Closed p4v4n closed 6 months ago
Thanks @p4v4n! CI is failing.
cc @kommen
I see why the CI is failing, but it seems to me it might be better to just depend on a explicit version of seq.el
instead of relying on whatever's available in Emacs. The way I see it either we have to downgrade to 2.20 in flycheck or set some explicit dep here as well, so it can be fetched from GNU ELPA if needed.
Also - might be good to look into abstracting away the test code duplication for the different major modes.
There are 2 issues here.
The downside to using a explicit version of seq.el is that every downstream package(depending on flycheck) needs to keep track of what version of seq flycheck is depending on and update regularly whenever flycheck updates the version upstream.
I forgot that clojure-ts-mode requires emacs29
And also need to install (from source?) the clojure treesitter gramma, right? I think it is already more convenient but haven't tried/looked into how this works now.
And also need to install (from source?) the clojure treesitter gramma, right? I think it is already more convenient but haven't tried/looked into how this works now.
Luckily its already handled by clojure-ts-mode
automatically with clojure-ts-ensure-grammars
set to true by default.
The tests are already passing on the emacs 29+ for this branch.
https://github.com/borkdude/flycheck-clj-kondo/actions/runs/7945808126/job/21697659833
I think we only need to make sure that the clojure-ts-mode tests are not run when emacs-version < 29.1.
Thanks!
Thanks @borkdude!
I was about to make a few more changes to this PR (like moving ts tests to a new file and other minor stuff). Will raise a separate PR later as it is not critical.
This is a follow up PR to https://github.com/borkdude/flycheck-clj-kondo/pull/26.
clojure-ts-mode
.