Use uid=20075619(cenv0899) gid=11125(cenv0899_) groups=11125(cenv0899…

…_),10031(OUCE_Linux_mistral),10044(linux-users),11180(cenv1030_) rather than

src/open_gira/direct_damages.py

@@ -304,7 +304,7 @@ def direct_damage(
 
     Args:
         exposure: Table containing exposed assets, likely edges split on a
-            raster grid (i.e. `edge_id` is not necessarily unique).
+            raster grid (i.e. `id` is not necessarily unique).
         damage_curves: Relationship between hazard intensity and damage
             fraction, keyed by `asset_type`.
         hazard_columns: Columns in `exposure` which denote hazard intensities.
@@ -313,7 +313,7 @@ def direct_damage(
 
     Returns:
         Direct damage fraction, rows are splits of edges.
-        Direct damage cost, rows are edges and `edge_id` should now be unique.
+        Direct damage cost, rows are edges and `id` should now be unique.
     """
 
     ##########################################################
@@ -382,11 +382,11 @@ def direct_damage(
     # join the other fields with the direct damage estimates
     logging.info("Unifying rasterised segments and summing damage costs")
 
-    # grouping on edge_id, sum all direct damage estimates to give a total dollar cost per edge
+    # grouping on id, sum all direct damage estimates to give a total dollar cost per edge
     direct_damages = pd.concat(
-        [direct_damages_only, damage_fraction["edge_id"]],
+        [direct_damages_only, damage_fraction["id"]],
         axis="columns"
-    ).set_index("edge_id")
+    ).set_index("id")
     grouped_direct_damages = direct_damages.groupby(direct_damages.index).sum()
 
-    return damage_fraction, grouped_direct_damages
+    return damage_fraction, grouped_direct_damages

workflow/transport-flood/event_set_direct_damages.py

@@ -82,21 +82,21 @@
 
     # lose columns like "cell_indicies" or rastered length measures that are specific to _rastered_ edges
     non_hazard_output_columns = list(set(non_hazard_columns) & set(unsplit.columns))
-    unsplit_subset = unsplit[non_hazard_output_columns].set_index("edge_id", drop=False)
+    unsplit_subset = unsplit[non_hazard_output_columns].set_index("id", drop=False)
 
     # rejoin direct damage cost estimates with geometry and metadata columns and write to disk
     # join on 'right' / grouped_direct_damages index to only keep rows we have damages for
     direct_damages = unsplit_subset.join(grouped_direct_damages, validate="one_to_one", how="right")
-    direct_damages["edge_id"] = direct_damages.index
+    direct_damages["id"] = direct_damages.index
     # we may not have calculated damages for every possible asset_type
     assert len(direct_damages) <= len(unsplit_subset)
-    assert "edge_id" in direct_damages.columns
+    assert "id" in direct_damages.columns
 
     # damage_fraction is on the split geometries, will have more rows
     assert len(damage_fraction) >= len(direct_damages)
 
     for dataframe in (damage_fraction, direct_damages):
-        assert "edge_id" in dataframe
+        assert "id" in dataframe
 
     logging.info(f"Writing out {damage_fraction.shape=} (per split geometry, event raster)")
     damage_fraction.to_parquet(damage_fraction_path)

workflow/transport-flood/return_period_direct_damages.py

@@ -130,15 +130,15 @@
 
     # lose columns like "cell_indicies" or rastered length measures that are specific to _rastered_ edges
     non_hazard_output_columns = list(set(non_hazard_columns) & set(unsplit.columns))
-    unsplit_subset = unsplit[non_hazard_output_columns].set_index("edge_id", drop=False)
+    unsplit_subset = unsplit[non_hazard_output_columns].set_index("id", drop=False)
 
     # rejoin direct damage cost estimates with geometry and metadata columns and write to disk
     # join on 'right' / grouped_direct_damages index to only keep rows we have damages for
     direct_damages = unsplit_subset.join(grouped_direct_damages, validate="one_to_one", how="right")
-    direct_damages["edge_id"] = direct_damages.index
+    direct_damages["id"] = direct_damages.index
     # we may not have calculated damages for every possible asset_type
     assert len(direct_damages) <= len(unsplit_subset)
-    assert "edge_id" in direct_damages.columns
+    assert "id" in direct_damages.columns
 
     expected_annual_damages_only = pd.DataFrame(data=expected_annual_damages, index=grouped_direct_damages.index)
     # rejoin expected annual damage cost estimates with geometry and metadata columns and write to disk
@@ -147,7 +147,7 @@
         unsplit_subset.join(expected_annual_damages_only, validate="one_to_one", how="right")
     )
     assert len(expected_annual_damages) <= len(unsplit_subset)
-    assert "edge_id" in expected_annual_damages.columns
+    assert "id" in expected_annual_damages.columns
 
     # combined the per return period and the integrated outputs into a single dataframe
     return_period_and_ead_damages = direct_damages.join(expected_annual_damages_only, validate="one_to_one")
@@ -161,7 +161,7 @@
     assert len(direct_damages) == len(expected_annual_damages)
 
     for dataframe in (damage_fraction, direct_damages, expected_annual_damages, return_period_and_ead_damages):
-        assert "edge_id" in dataframe
+        assert "id" in dataframe
 
     logging.info(f"Writing out {damage_fraction.shape=} (per split geometry, hazard RP map)")
     damage_fraction.to_parquet(damage_fraction_path)