search

SciML / Optimization.jl

Mathematical Optimization in Julia. Local, global, gradient-based and derivative-free. Linear, Quadratic, Convex, Mixed-Integer, and Nonlinear Optimization in one simple, fast, and differentiable interface.
https://docs.sciml.ai/Optimization/stable/
MIT License
691 stars 75 forks source link

Aqua + typos CI #639

Closed ArnoStrouwen closed 7 months ago

codecov[bot] commented 7 months ago

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Comparison is base (c178ba0) 0.00% compared to head (d893727) 7.74%.

Additional details and impacted files ```diff @@ Coverage Diff @@ ## master #639 +/- ## ========================================= + Coverage 0.00% 7.74% +7.74% ========================================= Files 12 40 +28 Lines 1141 2699 +1558 ========================================= + Hits 0 209 +209 - Misses 1141 2490 +1349 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

ArnoStrouwen commented 7 months ago

I'm guessing putting a compat for a weakdep is not backwards compatible with 1.6?

Vaibhavdixit02 commented 7 months ago

Looks like it yeah. We could add the packages, also to the extras but it doesn't feel nice to do it 😅

ArnoStrouwen commented 7 months ago

Can we just up the Julia compat to 1.8 or 1.9? e.g. NonlinearSolve has done so aswell.

ChrisRackauckas commented 7 months ago

Bump to v1.9

sethaxen commented 6 months ago

Was it really necessary to drop support for 3 Julia versions including the LTS? Optimization has 38 direct dependents. For those packages to e.g. support the new callback signature, they now need to drop support for all of those versions. Was the only reason for doing this to avoid adding the packages also to [extras]? That's the documented approach: https://pkgdocs.julialang.org/v1/creating-packages/#Conditional-loading-of-code-in-packages-(Extensions)

ChrisRackauckas commented 6 months ago

v1.10 will be the new LTS, so it's not dropping the LTS

sethaxen commented 6 months ago

v1.10 will be the new LTS

Is this documented somewhere? I couldn't find any statement or timeline on this.

so it's not dropping the LTS

Well, it already did drop the LTS and quite a few other versions, until there's a new one.

ChrisRackauckas commented 6 months ago

Are you volunteering to do that support for all of the SciML packages for the previous LTS (v1.6)? Our expectations are generally to have answers for each related question within about 12 hours and any major bugfixes within the week. Please let me know if you're willing to take on getting everything to support v1.6 down. I suspect it's about 40 hours to start given the requirements on NonlinearSolve.jl and LinearSolve.jl due to the LinearAlgebra changes in Base which have been substantial, then about 4 hours a week level of commitment. Usually we don't assume a new volunteer who hasn't done such contributions will suddenly take on that level of effort but if you're willing to do it I can show you where to start.

ArnoStrouwen commented 6 months ago

@sethaxen it has not been announced yet. From what I understand, the final decision on LTS will not be made until Julia 1.11 is out. But it is very likely 1.10 will be the new LTS.