Jack,
yea, interesting....got the same
BUT
i put a get_time call in a package/stored procedure
and got the problem again.
BUT
I'll put my pseudopods away for now and squawk
back to the list when/if i turn something up.
Thanks, everybody, for your cogent thoughts on this.
Ross
-----Original Message-----
From: Jacques Kilchoer
To: Multiple recipients of list ORACLE-L
Sent: 3/16/2001 9:25 PM
Subject: RE: For You Sticky Developers out there.....
>-----Original Message-----
>From: Mohan, Ross [ mailto:[EMAIL PROTECTED]
<mailto:[EMAIL PROTECTED]> ]
>
>get a version-specific glitch ( works in 816, fails in 805 )
>for a call to the DBMS_UTILITY.GET_TIME built-in:
>
>######################
>SQL> select dbms_utility.get_time from dual;
>select dbms_utility.get_time from dual
> *
>ERROR at line 1:
>ORA-06571: Function GET_TIME does not guarantee not to update database
>######################
>
>p.s. a simple assignment, like, for example
> time := DBMS_UTILITY.GET_TIME;
> fails as well.
>p.p.s I know how to use V$TIMER, so i am looking for
> a "why this doesn't
work"
> not a "How to fix it by using V$TIMER" answer.
I don't know what to tell you. I ran the following SQL*Plus statements
in Oracle 7.3.4, 8.0.5 and 8.1.6 without any errors:
SQL> var n number
SQL> begin
2 :n := dbms_utility.get_time ;
3 end ;
4 /
PL/SQL procedure successfully completed.
SQL> print n
N
----------
279430259
