Open dbatten5 opened 2 months ago
*neotest.Config.discovery*
Fields~
{enabled} `(boolean)`
{concurrent} `(integer)` Number of workers to parse files concurrently. 0
automatically assigns number based on CPU. Set to 1 if experiencing lag.
{filter_dir} `(nil)` | fun(name: string, rel_path: string, root: string):
boolean A function to filter directories when searching for test files.
Receives the name, path relative to project root and project root path
Checkout this
Already mentioned above, but yes, you can make neotest only query the opened buffer for tests, which likely will help you out here:
{
"nvim-neotest/neotest",
opts = {
-- See all config options with :h neotest.Config
discovery = {
-- Drastically improve performance in ginormous projects by
-- only AST-parsing the currently opened buffer.
enabled = false,
-- Number of workers to parse files concurrently.
-- A value of 0 automatically assigns number based on CPU.
-- Set to 1 if experiencing lag.
concurrent = 1,
},
running = {
-- Run tests concurrently when an adapter provides multiple commands to run.
concurrent = true,
},
summary = {
-- Enable/disable animation of icons.
animated = false,
},
},
}
I feel this kind of info should be more prominently explained in the main README.
NeoVim Version NVIM v0.10.0 Build type: Release LuaJIT 2.1.1716656478
Describe the bug Apologies for the vagueness but I'm not really sure how to pinpoint this issue.
I work in a very large monorepo at work. I've previously been able to run neotest DAP with
nvim-dap
without any issues to debug the application. Since upgradingneotest
to version that requiredneotest-nio
, whenever I try to run neotest with DAP strategy my neovim freezes and I end up having to close the tab in my terminal. I can see the neovim process at around 99% CPU. I'm not sure if this is an issue with neotest or nvim-dap or a combination of the two. I'm also not sure how to provide any useful information hereLogs The log file is 1.8m lines long. A huge amount the lines have this form:
I can also see that it's scanning through e.g.
node_modules
that are buried within the monorepo. Is there a config option I can be setting to to make sure I don't scan the entire monorepo here?My setup looks like this: