Fix for #420 (Conditionals must have a boolean result.) and basic tests for #425#426
Conversation
|
Could you have a look at the linting errors? |
628d668 to
612f2cf
Compare
removed old tests, used only mock server
removed pidfile as not necessary anymore
612f2cf to
cd8be11
Compare
|
@guenhter lint test passed 😃 |
| --- | ||
| galaxy_info: | ||
| author: Erik-jan Riemers | ||
| role_name: gitlab_runner |
There was a problem hiding this comment.
@riemers Changes in this file always ring alarm bells. Do you think this is again a problem with ansible galaxy? I mean the change should be done as the linter complained...
There was a problem hiding this comment.
It was needed for the molecule to use the role in the converge.yml playbook. It complained that the name wasn't set, as I remember.
I'm sorry, I missed that I used an "underscore" instead of a "dash" in the role name 🙈 I can revert it, remove the role_name check from the lint, and, as a role to test, use "{{ playbook_dir }}/../..". What do you think?
There was a problem hiding this comment.
Import by path works in converge.yml, but molecule still doesn't like the role name. I'm not sure how to fix it :(
|
Has to do with the new naming format I believe. But I support both formats on ansi Le galaxy. The new and the old I think. Has to do with underscore too perhap.
Verzonden vanuit Outlook voor iOS<https://aka.ms/o0ukef>
________________________________
Van: Günther Grill ***@***.***>
Verzonden: Wednesday, February 4, 2026 6:40:51 PM
Aan: riemers/ansible-gitlab-runner ***@***.***>
CC: Erik-jan Riemers ***@***.***>; Mention ***@***.***>
Onderwerp: Re: [riemers/ansible-gitlab-runner] Fix for #420 (Conditionals must have a boolean result.) and basic tests for #425 (PR #426)
@guenhter approved this pull request.
________________________________
In meta/main.yml<#426 (comment)>:
@@ -1,7 +1,8 @@
---
galaxy_info:
- author: Erik-jan Riemers
+ role_name: gitlab_runner
@riemers<https://github.com/riemers> Changes in this file always ring alarm bells. Do you think this is again a problem with ansible galaxy? I mean the change should be done as the linter complained...
—
Reply to this email directly, view it on GitHub<#426 (review)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AAFGQOPFTSDUGWPXJIJM42L4KIVKHAVCNFSM6AAAAACT6USALOVHI2DSMVQWIX3LMV43YUDVNRWFEZLROVSXG5CSMV3GSZLXHMZTONJRG44DGMRWHA>.
You are receiving this because you were mentioned.Message ID: ***@***.***>
|
|
@riemers do you think I can merge that change? |
|
We can always try and see. If it goes bad we can revert. But that we need tests is a given.
Verzonden vanuit Outlook voor iOS<https://aka.ms/o0ukef>
…________________________________
Van: Günther Grill ***@***.***>
Verzonden: Thursday, February 5, 2026 6:58:56 AM
Aan: riemers/ansible-gitlab-runner ***@***.***>
CC: Erik-jan Riemers ***@***.***>; Mention ***@***.***>
Onderwerp: Re: [riemers/ansible-gitlab-runner] Fix for #420 (Conditionals must have a boolean result.) and basic tests for #425 (PR #426)
[https://avatars.githubusercontent.com/u/16029571?s=20&v=4]guenhter left a comment (riemers/ansible-gitlab-runner#426)<#426 (comment)>
@riemers<https://github.com/riemers> do you think I can merge that change?
—
Reply to this email directly, view it on GitHub<#426 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AAFGQOO3BACHNOJ4M55KHST4KLL2BAVCNFSM6AAAAACT6USALOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTQNJRGI2DSNRRHA>.
You are receiving this because you were mentioned.Message ID: ***@***.***>
|
|
Nice. @joannakotula-enghouse thx for the effort |
|
It created riemers.gitlab_runner in ansible galaxy. We can always tell folks to move to that one. I have now 3 different versions, but i just update them all when i release something new anyway. 1 i have to remove yet, but lets worry on that later ;p |
|
also @guenhter now we need to do something with the contribution md files and github actions etc etc.. time is a problem we all have it seems ;P |
Fix is in the first commit - simple one line
Tests:
I've created basic tests with only one mode tested (
by_template) on Ubuntu 22.04, in both greenfield and brownfield.I leave it that way to make it easier for someone else to close (eventually) #425 - I don't have enough time.
To run the test, install requirements from
molecule/requirements.txtand run therun_tests.shscript.I'm not adding anything to README because:
CONTRIBUTING.mdfile (I've read a suggestion that it should be.github/CONTRIBUTING.mdfile ;) )