--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/csopraSplitInput.py	Wed Aug 28 14:53:30 2024 +0000
@@ -0,0 +1,50 @@
+import pandas as pd
+import os
+import shutil
+import math
+import sys
+
+def split_csv(input_folder, output_folder, max_chunck_size):
+    unit_path = input_folder + "/units.csv"
+    soil_path = input_folder + "/soil.csv"
+    crop_path = input_folder + "/crop.csv"
+
+    os.mkdir(output_folder)
+    
+    dfs = pd.read_csv(soil_path)
+    dfc = pd.read_csv(crop_path)
+
+    max_ID = dfs.loc[dfs['ID'].idxmax()]['ID']
+
+    step = max_chunck_size
+    
+    nb_split = math.ceil(max_ID / max_chunck_size)
+
+    print(max_ID)
+
+    print(nb_split)
+
+    for i in range(1,nb_split + 1):
+
+        dfSplit = pd.DataFrame()
+        dfSplit = dfs.loc[((dfs['ID'] >= (((i-1) * step)) + 1) & (dfs['ID'] <= ((((i-1) * step)) + step )))]
+
+        dfcSplit = pd.DataFrame()
+        dfcSplit = dfc.loc[((dfc['ID'] >= (((i-1) * step)) + 1) & (dfc['ID'] <= ((((i-1) * step)) + step )))]
+        
+        os.mkdir(os.path.join(output_folder, f"{input_folder}_{i}"))
+        shutil.copy(unit_path, os.path.join(output_folder, f"{input_folder}_{i}"))
+        dfSplit.to_csv(os.path.join(output_folder, f"{input_folder}_{i}", "soil.csv"), index=False)
+        dfcSplit.to_csv(os.path.join(output_folder, f"{input_folder}_{i}", "crop.csv"), index=False)
+        shutil.make_archive( os.path.join(output_folder,f"{input_folder}_{i}"), 'zip',  os.path.join(output_folder, f"{input_folder}_{i}"))
+
+# Example usage
+input_folder = "input"
+output_folder = "outputs"
+
+input_file = sys.argv[1]
+max_chunck_size = int(sys.argv[2])
+
+shutil.unpack_archive(input_file, input_folder, 'zip')
+
+split_csv(input_folder, output_folder, max_chunck_size)