address review comments

renku/core/models/datasets.py

@@ -894,24 +894,27 @@ def is_dataset_name_valid(name):
 
 def generate_default_name(dataset_title, dataset_version=None):
     """Get dataset name."""
+    max_length = 24
     # For compatibility with older versions use name as name
     # if it is valid; otherwise, use encoded name
     if is_dataset_name_valid(dataset_title):
         return dataset_title
 
     slug = get_slug(dataset_title)
-    slug = slug.lower()[:24]
+    name = slug[:max_length]
 
     if dataset_version:
-        version_slug = get_slug(dataset_version)
-        return f"{slug}_{version_slug}"
+        max_version_length = 10
+        version_slug = get_slug(dataset_version)[:max_version_length]
+        name = f"{name[:-(len(version_slug) + 1)]}_{version_slug}"
 
-    return slug
+    return name
 
 
 def get_slug(name):
     """Create a slug from name."""
-    no_space = re.sub(r"\s+", "_", name)
+    lower_case = name.lower()
+    no_space = re.sub(r"\s+", "_", lower_case)
     normalized = unicodedata.normalize("NFKD", no_space).encode("ascii", "ignore").decode("utf-8")
     no_invalid_characters = re.sub(r"[^a-zA-Z0-9._-]", "_", normalized)
     no_duplicates = re.sub(r"([._-])[._-]+", r"\1", no_invalid_characters)

tests/cli/test_integration_datasets.py

@@ -42,13 +42,13 @@
     [
         {
             "doi": "10.5281/zenodo.2658634",
-            "name": "pyndl_naive_discriminati_v0.6.4",
+            "name": "pyndl_naive_discr_v0.6.4",
             "creator": "Konstantin Sering, Marc Weitz, David-Elias Künstle, Lennart Schneider",
             "version": "v0.6.4",
         },
         {
             "doi": "10.7910/DVN/F4NUMR",
-            "name": "replication_data_for_cam_2.2",
+            "name": "replication_data_for_2.2",
             "creator": "James Druckman, Martin Kifer, Michael Parkin",
             "version": "2",
         },
@@ -171,7 +171,7 @@ def test_dataset_import_real_doi_warnings(runner, project, sleep_after):
 
     result = runner.invoke(cli, ["dataset", "ls"])
     assert 0 == result.exit_code, result.output + str(result.stderr_bytes)
-    assert "pyndl_naive_discriminat" in result.output
+    assert "pyndl_naive_discr_v0.7.2" in result.output
 
 
 @pytest.mark.parametrize(

tests/core/commands/test_dataset.py

@@ -262,6 +262,7 @@ def test_cannot_mutate_immutable_dataset():
 @pytest.mark.parametrize(
     "name, slug",
     [
+        ("UPPER-CASE", "upper-case"),
         (" a name  ", "a_name"),
         ("..non-alphanumeric start or end-_", "non-alphanumeric_start_or_end"),
         ("double ..__._- non-alphanumeric", "double_non-alphanumeric"),