Simplify and rename function

Author: tsmbland

src/muse/readers/csv.py

@@ -13,7 +13,7 @@
     "read_initial_market",
     "read_attribute_table",
     "read_regression_parameters",
-    "read_csv_outputs",
+    "read_presets",
 ]
 
 from collections.abc import Sequence
@@ -788,13 +788,13 @@ def read_regression_parameters(path: Union[str, Path]) -> xr.Dataset:
     return coeffs
 
 
-def read_csv_outputs(
+def read_presets(
     paths: Union[str, Path, Sequence[Union[str, Path]]],
     columns: str = "commodity",
-    indices: Sequence[str] = ("RegionName", "ProcessName", "Timeslice"),
+    indices: Sequence[str] = ("RegionName", "Timeslice"),
     drop: Sequence[str] = ("Unnamed: 0",),
 ) -> xr.Dataset:
-    """Read standard MUSE output files for consumption or supply."""
+    """Read consumption or supply files for preset sectors."""
     from re import match
 
     from muse.readers import camel_to_snake
@@ -816,12 +816,22 @@ def expand_paths(path):
     datas = {}
     for path in allfiles:
         data = pd.read_csv(path, low_memory=False)
-        index_columns = [u for u in indices if u in data.columns]
+        assert all(u in data.columns for u in indices)
+
+        # Legacy: drop ProcessName column and sum data (PR #448)
+        if "ProcessName" in data.columns:
+            data = (
+                data.drop(columns=["ProcessName"])
+                .groupby(list(indices))
+                .sum()
+                .reset_index()
+            )
+
         data = data.drop(columns=[k for k in drop if k in data.columns])
-        data.index = pd.MultiIndex.from_arrays([data[u] for u in index_columns])
+        data.index = pd.MultiIndex.from_arrays([data[u] for u in indices])
         data.index.name = "asset"
         data.columns.name = columns
-        data = data.drop(columns=list(index_columns))
+        data = data.drop(columns=list(indices))
 
         reyear = match(r"\S*.(\d{4})\S*\.csv", path.name)
         if reyear is None:
@@ -838,7 +848,7 @@ def expand_paths(path):
         .sortby("year")
         .fillna(0)
         .unstack("asset")
-        .rename({k: k.replace("Name", "").lower() for k in index_columns})
+        .rename({k: k.replace("Name", "").lower() for k in indices})
     )
 
     if "commodity" in result.coords:

src/muse/sectors/preset_sector.py

@@ -24,8 +24,8 @@ def factory(cls, name: str, settings: Any) -> PresetSector:
         from muse.commodities import CommodityUsage
         from muse.readers import (
             read_attribute_table,
-            read_csv_outputs,
             read_macro_drivers,
+            read_presets,
             read_regression_parameters,
             read_timeslice_shares,
             read_timeslices,
@@ -40,7 +40,7 @@ def factory(cls, name: str, settings: Any) -> PresetSector:
             getattr(sector_conf, "timeslice_levels", None)
         ).timeslice
         if getattr(sector_conf, "consumption_path", None) is not None:
-            consumption = read_csv_outputs(sector_conf.consumption_path)
+            consumption = read_presets(sector_conf.consumption_path)
             presets["consumption"] = consumption.assign_coords(timeslice=timeslice)
         elif getattr(sector_conf, "demand_path", None) is not None:
             presets["consumption"] = read_attribute_table(sector_conf.demand_path)
@@ -89,7 +89,7 @@ def factory(cls, name: str, settings: Any) -> PresetSector:
             )
 
         if getattr(sector_conf, "supply_path", None) is not None:
-            supply = read_csv_outputs(sector_conf.supply_path)
+            supply = read_presets(sector_conf.supply_path)
             supply.coords["timeslice"] = presets.timeslice
             presets["supply"] = supply
 
@@ -99,9 +99,9 @@ def factory(cls, name: str, settings: Any) -> PresetSector:
             getattr(sector_conf, "lcoe_path", None) is not None and "supply" in presets
         ):
             costs = (
-                read_csv_outputs(
+                read_presets(
                     sector_conf.lcoe_path,
-                    indices=("RegionName", "ProcessName"),
+                    indices=("RegionName",),
                     columns="timeslices",
                 )
                 * presets["supply"]
@@ -131,9 +131,6 @@ def factory(cls, name: str, settings: Any) -> PresetSector:
             [CommodityUsage.PRODUCT if u else CommodityUsage.OTHER for u in comm_usage],
         )
         presets = presets.set_coords("comm_usage")
-        if "process" in presets.dims:
-            presets = presets.sum("process")
-
         interpolation_mode = getattr(sector_conf, "interpolation_mode", "linear")
         return cls(presets, interpolation_mode=interpolation_mode, name=name)
 

tests/test_readers.py

@@ -673,11 +673,11 @@ def test_read_attribute_table(default_model):
     )
 
 
-def test_read_csv_outputs(default_model):
-    from muse.readers.csv import read_csv_outputs
+def test_read_presets(default_model):
+    from muse.readers.csv import read_presets
 
     path = default_model / "technodata" / "preset" / "*Consumption.csv"
-    data = read_csv_outputs(str(path))
+    data = read_presets(str(path))
 
     assert isinstance(data, xr.DataArray)
     assert data.dtype == np.float64