Merge pull request #23 from IGNF/fix/duplicatd_new_column

Overwrite "NEW_COLUMN" when it already exists in recipient file

Author: leavauchier

CHANGELOG.md

@@ -1,5 +1,8 @@
 # CHANGELOG
 
+## 1.4.2
+- fix: écrase la colonne "NEW_COLUMN" si elle existe dans le fichier receveur
+
 ## 1.4.1
 - fix lorsque les las donneurs ne sont pas de même version (1.2 et 1.4): on ne garde que les attributs communs
 

patchwork/patchwork.py

@@ -1,4 +1,5 @@
 import os
+import warnings
 from shutil import copy2
 from typing import List, Tuple
 
@@ -136,14 +137,16 @@ def get_complementary_points(
     )
 
     dfs_donor_points = []
-    
+
     if len(df_donor_info.index):
         donor_common_columns = get_common_donor_columns(df_donor_info)
         for index, row in df_donor_info.iterrows():
             with laspy.open(row["full_path"]) as donor_file:
                 raw_donor_points = donor_file.read().points
                 points_loc_gdf = gpd.GeoDataFrame(
-                    geometry=gpd.points_from_xy(raw_donor_points.x, raw_donor_points.y, raw_donor_points.z, crs=config.CRS)
+                    geometry=gpd.points_from_xy(
+                        raw_donor_points.x, raw_donor_points.y, raw_donor_points.z, crs=config.CRS
+                    )
                 )
                 footprint_gdf = gpd.GeoDataFrame(geometry=[row["geometry"]], crs=config.CRS)
                 points_in_footprint_gdf = points_loc_gdf.sjoin(footprint_gdf, how="inner", predicate="intersects")
@@ -190,7 +193,7 @@ def get_field_from_header(las_file: LasReader) -> List[str]:
 
 def test_field_exists(file_path: str, column: str) -> bool:
     output_file = laspy.read(file_path)
-    return column in get_field_from_header(output_file)
+    return column.lower() in get_field_from_header(output_file)
 
 
 def append_points(config: DictConfig, extra_points: pd.DataFrame):
@@ -217,19 +220,21 @@ def append_points(config: DictConfig, extra_points: pd.DataFrame):
 
     copy2(recipient_filepath, output_filepath)
 
-    # if we want a new column, we start by adding its name
+    # if we want a new column, we start by adding its name (or remove the existing column)
+
     if config.NEW_COLUMN:
+        output_las = laspy.read(output_filepath)
+
         if test_field_exists(recipient_filepath, config.NEW_COLUMN):
-            raise ValueError(
-                f"{config.NEW_COLUMN} already exists as \
-                             column name in {recipient_filepath}"
+            warnings.warn(
+                f"{config.NEW_COLUMN} already exists as a column name in {recipient_filepath}. Overwriting it."
             )
-        new_column_type = get_type(config.NEW_COLUMN_SIZE)
-        output_las = laspy.read(output_filepath)
+            output_las.remove_extra_dim(config.NEW_COLUMN)
+
         output_las.add_extra_dim(
             laspy.ExtraBytesParams(
                 name=config.NEW_COLUMN,
-                type=new_column_type,
+                type=get_type(config.NEW_COLUMN_SIZE),
                 description="Point origin: 0=initial las",
             )
         )

test/data/lidar_HD_decimated_with_origin_value/Semis_2022_0673_6362_LA93_IGN69_with_origin.laz

@@ -11,9 +11,9 @@
 import patchwork.constants as c
 from patchwork.patchwork import (
     append_points,
+    get_common_las_columns,
     get_complementary_points,
     get_field_from_header,
-    get_common_las_columns,
     get_selected_classes_points,
     get_type,
     patchwork,
@@ -434,6 +434,11 @@ def test_patchwork_default(tmp_path_factory, recipient_path, expected_nb_added_p
             True,
             0,
         ),  # No donor
+        (
+            "test/data/lidar_HD_decimated_with_origin_value/Semis_2022_0673_6362_LA93_IGN69_with_origin.laz",
+            True,
+            128675,
+        ),  # One donor. Origin dimension already in input file
     ],
 )
 def test_patchwork_with_origin(tmp_path_factory, recipient_path, donor_use_synthetic_points, expected_nb_added_points):
@@ -563,25 +568,40 @@ def test_patchwork_with_mount_points(tmp_path_factory, input_shp_path, recipient
         assert np.all(output_points.classification[output_points.Origin == 1] == 11)
         assert not np.any(output_points.classification[output_points.Origin == 0] == 11)
 
+
 @pytest.mark.parametrize(
     "las_liste",
     [
-        (["test/data/aveyron_lidarBD/data/NUALID_1-0_IAVEY_PTS_0673_6363_LAMB93_IGN69_20170519.laz", "test/data/aveyron_lidarBD/data/NUALID_1-0_IAVEY_PTS_0673_6364_LAMB93_IGN69_20170519.laz"]),
-        (['test/data/grand_geneve/grand_geneve_BD/data/NUALID_1-0_DS19RFAN_PTS_0963_6543_LAMB93_IGN69_20191002.laz', 'test/data/grand_geneve/grand_geneve_BD/data2/0963_6543.laz']),
+        (
+            [
+                "test/data/aveyron_lidarBD/data/NUALID_1-0_IAVEY_PTS_0673_6363_LAMB93_IGN69_20170519.laz",
+                "test/data/aveyron_lidarBD/data/NUALID_1-0_IAVEY_PTS_0673_6364_LAMB93_IGN69_20170519.laz",
+            ]
+        ),
+        (
+            [
+                "test/data/grand_geneve/grand_geneve_BD/data"
+                + "/NUALID_1-0_DS19RFAN_PTS_0963_6543_LAMB93_IGN69_20191002.laz",
+                "test/data/grand_geneve/grand_geneve_BD/data2/0963_6543.laz",
+            ]
+        ),
     ],
 )
 def test_get_common_las_columns(las_liste):
     common_columns = get_common_las_columns(las_liste)
-    assert all(col in common_columns for col in ["x", "y", "z", "classification", "gps_time", "intensity", "return_number", "number_of_returns"])
+    assert all(
+        col in common_columns
+        for col in ["x", "y", "z", "classification", "gps_time", "intensity", "return_number", "number_of_returns"]
+    )
 
 
 def test_patchwork_with_different_las_format(tmp_path_factory):
 
-    recipient_path=  "test/data/grand_geneve/lidar_HD_decimate/Semis_2021_0963_6543_LA93_IGN69_decimate.laz"
+    recipient_path = "test/data/grand_geneve/lidar_HD_decimate/Semis_2021_0963_6543_LA93_IGN69_decimate.laz"
     input_shp_path = "test/data/grand_geneve/geometry_GrandGeneve/zones.geojson"
     tmp_file_dir = tmp_path_factory.mktemp("data")
     tmp_output_las_name = "result_patchwork_different_las.laz"
-    
+
     tmp_output_indices_map_name = "result_patchwork_indices.tif"
     donor_class_translation = {2: 11, 9: 11}
 
@@ -594,7 +614,7 @@ def test_patchwork_with_different_las_format(tmp_path_factory):
                 f"filepath.SHP_DIRECTORY={os.path.dirname(input_shp_path)}",
                 f"filepath.SHP_NAME={os.path.basename(input_shp_path)}",
                 f"filepath.OUTPUT_DIR={tmp_file_dir}",
-                f"filepath.DONOR_SUBDIRECTORY=''",
+                "filepath.DONOR_SUBDIRECTORY=''",
                 f"filepath.OUTPUT_NAME={tmp_output_las_name}",
                 f"filepath.OUTPUT_INDICES_MAP_DIR={tmp_file_dir}",
                 f"filepath.OUTPUT_INDICES_MAP_NAME={tmp_output_indices_map_name}",
@@ -614,7 +634,7 @@ def test_patchwork_with_different_las_format(tmp_path_factory):
 
     with laspy.open(recipient_path) as recipient_file:
         recipient_points = recipient_file.read().points
-        
+
     with laspy.open(output_path) as las_file:
         output_points = las_file.read().points
         assert {n for n in las_file.header.point_format.dimension_names} == {
@@ -625,4 +645,4 @@ def test_patchwork_with_different_las_format(tmp_path_factory):
         assert np.sum(output_points.Origin == 1) == expected_nb_added_points
 
         assert np.all(output_points.classification[output_points.Origin == 1] == 11)
-        assert not np.any(output_points.classification[output_points.Origin == 0] == 11)
+        assert not np.any(output_points.classification[output_points.Origin == 0] == 11)

test/test_patchwork.py

@@ -1,4 +1,4 @@
-__version__ = "1.4.1"
+__version__ = "1.4.2"
 
 if __name__ == "__main__":
     print(__version__)