gitlab-org--gitlab-foss/app/services/ci/destroy_pipeline_service.rb

# frozen_string_literal: true

module Ci
  class DestroyPipelineService < BaseService
    def execute(pipeline)
      raise Gitlab::Access::AccessDeniedError unless can?(current_user, :destroy_pipeline, pipeline)

      Ci::ExpirePipelineCacheService.new.execute(pipeline, delete: true)

      pipeline.destroy!
    end
  end
end
Destroy pipeline in service Move all logic for destroying a Pipeline into a service so it's easily reusable. 2018-11-12 18:18:57 +00:00			`# frozen_string_literal: true`

			`module Ci`
			`class DestroyPipelineService < BaseService`
			`def execute(pipeline)`
Raise exception when user is not authorized 2018-11-13 16:27:00 +00:00			`raise Gitlab::Access::AccessDeniedError unless can?(current_user, :destroy_pipeline, pipeline)`
Destroy pipeline in service Move all logic for destroying a Pipeline into a service so it's easily reusable. 2018-11-12 18:18:57 +00:00
Properly expire all pipeline caches when pipeline is deleted When deleting a pipeline, only some of the cache structures were being expired, but not the full pipeline list. We have to synchronously schedule a pipeline cache expiration because the pipeline will be deleted if the Sidekiq expiration job picks it up. To do this, properly extract all the logic buried in the Sidekiq worker into a service, and then call the service. Closes https://gitlab.com/gitlab-org/gitlab-ce/issues/60469 2019-04-12 19:29:47 +00:00			`Ci::ExpirePipelineCacheService.new.execute(pipeline, delete: true)`
Clear pipeline status cache after destruction of pipeline `project.pipeline_status.has_status?` is cached, which can lead to Error 500s in the UI if the this condition is used to check whether a pipeline exists for a commit. We now expire the cache to ensure that the information is consistent. Closes https://gitlab.com/gitlab-org/gitlab-ce/issues/59453 2019-03-26 06:01:14 +00:00
Properly expire all pipeline caches when pipeline is deleted When deleting a pipeline, only some of the cache structures were being expired, but not the full pipeline list. We have to synchronously schedule a pipeline cache expiration because the pipeline will be deleted if the Sidekiq expiration job picks it up. To do this, properly extract all the logic buried in the Sidekiq worker into a service, and then call the service. Closes https://gitlab.com/gitlab-org/gitlab-ce/issues/60469 2019-04-12 19:29:47 +00:00			`pipeline.destroy!`
Destroy pipeline in service Move all logic for destroying a Pipeline into a service so it's easily reusable. 2018-11-12 18:18:57 +00:00			`end`
			`end`
			`end`