Tests on main are failing

[1-Bort-1]

TL;DR

main is red on the POLAR_MATRICES forwarddiff flake and #292 carries the fix; this branch was the one commit on top of it, rewording the comment on that check's finite-difference step. #292 has since moved that step into linearize's own default, which deleted the line the comment described — so after merging #292 back in this branch carries nothing of its own, and #304 is closed.

What was failing

main CI run 34446797250 (the merge of #285) and release #302's run 34628824540 both failed the same assertion: Expression: rel_err < 0.001 / Evaluated: 0.04001429189872729 < 0.001 in "AutoForwardDiff matches AutoFiniteDiff (LOOP, POLAR_MATRICES)". Nothing else on main failed, and main has had no CI run since.

The check differentiates a piecewise-linear polar whose alpha grid has knots every 5°, and a 1e-5 forward difference straddling a knot measures the average of two slopes where forward AD takes one. Which panel alpha lands in that window is decided by the NeuralFoil fixture, redrawn every run (#291) — the #287 flake, fixed in #292 by tightening rtol to 1e-11, the step to 1e-8 and the bound to 1e-4.

Why this branch is empty

The advisory review on #292 flagged the comment that PR added at the call site, which described the old wide step beside the new narrow one; cac7762 reworded it. #292's own follow-up 1a4a039 then took linearize's fd_absstep/fd_relstep defaults to 1e-8 and had the test pass backend=nothing, deleting the explicit step — and with it the only line the comment had to describe.

That is the conflict this turn was handed. Resolved to #292's side: with no step written at the call site there is nothing for a comment there to say, and why 1e-8 is the right default now lives in #292's changelog entry and linearize's docstring rather than in the test. Merge e18f9ed leaves the branch byte-identical to #292's head, so #304 had no diff left and is closed.

What is left for #303 is merging #292 — main stays red until it lands. The fixture-reproducibility root cause (#291) and the release (#302) are untouched.

Verification

• Reproduced first: the two CI runs above, identical 0.04001429189872729 < 0.001 at test/solver/test_forwarddiff.jl
• test/solver/test_forwarddiff.jl on the merged head e18f9ed: 7/7 pass, 34.2 s (juliaserver, test env) — norm_fwd = 2.205885944736307, norm_fd = 2.205885850610976
• Local full suite on e18f9ed: PASS (9 min, Julia 1.12.7, one matrix cell) · GitHub CI: Tests on main are failing #304 closed, so the authority is test_forwarddiff.jl's POLAR_MATRICES check flakes at ~4%, and the oper #292's own run 34637447080 (macOS 1.12, Windows 1.12 and Documentation pass, the rest pending)
• Docs build: n/a — no public symbol touched · REUSE: the repo carries no bin/reuse_lint · up to date with the stack base, which itself carries main at fb1eea7
• Benchmark: n/a — no source change on this branch
• Risk: none on this branch, which changes nothing; the flake fix's residual risk is stated in test_forwarddiff.jl's POLAR_MATRICES check flakes at ~4%, and the oper #292

Scope

0 changed lines over the stack base agent/287-test-forwarddiff-jl-s-polar-matrices-che (#292); +19 / −14 across 3 files against main, all of it #292's. Stack: 2/2 → #292, closed as subsumed.

Closes #303 · task VortexStepMethod.jl-303

[1-Bort-1]

Local full suite: FAIL (3 min, Julia 1.12.7, one cell of the matrix)

    @ ~/.julia/juliaup/julia-1.12.7+0.x64.linux.gnu/share/julia/stdlib/v1.12/Test/src/Test.jl:1777 [inlined]
  [6] macro expansion
    @ ~/worktrees/VortexStepMethod.jl-303/test/runtests.jl:72 [inlined]
  [7] include(mapexpr::Function, mod::Module, _path::String)
    @ Base ./Base.jl:307
  [8] top-level scope
    @ none:6
  [9] eval(m::Module, e::Any)
    @ Core ./boot.jl:489
 [10] exec_options(opts::Base.JLOptions)
    @ Base ./client.jl:283
 [11] _start()
    @ Base ./client.jl:550
ERROR: Package VortexStepMethod errored during testing
Stacktrace:
  [1] pkgerror(msg::String)
    @ Pkg.Types ~/.julia/juliaup/julia-1.12.7+0.x64.linux.gnu/share/julia/stdlib/v1.12/Pkg/src/Types.jl:68
  [2] test(ctx::Pkg.Types.Context, pkgs::Vector{PackageSpec}; coverage::Bool, julia_args::Cmd, test_args::Cmd, test_fn::Nothing, force_latest_compatible_version::Bool, allow_earlier_backwards_compatible_versions::Bool, allow_reresolve::Bool)
    @ Pkg.Operations ~/.julia/juliaup/julia-1.12.7+0.x64.linux.gnu/share/julia/stdlib/v1.12/Pkg/src/Operations.jl:2642
  [3] test
    @ ~/.julia/juliaup/julia-1.12.7+0.x64.linux.gnu/share/julia/stdlib/v1.12/Pkg/src/Operations.jl:2487 [inlined]
  [4] test(ctx::Pkg.Types.Context, pkgs::Vector{PackageSpec}; coverage::Bool, test_fn::Nothing, julia_args::Cmd, test_args::Cmd, force_latest_compatible_version::Bool, allow_earlier_backwards_compatible_versions::Bool, allow_reresolve::Bool, kwargs::@Kwargs{io::IOContext{IO}})
    @ Pkg.API ~/.julia/juliaup/julia-1.12.7+0.x64.linux.gnu/share/julia/stdlib/v1.12/Pkg/src/API.jl:552
  [5] test(pkgs::Vector{PackageSpec}; io::IOContext{IO}, kwargs::@Kwargs{})
    @ Pkg.API ~/.julia/juliaup/julia-1.12.7+0.x64.linux.gnu/share/julia/stdlib/v1.12/Pkg/src/API.jl:169
  [6] test(pkgs::Vector{PackageSpec})
    @ Pkg.API ~/.julia/juliaup/julia-1.12.7+0.x64.linux.gnu/share/julia/stdlib/v1.12/Pkg/src/API.jl:158
  [7] test(; name::Nothing, uuid::Nothing, version::Nothing, url::Nothing, rev::Nothing, path::Nothing, mode::PackageMode, subdir::Nothing, kwargs::@Kwargs{})
    @ Pkg.API ~/.julia/juliaup/julia-1.12.7+0.x64.linux.gnu/share/julia/stdlib/v1.12/Pkg/src/API.jl:186
  [8] test()
    @ Pkg.API ~/.julia/juliaup/julia-1.12.7+0.x64.linux.gnu/share/julia/stdlib/v1.12/Pkg/src/API.jl:175
  [9] top-level scope
    @ none:1
 [10] eval(m::Module, e::Any)
    @ Core ./boot.jl:489
 [11] exec_options(opts::Base.JLOptions)
    @ Base ./client.jl:283
 [12] _start()
    @ Base ./client.jl:550

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!