You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
While the Google 10k words source provided by coc.nvim is good, when I write really long fiction/technical/editorial works in LaTeX and/or Markdown, these commonly-used words quickly become insufficient. I am then forced to manually type out the full (potentially) long words. The coc-dictionary source also is not satisfactory.
Use BSD-look
The look utility is pre-installed on macOS, *BSD and pretty much every linux distribution (either pre-installed or available in repos), and since it is standard C-code, it can be even compiled for Windows natively (a cygwin package is already available anyway). With its stdin/stdout interface and a host of other simple programmable interfaces, it can serve as a fantastic English language completion utility. It uses the local dictionary installed in the OS, eg. in /usr/share/dict/words and allows for adding custom words to the dictionary (which may be useful for highly specialised technical/medical writing for example).
Alternatives Considered
Before coc.nvim, I was using deoplete and then ncm-2. They both have implemented look completion sources, which I was quite happy with.
This is a copy of issue 1689 at the coc.nvim repository. Opening it here since it doesn't look like that repository deals with sources.
Background
While the Google 10k words source provided by coc.nvim is good, when I write really long fiction/technical/editorial works in LaTeX and/or Markdown, these commonly-used words quickly become insufficient. I am then forced to manually type out the full (potentially) long words. The
coc-dictionary
source also is not satisfactory.Use BSD-look
The
look
utility is pre-installed on macOS, *BSD and pretty much every linux distribution (either pre-installed or available in repos), and since it is standard C-code, it can be even compiled for Windows natively (a cygwin package is already available anyway). With its stdin/stdout interface and a host of other simple programmable interfaces, it can serve as a fantastic English language completion utility. It uses the local dictionary installed in the OS, eg. in/usr/share/dict/words
and allows for adding custom words to the dictionary (which may be useful for highly specialised technical/medical writing for example).Alternatives Considered
Before coc.nvim, I was using deoplete and then ncm-2. They both have implemented
look
completion sources, which I was quite happy with.Additional context
Please retain google 10k source with a lower completion priority than
look
.The text was updated successfully, but these errors were encountered: