John
Your post managed to fire off a long unused neuron and I - you know, the
when-all-else-fails option - took a look in the manual, the one describing
what the emulator is supposed to be emulating.
It may have opened up the cerebral pathway that Paul Gilmartin expressed
himself earlier in tones which could be taken as derogatory regarding 3270
architecture.
Anyhow, here is what I thought I remembered from the 3174 Establishment
Controller Functional Description, GA23-0218-11:
http://publibfp.dhe.ibm.com/cgi-bin/bookmgr/BOOKS/cn7a7003/2.4.4.3
<quote>
2.4.4.3 Using Insert Mode with Null/Space Processing
Null/space processing has a feature that is especially useful in Insert
mode. Without null/space processing, when a character string is followed by
blanks to the end of a field, characters could be inserted only if those
blanks were actually nulls. Spaces at the end of a field had to be changed
to nulls before Insert could be used. With null/space processing, those
blanks at the end of a field can be either nulls or spaces. Insert mode
works with either.
When null/space processing is active and a character is inserted on an
unformatted screen, only the characters between the cursor and the end of
the screen are shifted to the right. Characters cannot be wrapped past the
end of the screen because null-to-space conversions have already been
performed on all nulls before the cursor. In fact, with null/space
processing active, all leading nulls and embedded nulls within the text are
actually converted to spaces before Insert mode is activated.
</quote>
In other words, converting useless blanks to nulls is *not* some wizard
wheeze dreamed up by an emulator programmer geek but simply an emulator
doing what it says on the box.
And, by the way, it is even more "insert-friendly" than Paul Gilmartin
wanted. I wonder if your emulator "feature" is as solicitous.
Chris Mason
----- Original Message -----
From: "John McKown" <joa...@swbell.net>
To: <ASSEMBLER-LIST@LISTSERV.UGA.EDU>
Sent: Tuesday, February 01, 2011 1:13 PM
Subject: Re: ISPF Nulls etc (was Re: ASSEMBLER-LIST Digest - 20 Jan 2011 to
21 Jan 2011 (#2011-11))
Though this is completed off-topic, I love the fact my my TN3270
emulator has an option which acts like trailing blanks are nulls. If a
field has trailing blanks on it and I am in insert mode within that
field, then trailing blanks just fall off the tail end of the field in
insert mode, just like with ISPF NULLS ON.
On Mon, 2011-01-31 at 20:39 -0700, Paul Gilmartin wrote:
On Jan 31, 2011, at 13:44, John Walker wrote:
> Say, I have been noticing over the last year or so that i FREQUENTLY
> end up with my edit setting set to NULLS ON. This is an annoying and
> aberrant setting for any non-PC or non-unix use, which is to say ALL
> mainframe applications and work. However, when I set it to off, it
> ends up being reset somehow by ISPF. It seems that sometimes I am not
> even in browse or edit and it still does it. How can I reset this
> obvious over-riding default which I dislike?
>
???
I couldn't _live_ with NULLS OFF. For example, when editing JCL
(surely a non-PC and non-unix use), if I have NULLS OFF and attempt
to insert a character, the keyboard locks and I must:
o Press RESET to unlock the keyboard.
o Press INSERT to reassert that mode.
o Move the cursor past the end of the line.
o Press ERASE EOF.
o Move the cursor back to the intended insertion point.
o Insert my character.
Yucch! (But wouldn't it be neat if pressing INSERT MODE
momentarily set NULLS ON?)
(Usually, I just set NULLS ON; do my insertion; and hope that
it doesn't end up being reset somehow by ISPF.)
-- gil
--
John McKown
Maranatha! <><
