The main situation in which you can use this wizard is where you have one or more existing COBOL data structures that you want to use as patterns for passing data into or for receiving data back from an invoked flow or an invoked nonterminal application. Rather than having to use the message definition editor to manually create a service flow project message for each COBOL data structure, you can use the Import COBOL Files wizard to automatically convert the COBOL structures to messages.
Also, you can optionally use the wizard to create one or more operations (program definitions) each of which can reference any of the new messages as an input message, as an output message, or as a fault message.
The Import COBOL Files wizard has the following requirements for importing COBOL source files from a remote z/OS® system:
If the remote file is mapped to a .cbl, .cob, or .ccp extension, then the remote file is assumed to be a complete COBOL program that can have dependencies on other files. If the COBOL program does in fact have dependencies on other remote files, then you should provide the information about the partitioned data sets (PDSs) where the copy books exist by setting the SYSLIB for the remote file (see Adding a source path to the SYSLIB tab).
If the remote file is mapped to .cpy, it is assumed to be a copy book consisting of only 01 data structures or 01 or 77 elementary data item definitions. If the copy book that you select has references to other copy books, then their contents must be listed in the copy book that you select.
You can change the default behavior of the remote file extension in the Preferences window. Follow theses steps:
In the left pane, select .
In the right pane, select the More COBOL options tab.
Normally the Import COBOL Files wizard assigns the same name to an imported data structure (such as ACCOUNTINFO) that the original data structure had in its original COBOL source file (ACCOUNTINFO).
However, if the data structure is named DFHCOMMAREA in the original COBOL source file, then the wizard generates a name for the imported data structure based on the pattern <program_id> + _DFHCOMMAREA, where <program_id> is the program ID of the program from which the data structure is extracted (for example QUERYACC_DFHCOMMAREA). This change is made to avoid name collisions between multiple data structures having the commonly used name DFHCOMMAREA.
You should be aware of the following minor restrictions: