-
Notifications
You must be signed in to change notification settings - Fork 72
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Update documents #295
Update documents #295
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## master #295 +/- ##
=======================================
Coverage 73.94% 73.94%
=======================================
Files 36 36
Lines 6687 6687
=======================================
Hits 4945 4945
Misses 1742 1742 ☔ View full report in Codecov by Sentry. |
Thanks! |
Don't know what these 'doctests' are... https://github.com/JuliaGraphics/Luxor.jl/actions/runs/7948031755/job/21697557018#step:6:12 |
Hmm, it should be fixed by this line.. I'll take a look. |
Ah, that was because of these lines in Luxor.jl/.github/workflows/ci.yml Lines 74 to 79 in 80a780b
I'll open a PR with julia-docdeploy workflow. |
Cool, thanks. To be honest, I thought you were just going to add a deprecation for a Point method ready for a 3.9 release - there seem to be many more changes happening now 🤔 |
Sorry for not having enough explanation! ImplementationI initially thought we could just deprecate and remove methods such as There are three possibilities for this situation.
My first question: Is it okay to keep the first one? DeprecationI initially thought we need just deprecations for methods like
My second question: Is it acceptable to avoid releasing DocumentationLuxor.jl v4 breaks code like I'm also working on JuliaDocs/Documenter.jl#2452, and I would like to keep Luxor.jl in the list. My third question: Do we hurry for the next release? Updating the docs may take time a bit. |
Although I personally think the Ref syntax is unintuitive and ugly, but if it’s more correct to remove it than to keep it I won’t argue… 😀 People complain loudly about correctness so let’s not give them anything to complain about. (I may add a new function that does what I want 😂) I don’t think the example page in the Documenter docs is worth worrying about. Luxor was a very early entry, but now there are thousands - including the official Julia docs. I think that the section could be removed. I think we can release v4 without adding deprecations to an existing release ? That’s what I’d do. 🤔 You’ve taken on a lot of work - thank you - take your time! |
Thank you for the reply!
Exactly 😂😂
Agreed! Let's release v4.0.0 after updating the docs! |
I found another fourth possibility for the issue. Prohibiting julia> getindex.("abc", 2:3) # Do not need `Ref` to `"abc"`
2-element Vector{Char}:
'b': ASCII/Unicode U+0062 (category Ll: Letter, lowercase)
'c': ASCII/Unicode U+0063 (category Ll: Letter, lowercase)
julia> "abc" .+ 1 # Do not return `"bcd"`. Note that `['a', 'b', 'c'] .+ 1 == ['b', 'c', 'd']`
ERROR: MethodError: no method matching +(::String, ::Int64)
Closest candidates are:
+(::Any, ::Any, ::Any, ::Any...)
@ Base operators.jl:587
+(::Base.CoreLogging.LogLevel, ::Integer)
@ Base logging.jl:131
+(::Complex{Bool}, ::Real)
@ Base complex.jl:320
... So, my current question is which one of these codes can be dropped?
I really don't have much preference about this, but I think we need to reconsider the interface before releasing v4. |
I can understand that ' Also I think It's easy to add methods that avoid broadcasting. For example these are easy to add and easier to understand than the function between(pt1::Point, pt2::Point, r::Real)
function between(pt1::Point, pt2::Point, r::AbstractRange)
function between(pt1::Point, pt2::Point, a::AbstractArray) |
Sorry for the late response. I now think we should drop support for |
jldoctest
docs/make.jl
format (indent etc.)