On Wed, 30 Aug 2023 22:43:06 +0000, Seymour J Metz wrote:
>No, that conflates things that happe ned at very different times.
>
I believe it confuses the reader by introducing a distinction where no
difference exists.
I'd prefer:
if after determining equivalent JCL <hyperlink to page 41> the
result is a single ampersand followed by a name, that names
a temporary data set or sysout. For example:
// SET VAL='&FOO'
...
//SYSUT2 DSN=&VAL, * Names a temporary data set with qualifier FOO.
>First, OS/360 had DSN=&foo for temporary datasets.
>
>S/360 added symbolic parameters on the PROC and EXEC statements. As part of
>that, &name is replaced with the value of the referenced parameter if it
>exists, and otherwise left as is. A double ampersand is replaced with a single
>ampersand. As long as you don't have foo=bar, DSN=&foo works like it always
>did, but you can code it DSN=&&foo just in case.
>
>Then, decades later, // SET came alon,, with similar rules..
>
>________________________________________
>From: Mike Schwab
>Sent: Wednesday, August 30, 2023 4:59 PM
>Subject: Re: Threading (was: LISTSERV Trivia: Deleting drafts?)
>
>On Wed, Aug 30, 2023 at 3:32 PM Paul Gilmartin wrote:
>
>> And I've seen no good explanation for "&&TEMPDSN"
>>
>S/360 started with &tempdsn.
>Later (S/360-370), // SET VAR=value was added.
>Created &&tempdsn is always &tempdsn.
>&var picks up the SET value and if none is specified it uses the name
>of the var.
--
gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
