Closed cottsay closed 3 years ago
Of course I think of something as I click approve... Does it make sense to only perform the sed for the affected mock version or do we trust that whatever solution is provided will change the line enough that this will not actually result in a text change.
Does it make sense to only perform the sed for the affected mock version or do we trust that whatever solution is provided will change the line enough that this will not actually result in a text change.
This is a good thought, but I'm moderately confident that this change will be safe even if another change solves the problem. The addMacro
function can only take a string for that parameter, so I think that slightly degraded performance would be the worst we'd see. Hopefully we'll see an upstream fix soon and can just drop this hack quickly.
Another thought - adding bash logic to detect the mock version and conditionally apply the sed is probably about as fragile as just YOLO-ing the hack unconditionally, considering how straightforward the change is.
This change does not require deployment, and should be consumed by all future builds.
Upstream issue: rpm-software-management/mock#753
This is a temporary hack to unblock jobs until the issue is fixed upstream.
Before: After: