Add support for rgb and indexed colors

[uttarayan21]

Added support for rgb and indexed colors by using cusom de/serialization of the tui::style::Color struct.

Source here https://git.uttarayan.me/uttarayan/color-parser-tui. I will publish to crates.io after a bit more testing.

[grtcdr]

LGTM.

(Renaming theme::theme to theme::base is a very welcome change)

[grtcdr]

Your crate sorta makes theme::color::MacchinaColor obsolete, should we remove ours?

[uttarayan21]

Hmm I haven't really looked at MacchinaColor closely I just Renamed MacchinaColor to Color and fixed the compiler errors, and it just worked.

[uttarayan21]

Actually now that I took a look it seems that MacchinaColor was a wrapper enum to use Color with clap. Now that it's not required, we can probably remove it.

[grtcdr]

Hmm I haven't really looked at MacchinaColor closely I just Renamed MacchinaColor to Color and fixed the compiler errors, and it just worked.

It implements Deserializer and Serializer, that's about it. And your crate implements that as well, so how should we proceed?

[uttarayan21]

There's a make_random_color function closure in main that needs to be manually rewritten I guess. Other than that I can remove it.

[grtcdr]

I've been playing around with strum and strum_macros which provide EnumString (providing from_str()) and EnumIter (providing iter()). The issue with using arg_enum! is it just doesn't allow enumerations that use constructors, which we need for Rgb(u8, u8, u8). For Enum::variants() we can use Enum::iter().collect().

[grtcdr]

Before I forget, this should probably be merged into v6.

[uttarayan21]

Removed MacchinaColor. Updated color-parser-tui with a module optional which adds support for Option<Color>

[grtcdr]

Are we ready to merge to the v6 branch?

I'd rather bundle up our new features into one major release this time.

[uttarayan21]

It's core logic is done. Ill just need to clean up the code a bit, add better docs, and add a few more tests.

I'll probably publish it on saturday after all is done.

[grtcdr]

Take your time, v6 is probably a long way from being ready.

[uttarayan21]

Is the name of the crate alright ?. I was thinking to name it similar to ansi-to-tui, and make it hex-to-tui or color-to-tui.

Any suggestions ?

[grtcdr]

If I were you, I'd name it the way I name most of my crates, in Esperanto. So "koloro" in this case.

[uttarayan21]

I mean if I made a standalone crate I might consider it. But this is just a helper crate of sorts, so I think it might be better to name it descriptive.

[grtcdr]

Then color-to-tui it is. I like that.

(as it does more than just hexadecimal colors)

[grtcdr]

Ready to merge into v6?

[uttarayan21]

Yes. Let me know if something needs to be changed.

[grtcdr]

As far as I can see, everything looks great :)

[uttarayan21]

Rebased to v6 branch.

[uttarayan21]

I think we need to show the parsing errors to the users. I can't figure out what is wrong with Hydrogen.toml

[grtcdr]

We really do, in the meantime, can you show me the file?

[uttarayan21]

# for a complete reference, visit
# https://github.com/Macchina-CLI/macchina/wiki/Theme-Documentation

spacing = 2
padding = 0
hide_ascii = false
hide_bar_delimiters = true
prefer_small_ascii = false
separator = ">"
key_color = "#ff6135"
separator_color = "#0ac365"
palette = "Light"

[bar]
glyph = "ߋ"
symbol_open = '['
symbol_close = ']'
visible = true

[box]
title = "aurc"
visible = true

[box.inner_margin]
x = 1
y = 0

[custom_ascii]
# color = "Green"
path = "~/.config/macchina/ascii/archlinux.ascii"

[randomize]
key_color = false
separator_color = false

[keys]
host = "Host"
kernel = "Kernel"
battery = "Battery"
os = "OS"
de = "DE"
wm = "WM"
distro = "Distro"
terminal = "Terminal"
shell = "Shell"
packages = "Packages"
uptime = "Uptime"
memory = "Memory"
machine = "Machine"
local_ip = "Local IP"
backlight = "Brightness"
resolution = "Resolution"
cpu_load = "CPU Load"
cpu = "CPU"

Its this one.
Meanwhile I'll try make another pr to show the parsing error

[grtcdr]

The theme parses and shows up correctly when building the usual macchina/main, but it fails on this specific branch/fork, did we deprecate something and not notice?

EDIT: Nope, no drastic changes that would cause something like this to happen.

Meanwhile I'll try make another pr to show the parsing error

That would help us out a lot.

[uttarayan21]

Oh I forgot to mention, I figured it out.
There were no #[serde(default)] in Block and Bar structs so It failed to derive with missing values.

[grtcdr]

Good catch :)

[uttarayan21]

I actually hadn't pushed the #[serde(default)] changes. I'll merge them in #208

[grtcdr]

I actually hadn't pushed the #[serde(default)] changes. I'll merge them in #208

Woops, alright :)