ENH: Add support for spatial aggregation arguments to exactextract (t…

…heroggy#159)

.gitignore

@@ -14,6 +14,7 @@ local/
 
 .vs/*
 .vscode/*
+.mypy_cache/*
 
 # Local History for Visual Studio Code
 .history/

cropclassification/preprocess/_timeseries_calc_openeo.py

@@ -90,7 +90,7 @@ def calculate_periodic_timeseries(
         id_column=conf.columns["id"],
         rasters_bands=images_bands,
         output_dir=timeseries_periodic_dir,
-        stats=["count", "mean", "median", "std", "min", "max"],  # type: ignore[arg-type]
+        stats=["count", "mean", "median", "std", "min", "max"],
         engine="pyqgis",
         nb_parallel=nb_parallel,
     )

cropclassification/util/zonal_stats_bulk/_zonal_stats_bulk_exactextract.py

@@ -274,10 +274,15 @@ def zonal_stats_band_tofile(
     for band in bands:
         index = raster_info.bands[band].band_index
         band_columns = include_cols.copy()
-        band_columns.extend([f"band_{index}_{stat}" for stat in stats])
+        band_columns.extend(
+            [f"band_{index}_{stat}" for stat in [stat.split("(")[0] for stat in stats]]
+        )
         band_stats_df = stats_df[band_columns].copy()
         band_stats_df.rename(
-            columns={f"band_{index}_{stat}": stat for stat in stats},
+            columns={
+                f"band_{index}_{stat}": stat
+                for stat in [stat.split("(")[0] for stat in stats]
+            },
             inplace=True,
         )
         # Add fid column to the beginning of the dataframe
@@ -288,6 +293,19 @@ def zonal_stats_band_tofile(
                 f"Write data for {len(band_stats_df.index)} parcels found to {output_paths[band]}"  # noqa: E501
             )
             if not output_paths[band].exists():
-                pdh.to_file(band_stats_df, output_paths[band], index=False)
+                # Write the info table to the output file
+                pdh.to_file(df=band_stats_df, path=output_paths[band], index=False)
+
+                # Write the parameters table to the output file
+                spatial_aggregation_args_df = pd.DataFrame(
+                    data=stats, columns=["stats"]
+                )
+                pdh.to_file(
+                    df=spatial_aggregation_args_df,
+                    path=output_paths[band],
+                    table_name="params",
+                    index=False,
+                    append=True,
+                )
 
     return output_paths

tests/test_zonal_stats_bulk.py

@@ -3,14 +3,28 @@
 import geofileops as gfo
 import pytest
 
+from cropclassification.helpers import config_helper as conf
 from cropclassification.helpers import pandas_helper as pdh
 from cropclassification.util import zonal_stats_bulk
 from cropclassification.util.zonal_stats_bulk._zonal_stats_bulk_pyqgis import HAS_QGIS
 from tests.test_helper import SampleData
 
 
-@pytest.mark.parametrize("engine", ["pyqgis", "rasterstats", "exactextract"])
-def test_zonal_stats_bulk(tmp_path, engine):
+@pytest.mark.parametrize(
+    "engine, stats",
+    [
+        ("pyqgis", ["mean", "count"]),
+        ("rasterstats", ["mean", "count"]),
+        (
+            "exactextract",
+            [
+                "mean(min_coverage_frac=0.5,coverage_weight=none)",
+                "count(min_coverage_frac=0.5,coverage_weight=none)",
+            ],
+        ),
+    ],
+)
+def test_zonal_stats_bulk(tmp_path, engine, stats):
     if engine == "pyqgis" and not HAS_QGIS:
         pytest.skip("QGIS is not available on this system.")
 
@@ -19,6 +33,16 @@ def test_zonal_stats_bulk(tmp_path, engine):
     test_dir = tmp_path / sample_dir.name
     shutil.copytree(sample_dir, test_dir)
 
+    # Read the configuration
+    config_paths = [
+        SampleData.config_dir / "cropgroup.ini",
+        SampleData.tasks_dir / "local_overrule.ini",
+    ]
+    conf.read_config(
+        config_paths=config_paths,
+        default_basedir=sample_dir,
+    )
+
     # Make sure the s2-agri input file was copied
     test_image_roi_dir = test_dir / SampleData.image_dir.name / SampleData.roi_name
     test_s1_asc_dir = test_image_roi_dir / SampleData.image_s1_asc_path.parent.name
@@ -34,7 +58,7 @@ def test_zonal_stats_bulk(tmp_path, engine):
         id_column="UID",
         rasters_bands=images_bands,
         output_dir=tmp_path,
-        stats=["mean", "count"],
+        stats=stats,
         engine=engine,
     )