Using the encoding approach would allow that functionality to be regression tested in Ansible-core against the existing test infrastructure, ie without needing to add z/OS machine support.
Look at the commits tab - Notice in the commit log there are a few ‘merge devel into xxx’. Those suggest a rebase gone bad. Someone with better knowledge may know how to fix this. I tend to give up on the PR and create a new, clean one off a current devel.
1 Like
I followed a “merge” strategy whereas it seems the community prefers a “rebase” strategy to bring branches up to speed. Let me go ahead and open up a new PR based on the latest devel as suggested.
To @gundalow’s point, the initial code changes to the builtin modules we are targeting will be platform independent, so it makes sense that running them through existing regression infrastructure will be sufficient.