On Mon, 16 Mar 2020 13:38:37 -0400, Bob Bridges wrote:

>Gil, you're probably going to say "yeah, that's not what I meant", but it sort 
>of does.  I mean, that's what the queued() function call will tell you, no?
> 
I believe not -- queued() works for PARSE PULL, not PARSE EXTERNAL.
(And I have a glitch in Regina's LINES() and STREAM() functions.)

>-----Original Message-----
>From: Paul Gilmartin
>Sent: Monday, March 16, 2020 12:31
>
>Why does the Rexx standard not provide a means for PARSE EXTERNAL
>to report end of input (e.g. SYSTSIN)?


On Mon, 16 Mar 2020 12:46:24 -0500, John McKown wrote:
>
>This is why I wish IBM would write a solution: so that other could use it
>pretty much anywhere without needing to get permission to install something
>from Github or CBTTAPE, which is unlikely. The only way to get ooREXX onto
>most PCs is to show Microsoft how it would make them tons of money. That
>seems to be _all_ that they care about, like most capitalistic companies, I
>guess. But M$, like most vendors, have NIH syndrome really bad. "Only we
>really know what & how to do anything!"
> 
On most UNIXen, (surprisingly including z/OS) one can use /dev/fd/number
with LINEIN() and CHARIN().

(IBM first provided /dev/fdnumber to circumvent the 256-byte limitation
of allocation (why a 256-byte limit when the TU descriptor is a halfword!?),
then overcame NIH to accommodate the prevalent /dev/fd/number.)

-- gil

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

Reply via email to