|
|
sybperl-l Archive
Up Prev Next
From: "Lee Falkenhagen" <falkenl at hotmail dot com>
Subject: Getting Row Counts using sql() and transaction management
Date: May 14 1998 3:13PM
I am using the sql command to execute several update statements based
upon data I read from a file. The file is a tab-delimited file that I
read in and create a update statement for each row based upon a key that
is defined from the first five values in the file.
Because each row in the file is a separate update statement, I need to
track how many rows I have processed. My idea is to declare a variable
and just increment it as each statement is processed.
My question is this: how does sql determine if an error occurs if you
are batching statements together. In theory, I should be able to use
begin tran and commit tran to make sure it is all or nothing. Do I make
separate sql calls with begin tran/rollback/commit?
Also, sql returns the result set in an array. Is there a way to get
@@rowcount from it. I saw in the document about DBCOUNT, but, that does
not seem to work with sql...
Or should I just use sql for select statements?
Thanks
Lee Falkenhagen
Certified Sybase Professional
falkenl@hotmail.com
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com
|