Change RPG compiler to require prototypes for exported procedures

IBM believes that the request described has been solved and is available with PTFs for 7.3 and 7.4.

There is a new REQPREXP (Require Prototypes for Exported Procedures) Control keyword. The default for this keyword is *NO, but you can also specify *WARN to cause the compiler to issue a severity-10 warning for missing prototypes or *REQUIRE to cause the compiler to issue a severity-30 error for missing prototypes.

You can also specify REQPROTO(*NO) for any exported procedure or for the Procedure Interface of a cycle-main procedure if you know that a prototype is never needed for that procedure.

There is also a new REQPREXP parameter for the CRTBNDRPG and CRTRPGMOD commands. If you do not obtain the PTF that provides the new parameter for the commands, you can still use the Control keyword aspect of the enhancement.

PTFs for 7.3:
- SI74612: TGTRLS(*CURRENT) compiler
- SI74590: New REQPREXP parameter for CRTBNDRPG and CRTRPGMOD. This PTF is not needed for the Control keyword aspect of this enhancement. Please be sure to read the special instructions in the cover letter before applying this PTF.

PTFs for 7.4:
- SI74613: TGTRLS(*CURRENT) compiler
- SI74614: TGTRLS(V7R3M0) compiler
- SI74591: New REQPREXP parameter for CRTBNDRPG and CRTRPGMOD. This PTF is not needed for the Control keyword aspect of this enhancement. Please be sure to read the special instructions in the cover letter before applying this PTF.

For more details, please see https://ibm.biz/rpgcafe_require_prototypes_for_exported_procedures

This request is still a candidate to be an enhancement for the RPG compiler.

IBM will use this request as input to planning but no commitment is made or implied. This request will be updated in the future if IBM implements it. IBM will use votes and comments from others in the community to help prioritize this request.

The current plan is that the compiler will provide a way for the RPG programmer to request a diagnostic message if there is an exported procedure without a prototype or if the main procedure does not have a prototype.

By default, no message will be issued.

A new H-spec keyword, such as REQEXTPR, will be added that will allow programmers to request that either a warning or error message be issued if there is no prototype for an exported procedure.

Additionally, there will be a new command parameter, such as REQEXTPR, for CRTRPGMOD and CRTBNDRPG. It will default to *NO, but it will support values *WARN and *REQUIRE. It will be possible to use CHGCMDDFT to change the default.

IBM agrees with the request and a solution appears to be desirable and feasible. IBM intends to provide a solution. However, IBM's plans may change and no commitment is made that a solution will be provided.

As discussed in the comments for this RFE, the current plan is that the compiler will issue a diagnostic message if there is an exported procedure without a prototype or if the main procedure does not have a prototype.

By default, it will be a sev-10 warning message.

A new H-spec keyword, such as DCLOPT(*REQEXTPR), will also be added that will cause a sev-30 message to be issued.

Also, there may be a P-spec keyword, such as REQPROTO(*NO), that programmers can code on a specific exported procedure to say that no prototype is required. This could be useful for Java Native methods, which are normally only called by Java code.

The COMMON Europe Advisory Council (CEAC) has reviewed this requirement and recommends that IBM view this as a MEDIUM priority requirement that should be addressed.

Background: The CEAC members have a broad range of experience in working with small and medium-sized IBM i customers. CEAC has a crucial role in working with IBM i development to help assess the value and impact of individual RFEs on the broader IBM i community and has therefore reviewed your RFE.

To find out how CEAC help to shape the future of IBM i, see CEAC @ ibm.biz/BdYSYj and the article "The Five Hottest IBM i RFEs Of The Quarter" at ibm.biz/BdYSZT

Therese Eaton – CEAC Program Manager, IBM

Maybe consider to also let the compiler automatically export the prototypes, so they only have to be declared in the procedure it self. That will make it consistent all the way around.

something like:

EXPPATH('/mypath/')
and/or
EXPSRC(*LIBL/QRPGLEHDR')

The CAAC has reviewed this requirement and recommends that IBM view this as a high priority requirement that is important to be addressed. The warning message and possible escalation would lead toward more bullet-proof use of the RPG language.

Background: The COMMON Americas Advisory Council (CAAC) members have a broad range of experience in working with small and medium-sized IBM i customers. CAAC has a key role in working with IBM i development to help assess the value and impact of individual RFEs on the broader IBM i community, and has therefore reviewed your RFE.

For more information about CAAC, see www.common.org/caac

For more details about CAAC's role with RFEs, see http://www.ibmsystemsmag.com/Blogs/i-Can/May-2017/COMMON-Americas-Advisory-Council-%28CAAC%29-and-RFEs/

Nancy Uthke-Schmucki - CAAC Program Manager

IBM has received the requirement and is evaluating it. IBM will provide a response after evaluation is complete.

Mark's suggestion is a good one indeed. I think that it would be nice warning while linking too, whether liked modules have not the required interface.

I like Mark's compromise suggestion. A warning could be added without breaking compatibility and the ability to escalate it to an error would be good.

I like Mark's compromise suggestion. A warning could be added without breaking compatibility and the ability to escalate it to an error would be good.

Hi Mark, that's a good idea for the compiler to add a warning message, since there would be no compatibility issues.

To escalate to an error, the compiler would add a new H spec keyword, plus possibly a new command parameter (but probably not a new OPTION, since it's impossible to change the command default for the OPTION parameter)

I think a meaningful compromise would be to have the ILE RPG IV compiler issue at least a "Warning" message when a Procedure Interface (PI spec) has the "EXPORT" keyword specified, but there is no corresponding Procedure declaration (PR spec) for that procedure.

Also, it might be nice to have the compiler escalate that Warning to an "Error" message (that halts the compile) if some new keyword is specified on the H-spec, or in the default H-specs in the data area RPGHSPEC, so that a customer site can make it their "shop standard" to require a PR-spec for any exported PI spec.

I think that would be a good step in the right direction.

@Bob_Cozzi, validating the call at bind time is much less likely to be implemented, since it would require changes by all the compilers as well as the binder. Indeed, it would need a different RFE.

Note that the compiler can't start requiring prototypes for exported procedures without some kind of option, since that would violate upward compatibility. So if this requirement is addressed, there would be an H spec option, plus possibly a command parameter.

Perhaps a more plausible option would be to validate the prototyped call at bind time. That is much more likely to be implemented than this request. But it might need to be a different RFE.