Mark unit tests with @pytest.mark.benchmark part 2 (#2924)

* Benchmark test_binstats_no_outgrid
* Benchmark test_blockmean_input_xyz and test_blockmean_input_xyz
* Benchmark test_blockmedian_input_table_matrix
* Benchmark test_load_libgmt
* Benchmark test_put_matrix and test_put_matrix_grid
* Benchmark test_put_strings
* Benchmark test_put_vector and test_put_vector_string_dtype
* Benchmark test_call_module and test_dataarray_to_matrix_works
* Benchmark four test_clib_virtualfile tests

Specifically test_virtual_file, test_virtualfile_from_data_required_z_matrix, test_virtualfile_from_vectors, and test_virtualfile_from_matrix

* Install pyarrow in benchmarks.yml
* Benchmark test_coast_region
* Benchmark test_colorbar_box
* Benchmark test_config_format_date_map
* Benchmark test_contour_matrix
* Benchmark load_remote_dataset_wrapper
* Benchmark test_earth_relief_holes
* Benchmark test_dimfilter_no_outgrid
* Benchmark test_figure_repr
* Benchmark test_filter1d_format
* Benchmark test_accessor_set_geographic_cartesian_roundtrip
* Benchmark test_virtualfile_from_vectors_one_string_or_object_column

Author: weiji14

.github/workflows/benchmarks.yml

@@ -58,7 +58,7 @@ jobs:
           echo $CONDA/bin >> $GITHUB_PATH
           conda install --solver=libmamba gmt=6.4.0 python=3.12 \
                         numpy pandas xarray netCDF4 packaging \
-                        geopandas pytest pytest-mpl
+                        geopandas pyarrow pytest pytest-mpl
           python -m pip install -U pytest-codspeed setuptools
 
       # Download cached remote files (artifacts) from GitHub

pygmt/tests/test_accessor.py

@@ -46,6 +46,7 @@ def test_accessor_set_pixel_registration():
     assert grid.gmt.registration == 1  # ensure changed to pixel registration
 
 
+@pytest.mark.benchmark
 def test_accessor_set_geographic_cartesian_roundtrip():
     """
     Check that we can set a grid to switch between the default Cartesian coordinate type

pygmt/tests/test_binstats.py

@@ -4,6 +4,7 @@
 from pathlib import Path
 
 import numpy.testing as npt
+import pytest
 from pygmt import binstats
 from pygmt.helpers import GMTTempFile
 
@@ -26,6 +27,7 @@ def test_binstats_outgrid():
         assert Path(tmpfile.name).stat().st_size > 0  # check that outgrid exists
 
 
+@pytest.mark.benchmark
 def test_binstats_no_outgrid():
     """
     Test binstats with no set outgrid.

pygmt/tests/test_blockm.py

@@ -45,6 +45,7 @@ def test_blockmean_input_table_matrix(array_func, dataframe):
     npt.assert_allclose(output.iloc[0], [245.888877, 29.978707, -384.0])
 
 
+@pytest.mark.benchmark
 def test_blockmean_input_xyz(dataframe):
     """
     Run blockmean by passing in x/y/z as input.
@@ -99,6 +100,7 @@ def test_blockmean_without_outfile_setting():
     npt.assert_allclose(output.iloc[0], [245.888877, 29.978707, -384.0])
 
 
+@pytest.mark.benchmark
 def test_blockmode_input_dataframe(dataframe):
     """
     Run blockmode by passing in a pandas.DataFrame as input.

pygmt/tests/test_blockmedian.py

@@ -31,6 +31,7 @@ def test_blockmedian_input_dataframe(dataframe):
     npt.assert_allclose(output.iloc[0], [245.88819, 29.97895, -385.0])
 
 
+@pytest.mark.benchmark
 def test_blockmedian_input_table_matrix(dataframe):
     """
     Run blockmedian using table input that is not a pandas.DataFrame but still a matrix.

pygmt/tests/test_clib.py

@@ -130,6 +130,7 @@ def test_destroy_session_fails():
     ses.destroy()
 
 
+@pytest.mark.benchmark
 def test_call_module():
     """
     Run a command to see if call_module works.
@@ -388,6 +389,7 @@ def test_write_data_fails():
                 )
 
 
+@pytest.mark.benchmark
 def test_dataarray_to_matrix_works():
     """
     Check that dataarray_to_matrix returns correct output.

pygmt/tests/test_clib_loading.py

@@ -62,6 +62,7 @@ def test_clib_names():
 
 ###############################################################################
 # Test load_libgmt
+@pytest.mark.benchmark
 def test_load_libgmt():
     """
     Test that loading libgmt works and doesn't crash.

pygmt/tests/test_clib_put_matrix.py

@@ -19,6 +19,7 @@ def fixture_dtypes():
     return "int8 int16 int32 int64 uint8 uint16 uint32 uint64 float32 float64".split()
 
 
+@pytest.mark.benchmark
 def test_put_matrix(dtypes):
     """
     Check that assigning a numpy 2-D array to a dataset works.
@@ -64,6 +65,7 @@ def test_put_matrix_fails():
                 lib.put_matrix(dataset=None, matrix=np.empty((10, 2)), pad=0)
 
 
+@pytest.mark.benchmark
 def test_put_matrix_grid(dtypes):
     """
     Check that assigning a numpy 2-D array to an ASCII and netCDF grid works.

pygmt/tests/test_clib_put_strings.py

@@ -9,6 +9,7 @@
 from pygmt.helpers import GMTTempFile
 
 
+@pytest.mark.benchmark
 def test_put_strings():
     """
     Check that assigning a numpy array of dtype str to a dataset works.

pygmt/tests/test_clib_put_vector.py

@@ -20,6 +20,7 @@ def fixture_dtypes():
     return "int8 int16 int32 int64 uint8 uint16 uint32 uint64 float32 float64".split()
 
 
+@pytest.mark.benchmark
 def test_put_vector(dtypes):
     """
     Check that assigning a numpy array to a dataset works.
@@ -97,6 +98,7 @@ def test_put_vector_mixed_dtypes(dtypes):
                 npt.assert_allclose(newy, y)
 
 
+@pytest.mark.benchmark
 def test_put_vector_string_dtype():
     """
     Passing string type vectors to a dataset.

pygmt/tests/test_clib_virtualfiles.py

@@ -47,6 +47,7 @@ def fixture_dtypes_pandas(dtypes):
     return tuple(dtypes_pandas)
 
 
+@pytest.mark.benchmark
 def test_virtual_file(dtypes):
     """
     Test passing in data via a virtual file with a Dataset.
@@ -122,6 +123,7 @@ def test_virtual_file_bad_direction():
                 pass
 
 
+@pytest.mark.benchmark
 @pytest.mark.parametrize(
     ("array_func", "kind"),
     [(np.array, "matrix"), (pd.DataFrame, "vector"), (xr.Dataset, "vector")],
@@ -202,6 +204,7 @@ def test_virtualfile_from_data_fail_non_valid_data(data):
             )
 
 
+@pytest.mark.benchmark
 def test_virtualfile_from_vectors(dtypes):
     """
     Test the automation for transforming vectors to virtual file dataset.
@@ -221,6 +224,7 @@ def test_virtualfile_from_vectors(dtypes):
             assert output == expected
 
 
+@pytest.mark.benchmark
 @pytest.mark.parametrize("dtype", [str, object])
 def test_virtualfile_from_vectors_one_string_or_object_column(dtype):
     """
@@ -290,6 +294,7 @@ def test_virtualfile_from_vectors_diff_size():
                 pass
 
 
+@pytest.mark.benchmark
 def test_virtualfile_from_matrix(dtypes):
     """
     Test transforming a matrix to virtual file dataset.

pygmt/tests/test_coast.py

@@ -6,6 +6,7 @@
 from pygmt.exceptions import GMTInvalidInput
 
 
+@pytest.mark.benchmark
 @pytest.mark.mpl_image_compare
 def test_coast_region():
     """

pygmt/tests/test_colorbar.py

@@ -5,6 +5,7 @@
 from pygmt import Figure
 
 
+@pytest.mark.benchmark
 @pytest.mark.mpl_image_compare
 def test_colorbar_box():
     """

pygmt/tests/test_config.py

@@ -64,6 +64,7 @@ def test_config_font_annot():
     return fig
 
 
+@pytest.mark.benchmark
 @pytest.mark.mpl_image_compare
 def test_config_format_date_map():
     """

pygmt/tests/test_contour.py

@@ -46,6 +46,7 @@ def test_contour_vec(region):
     return fig
 
 
+@pytest.mark.benchmark
 @pytest.mark.mpl_image_compare(filename="test_contour_matrix.png")
 @pytest.mark.parametrize(
     "array_func",

pygmt/tests/test_datasets_load_remote_datasets.py

@@ -19,6 +19,21 @@ def load_remote_dataset_wrapper(resolution="01d", region=None, registration=None
     )
 
 
+@pytest.mark.benchmark
+def test_load_remote_dataset_benchmark_with_region():
+    """
+    Benchmark loading a remote dataset with 'region'.
+    """
+    data = load_remote_dataset_wrapper(resolution="01d", region=[-10, 10, -5, 5])
+    assert data.name == "seafloor_age"
+    assert data.attrs["long_name"] == "age of seafloor crust"
+    assert data.attrs["cpt"] == "@earth_age.cpt"
+    assert data.attrs["units"] == "Myr"
+    assert data.attrs["horizontal_datum"] == "WGS84"
+    assert data.gmt.registration == 0
+    assert data.shape == (11, 21)
+
+
 def test_load_remote_dataset_invalid_resolutions():
     """
     Make sure _load_remote_dataset fails for invalid resolutions.

pygmt/tests/test_datasets_samples.py

@@ -180,6 +180,7 @@ def test_load_notre_dame_topography():
     assert data["z"].max() == 960
 
 
+@pytest.mark.benchmark
 def test_earth_relief_holes():
     """
     Check that the @earth_relief_20m_holes.grd dataset loads without errors.

pygmt/tests/test_dimfilter.py

@@ -59,6 +59,7 @@ def test_dimfilter_outgrid(grid, expected_grid):
         xr.testing.assert_allclose(a=temp_grid, b=expected_grid)
 
 
+@pytest.mark.benchmark
 def test_dimfilter_no_outgrid(grid, expected_grid):
     """
     Test the required parameters for dimfilter with no set outgrid.

pygmt/tests/test_figure.py

@@ -56,6 +56,7 @@ def test_figure_region_country_codes():
     npt.assert_allclose(fig.region, np.array([0.0, 360.0, -90.0, 90.0]))
 
 
+@pytest.mark.benchmark
 def test_figure_repr():
     """
     Make sure that figure output's PNG and HTML printable representations look ok.

pygmt/tests/test_filter1d.py

@@ -22,7 +22,7 @@ def fixture_data():
 
 def test_filter1d_no_outfile(data):
     """
-    Test filter1d with no set outgrid.
+    Test filter1d with no set outfile.
     """
     result = filter1d(data=data, filter_type="g5")
     assert result.shape == (671, 2)
@@ -79,13 +79,17 @@ def test_filter1d_outfile_incorrect_output_type(data):
             assert Path(tmpfile.name).stat().st_size > 0  # check that outfile exists
 
 
+@pytest.mark.benchmark
 def test_filter1d_format(data):
     """
     Test that correct formats are returned.
     """
     time_series_default = filter1d(data=data, filter_type="g5")
     assert isinstance(time_series_default, pd.DataFrame)
+    assert time_series_default.shape == (671, 2)
     time_series_array = filter1d(data=data, filter_type="g5", output_type="numpy")
     assert isinstance(time_series_array, np.ndarray)
+    assert time_series_array.shape == (671, 2)
     time_series_df = filter1d(data=data, filter_type="g5", output_type="pandas")
     assert isinstance(time_series_df, pd.DataFrame)
+    assert time_series_df.shape == (671, 2)