Standardize naming conventions (#63)

docs/src/manual/linear_systems.md

@@ -49,14 +49,14 @@ Here is a list of currently supported linear solvers:
 
 | Linear solver type | System | Backend | Method |
 |:-------------------|:-------|:--------|:-------|
-| [`Dense_SymPosDef`](@ref) | Normal equations | Dense / LAPACK | Cholesky
-| [`Cholmod_SymQuasDef`](@ref) | Augmented system | CHOLMOD | LDLᵀ
-| [`Cholmod_SymPosDef`](@ref) | Normal equations | CHOLMOD | Cholesky
-| [`LDLFact_SymQuasDef`](@ref) | Augmented system | [LDLFactorizations.jl](https://github.com/JuliaSmoothOptimizers/LDLFactorizations.jl) | LDLᵀ
-| [`KrylovSPDSolver`](@ref) | Normal equations | [Krylov.jl](https://github.com/JuliaSmoothOptimizers/Krylov.jl) | Krylov
-| [`KrylovSIDSolver`](@ref) | Augmented system[^1] | [Krylov.jl](https://github.com/JuliaSmoothOptimizers/Krylov.jl) | Krylov
-| [`KrylovSQDSolver`](@ref) | Augmented system[^1] | [Krylov.jl](https://github.com/JuliaSmoothOptimizers/Krylov.jl) | Krylov
-
-[^1]: [`KrylovSIDSolver`](@ref)s view the augmented system as a symmetric indefinite system,
-    while [`KrylovSQDSolver`](@ref)s exploit its 2x2 structure and quasi-definite property.
+| [`DenseSPD`](@ref) | Normal equations | Dense / LAPACK | Cholesky
+| [`CholmodSQD`](@ref) | Augmented system | CHOLMOD | LDLᵀ
+| [`CholmodSPD`](@ref) | Normal equations | CHOLMOD | Cholesky
+| [`LDLFactSQD`](@ref) | Augmented system | [LDLFactorizations.jl](https://github.com/JuliaSmoothOptimizers/LDLFactorizations.jl) | LDLᵀ
+| [`KrylovSPD`](@ref) | Normal equations | [Krylov.jl](https://github.com/JuliaSmoothOptimizers/Krylov.jl) | Krylov
+| [`KrylovSID`](@ref) | Augmented system[^1] | [Krylov.jl](https://github.com/JuliaSmoothOptimizers/Krylov.jl) | Krylov
+| [`KrylovSQD`](@ref) | Augmented system[^1] | [Krylov.jl](https://github.com/JuliaSmoothOptimizers/Krylov.jl) | Krylov
+
+[^1]: [`KrylovSID`](@ref)s view the augmented system as a symmetric indefinite system,
+    while [`KrylovSQD`](@ref)s exploit its 2x2 structure and quasi-definite property.
     See the reference documentation for more details.

docs/src/reference/kkt_solvers.md

@@ -44,7 +44,7 @@ CurrentModule = Tulip.KKT
 ## Dense/LAPACK
 
 ```@docs
-Dense_SymPosDef
+DenseSPD
 ```
 
 ## CHOLMOD
@@ -54,17 +54,17 @@ CholmodSolver
 ```
 
 ```@docs
-Cholmod_SymQuasDef
+CholmodSQD
 ```
 
 ```@docs
-Cholmod_SymPosDef
+CholmodSPD
 ```
 
 ## LDLFactorizations
 
 ```@docs
-LDLFact_SymQuasDef
+LDLFactSQD
 ```
 
 ## Krylov
@@ -75,13 +75,13 @@ LDLFact_SymQuasDef
 
 
 ```@docs
-KrylovSPDSolver
+KrylovSPD
 ```
 
 ```@docs
-KrylovSIDSolver
+KrylovSID
 ```
 
 ```@docs
-KrylovSQDSolver
+KrylovSQD
 ```

docs/src/reference/parameters.md

@@ -42,7 +42,7 @@ Numerical tolerances for the interior-point algorithm.
 | Parameter | Description | Default |
 |:----------|:------------|:--------|
 | `MatrixOptions` | See [`TLA.MatrixOptions`](@ref) | `SparseMatrixCSC`
-| `KKTOptions` | See [`KKT.SolverOptions`](@ref) | [`KKT.Cholmod_SymQuasDef`](@ref) for `Float64`, [`KKT.LDLFact_SymQuasDef`](@ref) otherwise |
+| `KKTOptions` | See [`KKT.SolverOptions`](@ref) | [`KKT.CholmodSQD`](@ref) for `Float64`, [`KKT.LDLFactSQD`](@ref) otherwise |
 | `BarrierPRegMin` | Minimum value of primal regularization | ``\sqrt{\epsilon}`` |
 | `BarrierDregMin` | Minimum value of dual regularization | ``\sqrt{\epsilon}``
 

src/Interfaces/MOI/MOI_wrapper.jl

@@ -53,17 +53,17 @@ end
     _bounds(s)
 
 """
-_bounds(s::MOI.EqualTo{Tv}) where{Tv} = s.value, s.value
-_bounds(s::MOI.LessThan{Tv}) where{Tv}  = Tv(-Inf), s.upper
-_bounds(s::MOI.GreaterThan{Tv}) where{Tv}  = s.lower, Tv(Inf)
-_bounds(s::MOI.Interval{Tv}) where{Tv}  = s.lower, s.upper
+_bounds(s::MOI.EqualTo{T}) where{T} = s.value, s.value
+_bounds(s::MOI.LessThan{T}) where{T}  = T(-Inf), s.upper
+_bounds(s::MOI.GreaterThan{T}) where{T}  = s.lower, T(Inf)
+_bounds(s::MOI.Interval{T}) where{T}  = s.lower, s.upper
 
-const SCALAR_SETS{Tv} = Union{
-    MOI.LessThan{Tv},
-    MOI.GreaterThan{Tv},
-    MOI.EqualTo{Tv},
-    MOI.Interval{Tv}
-} where{Tv}
+const SCALAR_SETS{T} = Union{
+    MOI.LessThan{T},
+    MOI.GreaterThan{T},
+    MOI.EqualTo{T},
+    MOI.Interval{T}
+} where{T}
 
 
 
@@ -76,12 +76,12 @@ const SCALAR_SETS{Tv} = Union{
 # ==============================================================================
 
 """
-    Optimizer{Tv}
+    Optimizer{T}
 
 Wrapper for MOI.
 """
-mutable struct Optimizer{Tv} <: MOI.AbstractOptimizer
-    inner::Model{Tv}
+mutable struct Optimizer{T} <: MOI.AbstractOptimizer
+    inner::Model{T}
 
     is_feas::Bool  # Model is feasibility problem if true
 
@@ -91,7 +91,7 @@ mutable struct Optimizer{Tv} <: MOI.AbstractOptimizer
     var_indices_moi::Vector{MOI.VariableIndex}
     var_indices::Dict{MOI.VariableIndex, Int}
     con_indices_moi::Vector{MOI.ConstraintIndex}
-    con_indices::Dict{MOI.ConstraintIndex{MOI.ScalarAffineFunction{Tv}, <:SCALAR_SETS{Tv}}, Int}
+    con_indices::Dict{MOI.ConstraintIndex{MOI.ScalarAffineFunction{T}, <:SCALAR_SETS{T}}, Int}
 
     # Variable and constraint names
     name2var::Dict{String, MOI.VariableIndex}
@@ -104,14 +104,14 @@ mutable struct Optimizer{Tv} <: MOI.AbstractOptimizer
     # Keep track of bound constraints
     var2bndtype::Dict{MOI.VariableIndex, Set{Type{<:MOI.AbstractScalarSet}}}
 
-    function Optimizer{Tv}(;kwargs...) where{Tv}
-        m = new{Tv}(
-            Model{Tv}(), false,
+    function Optimizer{T}(;kwargs...) where{T}
+        m = new{T}(
+            Model{T}(), false,
             # Variable and constraint counters
             0, 0, 
             # Index mapping
             MOI.VariableIndex[], Dict{MOI.VariableIndex, Int}(),
-            MOI.ConstraintIndex[], Dict{MOI.ConstraintIndex{MOI.ScalarAffineFunction, <:SCALAR_SETS{Tv}}, Int}(),
+            MOI.ConstraintIndex[], Dict{MOI.ConstraintIndex{MOI.ScalarAffineFunction, <:SCALAR_SETS{T}}, Int}(),
             # Name -> index mapping
             Dict{String, MOI.VariableIndex}(), Dict{String, MOI.ConstraintIndex}(),
             Dict{MOI.ConstraintIndex, String}(),  # Variable bounds tracking

src/Interfaces/MOI/attributes.jl

@@ -111,8 +111,8 @@ MOI.get(m::Optimizer, ::MOI.NumberOfVariables) = m.inner.pbdata.nvar
 #   ObjectiveFunctionType
 #
 MOI.get(
-    m::Optimizer{Tv}, ::MOI.ObjectiveFunctionType
-) where{Tv} = MOI.ScalarAffineFunction{Tv}
+    m::Optimizer{T}, ::MOI.ObjectiveFunctionType
+) where{T} = MOI.ScalarAffineFunction{T}
 
 #
 #   ObjectiveSense
@@ -140,15 +140,15 @@ end
 #
 #   ObjectiveValue
 #
-function MOI.get(m::Optimizer{Tv}, attr::MOI.ObjectiveValue) where{Tv}
+function MOI.get(m::Optimizer{T}, attr::MOI.ObjectiveValue) where{T}
     MOI.check_result_index_bounds(m, attr)
     return get_attribute(m.inner, ObjectiveValue())
 end
 
 #
 #   DualObjectiveValue
 #
-function MOI.get(m::Optimizer{Tv}, attr::MOI.DualObjectiveValue) where{Tv}
+function MOI.get(m::Optimizer{T}, attr::MOI.DualObjectiveValue) where{T}
     MOI.check_result_index_bounds(m, attr)
     return get_attribute(m.inner, DualObjectiveValue())
 end
@@ -161,11 +161,11 @@ MOI.get(m::Optimizer, ::MOI.RawSolver) = m.inner
 #
 #   RelativeGap
 #
-function MOI.get(m::Optimizer{Tv}, ::MOI.RelativeGap) where{Tv}
+function MOI.get(m::Optimizer{T}, ::MOI.RelativeGap) where{T}
     # TODO: dispatch a function call on m.inner
     zp = m.inner.solver.primal_bound_scaled
     zd = m.inner.solver.dual_bound_scaled
-    return (abs(zp - zd) / (Tv(1 // 10^6)) + abs(zd))
+    return (abs(zp - zd) / (T(1 // 10^6)) + abs(zd))
 end
 
 #