SciML / LinearSolve.jl

LinearSolve.jl: High-Performance Unified Interface for Linear Solvers in Julia. Easily switch between factorization and Krylov methods, add preconditioners, and all in one interface.
https://docs.sciml.ai/LinearSolve/stable/
Other
244 stars 52 forks source link

try documenter 1.0 upgrade #371

Closed ArnoStrouwen closed 12 months ago

codecov[bot] commented 1 year ago

Codecov Report

Merging #371 (daf7212) into main (5a25b7d) will increase coverage by 14.17%. The diff coverage is 5.00%.

@@             Coverage Diff             @@
##             main     #371       +/-   ##
===========================================
+ Coverage   20.01%   34.19%   +14.17%     
===========================================
  Files          14       23        +9     
  Lines        1444     1787      +343     
===========================================
+ Hits          289      611      +322     
- Misses       1155     1176       +21     
Files Changed Coverage Δ
ext/LinearSolveBlockDiagonalsExt.jl 0.00% <0.00%> (ø)
ext/LinearSolveKernelAbstractionsExt.jl 0.00% <0.00%> (ø)
ext/LinearSolveMKLExt.jl 0.00% <0.00%> (ø)
ext/LinearSolveMetalExt.jl 0.00% <ø> (ø)
src/appleaccelerate.jl 0.00% <0.00%> (ø)
src/extension_algs.jl 66.66% <ø> (+66.66%) :arrow_up:
src/LinearSolve.jl 93.84% <100.00%> (+6.15%) :arrow_up:

... and 12 files with indirect coverage changes

:mega: We’re building smart automated test selection to slash your CI/CD build times. Learn more

ArnoStrouwen commented 1 year ago

@mortenpi Could you have a look at the failure here: https://github.com/SciML/LinearSolve.jl/actions/runs/6208844246/job/16855431354?pr=371#step:5:11

mortenpi commented 1 year ago

So I don't think that's causing the error, but you will run into trouble with these at-eval blocks, which are returning a String. That's no longer allowed.

https://github.com/SciML/LinearSolve.jl/blob/82a4be503e25c766010c3fc790d5c0b23c682b99/docs/src/index.md?plain=1#L87-L115

Instead, this should be done with an object that has an text/html show method.

mortenpi commented 1 year ago

Alright, the main error is a Documenter regression. It actually comes from the (perfectly fine) use of the @doc_str macro here: https://github.com/SciML/SciMLBase.jl/blob/8d293bfc0b859b1ae870772aaf023d2a93f6eb45/src/problems/basic_problems.jl#L1