Pathmaker Programming Guide

Modifying the Database
Maintaining Pathmaker Applications
067868 Tandem Computers Incorporated 8–17
5. Regenerate the requester.
Confirm that the entries on the Requester Generation Data screen are current and
that Compile is Y . Press shifted F6 from the Requester Definition screen to
generate the requester. You must regenerate each requester that uses the changed
definition.
6. Update custom service code.
If the definition is sent to a service and if you change a field name or add a new
field, you must update the custom service code for every service that uses the
changed definition.
7. Regenerate servers.
If the definition is sent to a service, you must also regenerate servers that include
the service. Navigate to the Server Definition screen and press shifted F6 to
regenerate the server.
8. Move the SCREEN COBOL and server object code to the production subvolume.
If the application no longer resides in the testing environment, use SCUP to copy
the SCREEN COBOL code to the application’s SCREEN COBOL object library files
(POBJCOD, POBJDIR, and POBJSYM). See “Changing Screen Decorations or
Screen Layout” earlier in this section for instructions on using SCUP.
If you have modified many requesters, you might want to copy all the SCREEN
COBOL code to the application’s SCREEN COBOL object library files. The SCUP
command to do this is:
?COPY (*), $ vol.subvol .POBJ
Use FUP to move the server object code to the production subvolume.
9. Exit from the Pathmaker full screen interface.
10. Move the help text files to the production subvolume. See “Adding Help Text”
earlier in this section for instructions.
Modifying Records The procedure for modifying records is similar to the procedure for modifying
definitions, except that when you modify a DDL record, you must also update the file
that is associated with the record.
You can use the ?FUP command from the DDL compiler to create a FUP source file
with commands for creating the new file. If the file you are updating has data in it,
you must use extra care to load the new file correctly.
For example, assume that a record is being changed at the end of its structure and a
new field is being added to the end of the record. In this case, you can rename the old
file to protect it, create the new file using the FUP source file, and use the FUP LOAD
command to load the contents of the old file into the new file.
Or assume that the record is being changed in the middle of its structure; for example,
a field in the middle of the record definition is being lengthened. In this case, you can
rename the old file to protect it and create the new file using the FUP source file.