Re: RPG36 with OCL examples desired; 1) insert-a-record 2) update-that-record

On 29-Jul-2016 10:49 -0500, Barbara Morris wrote:

On 7/29/2016 9:54 AM, CRPence wrote:

... I want to be able to, probably in a separate program, is to
change the data of that row; i.e. effect what became available as
an UPDAT (Modify Existing Record) OpCode in a slightly more modern
RPG.

I think the difficulty will be in determining which row is "that
row" RPG36 doesn't support INFDS for DISK files, so I don't think
you'll have any way of knowing the RRN of the new row.

I presume there will be a need for only one row in the file, thus only that one row will need to be changed; similarly to just the one row changed [as every row changed in a one-row-file], I could just as well change every row in a multi-row file and still have achieved my goal of [at least] one required invocation of the DB update method.

Perhaps the issue under review will be found to require an indexed file, for which I will need to change the program to access the row with a CHAIN; I expect I will just not pursue the issue any further at that point, unless the interested party provide their own test-case for me to use.

You didn't say why you wanted the output to be done by System-36 RPG
... I assume you have some reason for that,

RPG36 and running in the S/36EE is a requirement; the defining parameters of the issue.

but just in case, I feel I should tell you that even if you have some
reason for running your program in the System-36 environment, the
program doesn't have to be written in a System-36 language. You can
LOAD+RUN a program written in any language.

Aware; but as noted, not an option. I intend to dispute a claim about effects specifically when using the S36Env RPG [a SRCTYPE(RPG36) member], run against program-described files within STRS36; thus the path I have chosen. FWiW: while the requirements are also using program-described files, I know a CRTPF RCDLEN() is near-enough identical for a non-keyed file, so I opted not for any use of BLDFILE [and BLDINDEX]; even if after I get my test-case working, I might replace test replacing CRTPF with BLDFILE to go all-in with the S/36 Environment.