Add get basis status for constraints and variables

[EdvinAblad]

Resolves issue jump-dev/MathOptInterface.jl#655 in MOI for Clp

[mlubin]

@odow Are the tests failing because of a recent LQOI change? https://travis-ci.org/JuliaOpt/Clp.jl/jobs/496747085#L540

@EdvinAblad This looks good but we need this functionality covered by at least one test. Doing so is a bit involved:

• Check out https://github.com/JuliaOpt/MathOptInterface.jl/blob/master/src/Test/contlinear.jl. If there's an an LP there with a unique optimal basis, then we can add on to that test, otherwise we need to create a new test problem.
• Add a field to the TestConfig struct for checking optimal bases. It could be basis::Bool.
• In an if config.basis block, test that the solver returns the expected basis.
• Try to cover MOI.BASIC, MOI.NONBASIC, MOI.NONBASIC_AT_LOWER, and MOI.NONBASIC_AT_UPPER. (SUPERBASIC doesn't happen for LPs.)
• I'll give you a break and won't ask you to add support for basis status to MOIU.MockOptimizer, but that would ideally happen if we wanted to cross our t's and dot our i's.
• Make sure Clp passes this new test.

[odow]

Urgh no MOI 8.2. @blegat can we just exclude the test or should it pass?

[blegat]

Solvers that use a CachingOptimizer in their tests don't have to excluded it, it just displays a warning. However, if the solver does not use a CachingOptimizer, it should exclude it.
So for Clp, it should be excluded

[EdvinAblad]

@mlubin I have a working (unpublished) test but two questions occurred:

1. Should I also implement the get for the "Interval Bridge" or is that a separate issue?
2. To me there is a redundancy in the nonbasic status. i.e., NON_BASIC_AT_LOWER, and NON_BASIC_AT_UPPER both imply NON_BASIC. E,g, if a variable is non_basis should I test NON_BASIC or the specific NON_BASIC_AT_LOWER. Should NON_BASIC be renamed to NON_BASIC_FIXED?

[blegat]

Should I also implement the get for the "Interval Bridge" or is that a separate issue?

For the interval bridge, it is an MOI issue: https://github.com/JuliaOpt/MathOptInterface.jl/blob/master/src/Bridges/intervalbridge.jl

[blegat]

I suppose we need to bump MOI to v0.8.3 in the REQUIRE file

[EdvinAblad]

I suppose we need to bump MOI to v0.8.3 in the REQUIRE file

Clp is indirectly dependent on MOI through the LQOI interface.
Should we temporarily add MOI v0.8.3 as a requirement (until there exist a LQOI version with this requirement)?

(It should "only" be the tests of Clp that require MOI v0.8.3, if this makes a difference)

[blegat]

If it's only the tests, it's less of a big deal but a user would still expects

] add Clp
] test Clp

to work so adding a MOI lower bound wouldn't hurt. LQOI doesn't require MOI v0.8.3 at the moment but if it does later we may remove the MOI lower bound here.

[EdvinAblad]

Is there something preventing this from being merged into master?

[mlubin]

Yes, I'm on vacation :). I have style comments that I may or may not be able to send out before next week.

…

On Sat, Mar 23, 2019, 14:27 EdvinAblad ***@***.***> wrote: Is there something preventing this from being merged into master? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#54 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABp0M7RKS-iZIg-x_dqyV3qGA01kdcvCks5vZmP8gaJpZM4bIcRU> .

[EdvinAblad]

Yes, I'm on vacation :). I have style comments that I may or may not be able to send out before next week.
…
On Sat, Mar 23, 2019, 14:27 EdvinAblad @.***> wrote: Is there something preventing this from being merged into master? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#54 (comment)>, or mute the thread https://github.com/notifications/unsubscribe-auth/ABp0M7RKS-iZIg-x_dqyV3qGA01kdcvCks5vZmP8gaJpZM4bIcRU .

Ok, there is really no rush, enjoy the vacation =)

[mlubin]

Thanks!