I don't know how directly related this is to your situation, but I was
SELECTing stream 1 (which would unknowingly fail because secondary was not
connected) and the OUTPUT after the SELECT would succeed, but actually write to
the primary output instead. So I had to do a STREAMSTATE to not do the write
at all.
/*...Determine if secondary output stream is defined/connected...*/
TRACE 'O'
'STREAMstate OUTPUT 1'; IF (rc<0) | (rc>8) THEN o1=0; ELSE o1=1
TRACE 'N'
/* SAY '@0182 o1='o1 */
DO nobs = 0 BY 1 /* Do until EOF */
'READTO record' /* Read from pipe */
IF (rc /= 0) THEN LEAVE
IF (...decision logic goes here...)
THEN DO
'SELECT OUTPUT 0' /* write accepts, assume always connected */
'OUTPUT' record
END
ELSE
IF (o1) THEN DO /* Write rejects only if secondary conn */
'SELECT OUTPUT 1'
'OUTPUT' record
END
END
-----Original Message-----
>From: Glenn Knickerbocker <[email protected]>
>Sent: Apr 16, 2010 10:11 AM
>To: [email protected]
>Subject: Re: [CMS-PIPELINES] Coding a user-written pipeline to understand NOT
>
>On Fri, 16 Apr 2010 08:51:26 -0500, Mark Wheeler wrote:
>>signal on error
>>
>>do forever
>> 'PEEKTO input_record' /* Peek at the input record */
>> if input_record = test_value then /* Simplistic */
>>
>> do
>> 'SELECT OUTPUT 0'; 'OUTPUT' input_record
>
>You're killing the stage as soon as you see a nonzero return code, but
>you've never tested whether the primary output is connected, so it dies
>as soon as you write to it. You can do just the same as you did for the
>alternate, testing with STREAMSTATE at the beginning and writing to it
>only if it was initially connected.
>
>¬R // Failing is not just for failures, it's \\ users.bestweb.net/~notr
>for everyone. Failures just have more experience. \\ iamlistener.com
