Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

add_constrained_variable with 2 sets #2574

Merged
merged 3 commits into from
Nov 6, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 17 additions & 0 deletions src/Test/test_model.jl
Original file line number Diff line number Diff line change
Expand Up @@ -1226,3 +1226,20 @@ function test_model_show(model::MOI.ModelLike, ::Config{T}) where {T}
@test sprint(show, model) isa String
return
end

function test_model_add_constrained_variable_tuple(
model::MOI.ModelLike,
::Config{T},
) where {T}
F = MOI.VariableIndex
set = (MOI.GreaterThan(zero(T)), MOI.LessThan(one(T)))
@requires MOI.supports_add_constrained_variable(model, typeof(set))
x, (c_l, c_u) = MOI.add_constrained_variable(model, set)
@test c_l == MOI.ConstraintIndex{F,MOI.GreaterThan{T}}(x.value)
@test c_u == MOI.ConstraintIndex{F,MOI.LessThan{T}}(x.value)
@test MOI.get(model, MOI.ConstraintFunction(), c_l) == x
@test MOI.get(model, MOI.ConstraintSet(), c_l) == set[1]
@test MOI.get(model, MOI.ConstraintFunction(), c_u) == x
@test MOI.get(model, MOI.ConstraintSet(), c_u) == set[2]
return
end
59 changes: 59 additions & 0 deletions src/variables.jl
Original file line number Diff line number Diff line change
Expand Up @@ -124,6 +124,65 @@ function add_constrained_variable(model::ModelLike, set::AbstractScalarSet)
return variable, constraint
end

"""
add_constrained_variable(
model::ModelLike,
set::Tuple{<:GreaterThan,<:LessThan},
)

A special-case method to add a scalar variable with a lower and upper bound.

This method should be implemented by optimizers which have native support for
adding a variable with bounds and which cannot performantly modify the variable
bounds after creation.

## Example

```jldoctest
julia> import MathOptInterface as MOI

julia> model = MOI.Utilities.Model{Float64}();

julia> set = (MOI.GreaterThan(1.0), MOI.LessThan(2.0));

julia> x, (c_l, c_u) = MOI.add_constrained_variable(model, set);

julia> c_l
MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.GreaterThan{Float64}}(1)

julia> c_u
MathOptInterface.ConstraintIndex{MathOptInterface.VariableIndex, MathOptInterface.LessThan{Float64}}(1)

julia> print(model)
Feasibility

Subject to:

VariableIndex-in-GreaterThan{Float64}
v[1] >= 1.0

VariableIndex-in-LessThan{Float64}
v[1] <= 2.0
```
"""
function add_constrained_variable(
model::ModelLike,
set::Tuple{<:GreaterThan,<:LessThan},
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@blegat are you okay with this signature?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I have a small preference for two separate arguments rather than the tuple. But if you both prefer the Tuple, let's do it!

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I feel like the two separate arguments just opens the door too much.

This feels more like a "set" composed of a tuple. Rather than a completely new method. It's more logically consistent with the rest of the API. And the output of the constraint index matches the input of the set. Tuple->Tuple.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I also think restricting to this one special case is a good start. We can always relax later if it turns out to be a good idea. But this is really the only case where we've had the need, and it's been years since we started MOI...

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Fair enough, lets go for it.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I continued to think about this all day, and I didn't come up with a reason not to do this.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

And the output of the constraint index matches the input of the set.

I agree, otherwise MOI.add_constrained_variable(model, set) should return vi, (ci,) but that would be breaking in addition to being annoying

)
set_1, set_2 = set
x, c_1 = add_constrained_variable(model, set_1)
c_2 = add_constraint(model, x, set_2)
return x, (c_1, c_2)
end

function supports_add_constrained_variable(
model::ModelLike,
::Type{Tuple{L,U}},
) where {L<:GreaterThan,U<:LessThan}
return supports_add_constrained_variable(model, L) &&
supports_constraint(model, VariableIndex, U)
end

"""
supports_add_constrained_variables(
model::ModelLike,
Expand Down
Loading