Fix GitDagBundle to support https (apache#46073)

airflow/config_templates/config.yml

@@ -2667,6 +2667,15 @@ dag_processor:
                     "repo_url": "[email protected]:example.com/my-dags.git",
                     "tracking_ref": "main",
                     "refresh_interval": 0
+            },
+            {
+                "name": "my-git-repo",
+                "classpath": "airflow.dag_processing.bundles.git.GitDagBundle",
+                "kwargs": {
+                    "subdir": "dags",
+                    "repo_url": "https://github.com/example/my-dags.git",
+                    "tracking_ref": "main",
+                    "refresh_interval": 0
             }
             ]
       default: >

airflow/dag_processing/bundles/git.py

@@ -140,9 +140,11 @@ def initialize(self) -> None:
             raise AirflowException(f"Connection {self.git_conn_id} doesn't have a host url")
         if isinstance(self.repo_url, os.PathLike):
             self._initialize()
-        elif not self.repo_url.startswith("git@") or not self.repo_url.endswith(".git"):
+        elif not (
+            self.repo_url.startswith("git@") or self.repo_url.startswith("https")
+        ) or not self.repo_url.endswith(".git"):
             raise AirflowException(
-                f"Invalid git URL: {self.repo_url}. URL must start with git@ and end with .git"
+                f"Invalid git URL: {self.repo_url}. URL must start with git@ or https and end with .git"
             )
         else:
             self._initialize()
@@ -236,6 +238,8 @@ def view_url(self, version: str | None = None) -> str | None:
             return None
         if url.startswith("git@"):
             url = self._convert_git_ssh_url_to_https(url)
+        if url.startswith("https"):
+            url = url.replace(".git", "")
         parsed_url = urlparse(url)
         host = parsed_url.hostname
         if not host:

tests/dag_processing/test_dag_bundles.py

@@ -399,7 +399,8 @@ def test_repo_url_validation_for_ssh(self, mock_hook, repo_url, session):
             tracking_ref=GIT_DEFAULT_BRANCH,
         )
         with pytest.raises(
-            AirflowException, match=f"Invalid git URL: {repo_url}. URL must start with git@ and end with .git"
+            AirflowException,
+            match=f"Invalid git URL: {repo_url}. URL must start with git@ or https and end with .git",
         ):
             bundle.initialize()
 
@@ -413,6 +414,7 @@ def test_repo_url_validation_for_ssh(self, mock_hook, repo_url, session):
                 "[email protected]:apache/airflow.git",
                 "https://myorg.github.com/apache/airflow/tree/0f0f0f",
             ),
+            ("https://github.com/apache/airflow.git", "https://github.com/apache/airflow/tree/0f0f0f"),
         ],
     )
     @mock.patch("airflow.dag_processing.bundles.git.Repo")