Tidy src/variable.jl (#583)

docs/src/examples/general_examples/basic_usage.jl

@@ -104,7 +104,7 @@ evaluate(y)
 #
 
 using GLPK
-x = Variable(4, :Int)
+x = Variable(4, IntVar)
 p = minimize(sum(x), x >= 0.5)
 solve!(p, GLPK.Optimizer; silent_solver = true)
 evaluate(x)

docs/src/examples/mixed_integer/binary_knapsack.jl

@@ -22,7 +22,7 @@ n = length(w)
 #-
 
 using Convex, GLPK
-x = Variable(n, :Bin)
+x = Variable(n, BinVar)
 problem = maximize(dot(p, x), dot(w, x) <= C)
 solve!(problem, GLPK.Optimizer)
 evaluate(x)

docs/src/examples/mixed_integer/n_queens.jl

@@ -6,7 +6,7 @@ include(aux("antidiag.jl"))
 
 n = 8
 # We encode the locations of the queens with a matrix of binary random variables.
-x = Variable((n, n), :Bin)
+x = Variable((n, n), BinVar)
 
 # Now we impose the constraints: at most one queen on any anti-diagonal, at most one queen on any diagonal, and we must have exactly one queen per row and per column.
 ## At most one queen on any anti-diagonal

docs/src/examples/mixed_integer/section_allocation.jl

@@ -7,15 +7,15 @@
 # 10,000 or a large number would mean the student can not attend that section).
 #
 # The goal will be to get an allocation matrix $X$, where $X_{ij} = 1$ if
-# student $i$ is assigned to section $j$ and $0$ otherwise. 
+# student $i$ is assigned to section $j$ and $0$ otherwise.
 
 using Convex, GLPK
 aux(str) = joinpath(@__DIR__, "aux_files", str) # path to auxiliary files
 
 # Load our preference matrix, `P`
 include(aux("data.jl"))
 
-X = Variable(size(P), :Bin)
+X = Variable(size(P), BinVar)
 
 # We want every student to be assigned to exactly one section. So, every row
 # must have exactly one non-zero entry. In other words, the sum of all the

src/deprecations.jl

@@ -49,3 +49,72 @@ function Base.in(x::AbstractExpr, y::Symbol)
     )
     return isposdef(x)
 end
+
+# Compatability with old `sets` model.
+#
+# Only dispatch to these methods when at least one set is given.
+function Variable(
+    size::Tuple{Int,Int},
+    sign::Sign,
+    set::Symbol,
+    sets::Symbol...,
+)
+    @warn(
+        "Using symbols in `Variable` constructor is deprecated. Use " *
+        "`Convex.BinVar` or `Convex.IntVar` instead.",
+        maxlog = 1,
+    )
+    sets = [set, sets...]
+    x = if :Bin in sets
+        Variable(size, sign, BinVar)
+    elseif :Int in sets
+        Variable(size, sign, IntVar)
+    else
+        Variable(size, sign, ContVar)
+    end
+    if :Semidefinite in sets
+        add_constraint!(x, x ⪰ 0)
+    end
+    return x
+end
+
+function Variable(sign::Sign, set::Symbol, sets::Symbol...)
+    return Variable((1, 1), sign, set, sets...)
+end
+
+function Variable(size::Tuple{Int,Int}, set::Symbol, sets::Symbol...)
+    return Variable(size, NoSign(), set, sets...)
+end
+
+function Variable(m::Int, set::Symbol, sets::Symbol...)
+    return Variable((m, 1), NoSign(), set, sets...)
+end
+
+function Variable(set::Symbol, sets::Symbol...)
+    return Variable((1, 1), NoSign(), set, sets...)
+end
+
+function ComplexVariable(size::Tuple{Int,Int}, set::Symbol, sets::Symbol...)
+    @warn(
+        "Using symbols in `ComplexVariable` constructor is deprecated. Use " *
+        "`isposdef(x)` to enforce semidefiniteness instead of `:Semidefinite.",
+        maxlog = 1,
+    )
+    sets = [set, sets...]
+    if :Bin in sets
+        throw(ArgumentError("Complex variables cannot be restricted to binary"))
+    elseif :Int in sets
+        throw(
+            ArgumentError("Complex variables cannot be restricted to integer"),
+        )
+    end
+    x = ComplexVariable(size)
+    if :Semidefinite in sets
+        add_constraint!(x, x ⪰ 0)
+    end
+    return x
+end
+
+function ComplexVariable(set::Symbol, sets::Symbol...)
+    return ComplexVariable((1, 1), set, sets...)
+end

src/problem_depot/problems/mip.jl

@@ -46,7 +46,7 @@ end
         @test p.optval ≈ 5 atol = atol rtol = rtol
     end
 
-    x = Variable(2, :Int)
+    x = Variable(2, IntVar)
     p = minimize(sum(x), x >= 4.3; numeric_type = T)
 
     if test
@@ -69,7 +69,7 @@ end
         @test p.optval ≈ 12 atol = atol rtol = rtol
     end
 
-    x = Variable(2, :Int)
+    x = Variable(2, IntVar)
     p = minimize(norm(x, 1), x[1] >= 4.3; numeric_type = T)
 
     if test
@@ -80,7 +80,7 @@ end
         @test p.optval ≈ 5 atol = atol rtol = rtol
     end
 
-    x = Variable(2, :Int)
+    x = Variable(2, IntVar)
     p = minimize(sum(x), x[1] >= 4.3, x >= 0; numeric_type = T)
 
     if test
@@ -91,7 +91,7 @@ end
         @test p.optval ≈ 5 atol = atol rtol = rtol
     end
 
-    x = Variable(2, :Int)
+    x = Variable(2, IntVar)
     p = minimize(sum(x), x >= 0.5; numeric_type = T)
 
     if test
@@ -110,7 +110,7 @@ end
     rtol,
     ::Type{T},
 ) where {T,test}
-    x = Variable(2, :Bin)
+    x = Variable(2, BinVar)
     p = minimize(sum(x), x >= 0.5; numeric_type = T)
 
     if test
@@ -121,7 +121,7 @@ end
         @test p.optval ≈ 2 atol = atol rtol = rtol
     end
 
-    x = Variable(2, :Bin)
+    x = Variable(2, BinVar)
     p = minimize(sum(x), x[1] >= 0.5, x >= 0; numeric_type = T)
 
     if test

src/problem_depot/problems/sdp.jl

@@ -6,7 +6,7 @@
     rtol,
     ::Type{T},
 ) where {T,test}
-    y = Variable((2, 2), :Semidefinite)
+    y = Semidefinite((2, 2))
     p = minimize(y[1, 1]; numeric_type = T)
 
     # @fact problem_vexity(p) --> ConvexVexity()
@@ -15,7 +15,7 @@
         @test p.optval ≈ 0 atol = atol rtol = rtol
     end
 
-    y = Variable((3, 3), :Semidefinite)
+    y = Semidefinite((3, 3))
     p = minimize(y[1, 1], y[2, 2] == 1; numeric_type = T)
 
     # @fact problem_vexity(p) --> ConvexVexity()
@@ -27,7 +27,7 @@
     # Solution is obtained as y[2,2] -> infinity
     # This test fails on Mosek. See
     # https://github.com/JuliaOpt/Mosek.jl/issues/29
-    # y = Variable((2, 2), :Semidefinite)
+    # y = Semidefinite((2, 2))
     # p = minimize(y[1, 1], y[1, 2] == 1; numeric_type = T)
 
     # # @fact problem_vexity(p) --> ConvexVexity()
@@ -43,7 +43,7 @@
         @test p.optval ≈ 1 atol = atol rtol = rtol
     end
 
-    y = Variable((3, 3), :Semidefinite)
+    y = Semidefinite((3, 3))
     p = minimize(tr(y), y[2, 1] <= 4, y[2, 2] >= 3; numeric_type = T)
 
     # @fact problem_vexity(p) --> ConvexVexity()