Original Publication Date: 2002-Jun-16
Included in the Prior Art Database: 2003-Jun-20
Described is a method by which an Application Programming Interface (API) Exit handler can deligate some or additional control of the running of exits to a "Control Exit". The following method is appropriate for the case when more than one exit can potentialy be run for each API call. For instances where there are more than one exit the collection is often described as a set of chained exits. This method is of particular relavence when there is also the concept of a session spanning a number of API calls. For example exits running on the API calls to a resource manager where an application might first connect to, perform some work on and finally disconnect from a particular resource. This method could be used to either avoid the need to build a complicated controlling mechanism into the exit handler or as a complimentary method of achiving the desired control. The exit mechanism itself is used to solve the problem through a "Control Exit". This control exit would typically be defined as the first exit in the chain of exits, and would, based upon the parameters with which it has been invoked, control whether to skip execution of the exits that follow it in the chain or to allow them to be executed. This allows for control from a user writen exit enabling flexibility that might not have otherwise been envisaged as a requirement in the design of the exit handler.