On Wed, 4 Aug 2010 11:45:47 +1000, Wayne Bickerdike wrote:
>
>Sort and compare is the lowest, KISS way to achieve this. The language
>is irrelevant for the process. I nominated REXX because of the simple
>integration with ISPF but obviously that has limits on the size of
>dataset that can be brought into memory etc.
>
I'm missing something. How does sort-and-compare differentiate
between File A:
Huey
Dewey
Louie
and File b:
Huey
Louie
Dewey
or, is this like an RDB where there's no intrinsic ordering of records?
In days of yore, the ISPF tutorial for SuperC described the algorithm.
I believe it computed a hash for each record for a manageable handle.
Then it found the longest sequence of identical records in the two
files (how?), eliminated that from both files, and iterated on the
remainder. Lots I don't remember.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
