diff --git a/ixmp/__init__.py b/ixmp/__init__.py
index acfc57117..a3c881884 100644
--- a/ixmp/__init__.py
+++ b/ixmp/__init__.py
@@ -69,8 +69,9 @@
 
 def __getattr__(name):
     if name == "utils":
-        import ixmp.util
+        # Import via the old name to trigger DeprecatedPathFinder
+        import ixmp.utils as util  # type: ignore [import-not-found]
 
-        return ixmp.util
+        return util
     else:
         raise AttributeError(name)
diff --git a/ixmp/backend/jdbc.py b/ixmp/backend/jdbc.py
index 7ba6b8b5d..637783bd1 100644
--- a/ixmp/backend/jdbc.py
+++ b/ixmp/backend/jdbc.py
@@ -4,6 +4,7 @@
 import re
 from collections import ChainMap
 from collections.abc import Iterable, Sequence
+from contextlib import contextmanager
 from copy import copy
 from functools import lru_cache
 from pathlib import Path, PurePosixPath
@@ -141,6 +142,15 @@ def _raise_jexception(exc, msg="unhandled Java exception: "):
     raise RuntimeError(msg) from None
 
 
+@contextmanager
+def _handle_jexception():
+    """Context manager form of :func:`_raise_jexception`."""
+    try:
+        yield
+    except java.Exception as e:
+        _raise_jexception(e)
+
+
 @lru_cache
 def _fixed_index_sets(scheme: str) -> Mapping[str, List[str]]:
     """Return index sets for items that are fixed in the Java code.
@@ -475,10 +485,8 @@ def get_scenario_names(self) -> Generator[str, None, None]:
 
     def get_scenarios(self, default, model, scenario):
         # List<Map<String, Object>>
-        try:
+        with _handle_jexception():
             scenarios = self.jobj.getScenarioList(default, model, scenario)
-        except java.IxException as e:
-            _raise_jexception(e)
 
         for s in scenarios:
             data = []
@@ -537,7 +545,7 @@ def read_file(self, path, item_type: ItemType, **kwargs):
         if path.suffix == ".gdx" and item_type is ItemType.MODEL:
             kw = {"check_solution", "comment", "equ_list", "var_list"}
 
-            if not isinstance(ts, Scenario):
+            if not isinstance(ts, Scenario):  # pragma: no cover
                 raise ValueError("read from GDX requires a Scenario object")
             elif set(kwargs.keys()) != kw:
                 raise ValueError(
@@ -556,10 +564,8 @@ def read_file(self, path, item_type: ItemType, **kwargs):
             if len(kwargs):
                 raise ValueError(f"extra keyword arguments {kwargs}")
 
-            try:
+            with _handle_jexception():
                 self.jindex[ts].readSolutionFromGDX(*args)
-            except java.Exception as e:
-                _raise_jexception(e)
 
             self.cache_invalidate(ts)
         else:
@@ -601,7 +607,7 @@ def write_file(self, path, item_type: ItemType, **kwargs):
         if path.suffix == ".gdx" and item_type is ItemType.SET | ItemType.PAR:
             if len(filters):
                 raise NotImplementedError("write to GDX with filters")
-            elif not isinstance(ts, Scenario):
+            elif not isinstance(ts, Scenario):  # pragma: no cover
                 raise ValueError("write to GDX requires a Scenario object")
 
             # include_var_equ=False -> do not include variables/equations in GDX
@@ -688,10 +694,8 @@ def init(self, ts, annotation):
 
         # Call either newTimeSeries or newScenario
         method = getattr(self.jobj, "new" + klass)
-        try:
+        with _handle_jexception():
             jobj = method(ts.model, ts.scenario, *args)
-        except java.IxException as e:  # pragma: no cover
-            _raise_jexception(e)
 
         self._index_and_set_attrs(jobj, ts)
 
@@ -703,12 +707,11 @@ def get(self, ts):
 
         # either getTimeSeries or getScenario
         method = getattr(self.jobj, "get" + ts.__class__.__name__)
-        try:
+
+        # Re-raise as a ValueError for bad model or scenario name, or other
+        with _handle_jexception():
             # Either the 2- or 3- argument form, depending on args
             jobj = method(*args)
-        except java.IxException as e:
-            # Re-raise as a ValueError for bad model or scenario name, or other
-            _raise_jexception(e)
 
         self._index_and_set_attrs(jobj, ts)
 
@@ -719,10 +722,8 @@ def del_ts(self, ts):
         self.gc()
 
     def check_out(self, ts, timeseries_only):
-        try:
+        with _handle_jexception():
             self.jindex[ts].checkOut(timeseries_only)
-        except java.IxException as e:
-            _raise_jexception(e)
 
     def commit(self, ts, comment):
         try:
@@ -1120,10 +1121,8 @@ def get_meta(
         if version is not None:
             version = java.Long(version)
 
-        try:
+        with _handle_jexception():
             meta = self.jobj.getMeta(model, scenario, version, strict)
-        except java.IxException as e:
-            _raise_jexception(e)
 
         return {entry.getKey(): _unwrap(entry.getValue()) for entry in meta.entrySet()}
 
@@ -1142,10 +1141,8 @@ def set_meta(
         for k, v in meta.items():
             jmeta.put(str(k), _wrap(v))
 
-        try:
+        with _handle_jexception():
             self.jobj.setMeta(model, scenario, version, jmeta)
-        except java.IxException as e:
-            _raise_jexception(e)
 
     def remove_meta(
         self,
diff --git a/ixmp/core/scenario.py b/ixmp/core/scenario.py
index dc539dea8..0ac2bcc78 100644
--- a/ixmp/core/scenario.py
+++ b/ixmp/core/scenario.py
@@ -478,7 +478,7 @@ def init_item(
 
         if idx_names and len(idx_names) != len(idx_sets):
             raise ValueError(
-                f"index names {repr(idx_names)} must have same length as index sets"
+                f"index names {repr(idx_names)} must have the same length as index sets"
                 f" {repr(idx_sets)}"
             )
 
diff --git a/ixmp/testing/resource.py b/ixmp/testing/resource.py
index c7e9da121..ee9a5f6fb 100644
--- a/ixmp/testing/resource.py
+++ b/ixmp/testing/resource.py
@@ -7,7 +7,7 @@
     import resource
 
     has_resource_module = True
-except ImportError:
+except ImportError:  # pragma: no cover
     # Windows
     has_resource_module = False
 
diff --git a/ixmp/tests/backend/test_jdbc.py b/ixmp/tests/backend/test_jdbc.py
index afd254e94..b17147ef6 100644
--- a/ixmp/tests/backend/test_jdbc.py
+++ b/ixmp/tests/backend/test_jdbc.py
@@ -201,6 +201,11 @@ def test_write_file(self, tmp_path, be):
         with pytest.raises(NotImplementedError):
             be.write_file(tmp_path / "test.csv", ixmp.ItemType.ALL, filters={})
 
+    # Specific to JDBCBackend
+    def test_gc(self, monkeypatch, be):
+        monkeypatch.setattr(ixmp.backend.jdbc, "_GC_AGGRESSIVE", True)
+        be.gc()
+
 
 def test_exceptions(test_mp):
     """Ensure that Python exceptions are raised for some actions."""
diff --git a/ixmp/tests/core/test_scenario.py b/ixmp/tests/core/test_scenario.py
index 8c6ec1d97..3eaa2dd50 100644
--- a/ixmp/tests/core/test_scenario.py
+++ b/ixmp/tests/core/test_scenario.py
@@ -147,7 +147,7 @@ def test_init_set(self, scen):
         with pytest.raises(ValueError, match="'foo' already exists"):
             scen.init_set("foo")
 
-    def test_init_par(self, scen):
+    def test_init_par(self, scen) -> None:
         scen = scen.clone(keep_solution=False)
         scen.check_out()
 
@@ -157,6 +157,10 @@ def test_init_par(self, scen):
         # Return type of idx_sets is still list
         assert scen.idx_sets("foo") == ["i", "j"]
 
+        # Mismatched sets and names
+        with pytest.raises(ValueError, match="must have the same length"):
+            scen.init_par("bar", idx_sets=("i", "j"), idx_names=("a", "b", "c"))
+
     def test_init_scalar(self, scen):
         scen2 = scen.clone(keep_solution=False)
         scen2.check_out()
@@ -243,14 +247,15 @@ def test_par(self, scen):
 
     def test_items0(self, scen):
         # Without filters
-        iterator = scen.items()
+        with pytest.warns(FutureWarning, match="par_data=False will be the default"):
+            iterator = scen.items()
 
         # next() can be called → an iterator was returned
         next(iterator)
 
         # Iterator returns the expected parameter names
         exp = ["a", "b", "d", "f"]
-        for i, (name, data) in enumerate(scen.items()):
+        for i, (name, data) in enumerate(scen.items(par_data=True)):
             # Name is correct in the expected order
             assert exp[i] == name
             # Data is one of the return types of .par()
@@ -260,7 +265,7 @@ def test_items0(self, scen):
         assert i == 3
 
         # With filters
-        iterator = scen.items(filters=dict(i=["seattle"]))
+        iterator = scen.items(filters=dict(i=["seattle"]), par_data=True)
         exp = [("a", 1), ("d", 3)]
         for i, (name, data) in enumerate(iterator):
             # Name is correct in the expected order
diff --git a/ixmp/tests/report/test_util.py b/ixmp/tests/report/test_util.py
new file mode 100644
index 000000000..47ff277ae
--- /dev/null
+++ b/ixmp/tests/report/test_util.py
@@ -0,0 +1,5 @@
+def test_import_rename_dims():
+    """RENAME_DIMS can be imported from .report.util, though defined in .common."""
+    from ixmp.report.util import RENAME_DIMS
+
+    assert isinstance(RENAME_DIMS, dict)
diff --git a/ixmp/tests/test_util.py b/ixmp/tests/test_util.py
index 4c9c8f1de..12c71a694 100644
--- a/ixmp/tests/test_util.py
+++ b/ixmp/tests/test_util.py
@@ -20,6 +20,13 @@ def test_import(self):
         ):
             import ixmp.reporting.computations  # type: ignore  # noqa: F401
 
+    @pytest.mark.filterwarnings("ignore")
+    def test_import1(self):
+        """utils can be imported from ixmp, but raises DeprecationWarning."""
+        from ixmp import utils
+
+        assert "diff" in dir(utils)
+
     def test_importerror(self):
         with pytest.warns(DeprecationWarning), pytest.raises(ImportError):
             import ixmp.reporting.foo  # type: ignore  # noqa: F401
diff --git a/ixmp/util/__init__.py b/ixmp/util/__init__.py
index 65d8ee29e..f6c1798a9 100644
--- a/ixmp/util/__init__.py
+++ b/ixmp/util/__init__.py
@@ -115,8 +115,8 @@ def diff(a, b, filters=None) -> Iterator[Tuple[str, pd.DataFrame]]:
     """
     # Iterators; index 0 corresponds to `a`, 1 to `b`
     items = [
-        a.items(filters=filters, type=ItemType.PAR),
-        b.items(filters=filters, type=ItemType.PAR),
+        a.items(filters=filters, type=ItemType.PAR, par_data=True),
+        b.items(filters=filters, type=ItemType.PAR, par_data=True),
     ]
     # State variables for loop
     name = ["", ""]