Julia stuff

[jbcaillau]

Profiling

@PierreMartinon

• Performance tips
• Profiling (native)
• Cthulhu
• BenchmarkTools
• ProfileView

[ocots]

Docstrings extensions:

• doc of the package
• need to be added into the Project.toml
• example

For instance:

using DocStringExtensions

"""
A short summary of `func`...

$(SIGNATURES)

where `x` and `y` should both be positive.

# Details

Some details about `func`...
"""
func(x, y) = x + y

[ocots]

@joseph-gergaud @PierreMartinon If you want to write some docstrings, you can use DocStringExtensions, cf. above.

[jbcaillau]

https://info.juliahub.com/en-us/tools-for-non-linear-mpc

[jbcaillau]

Struct equality test:
https://discourse.julialang.org/t/surprising-struct-equality-test/4890/8

[jbcaillau]

@ocots any idea where has the wiki page with info on registry add, dev... moved?

[ocots]

https://github.com/control-toolbox/control-toolbox.github.io/wiki/Dev-setup

[jbcaillau]

Sone knows the point in using x_1 instead of $x_1$ for Mardown math? See, e.g. this source file
and this result file.

[jbcaillau]

Nice Pluto notebooks (.jl, not .ipynb):

Invariant manifolds ¹

Going to Jupiter

Footnotes

1. see here for Pluto rendering on JuliaHub ↩

[jbcaillau]

Problem with conda install of julia:

• missing cert.pm (@j-l-s rings a bell 🙁?)
• problem with package install (HTTP, Plotly...)

AVOID, and use official release instead; see this post. Everything back to normal with 1.8.5 local install.

[ocots]

Taylor integration : https://docs.sciml.ai/TaylorIntegration/stable/common/

[ocots]

Doc and nice benchmark: https://juliasmoothoptimizers.github.io/Percival.jl/dev/

[jbcaillau]

On julia vs. markdown vs. notebooks files

Weave.jl

Literate.jl

[jbcaillau]

20 Awesome Julia Tips And Tricks For The Advantageous Programmer

Check this post

[j-l-s]

How to create "real" getters and setters (№17: “ object-oriented programming”)

[jbcaillau]

On no. 17: too convoluted. Julia is not o-oriented, dot!

[rveltz]

getters and setters are a must for big projects. Check Kwong, Tom. Hands-on Design Patterns and Best Practices with Julia: Proven Solutions to Common Problems in Software Design for Julia 1.x. Birmingham, UK: Packt Publishing, 2020.

[PierreMartinon]

A very long post indeed. I struggled to find any useful information in it, but that's probably me 😄

…

On Apr 6, 2023 at 17:00, Jean-Baptiste Caillau ***@***.***> wrote: 20 Awesome Julia Tips And Tricks For The Advantageous Programmer Check this post — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[jbcaillau]

Traits & more

JuliaLang/julia#2345 (comment)

https://invenia.github.io/blog/2019/10/30/julialang-features-part-1/

https://invenia.github.io/blog/2019/11/06/julialang-features-part-2/

https://github.com/mauro3/SimpleTraits.jl

https://github.com/jolin-io/WhereTraits.jl

https://github.com/PacktPublishing/Hands-on-Design-Patterns-and-Best-Practices-with-Julia

http://ucidatascienceinitiative.github.io/IntroToJulia/Html/DispatchDesigns

https://docs.julialang.org/en/v1/manual/methods/#Trait-based-dispatch-1

https://youtu.be/qhAe1PuWjRQ?si=9JWFxqvSfRDAUvO6

See also

Traits in rust

Example

From https://www.oxinabox.net/2018/10/03/Dispatch,-Traits-and-Metaprogramming-Over-Reflection.html#part-3-traits-falling-back-to-hasmethod

# The trait types
abstract type Scalarness end
struct Scalar <: Scalarness end
struct NonScalar <: Scalarness end

# The trait functions (extensible! contrary to unions)
scalarness(::Type) = Scalar() #fall-back, by default everything is scalar; could 
scalarness(::Type{<:AbstractArray}) = NonScalar()
scalarness(::Type{<:Tuple}) = NonScalar()
scalarness(::Type{<:AbstractString}) = NonScalar()

# Dispatching statically (values are types, known statically - nb: also works on small values such as booleans...)
aslist(x::T) where T = aslist(scalarness(T), x)
aslist(::Scalar, x) = [x]
aslist(::NonScalar, x) = x

[jbcaillau]

Perf. profiling in Julia

[rveltz]

I like also the idea of https://github.com/jolin-io/WhereTraits.jl To me, the bigger problem is the one of interfaces when you develop such big package as yours.

…

On Dec 28, 2023, at 9:39 AM, Jean-Baptiste Caillau ***@***.***> wrote: @rveltz <https://github.com/rveltz> anything more up to date / comprehensive on Julia traits than what is here <#22 (comment)>? — Reply to this email directly, view it on GitHub <#22 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAMEUBC5DHE6NW7GR23P2LTYLUV25AVCNFSM6AAAAAAVZW4SEOVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM3TSNRSHE2DA>. You are receiving this because you were mentioned.

[jbcaillau]

@rveltz Is it the disambiguation process (@traits_order) that you like in this package?

[jbcaillau]

@joseph-gergaud regarding an old discussion we had on Julia scoping, check this.

[jbcaillau]

@ocots on import vs. using, check this.

[ocots]

Blue: a Style Guide for Julia: This document specifies style conventions for Julia code. These conventions were created from a variety of sources including Python's PEP8, Julia's Notes for Contributors, and Julia's Style Guide.

[rveltz]

This is implemented in https://github.com/domluna/JuliaFormatter.jl This can be a lot of refactoring to comply to a style depending on your project status… I was told it is the basic thing to get PR from external users

…

On May 30, 2024, at 11:08 AM, Olivier Cots ***@***.***> wrote: Blue: a Style Guide for Julia <https://github.com/JuliaDiff/BlueStyle>: This document specifies style conventions for Julia code. These conventions were created from a variety of sources including Python's PEP8 <http://legacy.python.org/dev/peps/pep-0008/>, Julia's Notes for Contributors <https://github.com/JuliaLang/julia/blob/master/CONTRIBUTING.md>, and Julia's Style Guide <https://docs.julialang.org/en/v1/manual/style-guide/>. — Reply to this email directly, view it on GitHub <#22 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAMEUBDGHS4FQNR65TK4CILZE3T7LAVCNFSM6AAAAAAVZW4SEOVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TMMBTGQ3TQ>. You are receiving this because you were mentioned.

[jbcaillau]

thanks @rveltz (I remember you already pointed this style issue). For those on vscode ¹, these standards are taken into account in the current julia extension. Still interesting to try and run JuliaFormatter.jl on our code 🤞🏾...

Footnotes

1. Romain check this 🙂 ↩

[jbcaillau]

@ocots @joseph-gergaud on the scope of loops

[jbcaillau]

On the program:

• Th. 14:00, "Dude, where's my code?" (T. Holy)

[jbcaillau]

@ocots @PierreMartinon @joseph-gergaud we need a place to save howto's (create a package, a doc, CI info, etc.) There is this discussion, this wiki, and quite a few things scattered under the control-toolbox organisation. any idea on a good way to do this properly?

[ocots]

It would be better here.

[jbcaillau]

https://julialang.org/blog/2024/10/julia-1.11-highlights

[PierreMartinon]

Glad to see more performance improvements !

[amontoison]

https://github.com/LilithHafner/BasicAutoloads.jl

[ocots]

@jbcaillau https://juliareach.github.io

[jbcaillau]

@ocots @PierreMartinon @joseph-gergaud
https://www.researchgate.net/profile/Francis-Gagnon-2/publication/385825845_ModelPredictiveControljl_advanced_process_control_made_easy_in_Julia/links/6736e2adf255d572866b6b4a/ModelPredictiveControljl-advanced-process-control-made-easy-in-Julia.pdf?origin=publication_detail&_tp=eyJjb250ZXh0Ijp7ImZpcnN0UGFnZSI6InB1YmxpY2F0aW9uIiwicGFnZSI6InB1YmxpY2F0aW9uRG93bmxvYWQiLCJwcmV2aW91c1BhZ2UiOiJwdWJsaWNhdGlvbiJ9fQ

[jbcaillau]

@ocots @PierreMartinon @joseph-gergaud take a look at https://github.com/JuliaParallel/Dagger.jl (parallel computation)

[jbcaillau]

anyone tested that https://github.com/JuliaArrays/OffsetArrays.jl ? @ocots @PierreMartinon @rveltz @amontoison

[jbcaillau]

@ocots

• https://discourse.julialang.org/t/a-curated-list-of-gh-actions-that-improve-the-quality-of-your-package/134466/8
• https://github.com/JuliaBesties/BestieTemplate.jl