On Thu, 14 May 2020 14:58:27 +0000, Seymour J Metz wrote:
>The first thing I don't understand is the order of the code. Shouldn't the
>syscalls('ON') come before Address SYSCALL 'readdir 'BGZUSDIR' ls. lsst.'?
>OTOH, the code is looking where it should for the FD.
>
I suspect it doesn't matter. address SYSCALL command blind-dubs the TCB.
I'm quite suspicious of the commas. In a command, I'd expect
them to result in Rexx syntax errors.
But prior to SYSCALLS('ON') the SYSCALL constants are defined only in the
shell environment. I think he clarified to UNIX.
Context; Le bon Dieu est dans le détail. So:
SIGNAL ON NOVALUE /* Always, for debugging! */
PARSE SOURCE . /* Where are we? */
/* Then your sample code; post the trace. */
>TSO doesn't create file descriptors, so an FD of 0 is what I would expect
>calling the code from ISPF.
>
>________________________________________
>From: Michael Babcock
>Sent: Thursday, May 14, 2020 10:46 AM
>
>Sorry Yes Unix System Services.
>
>> ________________________________________
>> From: Michael Babcock
>> Sent: Thursday, May 14, 2020 8:53 AM
>>
>> In the USS REXX SYSCALL environment, is it normal for a command such as:
>>
>> open /u/xx35/test, o_rdwr+o_creat+o_trunc, 755
>>
>> To return a file descriptor of 0 (zero)?
>>
>> The reason I ask is that the subsequent write fails with
>>
>> RC 71 RSN 571011C
>>
>> Which basically says: file is not opened
>>
>> I thought fd 0 is normally STDIN.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
