Re: help! any known Pro*C arrayinsert indexing bugs??

I check the data for contraints prior to loading, this check is if the state of the databse changes by the time the data is loaded, I want to recover as efficiently as possible.

I have a related question...
Is there a data size limitation as well as an entry size limitiation? I feel that it was poorly worded in the Pro*C manual.

I am loading 2 VARCHAR arrays and I am doing a to_date() on an integer date array.

Even when I am not using the sqlca.sqlerrd[2] value to reload, I traverse the array doing selects until I find the row that does not exist then I reindex that way. But the insert keeps saying 0 rows inserted, and 0 rows get inserted, even though there is no violation on the current entry.

This has got to either be an Oracle bug, or there is a size limitation as well as an index length limitation(32,767)

Any ideas???

-Dan
dantill_at_nationalfraud.com

In article <3703ECD2.EB7E8B22_at_telesoft.com>,   Scott Dudley <scott_at_telesoft.com> wrote:
>
> Dan,
>
> I've written a large data mining app which uses arrays exclusively for
> insert/updates. It's now more than 2x faster than when I was doing singleton
> transactions. I encountered similar problems and elected to check referential
> constraints as I load the arrays. That way, if something is awry, I can
handle
> it without the insert/update puking.
>
> flaggedatbest_at_yahoo.com wrote:
>
> > Hello.
> >
> > I am using Pro*C version 8.0.5.1.0 and HPUX version 11.0.
> >
> > I have a C program that does array inserts. If the array insert fails
> > because of a constraint error, I reload the array by reindexing
> > the array (I copy the array to a temp array where entry 0 is the
> > entry after the one that Oracle flagged as a violation) I use
> > a 'for' clause to indicate the smaller array.
> >
> > I am using the sqlca.sqlerrd[2] value to determine how many records
> > were successfully loaded.
> >
> > I found that in my testing that sqlca.sqlerrd[2] did not contain the
> > correct index. Consistantly, if in the current batch that I was loading
> > there was a contraint violation, It always seemed to be off by 170 records.
> >
> > Is this a known problem???
> > Any advice appreciated!!
> >
> > -Dan
> > dantill_at_nationalfraud.com
> >
> > -----------== Posted via Deja News, The Discussion Network ==----------
> > http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
>
>

-----------== Posted via Deja News, The Discussion Network ==---------- http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own Received on Fri Apr 02 1999 - 15:40:22 CST