|
Back issue
|
Reusing a temporary dataset in the same step
For historical reasons, a passed temporary dataset is only intended to
be "received" once in the same job step. For example:
<![CDATA[
//S1 EXEC PGM=somepgm
//OUT1 DD DISP=(,PASS),UNIT=SYSDA,DSN=&&DSN,SPACE=....
//S2 EXEC PGM=otherpgm
//IN2 DD DISP=(OLD,PASS),DSN=&&DSN
//IN3 DD DISP=(OLD,PASS),DSN=&&DSN
]]>
The temporary dataset &&DSN will not be found for the second DD
statement IN3 in step S2 because it has already been used once by
DD statement IN2.
However, by referring to the information from IN2, it is possible
to use the same temporary dataset again in the same step:
<![CDATA[
//S1 EXEC PGM=somepgm
//OUT1 DD DISP=(,PASS),UNIT=SYSDA,DSN=&&DSN,SPACE=....
//S2 EXEC PGM=otherpgm
//IN2 DD DISP=(OLD,PASS),DSN=&&DSN
//IN3 DD DISP=(OLD,PASS),DSN=*.IN2,VOL=REF=*.IN2
]]>
This technique can also be used to refer to a temporary dataset
created within the same job step:
<![CDATA[
//S1 EXEC PGM=somepgm
//OUT1 DD DISP=(,PASS),UNIT=SYSDA,DSN=&&DSN,SPACE=....
//IN1 DD DISP=(OLD,PASS),DSN=*.OUT1,VOL=REF=*.OUT1
]]>
Here, the temporary dataset created on DD statement OUT1 can be reused
as an input in the same job step via DD statement IN1.
Other examples of the use of this technique can be found in the
samples of the use of the DFSORT SPLICE operator found in the document
linked from the page
Smart DFSORT Tricks
Feedback
Comments or suggestions can be
sent to the author.
Last Month's Issue
