Re: Data structure changes in QSYSINC?

Sorry if I wasn't clear.

The name changes didn't change the API. Any code that has already been
compiled using the old names still works even though the names are
different, because the data is in the same place. So any existing compiled
code should continue to work: object code compatibility.

Now, if you are trying to COMPILE source using the names from the old copy
book, but including the new copy book, the you compile will fail: source
code incompatibility.

Object incompatibility would have caused a great uproar. Source
incompatibility affects a much smaller market. That's why as programmers
we appreciate it so much, because IBM typically adheres to it even though
it's not strictly necessary.




On 4/24/2018 6:00 PM, Vernon Hamberg wrote:

Not sure I'm getting you here, Joe - in this case, the 7.1 version and
the 7.2+ versions are different in several subfield names - this would
break code that included this copy member - and example is QSQPO08 at 7.1
is QSQPRO at 7.2 - same position, same definition, different names.

I think the name change is the main thing - the data is in the same
location for what it is, like Privileges Option, CCSID Option, etc., The
name changes will result in "not defined" errors, right?

Bruce would never have done this, maybe someone rewrote his utility that
takes the C++ code and converts it to RPGLE.

Elsewhere in this member, B data types are changed to I - but the names
stay the same.

Cheers
Vern

7.1 change comments:

D*$A3= P9C39022 7D10 090327 HRGJAV: CCSID and privileges
D*$A4= D92303 090825 LAWILL Adding replace option
D*$A5= PFW579962 7D10 020131 ROCH: Adding obfuscation
D* option
D*$A6= PFW588646 7P10 120802 ROCH: Adding qualified name
D* option

7.2 change comments:

D*$A3= PFW566930 7D20 010803 ROCH: Obfuscation option
D*$A4= PFW606909 7D20 130731 mja: Adding additional
D* index option and index
D* instead of view option
D* and qualified name option

And tail end for 7.1

D QSQPO08 587 587
D* Privileges Option
D QSQCSIDO 588 588
D* CCSID Option
D QSQCORO 589 589
D* CrtOrReplace Option
D QSQOO11 590 590
D* Obfuscate Option
D QSQRSV100 591 591
D* Reserved1
D QSQRSV200 592 592
D* Reserved2
D QSQQNO 593 593
D* Qualified Name Option
D*QSQERVED02 594 594

And tail end for 7.2:

D QSQPRO 587 587
D* Privileges Option
D QSQCCO 588 588
D* CCSID Option
D QSQCRO 589 589
D* CrtOrReplace Option
D QSQOBO 590 590
D* Obfuscate Option
D QSQARO 591 591
D* Activate RCAC Option
D QSQMPO 592 592
D* MaskAndPermission Option
D QSQQNO 593 593
D* QualifiedName Option
D QSQAOO 594 594
D* AdditionalIndex Option
D QSQIVO 595 595
D* IndexInsteadofView Option
D*QSQERVED02 596 596

On 4/24/2018 4:51 PM, Joe Pluta wrote:

And that's what happens when you're worrying only about object
compatibility, not source compatibility.

If they changed the name of a field but left the attributes the same, no
compiled programs would fail. Only someone actually using the copy book
(that is, every programmer) would have a problem.


On 4/19/2018 6:44 PM, Bruce Vining wrote:

As the previous owner of QSYSINC I can safely say that is NOT
supposed to happen. There is no reason for the removal of fields
(which is not the same as saying the fields may no longer be used).

I don't think any fields were removed, but some of them have names
changed.

End result is, of course, the same.

david

--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/midrange-l.

Please contact support@xxxxxxxxxxxx for any subscription related
questions.

Help support midrange.com by shopping at amazon.com with our affiliate
link: http://amzn.to/2dEadiD