I would make pairs (MX-name, server) from the primary input, match those
against your table of server names. The pairs that don't match the server
table provide you the MX-name of a primary record to reject. That stream
can be used in 'not collate' to select the primary records that don't have
a reject. But I would need to sit down to make it does not stall.

Rob


On 24 June 2014 20:49, Bob Cronin <[email protected]> wrote:

> I have a primary stream consisting of records that look like this:
>
> MX-record-name mail-server1 mail-server2 ...
>
> I have a secondary stream consisting of all the unique mail-serverN names
> extracted from the primary stream for which A-records exist.
>
> I want to keep only those primary stream records for which ALL the
> mail-serverN names listed on that record are found in the secondary stream.
>
> I can't quite figure out how to approach this.
>
> As an example, say I have:
>
> d01mll96 d01mll96.ibm.com d01mc097.ibm.com
> d23m0072 d23m0072.ibm.com d23ml072.ibm.com
>
> on the primary stream and
>
> d01mll96.ibm.com
> d01mc097.ibm.com
> d23ml072.ibm.com
>
> on the secondary, I'd want just
>
> d01mll96 d01mll96.ibm.com d01mc097.ibm.com
>
> on the output (since d23m0072.ibm.com is not found on the secondary
> stream).
>
> Ideas?
> --
> bc
>

Reply via email to