Shmuel wrote:
>Whoosh!
I'm going to assume that's in reference to your sending a blank reply rather
than an appalling lack of comity.
>The whole point of Read Partition Query is so that software can detect what
features are supported. Had IBM implemented a Start Fiewld Contifuous buffer
order or a Protected character attribute, why would you expect SPF to use it
in a data stream targeted to a 3277? All of the new device support requires
the Extended Data Stream that came in with the 3274, and RP-Q was part of
that from Day One.
Yeah, no kidding about RP-Q. You're still missing my point, presumably
because I didn't spell it out in sufficient detail: if all this really lets
me do is add a character to a field, then it's of very limited value, AND
any application exploiting it has to dual-path the code and largely rethink
its data layout on the screen. Not a lot of bang for the buck.
Things like XEDIT would benefit the most, but even then I'm not convinced
you'd get more than one byte back per line. Would you really butt the PREFIX
area right up against the data:
0* * * Top of File * * *
1this is data
2* * * End of File * * *
or
=====* * * Top of File * * *
|...+....1....+....2...
=====this is data
=====* * * End of File * * *
or
00000* * * Top of File * * *
|...+....1....+....2....
00001this is data
00002* * * End of File * * *
?
That'sgoingtobeprettyhardtoread, especially if the data is numeric and you
use NUMBER ON:
00000* * * Top of File * * *
|...+....1....+....2...
0000112345678
00002* * * End of File * * *
While we can all come up with times when we'd like this ability, it's a
pretty narrow use case overall, I think. You generally-not always, of
course, but generally-need some visual way to tell the end of one field and
the start of the next. "But colors"-maybe. Again, narrow use case, and that
would reduce the utility of colors for other uses.
The place where I'd most want it is to do structured input kinds of things,
where you could use the TAB key more efficiently. I had an XEDIT macro
called ASMIN that I worked on for a while, that used SET RESERVED to make
hardware-tabbable input fields at columns 1, 10, 16, 40. This turned out to
be more of a pain than a benefit because of the attribute bytes getting in
the way (when you had a macro invocation, for example, that needed more than
five bytes starting at column 10. (Yes, I tinkered with making it 'smart'
such that "if column 15 is not blank, then concatenate rather than putting a
space between 15 and 16", but it just wasn't of enough value to be
worthwhile: it irritated me more than helped.)
I'm not saying it would be useless, I'm saying that I would imagine that
40++ years ago, it was discussed, and IBM correctly decided the value wasn't
there for the effort.
I've had an abiding love affair with 3270s for about 43 years now, have done
lots of 3270 datastream programming (including some major surgery on XMENU
back in the day). There are lots of things I wish had been different, but
this isn't one of them!
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
