Re: assertion 0 failed, 0.14.0

["Jason M. Felice" <jasonf@xxxxxxxxxxxxxxxx>]

On Wed, Dec 08, 1999 at 04:06:03PM -0500, pcunnane@learningco.com wrote:
>      Jay,
>      
>      It's not unusual for non-displayable characters to be sent to a 
>      terminal as display data.  This would happen, for example, when 
>      displaying raw physical file data, if the data contains packed decimal 
>      fields.  5250 terminals usually replace this data with a solid 
>      rectangle, to visually indicate non-printable characters.  If memory 
>      serves, these are all characters less than 0x40, and 0xFF.

True.  The problem is that there are escape codes (X'04') within the display
stream, as well as orders (X'0x' and X'1x') codes.  These instruct the 
terminal on how to behave.  Not all of the orders are used, but the tn5250
server can't just send arbitrary data with codes which are orders or escapes
and expect the terminal to understand what it means.

The real problem is that, as specified according to spec, we are (partially)
determining what is an order by zone-testing it.  The recent hacks allow
bytes which pass the order-test to be handled as data anyway.  Compound this
with the fact that our emulator doesn't implement (or even know about) all 
existing orders, as there are ones that are only sent to a 5294, for example,
and it becomes difficult to decide how a particular character should be treated.

I'm going to take a more in-depth look at this and analize some output from
the 400 as soon as I get this signed-field-transmit issue resolved.

-Jay 'Eraserhead' Felice

>      
>      HTH
>      
>      ____
>      Paul
>      
+---
| 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
+---