Re: tn5250 for DOS (fwd) -- LINUX5250

Wolfram, if you've succeeded in porting it to DOS, you've won $100 from
me.  I offered $100 to the first person to port it to DOS.  Send me
the code, and if it meets the requirements below, I'll send you the
prize.

I wrote:
Date: Fri, 30 Jul 1999 10:47:05 -0700 (PDT)
From: Seth Cohn <scohn@oregonmed.net>
Subject: DOS version of TN5250


> Actually if it weren't for the TCP/IP stack, I could be fairly certain
> that this would compile under DJGPP.  I know DJGPP has curses support
> and all. It wouldn't be a terrible stretch.

I figured not... but it's beyond me.  I'm a techie, and know enough
programming to read the code and make changes, but not write from scratch.

> > $100 to the first person to create a working DOS version...
> > including TCP/IP stack (you can use crynwr packet drivers)
> > 
> > If I could have a dos version tn5250, I'd be very very happy.
> > My boss would be even more so.  Be worth $100 to me just to give
> > this to him and amaze him at what open source can do.
> 
> Are you serious?

I am serious.  $100 for a DOS based answer. 

Conditions as follows:

size: I prefer it fitting on a boot floppy.  1.2 meg would be preferred
( a few older 1.2 drive out there).  If it's just not possible (why not?),
then I'd take a slightly larger answer... 1.44 or even small (20
meg) HDs...

tcp/ip: standard ethernet cards (as I said, crynwr drivers is fine)
I mostly need ne2000 or 3com card drivers.

5250:  I need to replace hardware terminals.  Color is fine, but not a
requirement.  Keymapping to match real 5250 is a requirement.  Full 5250
emulation, at least for the terminal end is needed.  I don't need printer
support (we use networked printers).  Stuff not doable in dos console is
ok (like rulers, odd characters, etc)

Open sourced answer:  GPL preferred, but I'm open to something less free.
No licensing hassles though, please.  I want 'free beer' at least.

>From searching usenet, I know others out there want answers like this...
I'd bet it's a common problem for As/400 using companies who have lots of
older 386/486 boxes, which runs windows like crap, and would love to not
have to buy more 5250 terminals (which are EXPENSIVE, a full PC system is
a better value these days) and they are SNA based (which is getting to be
more and more of a pain... SNA sucks, tcp/ip rocks.)

$100 for a working solution.  Out of my own pocket.  Worth at least that
much to me, just to see the look on his face.  I'll write a check when I
can go to one of my sites, insert a preconfigured floppy (with IP address,
gateway, etc) bring up an as/400 login, and the user can use it like a
5250 terminal, and I can walk away from it.

Seth

















On Fri, 1 Oct 1999, Michael Madore wrote:

> Hi,
> 
> Does this sound like the problem some of you are having logging on with a
> second session?  It would be funny if the emulator had to be ported to DOS
> to solve a problem we are having under Linux. 
> 
> Mike Madore
> 
> From: Wolfram Koehn <ratiosystem@t-online.de>
> Reply-To: Wolfram Koehn <wkoehn@ratiosystem.com>
> To: mmadore@blarg.net
> Subject: tn5250 for DOS
> 
> first of all I want to thank you for creating tn5250. For some time I was 
>searching for a 5250-emulation running under DOS. I found TN5250 via internet 
>but unfortunately it was designed for UNIX. So I thought about porting it to 
>MS-DOS. That should be pos
> 
> 
> sible, because you provided the source code, and it is written in C. Because 
>there is no curses for DOS, I had to emulate the needed curses-functions. 
>After resolving some minor Problems (e.g. the 12.51-version reached the 64k 
>structure-limit of my borlan
> 
> 
> d 4.5 compiler) I was happy to get it running. I have to admit that I don't 
>make use of the environment-accommodation, it makes no sense under DOS anyway. 
>If of some interest I can send you my curses-code. It is very simple and 
>straightforward, but it w
> 
> orks.
> 
> Now that I changed to the latest version (13.12) I couldn't detect any more 
>bugs except one: The dequeing of records in record.c works in the wrong 
>direction: tn5250_record_list_add should return the first in the queue (list) 
>instead of the last (record
> 
> ). This bug can lead to a blocking, if more than one telnet-record is sent in 
>a TCP-block. It always did, when logging into a second AS400-session and 
>getting the program-message screen. This problem
> didn't occur under Linux. It seems that Linux uses an
> 
>  other blocking-algorithm than the DOS-TCP/IP-stack does (I use the one from 
>FTP-Software).
> 
> Sincerely
> 
> Wolfram Koehn
> 
> 


+---
| This is the LINUX5250 Mailing List!
| To submit a new message, send your mail to LINUX5250@midrange.com.
| To subscribe to this list send email to LINUX5250-SUB@midrange.com.
| To unsubscribe from this list send email to LINUX5250-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---