|
Archives of the TeradataForumMessage Posted: Mon, 22 Dec 2008 @ 15:55:32 GMT
It would depend on the source of the value of the "count", be it row count, byte count, etc. If you are obtaining the count within the Unix script via some type of expression, then the count could be obtained and captured in a pair of variables, via: Generally CT1=table one count= count_value CT1='expression_to_get_table'_count'; use back tics to surround the count values CT2='expression_to_get_table2_count' To compare the values then use the shell script comparison techniques, which which will be slightly different depending on whether you are using Korn Shell, Bourne Shell, or C Shell. A variation on Bourne & Korn C IF [ $CT1 > $CT2 ] if [ $CT1 > $CT2 ] if [ $CT1 > $CT2 ] then THEN then do_action Shell_action do_action endif FI fi If you are allowed to run Unix scripts in the Unix node, then you can jump in and out of BTEQ to extract values. Please remember that the characters to remark out expressions is different for the Unix script "#", vs. any SQL you are running "--" or "/*...*/" or the two can collide when mixing SQL within Unix scripts. Hope this is a start. Quincy Leslie
| ||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||
Copyright 2016 - All Rights Reserved | ||||||||||||||||||||||||||||||||||||||||||||||||
Last Modified: 15 Jun 2023 | ||||||||||||||||||||||||||||||||||||||||||||||||