RE: INFDS layout for Display Files

[Buck Calabro <Buck.Calabro@xxxxxxxxxxxx>]

Nelson Smith wrote:

>Help!  I'm trying to understand what
>the manuals are telling me the data
>structure layout looks like for Display
>Files.  There is a common I/O area
>starting at position 241 that contains
>stuff like write-counts, read-counts, etc.
>and then a file-type specific area
>starting at 367 where the Aid Byte,
>cursor locations, etc are (for display files).

The RPG Reference section on File Exception/Errors has a decent explanation
of all this.  There are several feedback areas:

File feedback - RPG specific info about the file
Open feedback - Info updated at OPEN time
I/O feedback - Info updated at time of I/O operation,
               common to all file types
Device specific feedback - Info updated at time of I/O operation,
               common to the specific file associated with the DS
Get attributes feedback - Info filled by POST

>However, it seems that a POST operation goes
>back and overlays the common area starting at 241)
>with a completely different layout?  Am I reading
>that correctly?  Does anyone have a clear layout
>of what the INFDS looks like for display files only,
>and does it really change on a POST operation?

Yes, Yes, Yes and Yes.  Note that POST without a factor 1 specified is
different from POST and factor 1.  You probably want the latter.  The first
4 feedback areas mentioned above follow one another in memory.  The Get
Attributes area replaces the Device Specific area when POST occurs.

The aforementioned manual has the positions laid out and I know that several
people have posted a complete INFDS definition (search the archives
http://archive.midrange.com/index.shtml) but if you're still stuck after
reading the manual I'll post a display file-only example.

By the bye, I learnt this stuff on S/38 manuals but I am looking at the V5R1
books whilst I type this.  I think the V5R1 books are better than their
predecessors in this area.  If you don't find that to be true by all means
send in the reader reply card (better to light a candle than to curse the
darkness, eh?)

  --buck

"This box Rocks!" - related to 2002 NE IBM
Conference attendees by IBM's Stephanie Joy