict, str, None] = None,
+    force_download: bool = False,
+    etag_timeout: float = constants.DEFAULT_ETAG_TIMEOUT,
+    token: Union[bool, str, None] = None,
+    local_files_only: bool = False,
+    headers: Optional[dict[str, str]] = None,
+    endpoint: Optional[str] = None,
+    dry_run: Literal[False] = False,
+) -> str: ...
+
+
+@overload
+def hf_hub_download(
+    repo_id: str,
+    filename: str,
+    *,
+    subfolder: Optional[str] = None,
+    repo_type: Optional[str] = None,
+    revision: Optional[str] = None,
+    library_name: Optional[str] = None,
+    library_version: Optional[str] = None,
+    cache_dir: Union[str, Path, None] = None,
+    local_dir: Union[str, Path, None] = None,
+    user_agent: Union[dict, str, None] = None,
+    force_download: bool = False,
+    etag_timeout: float = constants.DEFAULT_ETAG_TIMEOUT,
+    token: Union[bool, str, None] = None,
+    local_files_only: bool = False,
+    headers: Optional[dict[str, str]] = None,
+    endpoint: Optional[str] = None,
+    dry_run: Literal[True] = True,
+) -> DryRunFileInfo: ...
+
+
+@overload
+def hf_hub_download(
+    repo_id: str,
+    filename: str,
+    *,
+    subfolder: Optional[str] = None,
+    repo_type: Optional[str] = None,
+    revision: Optional[str] = None,
+    library_name: Optional[str] = None,
+    library_version: Optional[str] = None,
+    cache_dir: Union[str, Path, None] = None,
+    local_dir: Union[str, Path, None] = None,
+    user_agent: Union[dict, str, None] = None,
+    force_download: bool = False,
+    etag_timeout: float = constants.DEFAULT_ETAG_TIMEOUT,
+    token: Union[bool, str, None] = None,
+    local_files_only: bool = False,
+    headers: Optional[dict[str, str]] = None,
+    endpoint: Optional[str] = None,
+    dry_run: bool = False,
+) -> Union[str, DryRunFileInfo]: ...
+
+
 @validate_hf_hub_args
 def hf_hub_download(
     repo_id: str,
@@ -782,7 +879,8 @@ def hf_hub_download(
     local_files_only: bool = False,
     headers: Optional[dict[str, str]] = None,
     endpoint: Optional[str] = None,
-) -> str:
+    dry_run: bool = False,
+) -> Union[str, DryRunFileInfo]:
     """Download a given file if it's not already present in the local cache.
 
     The new cache file layout looks like this:
@@ -857,9 +955,14 @@ def hf_hub_download(
             local cached file if it exists.
         headers (`dict`, *optional*):
             Additional headers to be sent with the request.
+        dry_run (`bool`, *optional*, defaults to `False`):
+            If `True`, perform a dry run without actually downloading the file. Returns a
+            [`DryRunFileInfo`] object containing information about what would be downloaded.
 
     Returns:
-        `str`: Local path of file or if networking is off, last version of file cached on disk.
+        `str` or [`DryRunFileInfo`]:
+            - If `dry_run=False`: Local path of file or if networking is off, last version of file cached on disk.
+            - If `dry_run=True`: A [`DryRunFileInfo`] object containing download information.
 
     Raises:
         [`~utils.RepositoryNotFoundError`]
@@ -929,6 +1032,7 @@ def hf_hub_download(
             cache_dir=cache_dir,
             force_download=force_download,
             local_files_only=local_files_only,
+            dry_run=dry_run,
         )
     else:
         return _hf_hub_download_to_cache_dir(
@@ -947,6 +1051,7 @@ def hf_hub_download(
             # Additional options
             local_files_only=local_files_only,
             force_download=force_download,
+            dry_run=dry_run,
         )
 
 
@@ -967,7 +1072,8 @@ def _hf_hub_download_to_cache_dir(
     # Additional options
     local_files_only: bool,
     force_download: bool,
-) -> str:
+    dry_run: bool,
+) -> Union[str, DryRunFileInfo]:
     """Download a given file to a cache folder, if not already present.
 
     Method should not be called directly. Please use `hf_hub_download` instead.
@@ -987,8 +1093,18 @@ def _hf_hub_download_to_cache_dir(
     # if user provides a commit_hash and they already have the file on disk, shortcut everything.
     if REGEX_COMMIT_HASH.match(revision):
         pointer_path = _get_pointer_path(storage_folder, revision, relative_filename)
-        if os.path.exists(pointer_path) and not force_download:
-            return pointer_path
+        if os.path.exists(pointer_path):
+            if dry_run:
+                return DryRunFileInfo(
+                    commit_hash=revision,
+                    file_size=os.path.getsize(pointer_path),
+                    filename=filename,
+                    is_cached=True,
+                    local_path=pointer_path,
+                    will_download=force_download,
+                )
+            if not force_download:
+                return pointer_path
 
     # Try to get metadata (etag, commit_hash, url, size) from the server.
     # If we can't, a HEAD request error is returned.
@@ -1031,8 +1147,18 @@ def _hf_hub_download_to_cache_dir(
             # Return pointer file if exists
             if commit_hash is not None:
                 pointer_path = _get_pointer_path(storage_folder, commit_hash, relative_filename)
-                if os.path.exists(pointer_path) and not force_download:
-                    return pointer_path
+                if os.path.exists(pointer_path):
+                    if dry_run:
+                        return DryRunFileInfo(
+                            commit_hash=commit_hash,
+                            file_size=os.path.getsize(pointer_path),
+                            filename=filename,
+                            is_cached=True,
+                            local_path=pointer_path,
+                            will_download=force_download,
+                        )
+                    if not force_download:
+                        return pointer_path
 
         # Otherwise, raise appropriate error
         _raise_on_head_call_error(head_call_error, force_download, local_files_only)
@@ -1045,6 +1171,17 @@ def _hf_hub_download_to_cache_dir(
     blob_path = os.path.join(storage_folder, "blobs", etag)
     pointer_path = _get_pointer_path(storage_folder, commit_hash, relative_filename)
 
+    if dry_run:
+        is_cached = os.path.exists(pointer_path) or os.path.exists(blob_path)
+        return DryRunFileInfo(
+            commit_hash=commit_hash,
+            file_size=expected_size,
+            filename=filename,
+            is_cached=is_cached,
+            local_path=pointer_path,
+            will_download=force_download or not is_cached,
+        )
+
     os.makedirs(os.path.dirname(blob_path), exist_ok=True)
     os.makedirs(os.path.dirname(pointer_path), exist_ok=True)
 
@@ -1124,7 +1261,8 @@ def _hf_hub_download_to_local_dir(
     cache_dir: str,
     force_download: bool,
     local_files_only: bool,
-) -> str:
+    dry_run: bool,
+) -> Union[str, DryRunFileInfo]:
     """Download a given file to a local folder, if not already present.
 
     Method should not be called directly. Please use `hf_hub_download` instead.
@@ -1139,13 +1277,23 @@ def _hf_hub_download_to_local_dir(
 
     # Local file exists + metadata exists + commit_hash matches => return file
     if (
-        not force_download
-        and REGEX_COMMIT_HASH.match(revision)
+        REGEX_COMMIT_HASH.match(revision)
         and paths.file_path.is_file()
         and local_metadata is not None
         and local_metadata.commit_hash == revision
     ):
-        return str(paths.file_path)
+        local_file = str(paths.file_path)
+        if dry_run:
+            return DryRunFileInfo(
+                commit_hash=revision,
+                file_size=os.path.getsize(local_file),
+                filename=filename,
+                is_cached=True,
+                local_path=local_file,
+                will_download=force_download,
+            )
+        if not force_download:
+            return local_file
 
     # Local file doesn't exist or commit_hash doesn't match => we need the etag
     (url_to_download, etag, commit_hash, expected_size, xet_file_data, head_call_error) = _get_metadata_or_catch_error(
@@ -1162,11 +1310,24 @@ def _hf_hub_download_to_local_dir(
 
     if head_call_error is not None:
         # No HEAD call but local file exists => default to local file
-        if not force_download and paths.file_path.is_file():
-            logger.warning(
-                f"Couldn't access the Hub to check for update but local file already exists. Defaulting to existing file. (error: {head_call_error})"
-            )
-            return str(paths.file_path)
+        if paths.file_path.is_file():
+            if dry_run or not force_download:
+                logger.warning(
+                    f"Couldn't access the Hub to check for update but local file already exists. Defaulting to existing file. (error: {head_call_error})"
+                )
+            local_path = str(paths.file_path)
+            if dry_run and local_metadata is not None:
+                return DryRunFileInfo(
+                    commit_hash=local_metadata.commit_hash,
+                    file_size=os.path.getsize(local_path),
+                    filename=filename,
+                    is_cached=True,
+                    local_path=local_path,
+                    will_download=force_download,
+                )
+            if not force_download:
+                return local_path
+
         # Otherwise => raise
         _raise_on_head_call_error(head_call_error, force_download, local_files_only)
 
@@ -1181,6 +1342,15 @@ def _hf_hub_download_to_local_dir(
         # etag matches => update metadata and return file
         if local_metadata is not None and local_metadata.etag == etag:
             write_download_metadata(local_dir=local_dir, filename=filename, commit_hash=commit_hash, etag=etag)
+            if dry_run:
+                return DryRunFileInfo(
+                    commit_hash=commit_hash,
+                    file_size=expected_size,
+                    filename=filename,
+                    is_cached=True,
+                    local_path=str(paths.file_path),
+                    will_download=False,
+                )
             return str(paths.file_path)
 
         # metadata is outdated + etag is a sha256
@@ -1192,6 +1362,15 @@ def _hf_hub_download_to_local_dir(
                 file_hash = sha_fileobj(f).hex()
             if file_hash == etag:
                 write_download_metadata(local_dir=local_dir, filename=filename, commit_hash=commit_hash, etag=etag)
+                if dry_run:
+                    return DryRunFileInfo(
+                        commit_hash=commit_hash,
+                        file_size=expected_size,
+                        filename=filename,
+                        is_cached=True,
+                        local_path=str(paths.file_path),
+                        will_download=False,
+                    )
                 return str(paths.file_path)
 
     # Local file doesn't exist or etag isn't a match => retrieve file from remote (or cache)
@@ -1210,8 +1389,28 @@ def _hf_hub_download_to_local_dir(
                 paths.file_path.parent.mkdir(parents=True, exist_ok=True)
                 shutil.copyfile(cached_path, paths.file_path)
             write_download_metadata(local_dir=local_dir, filename=filename, commit_hash=commit_hash, etag=etag)
+            if dry_run:
+                return DryRunFileInfo(
+                    commit_hash=commit_hash,
+                    file_size=expected_size,
+                    filename=filename,
+                    is_cached=True,
+                    local_path=str(paths.file_path),
+                    will_download=False,
+                )
             return str(paths.file_path)
 
+    if dry_run:
+        is_cached = paths.file_path.is_file()
+        return DryRunFileInfo(
+            commit_hash=commit_hash,
+            file_size=expected_size,
+            filename=filename,
+            is_cached=is_cached,
+            local_path=str(paths.file_path),
+            will_download=force_download or not is_cached,
+        )
+
     # Otherwise, let's download the file!
     with WeakFileLock(paths.lock_path):
         paths.file_path.unlink(missing_ok=True)  # delete outdated file first

huggingface_hub/hf_api.py

@@ -60,7 +60,7 @@ from ._commit_api import (
     _warn_on_overwriting_operations,
 )
 from ._inference_endpoints import InferenceEndpoint, InferenceEndpointType
-from ._jobs_api import JobInfo, ScheduledJobInfo, _create_job_spec
+from ._jobs_api import JobInfo, JobSpec, ScheduledJobInfo, _create_job_spec
 from ._space_api import SpaceHardware, SpaceRuntime, SpaceStorage, SpaceVariable
 from ._upload_large_folder import upload_large_folder_internal
 from .community import (
@@ -79,7 +79,7 @@ from .errors import (
     RepositoryNotFoundError,
     RevisionNotFoundError,
 )
-from .file_download import HfFileMetadata, get_hf_file_metadata, hf_hub_url
+from .file_download import DryRunFileInfo, HfFileMetadata, get_hf_file_metadata, hf_hub_url
 from .repocard_data import DatasetCardData, ModelCardData, SpaceCardData
 from .utils import (
     DEFAULT_IGNORE_PATTERNS,
@@ -466,11 +466,15 @@ class WebhookWatchedItem:
 class WebhookInfo:
     """Data structure containing information about a webhook.
 
+    One of `url` or `job` is specified, but not both.
+
     Attributes:
         id (`str`):
             ID of the webhook.
-        url (`str`):
+        url (`str`, *optional*):
             URL of the webhook.
+        job (`JobSpec`, *optional*):
+            Specifications of the Job to trigger.
         watched (`list[WebhookWatchedItem]`):
             List of items watched by the webhook, see [`WebhookWatchedItem`].
         domains (`list[WEBHOOK_DOMAIN_T]`):
@@ -482,7 +486,8 @@ class WebhookInfo:
     """
 
     id: str
-    url: str
+    url: Optional[str]
+    job: Optional[JobSpec]
     watched: list[WebhookWatchedItem]
     domains: list[constants.WEBHOOK_DOMAIN_T]
     secret: Optional[str]
@@ -5110,6 +5115,42 @@ class HfApi:
             endpoint=self.endpoint,
         )
 
+    @overload
+    def hf_hub_download(
+        self,
+        repo_id: str,
+        filename: str,
+        *,
+        subfolder: Optional[str] = None,
+        repo_type: Optional[str] = None,
+        revision: Optional[str] = None,
+        cache_dir: Union[str, Path, None] = None,
+        local_dir: Union[str, Path, None] = None,
+        force_download: bool = False,
+        etag_timeout: float = constants.DEFAULT_ETAG_TIMEOUT,
+        token: Union[bool, str, None] = None,
+        local_files_only: bool = False,
+        dry_run: Literal[False] = False,
+    ) -> str: ...
+
+    @overload
+    def hf_hub_download(
+        self,
+        repo_id: str,
+        filename: str,
+        *,
+        subfolder: Optional[str] = None,
+        repo_type: Optional[str] = None,
+        revision: Optional[str] = None,
+        cache_dir: Union[str, Path, None] = None,
+        local_dir: Union[str, Path, None] = None,
+        force_download: bool = False,
+        etag_timeout: float = constants.DEFAULT_ETAG_TIMEOUT,
+        token: Union[bool, str, None] = None,
+        local_files_only: bool = False,
+        dry_run: Literal[True],
+    ) -> DryRunFileInfo: ...
+
     @validate_hf_hub_args
     def hf_hub_download(
         self,
@@ -5125,7 +5166,8 @@ class HfApi:
         etag_timeout: float = constants.DEFAULT_ETAG_TIMEOUT,
         token: Union[bool, str, None] = None,
         local_files_only: bool = False,
-    ) -> str:
+        dry_run: bool = False,
+    ) -> Union[str, DryRunFileInfo]:
         """Download a given file if it's not already present in the local cache.
 
         The new cache file layout looks like this:
@@ -5192,9 +5234,14 @@ class HfApi:
             local_files_only (`bool`, *optional*, defaults to `False`):
                 If `True`, avoid downloading the file and return the path to the
                 local cached file if it exists.
+            dry_run (`bool`, *optional*, defaults to `False`):
+                If `True`, perform a dry run without actually downloading the file. Returns a
+                [`DryRunFileInfo`] object containing information about what would be downloaded.
 
         Returns:
-            `str`: Local path of file or if networking is off, last version of file cached on disk.
+            `str` or [`DryRunFileInfo`]:
+                - If `dry_run=False`: Local path of file or if networking is off, last version of file cached on disk.
+                - If `dry_run=True`: A [`DryRunFileInfo`] object containing download information.
 
         Raises:
             [`~utils.RepositoryNotFoundError`]
@@ -8815,6 +8862,7 @@ class HfApi:
             >>> print(webhook)
             WebhookInfo(
                 id="654bbbc16f2ec14d77f109cc",
+                job=None,
                 watched=[WebhookWatchedItem(type="user", name="julien-c"), WebhookWatchedItem(type="org", name="HuggingFaceH4")],
                 url="https://webhook.site/a2176e82-5720-43ee-9e06-f91cb4c91548",
                 secret="my-secret",
@@ -8834,7 +8882,8 @@ class HfApi:
 
         webhook = WebhookInfo(
             id=webhook_data["id"],
-            url=webhook_data["url"],
+            url=webhook_data.get("url"),
+            job=JobSpec(**webhook_data["job"]) if webhook_data.get("job") else None,
             watched=watched_items,
             domains=webhook_data["domains"],
             secret=webhook_data.get("secret"),
@@ -8884,7 +8933,8 @@ class HfApi:
         return [
             WebhookInfo(
                 id=webhook["id"],
-                url=webhook["url"],
+                url=webhook.get("url"),
+                job=JobSpec(**webhook["job"]) if webhook.get("job") else None,
                 watched=[WebhookWatchedItem(type=item["type"], name=item["name"]) for item in webhook["watched"]],
                 domains=webhook["domains"],
                 secret=webhook.get("secret"),
@@ -8897,7 +8947,8 @@ class HfApi:
     def create_webhook(
         self,
         *,
-        url: str,
+        url: Optional[str] = None,
+        job_id: Optional[str] = None,
         watched: list[Union[dict, WebhookWatchedItem]],
         domains: Optional[list[constants.WEBHOOK_DOMAIN_T]] = None,
         secret: Optional[str] = None,
@@ -8905,9 +8956,15 @@ class HfApi:
     ) -> WebhookInfo:
         """Create a new webhook.
 
+        The webhook can either send a payload to a URL, or trigger a Job to run on Hugging Face infrastructure.
+        This function should be called with one of `url` or `job_id`, but not both.
+
         Args:
             url (`str`):
                 URL to send the payload to.
+            job_id (`str`):
+                ID of the source Job to trigger with the webhook payload in the environment variable WEBHOOK_PAYLOAD.
+                Additional environment variables are available for convenience: WEBHOOK_REPO_ID, WEBHOOK_REPO_TYPE and WEBHOOK_SECRET.
             watched (`list[WebhookWatchedItem]`):
                 List of [`WebhookWatchedItem`] to be watched by the webhook. It can be users, orgs, models, datasets or spaces.
                 Watched items can also be provided as plain dictionaries.
@@ -8925,6 +8982,8 @@ class HfApi:
                 Info about the newly created webhook.
 
         Example:
+
+            Create a webhook that sends a payload to a URL
             ```python
             >>> from huggingface_hub import create_webhook
             >>> payload = create_webhook(
@@ -8937,6 +8996,43 @@ class HfApi:
             WebhookInfo(
                 id="654bbbc16f2ec14d77f109cc",
                 url="https://webhook.site/a2176e82-5720-43ee-9e06-f91cb4c91548",
+                job=None,
+                watched=[WebhookWatchedItem(type="user", name="julien-c"), WebhookWatchedItem(type="org", name="HuggingFaceH4")],
+                domains=["repo", "discussion"],
+                secret="my-secret",
+                disabled=False,
+            )
+            ```
+
+            Run a Job and then create a webhook that triggers this Job
+            ```python
+            >>> from huggingface_hub import create_webhook, run_job
+            >>> job = run_job(
+            ...     image="ubuntu",
+            ...     command=["bash", "-c", r"echo An event occured in $WEBHOOK_REPO_ID: $WEBHOOK_PAYLOAD"],
+            ... )
+            >>> payload = create_webhook(
+            ...     watched=[{"type": "user", "name": "julien-c"}, {"type": "org", "name": "HuggingFaceH4"}],
+            ...     job_id=job.id,
+            ...     domains=["repo", "discussion"],
+            ...     secret="my-secret",
+            ... )
+            >>> print(payload)
+            WebhookInfo(
+                id="654bbbc16f2ec14d77f109cc",
+                url=None,
+                job=JobSpec(
+                    docker_image='ubuntu',
+                    space_id=None,
+                    command=['bash', '-c', 'echo An event occured in $WEBHOOK_REPO_ID: $WEBHOOK_PAYLOAD'],
+                    arguments=[],
+                    environment={},
+                    secrets=[],
+                    flavor='cpu-basic',
+                    timeout=None,
+                    tags=None,
+                    arch=None
+                ),
                 watched=[WebhookWatchedItem(type="user", name="julien-c"), WebhookWatchedItem(type="org", name="HuggingFaceH4")],
                 domains=["repo", "discussion"],
                 secret="my-secret",
@@ -8946,9 +9042,19 @@ class HfApi:
         """
         watched_dicts = [asdict(item) if isinstance(item, WebhookWatchedItem) else item for item in watched]
 
+        post_webhooks_json = {"watched": watched_dicts, "domains": domains, "secret": secret}
+        if url is not None and job_id is not None:
+            raise ValueError("Set `url` or `job_id` but not both.")
+        elif url is not None:
+            post_webhooks_json["url"] = url
+        elif job_id is not None:
+            post_webhooks_json["jobSourceId"] = job_id
+        else:
+            raise ValueError("Missing argument for webhook: `url` or `job_id`.")
+
         response = get_session().post(
             f"{constants.ENDPOINT}/api/settings/webhooks",
-            json={"watched": watched_dicts, "url": url, "domains": domains, "secret": secret},
+            json=post_webhooks_json,
             headers=self._build_hf_headers(token=token),
         )
         hf_raise_for_status(response)
@@ -8957,7 +9063,8 @@ class HfApi:
 
         webhook = WebhookInfo(
             id=webhook_data["id"],
-            url=webhook_data["url"],
+            url=webhook_data.get("url"),
+            job=JobSpec(**webhook_data["job"]) if webhook_data.get("job") else None,
             watched=watched_items,
             domains=webhook_data["domains"],
             secret=webhook_data.get("secret"),
@@ -9013,6 +9120,7 @@ class HfApi:
             >>> print(updated_payload)
             WebhookInfo(
                 id="654bbbc16f2ec14d77f109cc",
+                job=None,
                 url="https://new.webhook.site/a2176e82-5720-43ee-9e06-f91cb4c91548",
                 watched=[WebhookWatchedItem(type="user", name="julien-c"), WebhookWatchedItem(type="org", name="HuggingFaceH4")],
                 domains=["repo"],
@@ -9036,7 +9144,8 @@ class HfApi:
 
         webhook = WebhookInfo(
             id=webhook_data["id"],
-            url=webhook_data["url"],
+            url=webhook_data.get("url"),
+            job=JobSpec(**webhook_data["job"]) if webhook_data.get("job") else None,
             watched=watched_items,
             domains=webhook_data["domains"],
             secret=webhook_data.get("secret"),
@@ -9068,6 +9177,7 @@ class HfApi:
             >>> enabled_webhook
             WebhookInfo(
                 id="654bbbc16f2ec14d77f109cc",
+                job=None,
                 url="https://webhook.site/a2176e82-5720-43ee-9e06-f91cb4c91548",
                 watched=[WebhookWatchedItem(type="user", name="julien-c"), WebhookWatchedItem(type="org", name="HuggingFaceH4")],
                 domains=["repo", "discussion"],
@@ -9087,7 +9197,8 @@ class HfApi:
 
         webhook = WebhookInfo(
             id=webhook_data["id"],
-            url=webhook_data["url"],
+            url=webhook_data.get("url"),
+            job=JobSpec(**webhook_data["job"]) if webhook_data.get("job") else None,
             watched=watched_items,
             domains=webhook_data["domains"],
             secret=webhook_data.get("secret"),
@@ -9120,6 +9231,7 @@ class HfApi:
             WebhookInfo(
                 id="654bbbc16f2ec14d77f109cc",
                 url="https://webhook.site/a2176e82-5720-43ee-9e06-f91cb4c91548",
+                jon=None,
                 watched=[WebhookWatchedItem(type="user", name="julien-c"), WebhookWatchedItem(type="org", name="HuggingFaceH4")],
                 domains=["repo", "discussion"],
                 secret="my-secret",
@@ -9138,7 +9250,8 @@ class HfApi:
 
         webhook = WebhookInfo(
             id=webhook_data["id"],
-            url=webhook_data["url"],
+            url=webhook_data.get("url"),
+            job=JobSpec(**webhook_data["job"]) if webhook_data.get("job") else None,
             watched=watched_items,
             domains=webhook_data["domains"],
             secret=webhook_data.get("secret"),