On Thu, 7 Jul 2022 08:17:11 -0700, Charles Mills wrote:
>
>Yes, the blank is apparently a factor (as you note). But I think it has to be 
>more than just a simple naïve scan (TRT or otherwise; TRT is just a software 
>loop written in micro- or millicode) for a blank.
> 
I'll assume (and OCO won't dispute me) that there are several tables,
selected by parse state, and either the designer never expected a
blank in that particular operand or a coder chose the wrong table.

>If it is a bug IBM is not going to fix it in time to get his task started, so 
>a workaround is the key.
> 
But if it's not reported it remains a pitfall for the next guy.

-- 
gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to