[jira] [Resolved] (JENA-2052) run github action maven.yml also for windows server and macos

["Andy Seaborne (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/JENA-2052?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Andy Seaborne resolved JENA-2052.
---------------------------------
    Fix Version/s: Jena 4.0.0
         Assignee: Andy Seaborne
       Resolution: Done

Added MacOS - Windows may run into limited on disk space for a general (free tier) user so it needs investigating.


> run github action maven.yml also for windows server and macos
> -------------------------------------------------------------
>
>                 Key: JENA-2052
>                 URL: https://issues.apache.org/jira/browse/JENA-2052
>             Project: Apache Jena
>          Issue Type: New Feature
>          Components: Build
>            Reporter: Øyvind Gjesdal
>            Assignee: Andy Seaborne
>            Priority: Minor
>             Fix For: Jena 4.0.0
>
>
> I've looked at the documentation for github actions [https://docs.github.com/en/actions/reference/workflow-syntax-for-github-actions#example-running-with-more-than-one-operating-system] and have wanted to try out running jobs on multiple OS'es.
>  
> Incoming PR (WIP) if it is of interest :) I've not used jira/github before, but I think I should check out a branch named by the Jena issue, created from this request?
>  
> Both windows and macos currently report failing tests from main branch:
> (under Build with maven)
> windows: [https://github.com/OyvindLGjesdal/jena/runs/1930240867?check_suite_focus=true]
> macos: [https://github.com/OyvindLGjesdal/jena/runs/1930240890?check_suite_focus=true]
>  
> I've not found out yet how to continue the CI tasks in the other OS'es if one of them fails, while still returning an error on completion. The second commit completes green when the continue on error: true is set, however there is no indication they failed, without viewing the job.
>  
> I thought maybe "if: sucess() | failure()" would work, but it seems the other jobs are still canceled, when the first one fails. I'll see if I can read a bit more to understand how this works and if it is possible.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)