Lacking sections about string and unicode

[Jiehong]

Hello,

First of all, thank you for this guide!

I'm new to Zig, and I cannot see a section in the guide about strings and unicode.

What I understood was that ascii is handled as []u8, but unicode quickly became a question mark as soon as I wrote in French and Chinese.

That's where things are unclear and not talked about in the doc.

For example, I was surprised by this not printing the right characters:

    const data = "abc好";
    for (data) |character| {
        std.debug.print("Here is a '{u}'\n", .{character});
    }

Checking online, there is an old lib lacking a maintainer: https://codeberg.org/dude_the_builder/zigstr

So, perhaps it would be nice to have a sections about strings in general and how to handle unicode by default?

[pfgithub]

for iterates by bytes, "好" is one grapheme cluster composed of one codepoint composed of three bytes.

Zig itself supports iteration by bytes (for ("str")), and codepoints (std.unicode.Utf8View.iterator()), however grapheme clusters require internationalization data that is updated yearly and the zig standard library doesn't include that.

If you iterate by codepoints, it will work for some strings but not others ("abc好" -> "a, b, c, 好" works, but "äbc" will become "a, ◌̈, b, c". and "äbc" will become "ä, b, c" because it's NFC normalized rather than NFD normalized)

zg can do grapheme clustering and normalization for when they're needed: https://codeberg.org/atman/zg

[okmanideep]

A Strings section in Language should be very helpful. Not sure if []const u8 being the default type when someone defines const str = "hello"; is intuitive enough.

Zig's concatenation and raw strings features also need some showcase.