The issue of Korean character fragmentation in Inky

[cheng80]

When using Inkle Studio’s narrative scripting language editor, Inky, on a Mac, Korean characters written in Unicode often appear fragmented, making it impossible to display them correctly.

This issue seems to stem from Ink utilizing the NFD (Normalization Form Canonical Decomposition) method for Unicode normalization, rather than NFC (Normalization Form Canonical Composition). To display Korean characters correctly, the NFC normalization form must be used.

[rhizoome]

I tried to reproduce this with

->hangul겺겻겼test->

=== hangul겺겻겼test ===
hangul겺겻겼test
->->

but the above works. Can you send a file that fails?

Also

1. You should probably submit the issue to: https://github.com/inkle/inky
2. Autocompletion for unicode is actually not supported since the version of ace editor used in inky does not yet support unicode regex. Because of that I'd recommend using ascii only for symbols like knots and variables.

[rhizoome]

The tree-siitter-ink project should support unicode fully #934, but I need to implement a language-server based on tree-sitter-ink before it is a full IDE:

[ffdd270]

This is input issue : Check this video.

2025-02-14.2.31.57.mov

[ffdd270]

This is the Korean input result I expected.

2025-02-14.2.34.36.mov

[cheng80]

That's right. That's the problem. So currently, I work in VS code and only check the results and errors in the Inky editor.

[rhizoome]

Ok. I tested with the virtual unicode entry system in MacOS. It seems to be very different to you using a real korean keyboard. Something like that might be fixable by updating Electron and AceJS. If you want try yourself it isn‘t too difficult.

https://github.com/inkle/inky?tab=readme-ov-file#help-develop-inky