From 2a0177902210d3cb5ce982f9cc114c5abb852f30 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Simon=20H=C3=B8xbro=20Hansen?= <simon.hansen@me.com>
Date: Wed, 31 Jan 2024 18:08:33 +0100
Subject: [PATCH 01/14] Pin pytest for now.

---
 setup.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/setup.py b/setup.py
index 47ca719bf..bd4bdc871 100644
--- a/setup.py
+++ b/setup.py
@@ -51,7 +51,7 @@
         'nbsmoke[verify] >0.5',
         'netcdf4',
         'pyarrow',
-        'pytest',
+        'pytest <8',  # Fails lint with IPynbFile is deprecated
         'pytest-benchmark',
         'pytest-cov',
         'rasterio',

From af10be7a099b2eb777979ac3259f516e01de6984 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Simon=20H=C3=B8xbro=20Hansen?= <simon.hansen@me.com>
Date: Wed, 31 Jan 2024 18:18:32 +0100
Subject: [PATCH 02/14] Update to ddf =
 ddf.repartition(npartitions=npartitions)

---
 datashader/tests/test_dask.py | 114 +++++++++++++++++-----------------
 1 file changed, 57 insertions(+), 57 deletions(-)

diff --git a/datashader/tests/test_dask.py b/datashader/tests/test_dask.py
index 38778c039..2c84021d1 100644
--- a/datashader/tests/test_dask.py
+++ b/datashader/tests/test_dask.py
@@ -124,7 +124,7 @@ def test_gpu_dependencies():
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_count(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     out = xr.DataArray(np.array([[5, 5], [5, 5]], dtype='i4'),
                        coords=coords, dims=dims)
@@ -140,7 +140,7 @@ def test_count(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_any(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     out = xr.DataArray(np.array([[True, True], [True, True]]),
                        coords=coords, dims=dims)
@@ -155,7 +155,7 @@ def test_any(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_sum(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     out = xr.DataArray(
         values(df_pd.i32).reshape((2, 2, 5)).sum(axis=2, dtype='f8').T,
@@ -173,7 +173,7 @@ def test_sum(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_first(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     out = xr.DataArray([[0, 10], [5, 15]], coords=coords, dims=dims)
     assert_eq_xr(c.points(ddf, 'x', 'y', ds.first('i32')), out)
@@ -185,7 +185,7 @@ def test_first(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_last(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     out = xr.DataArray([[4, 14], [9, 19]], coords=coords, dims=dims)
     assert_eq_xr(c.points(ddf, 'x', 'y', ds.last('i32')), out)
@@ -197,7 +197,7 @@ def test_last(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_min(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     out = xr.DataArray(
         values(df_pd.i64).reshape((2, 2, 5)).min(axis=2).astype('f8').T,
@@ -211,7 +211,7 @@ def test_min(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_max(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     out = xr.DataArray(
         values(df_pd.i64).reshape((2, 2, 5)).max(axis=2).astype('f8').T,
@@ -225,7 +225,7 @@ def test_max(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_min_row_index(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     out = xr.DataArray([[0, 10], [5, 15]], coords=coords, dims=dims)
     assert_eq_xr(c.points(ddf, 'x', 'y', ds._min_row_index()), out)
@@ -234,7 +234,7 @@ def test_min_row_index(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_max_row_index(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     out = xr.DataArray([[4, 14], [9, 19]], coords=coords, dims=dims)
     assert_eq_xr(c.points(ddf, 'x', 'y', ds._max_row_index()), out)
@@ -243,7 +243,7 @@ def test_max_row_index(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_min_n(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[-3, -1, 0, 4, nan, nan], [-13, -11, 10, 12, 14, nan]],
                          [[-9, -7, -5, 6, 8, nan], [-19, -17, -15, 16, 18, nan]]])
@@ -258,7 +258,7 @@ def test_min_n(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_max_n(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[4, 0, -1, -3, nan, nan], [14, 12, 10, -11, -13, nan]],
                          [[8, 6, -5, -7, -9, nan], [18, 16, -15, -17, -19, nan]]])
@@ -273,7 +273,7 @@ def test_max_n(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_min_n_row_index(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[0, 1, 2, 3, 4, -1], [10, 11, 12, 13, 14, -1]],
                          [[5, 6, 7, 8, 9, -1], [15, 16, 17, 18, 19, -1]]])
@@ -288,7 +288,7 @@ def test_min_n_row_index(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_max_n_row_index(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[4, 3, 2, 1, 0, -1], [14, 13, 12, 11, 10, -1]],
                          [[9, 8, 7, 6, 5, -1], [19, 18, 17, 16, 15, -1]]])
@@ -303,7 +303,7 @@ def test_max_n_row_index(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_first_n(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[0, -1, -3, 4, nan, nan], [10, -11, 12, -13, 14, nan]],
                          [[-5, 6, -7, 8, -9, nan], [-15, 16, -17, 18, -19, nan]]])
@@ -319,7 +319,7 @@ def test_first_n(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_last_n(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[4, -3, -1, 0, nan, nan], [14, -13, 12, -11, 10, nan]],
                          [[-9, 8, -7, 6, -5, nan], [-19, 18, -17, 16, -15, nan]]])
@@ -334,7 +334,7 @@ def test_last_n(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_count(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol = np.array([[[2, 1, 1, 1], [1, 1, 2, 1]], [[1, 2, 1, 1], [1, 1, 1, 2]]], dtype=np.uint32)
     assert_eq_ndarray(c.points(ddf, 'x', 'y', ds.by('cat2')).data, sol)
@@ -347,7 +347,7 @@ def test_categorical_count(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_min(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol_int = np.array([[[0, 1, 2, 3], [12, 13, 10, 11]], [[8, 5, 6, 7], [16, 17, 18, 15]]],
                        dtype=np.float64)
@@ -361,7 +361,7 @@ def test_categorical_min(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_max(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol_int = np.array([[[4, 1, 2, 3], [12, 13, 14, 11]], [[8, 9, 6, 7], [16, 17, 18, 19]]],
                        dtype=np.float64)
@@ -375,7 +375,7 @@ def test_categorical_max(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_min_n(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[[0, 4, nan], [1, nan, nan], [nan, nan, nan], [3, nan, nan]],
                           [[12, nan, nan], [13, nan, nan], [10, 14, nan], [11, nan, nan]]],
@@ -393,7 +393,7 @@ def test_categorical_min_n(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_max_n(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[[4, 0, nan], [1, nan, nan], [nan, nan, nan], [3, nan, nan]],
                           [[12, nan, nan], [13, nan, nan], [14, 10, nan], [11, nan, nan]]],
@@ -411,7 +411,7 @@ def test_categorical_max_n(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_min_row_index(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[0, 1, 2, 3], [12, 13, 10, 11]], [[8, 5, 6, 7], [16, 17, 18, 15]]])
     agg = c.points(ddf, 'x', 'y', ds.by('cat2', ds._min_row_index()))
@@ -421,7 +421,7 @@ def test_categorical_min_row_index(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_max_row_index(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[4, 1, 2, 3], [12, 13, 14, 11]], [[8, 9, 6, 7], [16, 17, 18, 19]]])
     agg = c.points(ddf, 'x', 'y', ds.by('cat2', ds._max_row_index()))
@@ -431,7 +431,7 @@ def test_categorical_max_row_index(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_min_n_row_index(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[[0, 4, -1], [1, -1, -1], [2, -1, -1], [3, -1, -1]],
                           [[12, -1, -1], [13, -1, -1], [10, 14, -1], [11, -1, -1]]],
@@ -449,7 +449,7 @@ def test_categorical_min_n_row_index(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_max_n_row_index(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[[4, 0, -1], [1, -1, -1], [2, -1, -1], [3, -1, -1]],
                           [[12, -1, -1], [13, -1, -1], [14, 10, -1], [11, -1, -1]]],
@@ -467,7 +467,7 @@ def test_categorical_max_n_row_index(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_first(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[0, -1, nan, -3],
                           [12, -13, 10, -11]],
@@ -481,7 +481,7 @@ def test_categorical_first(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_last(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[4, -1, nan, -3],
                           [12, -13, 14, -11]],
@@ -495,7 +495,7 @@ def test_categorical_last(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_first_n(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[[0, 4, nan], [-1, nan, nan], [nan, nan, nan], [-3, nan, nan]],
                           [[12, nan, nan], [-13, nan, nan], [10, 14, nan], [-11, nan, nan]]],
@@ -513,7 +513,7 @@ def test_categorical_first_n(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_last_n(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     solution = np.array([[[[4, 0, nan], [-1, nan, nan], [nan, nan, nan], [-3, nan, nan]],
                           [[12, nan, nan], [-13, nan, nan], [14, 10, nan], [-11, nan, nan]]],
@@ -533,7 +533,7 @@ def test_categorical_last_n(ddf, npartitions):
 def test_where_max(ddf, npartitions):
     # Important to test with npartitions > 2 to have multiple combination stages.
     # Identical results to equivalent pandas test.
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     out = xr.DataArray([[16, 6], [11, 1]], coords=coords, dims=dims)
     assert_eq_xr(c.points(ddf, 'x', 'y', ds.where(ds.max('i32'), 'reverse')), out)
@@ -554,7 +554,7 @@ def test_where_max(ddf, npartitions):
 def test_where_min(ddf, npartitions):
     # Important to test with npartitions > 2 to have multiple combination stages.
     # Identical results to equivalent pandas test.
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     out = xr.DataArray([[20, 10], [15, 5]], coords=coords, dims=dims)
     assert_eq_xr(c.points(ddf, 'x', 'y', ds.where(ds.min('i32'), 'reverse')), out)
@@ -575,7 +575,7 @@ def test_where_min(ddf, npartitions):
 def test_where_max_n(ddf, npartitions):
     # Important to test with npartitions > 2 to have multiple combination stages.
     # Identical results to equivalent pandas test.
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol_rowindex = np.array([[[ 4,  0,  1,  3, -1, -1],
                               [14, 12, 10, 11, 13, -1]],
@@ -608,7 +608,7 @@ def test_where_max_n(ddf, npartitions):
 def test_where_min_n(ddf, npartitions):
     # Important to test with npartitions > 2 to have multiple combination stages.
     # Identical results to equivalent pandas test.
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol_rowindex = np.array([[[3,  1,  0,  4, -1, -1],
                               [13, 11, 10, 12, 14, -1]],
@@ -639,7 +639,7 @@ def test_where_min_n(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_where_first(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     # Note reductions like ds.where(ds.first('i32'), 'reverse') are supported,
     # but the same results can be achieved using the simpler ds.first('reverse')
@@ -660,7 +660,7 @@ def test_where_first(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_where_last(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     # Note reductions like ds.where(ds.last('i32'), 'reverse') are supported,
     # but the same results can be achieved using the simpler ds.last('reverse')
@@ -683,7 +683,7 @@ def test_where_last(ddf, npartitions):
 def test_where_first_n(ddf, npartitions):
     # Important to test with npartitions > 2 to have multiple combination stages.
     # Identical results to equivalent pandas test.
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol_rowindex = np.array([[[ 0,  1,  3,  4, -1, -1],
                               [10, 11, 12, 13, 14, -1]],
@@ -716,7 +716,7 @@ def test_where_first_n(ddf, npartitions):
 def test_where_last_n(ddf, npartitions):
     # Important to test with npartitions > 2 to have multiple combination stages.
     # Identical results to equivalent pandas test.
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol_rowindex = np.array([[[ 4,  3,  1,  0, -1, -1],
                               [14, 13, 12, 11, 10, -1]],
@@ -747,7 +747,7 @@ def test_where_last_n(ddf, npartitions):
 @pytest.mark.parametrize('ddf', [_ddf])
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_summary_by(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
 
     # summary(by)
@@ -785,7 +785,7 @@ def test_summary_by(ddf, npartitions):
 def test_summary_where_n(ddf, npartitions):
     # Important to test with npartitions > 2 to have multiple combination stages.
     # Identical results to equivalent pandas test.
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol_min_n_rowindex = np.array([[[3,  1,  0,  4, -1],
                                     [13, 11, 10, 12, 14]],
@@ -821,7 +821,7 @@ def test_summary_where_n(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_mean(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     out = xr.DataArray(
         values(df_pd.i32).reshape((2, 2, 5)).mean(axis=2, dtype='f8').T,
@@ -838,7 +838,7 @@ def test_mean(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_var(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     out = xr.DataArray(
         values(df_pd.i32).reshape((2, 2, 5)).var(axis=2, dtype='f8').T,
@@ -855,7 +855,7 @@ def test_var(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_std(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     out = xr.DataArray(
         values(df_pd.i32).reshape((2, 2, 5)).std(axis=2, dtype='f8').T,
@@ -872,7 +872,7 @@ def test_std(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_count_cat(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol = np.array([[[5, 0, 0, 0],
                      [0, 0, 5, 0]],
@@ -930,7 +930,7 @@ def test_count_cat(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_sum(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol = np.array([[[ 10, nan, nan, nan],
                      [nan, nan,  60, nan]],
@@ -977,7 +977,7 @@ def test_categorical_sum_binning(ddf, npartitions):
         pytest.skip(
             "The categorical binning of 'sum' reduction is yet supported on the GPU"
         )
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol = np.array([[[8.0,  nan,  nan,  nan],
                      [nan,  nan, 60.0,  nan]],
@@ -1000,7 +1000,7 @@ def test_categorical_sum_binning(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_mean(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol = np.array([[[  2, nan, nan, nan],
                      [nan, nan,  12, nan]],
@@ -1036,7 +1036,7 @@ def test_categorical_mean_binning(ddf, npartitions):
         pytest.skip(
             "The categorical binning of 'mean' reduction is yet supported on the GPU"
         )
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol = np.array([[[  2, nan, nan, nan],
                      [nan, nan,  12, nan]],
@@ -1057,7 +1057,7 @@ def test_categorical_mean_binning(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_var(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol = np.array([[[ 2.5,  nan,  nan,  nan],
                      [ nan,  nan,   2.,  nan]],
@@ -1099,7 +1099,7 @@ def test_categorical_var(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_categorical_std(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol = np.sqrt(np.array([
         [[ 2.5,  nan,  nan,  nan],
@@ -1143,7 +1143,7 @@ def test_categorical_std(ddf, npartitions):
 @pytest.mark.parametrize('ddf', ddfs)
 @pytest.mark.parametrize('npartitions', [1, 2, 3, 4])
 def test_multiple_aggregates(ddf, npartitions):
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     agg = c.points(ddf, 'x', 'y',
                    ds.summary(f64_std=ds.std('f64'),
@@ -2365,7 +2365,7 @@ def test_canvas_size():
 def test_dataframe_dtypes(ddf, npartitions):
     # Issue #1235.
     ddf['dates'] = pd.Series(['2007-07-13']*20, dtype='datetime64[ns]')
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     ds.Canvas(2, 2).points(ddf, 'x', 'y', ds.count())
 
@@ -2408,7 +2408,7 @@ def test_dask_categorical_counts(on_gpu):
 def test_categorical_where_max(ddf, npartitions):
     # Important to test with npartitions > 2 to have multiple combination stages.
     # Identical results to equivalent pandas test.
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol_rowindex = xr.DataArray([[[4, 1, -1, 3], [12, 13, 14, 11]],
                                  [[8, 5, 6, 7], [16, 17, 18, 15]]],
@@ -2430,7 +2430,7 @@ def test_categorical_where_max(ddf, npartitions):
 def test_categorical_where_min(ddf, npartitions):
     # Important to test with npartitions > 2 to have multiple combination stages.
     # Identical results to equivalent pandas test.
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol_rowindex = xr.DataArray([[[0, 1, -1, 3], [12, 13, 10, 11]],
                                  [[8, 9, 6, 7], [16, 17, 18, 19]]],
@@ -2452,7 +2452,7 @@ def test_categorical_where_min(ddf, npartitions):
 def test_categorical_where_first(ddf, npartitions):
     # Important to test with npartitions > 2 to have multiple combination stages.
     # Identical results to equivalent pandas test.
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol_rowindex = xr.DataArray([[[0, 1, -1, 3], [12, 13, 10, 11]],
                                  [[8, 5, 6, 7], [16, 17, 18, 15]]],
@@ -2474,7 +2474,7 @@ def test_categorical_where_first(ddf, npartitions):
 def test_categorical_where_last(ddf, npartitions):
     # Important to test with npartitions > 2 to have multiple combination stages.
     # Identical results to equivalent pandas test.
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol_rowindex = xr.DataArray([[[4, 1, -1, 3], [12, 13, 14, 11]],
                                  [[8, 9, 6, 7], [16, 17, 18, 19]]],
@@ -2496,7 +2496,7 @@ def test_categorical_where_last(ddf, npartitions):
 def test_categorical_where_max_n(ddf, npartitions):
     # Important to test with npartitions > 2 to have multiple combination stages.
     # Identical results to equivalent pandas test.
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol_rowindex = xr.DataArray(
         [[[[4, 0, -1], [1, -1, -1], [-1, -1, -1], [3, -1, -1]],
@@ -2534,7 +2534,7 @@ def test_categorical_where_max_n(ddf, npartitions):
 def test_categorical_where_min_n(ddf, npartitions):
     # Important to test with npartitions > 2 to have multiple combination stages.
     # Identical results to equivalent pandas test.
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol_rowindex = xr.DataArray(
         [[[[0, 4, -1], [1, -1, -1], [-1, -1, -1], [3, -1, -1]],
@@ -2572,7 +2572,7 @@ def test_categorical_where_min_n(ddf, npartitions):
 def test_categorical_where_first_n(ddf, npartitions):
     # Important to test with npartitions > 2 to have multiple combination stages.
     # Identical results to equivalent pandas test.
-    ddf = ddf.repartition(npartitions)
+    ddf = ddf.repartition(npartitions=npartitions)
     assert ddf.npartitions == npartitions
     sol_rowindex = xr.DataArray(
         [[[[0, 4, -1], [1, -1, -1], [-1, -1, -1], [3, -1, -1]],

From 38e6964d9149c845120bf859cd300bfb7b8052fc Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Simon=20H=C3=B8xbro=20Hansen?= <simon.hansen@me.com>
Date: Wed, 31 Jan 2024 18:25:43 +0100
Subject: [PATCH 03/14] Ignore test_ffil_limit_area

---
 datashader/tests/test_datatypes.py | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/datashader/tests/test_datatypes.py b/datashader/tests/test_datatypes.py
index 243cee870..efa945846 100644
--- a/datashader/tests/test_datatypes.py
+++ b/datashader/tests/test_datatypes.py
@@ -873,6 +873,11 @@ def test_fillna_no_op_returns_copy(self):
     def test_fillna_series_method(self):
         pass
 
+    @pytest.mark.skip(reason="Can't fill with nested sequences")
+    def test_ffill_limit_area(self):
+        # Added in Pandas 2.2
+        pass
+
 
 class TestRaggedReshaping(eb.BaseReshapingTests):
     @pytest.mark.skip(reason="__setitem__ not supported")

From 6fb8e3d125f975c319884625474ee79f5238dc70 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Simon=20H=C3=B8xbro=20Hansen?= <simon.hansen@me.com>
Date: Wed, 31 Jan 2024 19:06:13 +0100
Subject: [PATCH 04/14] Work-around Pandas 2.2 changes for Extension

---
 datashader/tests/test_datatypes.py | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/datashader/tests/test_datatypes.py b/datashader/tests/test_datatypes.py
index efa945846..591ce37f0 100644
--- a/datashader/tests/test_datatypes.py
+++ b/datashader/tests/test_datatypes.py
@@ -835,6 +835,12 @@ def test_sort_values_frame(self):
     def test_where_series(self):
         pass
 
+    @pytest.mark.xfail(reason="not currently supported")
+    def test_duplicated(self):
+        # Added in Pandas 2.2
+        # https://github.com/pandas-dev/pandas/pull/55255
+        pass
+
 class TestRaggedPrinting(eb.BasePrintingTests):
     @pytest.mark.skip(reason="Can't autoconvert ragged array to numpy array")
     def test_dataframe_repr(self):
@@ -891,3 +897,11 @@ def test_transpose(self):
     @pytest.mark.skip(reason="transpose with numpy array elements seems not supported")
     def test_transpose_frame(self):
         pass
+
+    @pytest.mark.skipif(Version(pd.__version__) >= Version("2.2"), reason="Stopped working in Pandas 2.2")
+    def test_merge_on_extension_array(self, data):
+        super().test_merge_on_extension_array(data)
+
+    @pytest.mark.skipif(Version(pd.__version__) >= Version("2.2"), reason="Stopped working in Pandas 2.2")
+    def test_merge_on_extension_array_duplicates(self, data):
+        super().test_merge_on_extension_array_duplicates(data)

From 2639541b48f837f2da3a1aa2cc4c20980d22ed1e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Simon=20H=C3=B8xbro=20Hansen?= <simon.hansen@me.com>
Date: Wed, 31 Jan 2024 19:18:01 +0100
Subject: [PATCH 05/14] Fix lint

---
 datashader/tests/test_datatypes.py | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/datashader/tests/test_datatypes.py b/datashader/tests/test_datatypes.py
index 591ce37f0..6a5378a24 100644
--- a/datashader/tests/test_datatypes.py
+++ b/datashader/tests/test_datatypes.py
@@ -898,10 +898,14 @@ def test_transpose(self):
     def test_transpose_frame(self):
         pass
 
-    @pytest.mark.skipif(Version(pd.__version__) >= Version("2.2"), reason="Stopped working in Pandas 2.2")
+    @pytest.mark.skipif(
+        Version(pd.__version__) >= Version("2.2"), reason="Stopped working in Pandas 2.2"
+    )
     def test_merge_on_extension_array(self, data):
         super().test_merge_on_extension_array(data)
 
-    @pytest.mark.skipif(Version(pd.__version__) >= Version("2.2"), reason="Stopped working in Pandas 2.2")
+    @pytest.mark.skipif(
+        Version(pd.__version__) >= Version("2.2"), reason="Stopped working in Pandas 2.2"
+    )
     def test_merge_on_extension_array_duplicates(self, data):
         super().test_merge_on_extension_array_duplicates(data)

From 5468a0d94b028e9bd8de95b16a4f501f5944372e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Simon=20H=C3=B8xbro=20Hansen?= <simon.hansen@me.com>
Date: Thu, 1 Feb 2024 09:12:22 +0100
Subject: [PATCH 06/14] Move pytest setting to pyproject.toml

---
 pyproject.toml | 15 +++++++++++++++
 tox.ini        |  8 --------
 2 files changed, 15 insertions(+), 8 deletions(-)

diff --git a/pyproject.toml b/pyproject.toml
index 0528dfc7c..b74125734 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -18,3 +18,18 @@ target-version = "py39"
 
 [tool.ruff.per-file-ignores]
 "test_mpl_ext.py" = ["E402"]  # Module level import not at top of file
+
+[tool.pytest.ini_options]
+addopts = ["--pyargs", "--doctest-modules", "--doctest-ignore-import-errors", "--strict-config", "--strict-markers", "--color=yes"]
+norecursedirs = 'doc .git dist build _build .ipynb_checkpoints'
+minversion = "7"
+xfail_strict = true
+log_cli_level = "INFO"
+nbsmoke_skip_run = [ # skipping any notebooks that require extra deps
+    '.*tiling.ipynb$',
+    '.*streaming-aggregation.ipynb$',
+    '.*8_Geography.ipynb$',
+]
+filterwarnings = [
+    "ignore:Passing a (SingleBlockManager|BlockManager) to (Series|GeoSeries|DataFrame|GeoDataFrame) is deprecated:DeprecationWarning",  # https://github.com/holoviz/spatialpandas/issues/137
+]
diff --git a/tox.ini b/tox.ini
index 678860750..94e305b02 100644
--- a/tox.ini
+++ b/tox.ini
@@ -71,14 +71,6 @@ deps = unit: {[_unit]deps}
        examples_extra: {[_examples_extra]deps}
        all: {[_all]deps}
 
-[pytest]
-addopts = -v --pyargs --doctest-modules --doctest-ignore-import-errors
-norecursedirs = doc .git dist build _build .ipynb_checkpoints
-# skipping any notebooks that require extra deps
-nbsmoke_skip_run = .*tiling.ipynb$
-                   .*streaming-aggregation.ipynb$
-                   .*8_Geography.ipynb$
-
 [flake8]
 include = *.py
 # run_tests.py is generated by conda build, which appears to have a

From bc9c4095453559481d9e6ab8ffd45b771bb487b7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Simon=20H=C3=B8xbro=20Hansen?= <simon.hansen@me.com>
Date: Thu, 1 Feb 2024 09:38:20 +0100
Subject: [PATCH 07/14] Fix mix warnings in test suite

---
 datashader/tests/test_geopandas.py | 6 +++---
 datashader/tests/test_pandas.py    | 3 ++-
 datashader/tests/test_pipeline.py  | 2 +-
 3 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/datashader/tests/test_geopandas.py b/datashader/tests/test_geopandas.py
index 29b04b8b1..a039dddf3 100644
--- a/datashader/tests/test_geopandas.py
+++ b/datashader/tests/test_geopandas.py
@@ -200,7 +200,7 @@ def test_lines_spatialpandas(geom_type, explode, use_boundary, npartitions):
 def test_points_geopandas(geom_type):
     df = geopandas.read_file(geodatasets.get_path("nybb"))
 
-    df["geometry"] = df["geometry"].sample_points(100, seed=93814)  # multipoint
+    df["geometry"] = df["geometry"].sample_points(100, rng=93814)  # multipoint
     if geom_type == "point":
         df = df.explode(index_parts=False)  # Multipoint -> point.
     unique_geom_type = df["geometry"].geom_type.unique()
@@ -218,7 +218,7 @@ def test_points_geopandas(geom_type):
 def test_points_dask_geopandas(geom_type, npartitions):
     df = geopandas.read_file(geodatasets.get_path("nybb"))
 
-    df["geometry"] = df["geometry"].sample_points(100, seed=93814)  # multipoint
+    df["geometry"] = df["geometry"].sample_points(100, rng=93814)  # multipoint
     if geom_type == "point":
         df = df.explode(index_parts=False)  # Multipoint -> point.
     unique_geom_type = df["geometry"].geom_type.unique()
@@ -240,7 +240,7 @@ def test_points_dask_geopandas(geom_type, npartitions):
 def test_points_spatialpandas(geom_type, npartitions):
     df = geopandas.read_file(geodatasets.get_path("nybb"))
 
-    df["geometry"] = df["geometry"].sample_points(100, seed=93814)  # multipoint
+    df["geometry"] = df["geometry"].sample_points(100, rng=93814)  # multipoint
     if geom_type == "point":
         df = df.explode(index_parts=False)  # Multipoint -> point.
     unique_geom_type = df["geometry"].geom_type.unique()
diff --git a/datashader/tests/test_pandas.py b/datashader/tests/test_pandas.py
index 9ea724e2f..2133e5cdd 100644
--- a/datashader/tests/test_pandas.py
+++ b/datashader/tests/test_pandas.py
@@ -2734,7 +2734,8 @@ def test_line_antialias():
     numerator = np.nan_to_num(line_antialias_sol_0_intersect) + np.nan_to_num(line_antialias_sol_1)
     denom = np.nan_to_num(line_antialias_sol_count_ignore_aa_0) + \
         np.nan_to_num(line_antialias_sol_count_ignore_aa_1)
-    sol = 3*numerator / denom
+    with np.errstate(invalid='ignore'):  # Dividing by zero is expected
+        sol = 3*numerator / denom
     assert_eq_ndarray(agg.data, sol, close=True)
 
     agg = cvs.line(agg=ds._min_row_index(), **kwargs)
diff --git a/datashader/tests/test_pipeline.py b/datashader/tests/test_pipeline.py
index c099524c4..c8fbc5938 100644
--- a/datashader/tests/test_pipeline.py
+++ b/datashader/tests/test_pipeline.py
@@ -9,7 +9,7 @@
 df = pd.DataFrame({'x': np.array(([0.] * 10 + [1] * 10)),
                    'y': np.array(([0.] * 5 + [1] * 5 + [0] * 5 + [1] * 5)),
                    'f64': np.arange(20, dtype='f8')})
-df.f64.iloc[2] = np.nan
+df.loc['f64', 2] = np.nan
 
 cvs = ds.Canvas(plot_width=2, plot_height=2, x_range=(0, 1), y_range=(0, 1))
 cvs10 = ds.Canvas(plot_width=10, plot_height=10, x_range=(0, 1), y_range=(0, 1))

From 51997fceec62e4237d1473cc89686a2170880aaa Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Simon=20H=C3=B8xbro=20Hansen?= <simon.hansen@me.com>
Date: Thu, 1 Feb 2024 10:17:02 +0100
Subject: [PATCH 08/14] Make xfail run

---
 datashader/tests/test_datatypes.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/datashader/tests/test_datatypes.py b/datashader/tests/test_datatypes.py
index 6a5378a24..9f006a3d4 100644
--- a/datashader/tests/test_datatypes.py
+++ b/datashader/tests/test_datatypes.py
@@ -836,10 +836,10 @@ def test_where_series(self):
         pass
 
     @pytest.mark.xfail(reason="not currently supported")
-    def test_duplicated(self):
+    def test_duplicated(self, data):
         # Added in Pandas 2.2
         # https://github.com/pandas-dev/pandas/pull/55255
-        pass
+        super().test_duplicated(data)
 
 class TestRaggedPrinting(eb.BasePrintingTests):
     @pytest.mark.skip(reason="Can't autoconvert ragged array to numpy array")

From 001446ab897e4d5bbed1857d8b20ae87181cb4ab Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Simon=20H=C3=B8xbro=20Hansen?= <simon.hansen@me.com>
Date: Thu, 1 Feb 2024 10:18:42 +0100
Subject: [PATCH 09/14] Try commenting out fastparquet for slow windows solve

---
 setup.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/setup.py b/setup.py
index bd4bdc871..946b0eb5b 100644
--- a/setup.py
+++ b/setup.py
@@ -44,7 +44,7 @@
     'tests': geopandas + [
         'codecov',
         'geodatasets',
-        'fastparquet',  # optional dependency
+        # 'fastparquet',  # optional dependency
         'flake8',
         'nbconvert',
         'nbformat',

From db5a9027c6f56e6d85c9bad6eaade1b86693d05a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Simon=20H=C3=B8xbro=20Hansen?= <simon.hansen@me.com>
Date: Thu, 1 Feb 2024 12:06:31 +0100
Subject: [PATCH 10/14] Update nbsmoke_skip_run

---
 pyproject.toml | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/pyproject.toml b/pyproject.toml
index b74125734..8e26d6b15 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -25,11 +25,8 @@ norecursedirs = 'doc .git dist build _build .ipynb_checkpoints'
 minversion = "7"
 xfail_strict = true
 log_cli_level = "INFO"
-nbsmoke_skip_run = [ # skipping any notebooks that require extra deps
-    '.*tiling.ipynb$',
-    '.*streaming-aggregation.ipynb$',
-    '.*8_Geography.ipynb$',
-]
+# skipping any notebooks that require extra deps
+nbsmoke_skip_run = ".*tiling.ipynb$\n.*streaming-aggregation.ipynb$\n.*8_Geography.ipynb$"
 filterwarnings = [
     "ignore:Passing a (SingleBlockManager|BlockManager) to (Series|GeoSeries|DataFrame|GeoDataFrame) is deprecated:DeprecationWarning",  # https://github.com/holoviz/spatialpandas/issues/137
 ]

From 03adc9a6e26fa4d539325817038f1f74585b9b46 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Simon=20H=C3=B8xbro=20Hansen?= <simon.hansen@me.com>
Date: Thu, 1 Feb 2024 12:08:23 +0100
Subject: [PATCH 11/14] Update setup.py

---
 setup.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/setup.py b/setup.py
index 946b0eb5b..55cc24378 100644
--- a/setup.py
+++ b/setup.py
@@ -44,7 +44,6 @@
     'tests': geopandas + [
         'codecov',
         'geodatasets',
-        # 'fastparquet',  # optional dependency
         'flake8',
         'nbconvert',
         'nbformat',

From b2ff68c67c9ba4fc6874b5f14b0f01167a86b514 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Simon=20H=C3=B8xbro=20Hansen?= <simon.hansen@me.com>
Date: Fri, 9 Feb 2024 13:43:32 +0100
Subject: [PATCH 12/14] Ignore dask geopandas warning for now

---
 pyproject.toml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/pyproject.toml b/pyproject.toml
index 8e26d6b15..83954127c 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -29,4 +29,5 @@ log_cli_level = "INFO"
 nbsmoke_skip_run = ".*tiling.ipynb$\n.*streaming-aggregation.ipynb$\n.*8_Geography.ipynb$"
 filterwarnings = [
     "ignore:Passing a (SingleBlockManager|BlockManager) to (Series|GeoSeries|DataFrame|GeoDataFrame) is deprecated:DeprecationWarning",  # https://github.com/holoviz/spatialpandas/issues/137
+    "ignore:Accessing the underlying geometries through the `.data`:DeprecationWarning:dask_geopandas.core",  # https://github.com/geopandas/dask-geopandas/issues/264
 ]

From b466c2fa3b6e849a53aa80580e31714e8a4e8aed Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Simon=20H=C3=B8xbro=20Hansen?= <simon.hansen@me.com>
Date: Sat, 10 Feb 2024 10:12:17 +0100
Subject: [PATCH 13/14] Update skipif to only be valid for pandas 2.2.0

---
 datashader/tests/test_datatypes.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/datashader/tests/test_datatypes.py b/datashader/tests/test_datatypes.py
index 9f006a3d4..7cbd891ad 100644
--- a/datashader/tests/test_datatypes.py
+++ b/datashader/tests/test_datatypes.py
@@ -899,13 +899,13 @@ def test_transpose_frame(self):
         pass
 
     @pytest.mark.skipif(
-        Version(pd.__version__) >= Version("2.2"), reason="Stopped working in Pandas 2.2"
+        Version(pd.__version__) == Version("2.2.0"), reason="Regression in Pandas 2.2"
     )
     def test_merge_on_extension_array(self, data):
         super().test_merge_on_extension_array(data)
 
     @pytest.mark.skipif(
-        Version(pd.__version__) >= Version("2.2"), reason="Stopped working in Pandas 2.2"
+        Version(pd.__version__) == Version("2.2.0"), reason="Regression in Pandas 2.2"
     )
     def test_merge_on_extension_array_duplicates(self, data):
         super().test_merge_on_extension_array_duplicates(data)

From 8d3ada093ff5e84c68236c1cb0bdf8f75a364dba Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Simon=20H=C3=B8xbro=20Hansen?= <simon.hansen@me.com>
Date: Mon, 12 Feb 2024 12:01:04 +0100
Subject: [PATCH 14/14] Add NotImplementedError for duplicated.

---
 datashader/datatypes.py | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/datashader/datatypes.py b/datashader/datatypes.py
index 11e0f356a..f3d13b6e8 100644
--- a/datashader/datatypes.py
+++ b/datashader/datatypes.py
@@ -649,6 +649,10 @@ def __array__(self, dtype=None):
         dtype = np.dtype(object) if dtype is None else np.dtype(dtype)
         return np.asarray(self.tolist(), dtype=dtype)
 
+    def duplicated(self, *args, **kwargs):
+        msg = "duplicated is not implemented for RaggedArray"
+        raise NotImplementedError(msg)
+
 
 @jit(nopython=True, nogil=True)
 def _eq_ragged_ragged(start_indices1,