Extensible general workflow framework system

Abstract

The embodiment of the invention discloses an extensible general workflow framework system. The extensible general workflow framework system comprises a schedulable workflow, a schedulable work item, a first scheduling strategy module and a second scheduling strategy module. The first scheduling strategy module is used for customizing a scheduling strategy for scheduling the work items; the second scheduling strategy module is used for customizing a scheduling strategy for scheduling the workflow; the work items comprise executed work content and a context object, the executed work content is customized through a function pointer, the context object realizes self-definition of context data through a class template, and each work item in the workflow forms a directed graph. According to the system, the workflow framework supports data transmission among the work items, the scheduling strategy of the work items supports user customization, and the system has the general expansibility of the workflow.

Description

technical field [0001] The invention relates to the technical field of software testing, in particular to an expandable general workflow framework system. Background technique [0002] The existing open source framework of workflow has realized the functions of workflow establishment and scheduling, among which the scheduling design and implementation is divided into two levels: workflow and work item. Workflow can contain several work items, and work items include specific Contents of execution Job work item and definition condition and move to Condition work item. [0003] Each Job work item does one thing, and this thing is customized through a function pointer. The type of the function pointer is fixed as: std::function<void()>, that is, the execution content with no parameters and no return value. Each Condition work item is used to define the condition and direction, that is, which Job work item to schedule when the condition is met. When there are multiple Job ...

AI Technical Summary

Problems solved by technology

[0004] However, in the current workflow framework system, the data transferred between the work content nodes that each work item actually needs to execute in the workflow is defined by discrete global variables in the framework. Such a framework lacks guidance for users; The flow scheduling strategy cannot be customized, which cannot meet the complex business environment and loses the core expansion requirements of the workflow. The current scheduling scheme in this framework only allows scheduling of one direct subordinate when using defined conditions and moving to Condition work items. The content Job work items that are specifically executed downstream, and in many business situations, all or multiple content Job work items directly under the specific execution of the downstream need to be executed synchronously

It does not meet the application requirements in the field of semiconductor testing, that is, the workflow framework does not support data transfer between work items, the scheduling strategy of work items does not support user customization, and does not have the general scalability of workflow

Images