Closed jayrajput closed 1 month ago
How can I re-trigger the CI for linting?
Btw, I did checked local linting and that is working fine.
jr@Jays-Mac-mini ekg % eldev lint
eldev lint
Running linter `doc'
File `ekg-auto-save.el': no warnings
File `ekg-denote.el': no warnings
File `ekg-embedding.el': no warnings
File `ekg-llm.el': no warnings
File `ekg-logseq.el': no warnings
File `ekg-org-roam.el': no warnings
File `ekg-test-utils.el': no warnings
File `ekg.el': no warnings
Running linter `package'
[1/1] Installing package `package-lint' (0.23) from `melpa-stable'...
File `ekg-auto-save.el': no warnings
File `ekg-denote.el': no warnings
File `ekg-embedding.el': no warnings
File `ekg-llm.el': no warnings
File `ekg-logseq.el': no warnings
File `ekg-org-roam.el': no warnings
File `ekg-test-utils.el': no warnings
File `ekg.el': no warnings
Running linter `re'
[1/2] Installing package `xr' (1.25) from `gnu'...
[2/2] Installing package `relint' (1.24) from `gnu'...
File `ekg-auto-save.el': no warnings
File `ekg-denote.el': no warnings
File `ekg-embedding.el': no warnings
File `ekg-llm.el': no warnings
File `ekg-logseq.el': no warnings
File `ekg-org-roam.el': no warnings
File `ekg-test-utils.el': no warnings
File `ekg.el': no warnings
Linters have no complaints
jr@Jays-Mac-mini ekg %
I can see linting working fine in the Checks
, but now the tests are failing. Looking more into the test failure.
Interestingly the test are working fine when I am running them using ert
.
Selector: t
Passed: 8
Failed: 0
Skipped: 0
Total: 8/8
Started at: 2024-07-21 21:21:35+0530
Finished.
Finished at: 2024-07-21 21:21:39+0530
........
Thanks for these fixes! For retrying, you can click on "Details" and then there should be a button at the top to "Re-Run Jobs". You can also install Eldev locally and run that way, which is a good way to make sure you have fixed everything before checkin.
Would you like me to merge this in now, or would you prefer to get a clean CI check first?
I am running eldev locally. Linting passes, but the test is failing. Although ERT is working. I am not able to figure out why eldev tests are failing. Looking for some guidance on how to debug eldev failure.
I took a look, and it looks like the problem is that, in denote's current codebase, denote-sluggify
takes two arguments, and you only are using one.
I fixed both linting and regression tests (CI also showed that is passed). Although I realised that there is a dependency on denote package and whenever denote changes, the code breaks. Is there a way to make things more robust, by pinning down a version of denote. It took me sometime to realise that the ert
was working fine as it was using an older version of denote whereas eldev
failed as it was rightly pulling the latest denote (??).
There is a way to pin the denote version, but only if we have it as an actual dependency of the project. Since we don't want to do that (because it isn't a real dependency, it's optional), we should just document what version of Denote we're expecting at a minimum in the file.
Updated the documentation string to fix following linting errors (see during code checkin pipeline execution)