Replies: 2 comments
-
thought about it some more and i think at least the use of an option (which i'm sure exists, I just have to find it) to make the linter only run on file save also helps address poorly performing linters. So I dunno... If this was an issue I would just close this before someone else closes it for me at this point. But it's a discussion so I'll leave it open for input since I welcome that in all forms. |
Beta Was this translation helpful? Give feedback.
-
You can check the documentation for instructions on how to configure source-specific timeouts as well as how to make diagnostic sources run only on save. As for the multiple processes issue: I describe the cause and a potential fix in #596. Because of the way that null-ls works, I doubt slow sources like this will ever provide an ideal experience, but there are some improvements to be made if someone is willing to push them through. |
Beta Was this translation helpful? Give feedback.
-
I'm playing with cspell, and I found that it does work out of the box:
but in most of my code projects, it runs out its allotted 5s timeout:
This is unsurprising given that cspell took an entire second to process just a single 183-line yaml file; cspell clearly needs some optimization,
but until that happens, I'm interested in knowing what the correct approach would be to at least change this thing to call cspell on individual code files.Given the observed performance it's probably going to use too much cpu to justify on my macbook but I can use it on my server machine.Questions
I can only assume that trying to hack-change the call to only run on the file being edited may change the output format and break things... what's the approach here?So i looked at it again once I noticed it did run on my plugins.lua file but definitely cannot make the deadline on my init.lua file, yes it is already running on a per-file basis not on each file in a dir, since the command call does specifystdin
there. OKCSpell is the first tool i've encountered that brings these questions to the forefront, but these questions apply equally to all tools that have long runtimes.
Beta Was this translation helpful? Give feedback.
All reactions