RE: mi access to a dtaara

Steve,

Rather than update a DTAARA object to include the characteristics of a user
space (locking, APIs, User Domain, etc.) why not use a user space?

BTW, I agree with the idea of having this list be a forum for anything
architectural. Midrange-L is a great list, and I'm sure we're all on it, but
I want to make sure it doesn't get into the low-level stuff, that is what
this list is for. We've all heard the 7 words you can't say on television,
well I propose the following list of 7 phrases, if you are using them in
your post you should probably be here, not midrange-l:

Space/System Pointer
materialize
MI/NMI
architecture
Machine Context
program template
and finally... Alexei Pytel <G>

-Walden

-----Original Message-----
From: Steve Richter [mailto:srichter@AutoCoder.com]
Sent: Thursday, May 31, 2001 3:30 AM
To: MI400@midrange.com
Subject: Re: mi access to a dtaara


>
>This is no longer an MI question but I'll answer it here.  Further
discussion
>on these specific questions should be directed to the MIDRANGE-L, RPG400-L,
or
>C400-L lists.


I think this list should be more loosely defined as a forum for discussion
of the architecture, technical issues that span all languages, anything in
the system api reference, ...   But I am new to the list and am asking more
than answering, so I will defer to the opinions of others.

I am writing a computer language for the as400, so my main purpose in posing
questions is to get the info that I need.

but I find this topic interesting just in an academic sense.

Why is there not a chgdtaara api?
     ibm has had a long enough time to decide to do it
     and it seems simple enough to do
     and the fact that rpg and C call their own run time pgms, each with
limitations,  to do it shows that there is a need for a unified, full
functional api.
    and rpg ile is a bit limited without it because IN and OUT cannot be
used in a sub procedure.

And yet, no api.  And there is also no materialize dtaara defn api.

The dtaara does seem to have its design faults. It looks like it is just a
space with a data portion and an object defn portion.  This forces the
object into the system domain so it can be protected from an out of control
spcptr.  Any kind of granular locking of the dtaara ( lock a location in the
dtaara ) is prevented by the design.    A better way to do it would have
been a dual object design. one object holds the defn, the 2nd hold the data.
this would allow a spcptr to be set to the dtaara data. a space lock could
be applied.

can ibm not make up its mind on whether it wants to introduce a new  dtaara
type object or not?

Steve Richter


+---
| This is the MI Programmers Mailing List!
| To submit a new message, send your mail to MI400@midrange.com.
| To subscribe to this list send email to MI400-SUB@midrange.com.
| To unsubscribe from this list send email to MI400-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: dr2@cssas400.com
+---
+---
| This is the MI Programmers Mailing List!
| To submit a new message, send your mail to MI400@midrange.com.
| To subscribe to this list send email to MI400-SUB@midrange.com.
| To unsubscribe from this list send email to MI400-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: dr2@cssas400.com
+---