fix(dataset): set isBasedOn for renku datasets (#1617)

(cherry picked from commit 3aee6b8)

Author: mohammad-alisafaee

renku/core/commands/dataset.py

@@ -160,7 +160,6 @@ def add_file(
     sources=(),
     destination="",
     ref=None,
-    with_metadata=None,
     urlscontext=contextlib.nullcontext,
     commit_message=None,
     progress=None,
@@ -178,7 +177,6 @@ def add_file(
         sources=sources,
         destination=destination,
         ref=ref,
-        with_metadata=with_metadata,
         urlscontext=urlscontext,
         progress=progress,
         interactive=interactive,
@@ -257,8 +255,6 @@ def _add_to_dataset(
                     click.echo(WARNING + msg)
 
             if with_metadata:
-                for file_ in dataset.files:
-                    file_.based_on = None
                 # dataset has the correct list of files
                 with_metadata.files = dataset.files
                 with_metadata.url = dataset._id
@@ -306,7 +302,7 @@ def file_unlink(client, name, include, exclude, interactive=False, yes=False, co
             (
                 "include or exclude filters not found.\n"
                 "Check available filters with `renku dataset unlink --help`\n"
-                "Hint: `renku dataset unlink mydataset -I myfile`"
+                "Hint: `renku dataset unlink my-dataset -I path`"
             )
         )
 

renku/core/models/datasets.py

@@ -346,16 +346,16 @@ class Dataset(Entity, CreatorMixin, ReferenceMixin):
 
     EDITABLE_FIELDS = [
         "creators",
+        "date_created",
         "date_published",
         "description",
+        "files",
         "in_language",
         "keywords",
         "license",
         "title",
         "url",
         "version",
-        "date_created",
-        "files",
     ]
 
     _id = attr.ib(default=None, kw_only=True)
@@ -476,6 +476,8 @@ def update_metadata(self, other_dataset):
             if val:
                 setattr(self, field_, val)
 
+        self.same_as = other_dataset.same_as
+
         return self
 
     def update_files(self, files):

tests/cli/test_integration_datasets.py

@@ -312,9 +312,13 @@ def test_import_renku_dataset_preserves_directory_hierarchy(runner, project, cli
     assert 0 == runner.invoke(cli, ["dataset", "import", "--yes", "--name", "remote", url]).exit_code
 
     dataset = client.load_dataset("remote")
-    assert (client.path / dataset.data_dir / "README.md").exists()
-    assert (client.path / dataset.data_dir / "python" / "data" / "README.md").exists()
-    assert (client.path / dataset.data_dir / "r" / "data" / "README.md").exists()
+    paths = ["README.md", os.path.join("python", "data", "README.md"), os.path.join("r", "data", "README.md")]
+
+    for path in paths:
+        assert (client.path / dataset.data_dir / path).exists()
+        file_ = dataset.find_file(dataset.data_dir / path)
+        assert file_.based_on
+        assert file_.based_on.path.endswith(path)
 
 
 @pytest.mark.integration