Enable FutureWarnings/DeprecationWarnings as errors for dask_cudf (#1…

…5634)

Part of rapidsai/build-planning#26

Authors:
  - Matthew Roeschke (https://github.com/mroeschke)
  - GALI PREM SAGAR (https://github.com/galipremsagar)

Approvers:
  - GALI PREM SAGAR (https://github.com/galipremsagar)
  - Bradley Dice (https://github.com/bdice)
  - Lawrence Mitchell (https://github.com/wence-)

URL: #15634

python/cudf/cudf/core/index.py

@@ -1119,14 +1119,26 @@ def _concat(cls, objs):
             assert (
                 PANDAS_LT_300
             ), "Need to drop after pandas-3.0 support is added."
-            warnings.warn(
+            warning_msg = (
                 "The behavior of array concatenation with empty entries is "
                 "deprecated. In a future version, this will no longer exclude "
                 "empty items when determining the result dtype. "
                 "To retain the old behavior, exclude the empty entries before "
-                "the concat operation.",
-                FutureWarning,
+                "the concat operation."
             )
+            # Warn only if the type might _actually_ change
+            if len(non_empties) == 0:
+                if not all(objs[0].dtype == index.dtype for index in objs[1:]):
+                    warnings.warn(warning_msg, FutureWarning)
+            else:
+                common_all_type = find_common_type(
+                    [index.dtype for index in objs]
+                )
+                common_non_empty_type = find_common_type(
+                    [index.dtype for index in non_empties]
+                )
+                if common_all_type != common_non_empty_type:
+                    warnings.warn(warning_msg, FutureWarning)
         if all(isinstance(obj, RangeIndex) for obj in non_empties):
             result = _concat_range_index(non_empties)
         else:

python/cudf/cudf/tests/test_index.py

@@ -1039,7 +1039,9 @@ def test_index_append(data, other):
         (len(data) == 0 or len(other) == 0) and pd_data.dtype != pd_other.dtype
     ):
         expected = pd_data.append(pd_other)
-    with expect_warning_if(len(data) == 0 or len(other) == 0):
+    with expect_warning_if(
+        (len(data) == 0 or len(other) == 0) and gd_data.dtype != gd_other.dtype
+    ):
         actual = gd_data.append(gd_other)
     if len(data) == 0 and len(other) == 0:
         # Pandas default dtype to "object" for empty list
@@ -1237,7 +1239,10 @@ def test_index_append_list(data, other):
         and (any(d.dtype != data.dtype for d in other))
     ):
         expected = pd_data.append(pd_other)
-    with expect_warning_if(len(data) == 0 or any(len(d) == 0 for d in other)):
+    with expect_warning_if(
+        (len(data) == 0 or any(len(d) == 0 for d in other))
+        and (any(d.dtype != data.dtype for d in other))
+    ):
         actual = gd_data.append(gd_other)
 
     assert_eq(expected, actual)
@@ -2817,8 +2822,7 @@ def test_index_methods(index, func):
 
     if func == "append":
         expected = pidx.append(other=pidx)
-        with expect_warning_if(len(gidx) == 0):
-            actual = gidx.append(other=gidx)
+        actual = gidx.append(other=gidx)
     else:
         expected = getattr(pidx, func)()
         actual = getattr(gidx, func)()

python/dask_cudf/dask_cudf/io/tests/test_json.py

@@ -84,9 +84,8 @@ def test_read_json_nested(tmp_path):
         }
     )
     kwargs = dict(orient="records", lines=True)
-    with tmp_path / "data.json" as f, dask.config.set(
-        {"dataframe.convert-string": False}
-    ):
+    f = tmp_path / "data.json"
+    with dask.config.set({"dataframe.convert-string": False}):
         df.to_json(f, **kwargs)
         # Ensure engine='cudf' is tested.
         actual = dask_cudf.read_json(f, engine="cudf", **kwargs)

python/dask_cudf/dask_cudf/tests/test_accessor.py

@@ -543,7 +543,7 @@ def test_struct_explode(data):
 
 
 def test_tz_localize():
-    data = Series(date_range("2000-04-01", "2000-04-03", freq="H"))
+    data = Series(date_range("2000-04-01", "2000-04-03", freq="h"))
     expect = data.dt.tz_localize(
         "US/Eastern", ambiguous="NaT", nonexistent="NaT"
     )
@@ -560,8 +560,8 @@ def test_tz_localize():
 @pytest.mark.parametrize(
     "data",
     [
-        date_range("2000-04-01", "2000-04-03", freq="H").tz_localize("UTC"),
-        date_range("2000-04-01", "2000-04-03", freq="H").tz_localize(
+        date_range("2000-04-01", "2000-04-03", freq="h").tz_localize("UTC"),
+        date_range("2000-04-01", "2000-04-03", freq="h").tz_localize(
             "US/Eastern"
         ),
     ],

python/dask_cudf/dask_cudf/tests/test_groupby.py

@@ -233,7 +233,7 @@ def test_groupby_split_out(split_out, column):
     gddf = dask_cudf.from_cudf(gdf, npartitions=3)
 
     ddf_result = (
-        ddf.groupby(column)
+        ddf.groupby(column, observed=True)
         .a.mean(split_out=split_out)
         .compute()
         .sort_values()
@@ -368,10 +368,10 @@ def test_groupby_dropna_dask(dropna, by):
 
     if dropna is None:
         dask_cudf_result = gddf.groupby(by).e.sum()
-        dask_result = ddf.groupby(by).e.sum()
+        dask_result = ddf.groupby(by, observed=True).e.sum()
     else:
         dask_cudf_result = gddf.groupby(by, dropna=dropna).e.sum()
-        dask_result = ddf.groupby(by, dropna=dropna).e.sum()
+        dask_result = ddf.groupby(by, dropna=dropna, observed=True).e.sum()
 
     dd.assert_eq(dask_cudf_result, dask_result)
 
@@ -505,7 +505,7 @@ def test_groupby_reset_index_dtype():
     a = df.groupby("a").agg({"b": ["count"]})
 
     assert a.index.dtype == "int8"
-    assert a.reset_index().dtypes[0] == "int8"
+    assert a.reset_index().dtypes.iloc[0] == "int8"
 
 
 def test_groupby_reset_index_names():
@@ -563,7 +563,7 @@ def test_groupby_categorical_key():
     # (See: https://github.com/dask/dask/issues/9515)
     expect = (
         ddf.compute()
-        .groupby("name", sort=True)
+        .groupby("name", sort=True, observed=True)
         .agg({"x": ["mean", "max"], "y": ["mean", "count"]})
     )
     dd.assert_eq(expect, got)

python/dask_cudf/dask_cudf/tests/test_join.py

@@ -66,8 +66,12 @@ def test_join_inner(left_nrows, right_nrows, left_nkeys, right_nkeys):
     def gather(df, grows):
         grows[df["x"].values[0]] = (set(df.al), set(df.ar))
 
-    expect.reset_index().groupby("x").apply(partial(gather, grows=expect_rows))
-    expect.reset_index().groupby("x").apply(partial(gather, grows=got_rows))
+    expect.reset_index().groupby("x")[["x", "al", "ar"]].apply(
+        partial(gather, grows=expect_rows)
+    )
+    expect.reset_index().groupby("x")[["x", "al", "ar"]].apply(
+        partial(gather, grows=got_rows)
+    )
 
     assert got_rows == expect_rows
 
@@ -127,9 +131,13 @@ def gather(df, grows):
 
         grows[df["x"].values[0]] = (cola, colb)
 
-    expect.reset_index().groupby("x").apply(partial(gather, grows=expect_rows))
+    expect.reset_index().groupby("x")[["x", "al", "ar"]].apply(
+        partial(gather, grows=expect_rows)
+    )
 
-    expect.reset_index().groupby("x").apply(partial(gather, grows=got_rows))
+    expect.reset_index().groupby("x")[["x", "al", "ar"]].apply(
+        partial(gather, grows=got_rows)
+    )
 
     for k in expect_rows:
         np.testing.assert_array_equal(expect_rows[k][0], got_rows[k][0])

python/dask_cudf/pyproject.toml

@@ -107,3 +107,13 @@ skip = [
     "build",
     "dist",
 ]
+
+[tool.pytest.ini_options]
+filterwarnings = [
+    "error::FutureWarning",
+    "error::DeprecationWarning",
+    "ignore:create_block_manager_from_blocks is deprecated and will be removed in a future version. Use public APIs instead.:DeprecationWarning",
+    # https://github.com/dask/partd/blob/main/partd/pandas.py#L198
+    "ignore:Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.:DeprecationWarning",
+    "ignore:String support for `aggregate_files` is experimental. Behavior may change in the future.:FutureWarning:dask",
+]