db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Knut Anders Hatlen <knut.hat...@oracle.com>
Subject Re: [jira] Commented: (DERBY-4772) Data truncation error with XPLAIN-functionality enabled
Date Thu, 19 Aug 2010 22:48:38 GMT
Hi Stephan,

Thanks for the update. Some comments inline.

Stephan van Loendersloot <stephan@republika.nl> writes:

> Hi Knut Anders,
>
> I reported that issue (DERBY-4772) and I've been investigating it for
> a while, but haven't been able to come up with default values for the
> new column sizes, since they seem to vary quite a lot.
>
> Since then I've been using way too large values, just to make things work.
>
> I'll reinvestigate during the weekend to see what reasonable values I
> can come up with. However it would be nice if someone can shed some
> more light on it.

Perhaps it would make sense to use the maximum VARCHAR length (32672)?
The only downside I can think of, is that the query results may be
harder to read in an IJ session because IJ will reserve more room for
each column when the maximum length is increased. But even with the
current column sizes, you cannot expect a result to be easy to read if
it has more than one or two of these columns, so it's probably not much
of an issue.

> Also, it may be worthwile to merge DERBY-4772 and DERBY-4673.

Yes, I think we could close 4772 as a duplicate of 4673.

> If the community and I can come to an agreement on the new column
> sizes, I'll post a patch for both issues after that.

Another option, if it is difficult to pick a good max value, is to
truncate values that exceed the maximum length before attempting to
insert them into the XPLAIN tables. I don't have any strong preference,
and would be happy with the solution either way.

-- 
Knut Anders

Mime
View raw message