pyarrow: Check compatibility of pyarrow-backed pandas objects with numeric dtypes

.github/workflows/ci_tests.yaml

@@ -71,7 +71,7 @@ jobs:
             optional-packages: ''
           - python-version: '3.12'
             numpy-version: '1.26'
-            optional-packages: ' contextily geopandas ipython rioxarray sphinx-gallery'
+            optional-packages: ' contextily geopandas ipython pyarrow rioxarray sphinx-gallery'
 
     timeout-minutes: 30
     defaults:

.github/workflows/ci_tests_dev.yaml

@@ -153,7 +153,7 @@ jobs:
           python -m pip install --pre --prefer-binary \
                         --extra-index https://pypi.anaconda.org/scientific-python-nightly-wheels/simple \
                         numpy pandas xarray netCDF4 packaging \
-                        build contextily dvc geopandas ipython rioxarray \
+                        build contextily dvc geopandas ipython pyarrow rioxarray \
                         'pytest>=6.0' pytest-cov pytest-doctestplus pytest-mpl \
                         sphinx-gallery
 

.github/workflows/ci_tests_legacy.yaml

@@ -72,6 +72,7 @@ jobs:
             contextily
             geopandas
             ipython
+            pyarrow
             rioxarray
             sphinx-gallery
             build

doc/install.rst

@@ -107,6 +107,15 @@ The following are optional dependencies:
 * `GeoPandas <https://geopandas.org>`__: For using and plotting GeoDataFrame objects.
 * `RioXarray <https://corteva.github.io/rioxarray>`__: For saving multi-band rasters to GeoTIFFs.
 
+.. note::
+
+    If you have `PyArrow <https://arrow.apache.org/docs/python/index.html>`__
+    installed, PyGMT does have some initial support for ``pandas.Series`` and
+    ``pandas.DataFrame`` objects with Apache Arrow-backed arrays. Specifically,
+    only uint/int/float dtypes are supported for now. Support for datetime and
+    string Arrow dtypes are still works in progress. For more details, see
+    `issue #2800 <https://github.com/GenericMappingTools/pygmt/issues/2800>`__.
+
 Installing GMT and other dependencies
 -------------------------------------
 

pygmt/tests/test_clib_virtualfiles.py

@@ -2,6 +2,7 @@
 Test the C API functions related to virtual files.
 """
 import os
+from importlib.util import find_spec
 from itertools import product
 
 import numpy as np
@@ -321,16 +322,21 @@ def test_virtualfile_from_matrix_slice(dtypes):
 
 def test_virtualfile_from_vectors_pandas(dtypes):
     """
-    Pass vectors to a dataset using pandas Series.
+    Pass vectors to a dataset using pandas.Series, checking both numpy and
+    pyarrow dtypes.
     """
     size = 13
+    if find_spec("pyarrow") is not None:
+        dtypes.extend([f"{dtype}[pyarrow]" for dtype in dtypes])
+
     for dtype in dtypes:
         data = pd.DataFrame(
             data={
-                "x": np.arange(size, dtype=dtype),
-                "y": np.arange(size, size * 2, 1, dtype=dtype),
-                "z": np.arange(size * 2, size * 3, 1, dtype=dtype),
-            }
+                "x": np.arange(size),
+                "y": np.arange(size, size * 2, 1),
+                "z": np.arange(size * 2, size * 3, 1),
+            },
+            dtype=dtype,
         )
         with clib.Session() as lib:
             with lib.virtualfile_from_vectors(data.x, data.y, data.z) as vfile:

pygmt/tests/test_geopandas.py

@@ -5,6 +5,7 @@
 import pandas as pd
 import pytest
 from pygmt import Figure, info, makecpt, which
+from pygmt.helpers.testing import skip_if_no
 
 gpd = pytest.importorskip("geopandas")
 shapely = pytest.importorskip("shapely")
@@ -161,6 +162,24 @@ def test_geopandas_plot3d_non_default_circle():
         "int64",
         pd.Int32Dtype(),
         pd.Int64Dtype(),
+        pytest.param(
+            "int32[pyarrow]",
+            marks=[
+                skip_if_no(package="pyarrow"),
+                pytest.mark.xfail(
+                    reason="geopandas doesn't support writing columns with pyarrow dtypes to OGR_GMT yet."
+                ),
+            ],
+        ),
+        pytest.param(
+            "int64[pyarrow]",
+            marks=[
+                skip_if_no(package="pyarrow"),
+                pytest.mark.xfail(
+                    reason="geopandas doesn't support writing columns with pyarrow dtypes to OGR_GMT yet."
+                ),
+            ],
+        ),
     ],
 )
 @pytest.mark.mpl_image_compare(filename="test_geopandas_plot_int_dtypes.png")

pygmt/tests/test_info.py

@@ -12,6 +12,7 @@
 import xarray as xr
 from pygmt import info
 from pygmt.exceptions import GMTInvalidInput
+from pygmt.helpers.testing import skip_if_no
 
 TEST_DATA_DIR = os.path.join(os.path.dirname(__file__), "data")
 POINTS_DATA = os.path.join(TEST_DATA_DIR, "points.txt")
@@ -74,16 +75,27 @@ def test_info_2d_list():
     assert output == expected_output
 
 
-def test_info_series():
+@pytest.mark.parametrize(
+    "dtype",
+    ["int64", pytest.param("int64[pyarrow]", marks=skip_if_no(package="pyarrow"))],
+)
+def test_info_series(dtype):
     """
     Make sure info works on a pandas.Series input.
     """
-    output = info(pd.Series(data=[0, 4, 2, 8, 6]))
+    output = info(pd.Series(data=[0, 4, 2, 8, 6], dtype=dtype))
     expected_output = "<vector memory>: N = 5 <0/8>\n"
     assert output == expected_output
 
 
-def test_info_dataframe():
+@pytest.mark.parametrize(
+    "dtype",
+    [
+        "float64",
+        pytest.param("float64[pyarrow]", marks=skip_if_no(package="pyarrow")),
+    ],
+)
+def test_info_dataframe(dtype):
     """
     Make sure info works on pandas.DataFrame inputs.
     """