Road to 1.0

[timholy]

I'd like to move the whole JuliaImages stack to version numbers >= 1.0 (see JuliaImages/Images.jl#825). Let's use this issue to collect plans for breaking changes in the near future.

Issues/PRs on my agenda:

• Modify strictness of color parser (#357) #360
• [RFC] Order of alpha in hex notation of TransparentColor (#AARRGGBB or #RRGGBBAA) #353
• possibly MSC() is strange #349, but I worry that will take a while and hence might be 2.0 material
• Julia 0.6 TODO: write awesome promotion rules #273 (ugh)

[kimikage]

• A part of Conversion methods should be organized #354?

Actually, I have forgotten how far I have worked 😅, but somethings should have already been achieved.
master...kimikage:cnvt

However, there are few breaking changes other than the bug fixes. In regards to the white point wp, it follows the current interface (convert(::Type{C}, c::Color, wp::XYZ)), but it is necessary to discuss how white points should be handled.

[timholy]

You have done wonders for these packages.

"Cleanups" can be done after 1.0 if they don't break the public API. If we grossly change the result of a white-point aware conversion, though, that is certainly breaking.

[kimikage]

• [RFC] Add support for parsing 8-digit and 4-digit hex notations #371 following [RFC] Order of alpha in hex notation of TransparentColor (#AARRGGBB or #RRGGBBAA) #353

[kimikage]

CC @cormullion

• Tidy the logo up (cf. https://github.com/JuliaLang/julia-logo-graphics)

It may be useful for the testing to add "julia_blue", "julia_green", "julia_purple" and "julia_red" to the named colors. I also feel that "nationalism" is too much, though. 😄

[cormullion]

No problem. What do you mean by "I also feel that "nationalism" is too much, though."?

[kimikage]

Adding "julia_blue" etc. to the named colors is not necessarily a public interest.
"Colors of the Julia users, by the Julia users, for the Julia users!"

[cormullion]

Ah, I see. :) Anyway, probably not many people use them (apart from me, I expect)...

[kimikage]

• Fatal performance degradation due to precompilation or inference bugs #496

[jonathanBieler]

If there breaking changes in master currently ? I wanted to use some of the XYZ to RGB conversion but last release was before it was added (27 Apr 2021). Would it be possible to make a minor release before 1.0 ?

[timholy]

Yeah, we need to do something about this. I am pretty swamped with TTFX work until at least mid-October, unfortunately. If you'd be willing to figure out what to do perhaps we could review your work and then make a release.

[jonathanBieler]

I had a look at the merged PRs and there's a couple labelled as breaking since last release :

• Relax color parser based on the current and upcoming CSS specs #478
• Change the default alpha position for hex(::ColorAlpha) #479
• Use Float32 to calculate DE_2000 for low precision colors #494
• Use Float32 to calculate colordiff for low precision colors #504

Otherwise it's mostly performance & accuracy improvements, with a few new features (polar_to_cartesian, weighted_color_mean, ...). I don't see serious open issues either, so it seems reasonable to make a new release. But maybe @kimikage has a different opinion ?

[stevengj]

Can we transition this package to 1.0 already? The problem with 0.xx releases is that they provide strictly less information than 1.x.y release — semver assumes that every 0.xx release is breaking.'

Version numbers are cheap. 30+ packages depend on this package. It's been out for ages, and the API hasn't changed much for a while.