[WQ] Drop duplicate boundaries instead of making them unique. (#1917)

Found while debugging the LHM model. A tracer was added, but didn't show
up in Delwaq because the boundary list (bndlist) was shifted after
duplicate UserDemand edges.

Also added more information to the generated.csv files for debugging.

python/ribasim/ribasim/delwaq/generate.py

@@ -333,8 +333,15 @@ def generate(
 
     # Write topology to delwaq pointer file
     pointer = pd.DataFrame(G.edges(), columns=["from_node_id", "to_node_id"])
-    pointer.to_csv(output_path / "network.csv", index=False)  # not needed
     write_pointer(output_path / "ribasim.poi", pointer)
+    pointer["riba_edge_id"] = [e[2] for e in G.edges.data("id")]
+    pointer["riba_from_node_id"] = pointer["from_node_id"].map(
+        {v: k for k, v in node_mapping.items()}
+    )
+    pointer["riba_to_node_id"] = pointer["to_node_id"].map(
+        {v: k for k, v in node_mapping.items()}
+    )
+    pointer.to_csv(output_path / "network.csv", index=False)  # not needed
 
     total_segments = len(basin_mapping)
     total_exchanges = len(pointer)
@@ -363,6 +370,7 @@ def generate(
     flows.loc[m, "flow_rate"] = flows.loc[m, "flow_rate"] * -1
 
     # Map edge_id to the new edge_id and merge any duplicate flows
+    flows["riba_edge_id"] = flows["edge_id"]
     flows["edge_id"] = flows["edge_id"].map(edge_mapping)
     flows.dropna(subset=["edge_id"], inplace=True)
     flows["edge_id"] = flows["edge_id"].astype("int32")
@@ -389,7 +397,7 @@ def generate(
     nflows.sort_values(by=["time", "edge_id"], inplace=True)
     nflows.to_csv(output_path / "flows.csv", index=False)  # not needed
     nflows.drop(
-        columns=["edge_id"],
+        columns=["edge_id", "riba_edge_id"],
         inplace=True,
     )
     write_flows(output_path / "ribasim.flo", nflows, timestep)
@@ -400,12 +408,13 @@ def generate(
     # Write volumes to Delwaq format
     basins.drop(columns=["level"], inplace=True)
     volumes = basins[["time", "node_id", "storage"]]
+    volumes["riba_node_id"] = volumes["node_id"]
     volumes.loc[:, "node_id"] = (
         volumes["node_id"].map(basin_mapping).astype(pd.Int32Dtype())
     )
     volumes = volumes.sort_values(by=["time", "node_id"])
     volumes.to_csv(output_path / "volumes.csv", index=False)  # not needed
-    volumes.drop(columns=["node_id"], inplace=True)
+    volumes.drop(columns=["node_id", "riba_node_id"], inplace=True)
     write_volumes(output_path / "ribasim.vol", volumes, timestep)
     write_volumes(
         output_path / "ribasim.vel", volumes, timestep
@@ -479,24 +488,9 @@ def generate(
     bnd["node_id"] = [G.nodes(data="id")[bid] for bid in bnd["bid"]]
     bnd["fid"] = list(map(_boundary_name, bnd["node_id"], bnd["node_type"]))
     bnd["comment"] = ""
+    bnd.to_csv(output_path / "bndlist.csv", index=False)
     bnd = bnd[["fid", "comment", "node_type"]]
-
-    def prefix(d):
-        """Replace duplicate boundary names with unique names.
-
-        As the string length is fixed, we replace the first
-        character with a unique number.
-        """
-        n = len(d)
-        if n == 1:
-            return d
-        elif n == 2:
-            logger.debug(f"Renaming duplicate boundaries {d.iloc[0]}")
-            return [str(i) for i in range(n)] + d.str.lstrip("U")
-        else:
-            raise ValueError("Found boundary with more than 2 duplicates.")
-
-    bnd["fid"] = bnd.groupby("fid").fid.transform(prefix)
+    bnd.drop_duplicates(subset="fid", inplace=True)
     assert bnd["fid"].is_unique
 
     bnd.to_csv(