Re: Cursor positioning in latest CVS version -- LINUX5250

I'm glad cursor positioning has come up again. I'm not a masochist, just
curious.

I asked about cursor positioning before for a Java 5250 API I was writing
and got no response. So I took a look at the code (version 0.16.5) and
noticed
that on line 836 of session.c in the tn5250_session_write_to_display, is
the following:
if (end_y!=0xff && end_x!=0xff)
  tn5250_display_set_cursor(This->display, end_y, end_x);

That, as far as I can see, will never be true, as end_y and end_x are never
changed after they are assigned. Now (strangely) this seems to work so I
guess we can chalk it up to a lucky accident, or leftover code. But I would
still like to know, in English (not C or Java (or Delphi)), how have people
interpreted the rules for cursor positioning.
I ask because IBM's documentation for this is ambiguous (to be kind).
For example, they say:

a) "If the WTD command in valid, the cursor moves to one of three
locations": the IC if the 2nd CC bit 1 is 1, the first non-bypass field, or
row 1, col 1.
But they don't say how this affects the home position, or how long to keep
a pending IC around for, if it's not invoked after the WTD it was found in.
Or when does an IC change the home position? When it's received or when
it's invoked?
Also, they later mention that

b) a locked state and the 2nd CC bit 3 =1 is required to move the cursor.

So is it a) a valid WTD that moves the cursor? or an "unlocking" that moves
the cursor.

Then we have (in tn5250)
TN5250_SESSION_CTL_UNLOCK=0x08 used where (as I understand from IBM's
reference) TN5250_SESSION_CTL_IC_ULOCK=0x40 should be used. Or maybe the
OR-ed value should be used? There's a comment there (session.h line 74) )/*
??? - Not in my (older) spec */. Which older spec?

I think the "???" sums this all up pretty well (at least for me).

Can we please discuss this a little. I've got the 5394 Remote Control Unit
User's Guide and Functions Reference (Release 1 and 2), the IBM 5250
Information Display System Functions Reference Manual,  and whatever is on
the web.

Thanks, for tn5250. The app and the source.
-Mark



                      Frank Richter
                      <frichter@esda.co        To:       linux5250@midrange.com
                      m>                       cc:
                      Sent by:                 Subject:  [LINUX5250] Cursor 
positioning in latest CVS version
                      linux5250-admin@m
                      idrange.com


                      08/14/2002 07:36
                      AM
                      Please respond to
                      linux5250






Hi everybody,

I just tested the newest version of tn5250 from CVS. Macros work very
fine, thank you.

If I did not break things locally, there is probably something wrong
with the latest cursor positioning changes mentioned in the Changelog.

The cursor is allways stuck in the upper left corner, even on the logon
screen. I have to press TAB, to move it to the first input-enabled field.
If I reverse the patch in function tn5250_wtd_context_convert_nosrc
in wtd.c, which shifts IC after SOH, the cursor is positioned to the
right place again, at least with my applications.
Our AS400 is at V4R3, but that should make no difference here.

--
Esda Feinstrumpffabrik GmbH
Frank Richter, Leiter EDV
Hauptstr. 76, D-09392 Auerbach
E-Mail: frichter@esda.com
_______________________________________________
This is the Linux 5250 Development Project (LINUX5250) mailing list
To post a message email: LINUX5250@midrange.com
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/cgi-bin/listinfo/linux5250
or email: LINUX5250-request@midrange.com
Before posting, please take a moment to review the archives
at http://archive.midrange.com/linux5250.