Merge pull request #64 from astronomy-commons/issue/50/plot

Update columns of generated partition_info file.

src/hipscat/io/write_metadata.py

@@ -106,8 +106,25 @@ def write_partition_info(catalog_parameters, destination_pixel_map: dict):
         catalog_parameters.catalog_base_dir
     )
     data_frame = pd.DataFrame(destination_pixel_map.keys())
-    data_frame.columns = ["order", "pixel", "num_objects"]
-    data_frame = data_frame.astype(int)
+
+    # Set column names and add a directory column.
+    data_frame.columns = [
+        "Norder",
+        "Npix",
+        "num_objects",
+    ]
+    data_frame["Dir"] = int(data_frame["Npix"] / 10_000) * 10_000
+
+    # Reorder the columns to match full path, and force to integer types.
+    data_frame = data_frame[
+        [
+            "Norder",
+            "Dir",
+            "Npix",
+            "num_objects",
+        ]
+    ].astype(int)
+
     file_io.write_dataframe_to_csv(data_frame, partition_info_pointer, index=False)
 
 

tests/hipscat/io/test_write_metadata.py

@@ -106,8 +106,8 @@ def test_write_provenance_info(assert_text_file_matches, tmp_path, basic_catalog
 def test_write_partition_info(assert_text_file_matches, tmp_path, basic_catalog_info):
     """Test that we accurately write out the individual partition stats"""
     expected_lines = [
-        "order,pixel,num_objects",
-        "0,11,131",
+        "Norder,Dir,Npix,num_objects",
+        "0,0,11,131",
     ]
     pixel_map = {tuple([0, 11, 131]): [44, 45, 46]}
     io.write_partition_info(basic_catalog_info, pixel_map)
@@ -121,8 +121,8 @@ def test_write_partition_info_float(
     """Test that we accurately write out the individual partition stats
     even when the input is floats instead of ints"""
     expected_lines = [
-        "order,pixel,num_objects",
-        "0,11,131",
+        "Norder,Dir,Npix,num_objects",
+        "0,0,11,131",
     ]
     pixel_map = {tuple([0.0, 11.0, 131.0]): [44.0, 45.0, 46.0]}
     io.write_partition_info(basic_catalog_info, pixel_map)