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
Currently the plugin only supports words of the dictionary. This unfortunately leads to a lot of false positives when using this LSP on specific types of files.
For example,
In lua, when using luaCATS this causes issues when using these annotations.
---@param---@nodiscard---@vararg
As quite a few repositories use release-please to generate a CHANGELOG.md file. The files title is CHANGELOG.
Unfortunately, this gets picked up by this LSP as an incorrect spelling. I think the LSP should ignore it.
This also introduces a new issue as commit messages usually look something like this.
feat(renderer): Added support for custom renderer
This also gets picked up by the LSP. This can lead to a LOT of false positives. I have almost 1200+ in 1000 line file.
It would be nice if the LSP could understand words used in editors(in my case Neovim).
For example,
extmark
autocmd
augroup
But this can be ignored.
As the scope of this is WAY TOO big(as the LSP by default activates in quite a few filetypes) this can either be solved by,
PRs(most likely from users) adding dictonaries for specific languages.
Adding pattern(s) that the LSP would ignore.
But you should decide what will be the best solution.
The text was updated successfully, but these errors were encountered:
The problem
1200+ diagnostics from Harper in a markdown file
Currently the plugin only supports words of the dictionary. This unfortunately leads to a lot of false positives when using this LSP on specific types of files.
For example,
In lua, when using luaCATS this causes issues when using these annotations.
As quite a few repositories use
release-please
to generate aCHANGELOG.md
file. The files title isCHANGELOG
.Unfortunately, this gets picked up by this LSP as an incorrect spelling. I think the LSP should ignore it.
This also introduces a new issue as commit messages usually look something like this.
This also gets picked up by the LSP. This can lead to a LOT of false positives. I have almost 1200+ in 1000 line file.
It would be nice if the LSP could understand words used in editors(in my case
Neovim
).For example,
But this can be ignored.
As the scope of this is WAY TOO big(as the LSP by default activates in quite a few filetypes) this can either be solved by,
PRs(most likely from users) adding dictonaries for specific languages.
Adding pattern(s) that the LSP would ignore.
But you should decide what will be the best solution.
The text was updated successfully, but these errors were encountered: