Joined: 07 Dec 2007 Posts: 2205 Location: San Jose
HameedAli wrote:
I don't understand the significance of the CNTL cards.
how 4,10 refers to INB not INA and the (81,1,A)SORTED,NOSEQCK
HameedAli,
Since your files did not have a key to match from both files, we pad an arbitrary value of '1' in both files at the end of the record and use that as a matching key. Since we are padding the extra byte and we know it is sorted, so there is no need for sorting during match. SORTED,NOSEQCK tells DFSORT not to check the order of the records which can improve performance.
HameedAli wrote:
Is it possible to SORT while using JOINKEYS, I tried SORT FIELDS(1,3,CH,A). The step went fine, but the output wasn't sorted.
Yes it is possible. Replace the OPTION COPY statement with SORT FIELDS=(1,3,CH,A).
I am assuming that you did not remove the OPTION COPY statement and just added SORT statement which DFSORT ignored as OPTION COPY was used.