refactor: BLT sort equations and variables in tearing #3681
Conversation
| @@ -182,7 +183,7 @@ end | |||
| prob = ODEProblem(sys, [], (0.0, 1.0)) | |||
| sol = solve(prob) | |||
| @test SciMLBase.successful_retcode(sol) | |||
| @test sol.u[1] ≈ [1.0, 1.0, 0.5, 0.5] | |||
| @test sol[x, 1] ≈ [1.0, 1.0, 0.5, 0.5] | |||
There was a problem hiding this comment.
x is an array symbolic
@variables x(t)[1:4] y(t)[1:2]
|
Does this apply to all system types, e.g. ODEs? |
|
Yes, all system types will have their equations sorted in BLT order. |
AayushSabharwal
force-pushed
the
as/better-sccs
branch
from
4d219b9 to
3e124da
Compare
|
well there is only one system type 😅 |
|
Yeah, I guess better phrasing is "all types of systems" |
|
Ok, that was what I hoped and suspected, great! 😃 Are there any current plans to extend the strongly coupled logic to other problems as well, such as |
|
The FMI test fails because it relies on variable order. I have a fix locally and verified that it works. Will let the rest of CI run, and if it passes I'll just push and merge. |
AayushSabharwal
force-pushed
the
as/better-sccs
branch
2 times, most recently
from
ac2da4b to
e45ae8a
Compare
…onlinearProblem`
AayushSabharwal
force-pushed
the
as/better-sccs
branch
from
e45ae8a to
26437e2
Compare
This turned out to be more work than I thought, but it works!