Stronger 'Basin / profile' validation (#1486)

Fixes #279.
Follow up of #1469.

Most rules were already in place, this just adds a bit more validation.
I moved the validation out of the constructor to be more in line with
the other validation taking place.

core/src/parameter.jl

@@ -186,44 +186,6 @@ struct Basin{T, C, V1, V2, V3} <: AbstractParameterNode
     demand::Vector{Float64}
     # Data source for parameter updates
     time::StructVector{BasinTimeV1, C, Int}
-
-    function Basin(
-        node_id,
-        inflow_ids,
-        outflow_ids,
-        vertical_flux_from_input::V1,
-        vertical_flux::V2,
-        vertical_flux_prev::V3,
-        vertical_flux_integrated::V3,
-        vertical_flux_bmi::V3,
-        current_level::T,
-        current_area::T,
-        area,
-        level,
-        storage,
-        demand,
-        time::StructVector{BasinTimeV1, C, Int},
-    ) where {T, C, V1, V2, V3}
-        is_valid = valid_profiles(node_id, level, area)
-        is_valid || error("Invalid Basin / profile table.")
-        return new{T, C, V1, V2, V3}(
-            node_id,
-            inflow_ids,
-            outflow_ids,
-            vertical_flux_from_input,
-            vertical_flux,
-            vertical_flux_prev,
-            vertical_flux_integrated,
-            vertical_flux_bmi,
-            current_level,
-            current_area,
-            area,
-            level,
-            storage,
-            demand,
-            time,
-        )
-    end
 end
 
 """

core/src/read.jl

@@ -571,6 +571,11 @@ function Basin(db::DB, config::Config, graph::MetaGraph, chunk_sizes::Vector{Int
 
     node_id = NodeID.(NodeType.Basin, node_id)
 
+    is_valid = valid_profiles(node_id, level, area)
+    if !is_valid
+        error("Invalid Basin / profile table.")
+    end
+
     return Basin(
         Indices(node_id),
         [collect(inflow_ids(graph, id)) for id in node_id],

core/src/validation.jl

@@ -186,16 +186,20 @@ end
 Check whether the profile data has no repeats in the levels and the areas start positive.
 """
 function valid_profiles(
-    node_id::Indices{NodeID},
+    node_id::Vector{NodeID},
     level::Vector{Vector{Float64}},
     area::Vector{Vector{Float64}},
 )::Bool
     errors = false
-
     for (id, levels, areas) in zip(node_id, level, area)
+        n = length(levels)
+        if n < 2
+            errors = true
+            @error "$id profile must have at least two data points, got $n."
+        end
         if !allunique(levels)
             errors = true
-            @error "$id has repeated levels, this cannot be interpolated."
+            @error "$id profile has repeated levels, this cannot be interpolated."
         end
 
         if areas[1] <= 0
@@ -206,9 +210,9 @@ function valid_profiles(
             )
         end
 
-        if areas[end] < areas[end - 1]
+        if any(diff(areas) .< 0.0)
             errors = true
-            @error "$id profile cannot have decreasing area at the top since extrapolating could lead to negative areas."
+            @error "$id profile cannot have decreasing areas."
         end
     end
     return !errors

core/test/validation_test.jl

@@ -1,29 +1,26 @@
 @testitem "Basin profile validation" begin
-    using Dictionaries: Indices
     using Ribasim: NodeID, valid_profiles, qh_interpolation, ScalarInterpolation
     using Logging
     using StructArrays: StructVector
 
-    node_id = Indices([NodeID(:Basin, 1)])
+    node_id = [NodeID(:Basin, 1)]
     level = [[0.0, 0.0, 1.0]]
     area = [[0.0, 100.0, 90]]
 
     logger = TestLogger(; min_level = Debug)
     with_logger(logger) do
         @test !valid_profiles(node_id, level, area)
     end
-
     @test length(logger.logs) == 3
     @test logger.logs[1].level == Error
     @test logger.logs[1].message ==
-          "Basin #1 has repeated levels, this cannot be interpolated."
+          "Basin #1 profile has repeated levels, this cannot be interpolated."
     @test logger.logs[2].level == Error
     @test logger.logs[2].message ==
           "Basin #1 profile cannot start with area <= 0 at the bottom for numerical reasons."
     @test logger.logs[2].kwargs[:area] == 0
     @test logger.logs[3].level == Error
-    @test logger.logs[3].message ==
-          "Basin #1 profile cannot have decreasing area at the top since extrapolating could lead to negative areas."
+    @test logger.logs[3].message == "Basin #1 profile cannot have decreasing areas."
 
     table = StructVector(; flow_rate = [0.0, 0.1], level = [1.0, 2.0], node_id = [5, 5])
     itp = qh_interpolation(NodeID(:TabulatedRatingCurve, 5), table)

docs/core/usage.qmd

@@ -293,13 +293,13 @@ The profile table defines the physical dimensions of the storage reservoir of ea
 column    | type    | unit         | restriction
 --------- | ------- | ------------ | -----------
 node_id   | Int32   | -            | sorted
-area      | Float64 | $m^2$        | non-negative, per node_id: start positive and increasing
+area      | Float64 | $m^2$        | non-negative, per node_id: start positive and not decreasing
 level     | Float64 | $m$          | per node_id: increasing
 
 The level is the level at the basin outlet. All levels are defined in meters above a datum
 that is the same for the entire model. An example of the first 5 rows of such a table is
 given below. The first 4 rows define the profile of ID `2`. The number of rows can vary
-per ID. Using a very large number of rows may impact performance.
+per ID, and must be at least 2. Using a very large number of rows may impact performance.
 
 node_id |   area | level
 ------- |------- |-------