ENH Deprecate clone_from behavior

src/huggingface_hub/repository.py

@@ -5,6 +5,7 @@
 import tempfile
 import threading
 import time
+import warnings
 from contextlib import contextmanager
 from pathlib import Path
 from typing import Callable, Dict, Iterator, List, Optional, Tuple, Union
@@ -655,6 +656,12 @@ def clone_from(self, repo_url: str, use_auth_token: Union[bool, str, None] = Non
                                 " on Hugging Face Hub."
                             )
                         else:
+                            warnings.warn(
+                                "Creating a repository through clone_from is deprecated"
+                                "will be removed in v0.10.",
+                                FutureWarning,
+                            )
+
                             self.client.create_repo(
                                 repo_id=repo_id,
                                 token=token,

tests/test_repocard.py

@@ -204,9 +204,12 @@ def setUpClass(cls):
     def setUp(self) -> None:
         self.repo_path = Path(tempfile.mkdtemp())
         self.REPO_NAME = repo_name()
+        url = self._api.create_repo(
+            repo_id=f"{USER}/{self.REPO_NAME}", token=self._token
+        )
         self.repo = Repository(
             self.repo_path / self.REPO_NAME,
-            clone_from=f"{USER}/{self.REPO_NAME}",
+            clone_from=url,
             use_auth_token=self._token,
             git_user="ci",
             git_email="ci@dummy.com",

tests/test_repository.py

@@ -81,16 +81,19 @@ def setUp(self):
         logger.info(
             f"Does {WORKING_REPO_DIR} exist: {os.path.exists(WORKING_REPO_DIR)}"
         )
+
         self.REPO_NAME = repo_name()
-        self._repo_url = self._api.create_repo(
-            repo_id=self.REPO_NAME, token=self._token
-        )
-        self._api.upload_file(
-            path_or_fileobj=BytesIO(b"some initial binary data: \x00\x01"),
-            path_in_repo="random_file.txt",
-            repo_id=f"{USER}/{self.REPO_NAME}",
-            token=self._token,
-        )
+        # skip creation of repository for clone from test
+        if self.shortDescription() != "SKIP CREATION":
+            self._repo_url = self._api.create_repo(
+                repo_id=self.REPO_NAME, token=self._token
+            )
+            self._api.upload_file(
+                path_or_fileobj=BytesIO(b"some initial binary data: \x00\x01"),
+                path_in_repo="random_file.txt",
+                repo_id=f"{USER}/{self.REPO_NAME}",
+                token=self._token,
+            )
 
     def tearDown(self):
         try:
@@ -140,6 +143,19 @@ def test_clone_from_space(self):
                 use_auth_token=self._token,
             )
 
+    def test_clone_from_deprecation_warning(self):
+        """SKIP CREATION"""
+        with pytest.raises(
+            FutureWarning,
+            match="Creating a repository through clone_from is deprecated*",
+        ):
+            Repository(
+                WORKING_REPO_DIR,
+                clone_from=f"{USER}/{uuid.uuid4()}",
+                use_auth_token=self._token,
+                repo_type="dataset",
+            )
+
     def test_clone_from_model(self):
         temp_repo_url = self._api.create_repo(
             repo_id=f"{self.REPO_NAME}-temp", token=self._token, repo_type="model"
@@ -437,9 +453,15 @@ def test_clone_with_endpoint(self):
 
     @retry_endpoint
     def test_clone_with_repo_name_and_org(self):
+        repo_url = self._api.create_repo(
+            token=self._token,
+            repo_id=f"valid_org/{self.REPO_NAME}",
+            repo_type="dataset",
+        )
+
         clone = Repository(
             f"{WORKING_REPO_DIR}/{self.REPO_NAME}",
-            clone_from=f"valid_org/{self.REPO_NAME}",
+            clone_from=repo_url,
             use_auth_token=self._token,
             git_user="ci",
             git_email="ci@dummy.com",
@@ -455,7 +477,7 @@ def test_clone_with_repo_name_and_org(self):
 
         Repository(
             f"{WORKING_REPO_DIR}/{self.REPO_NAME}",
-            clone_from=f"valid_org/{self.REPO_NAME}",
+            clone_from=repo_url,
             use_auth_token=self._token,
             git_user="ci",
             git_email="ci@dummy.com",
@@ -1698,6 +1720,11 @@ def test_clone_with_endpoint(self):
 
     @retry_endpoint
     def test_clone_with_repo_name_and_org(self):
+        self._api.create_repo(
+            token=self._token,
+            repo_id=f"valid_org/{self.REPO_NAME}",
+            repo_type="dataset",
+        )
         clone = Repository(
             f"{WORKING_DATASET_DIR}/{self.REPO_NAME}",
             clone_from=f"valid_org/{self.REPO_NAME}",
@@ -1761,7 +1788,7 @@ def test_clone_with_repo_name_and_no_namespace(self):
         self.assertRaises(
             OSError,
             Repository,
-            f"{WORKING_DATASET_DIR}/{self.REPO_NAME}",
+            f"{WORKING_REPO_DIR}/{self.REPO_NAME}",
             clone_from=self.REPO_NAME,
             repo_type="dataset",
             use_auth_token=self._token,

tests/test_snapshot_download.py

@@ -36,9 +36,10 @@ def setUp(self) -> None:
         if os.path.exists(REPO_NAME):
             shutil.rmtree(REPO_NAME, onerror=set_write_permission_and_retry)
         logger.info(f"Does {REPO_NAME} exist: {os.path.exists(REPO_NAME)}")
+        url = self._api.create_repo(repo_id=f"{USER}/{REPO_NAME}", token=self._token)
         repo = Repository(
             REPO_NAME,
-            clone_from=f"{USER}/{REPO_NAME}",
+            clone_from=url,
             use_auth_token=self._token,
             git_user="ci",
             git_email="ci@dummy.com",