Post-Module & Post-Stage execution
Closed this issue · 1 comments
Right now MPL supports only post-pipeline execution steps, but seems it will be useful to introduce new mechanisms to have post-module & post-stage executions.
Example: to collect reports or artifacts before the stage-agent will be destroyed even if there is an error happened during module/stage execution.
Anyone else see benefits of such feature?
Ok, seems it could be implemented relatively easy, but there is no need to post-stage execution (could be supported using already implemented MPLPostStep and stage post).
So, I will add additional step MPLModulePostStep(closure) that will be bound to the current module (don't want to add another interface for potential cross-module communication). MPLManager interface will support any module name and you can use it with your own step in the nested library.