Text completion does not include words with UTF-8 characters that take up more that a single byte

[AntonC9018]

FAQ

• [X] I have checked the FAQ and it didn't resolve my problem.

Announcement

• [X] I have checked Breaking change announcement.

Minimal reproducible full config

A default config should repro this. I haven't changed anything that has to do with this.

Description

Here I have a piece of text attăăăă with a few Romanian letters that take up 2 bytes each. As you can see, the completion does not include this word as a completion option.

If I do normal a's instead, it does offer that as an option.

Steps to reproduce

• Type a word that has e.g. Romanian letters in it (but I suspect that this is the case for any multibyte characters);
• Type e.g. the first letter of that word;
• Request buffer completion.

Expected behavior

The word is shown as an option.

Actual behavior

It's not.

Additional context

Vim's <C-p> word completion does work with this.

[Shougo]

Please see this.

https://github.com/hrsh7th/nvim-cmp/issues/978