Subject | Re: [ib-support] views unreliable? |
---|---|
Author | Claudio Valderrama C. |
Post date | 2001-12-07T07:00:30Z |
""Rob Schuff"" <rob@...> wrote in message
news:002501c17e7f$0b0507f0$0401a8c0@picard...
free, at any time the show is running.
company. Just search our bug tracker and you will find at least 10 serious
issues. As it's not enough, Dmitry Kuzmenko logged two more today. The only
serious one I remember I fixed (shown among closed bugs) is that one that
came "only" from IB4: joins of views using indices in the underlying tables
kill the engine at prepare time, no need to run the query. Several of those
bugs come from 1994 and never were fixed.
I use views in my applications. If I happen to run in one of the bugs, I
will find a way to restructure my SQL view so it doesn't fail. It's like
running near the nest of very aggressive bees.
the new "virtual behavior" for views, please take the sources and branch a
new fork of the original product. Questions about the philosophical
foundation for the new behavior can be sent to Jason Wharton. <g> I think he
will copy/paste one of his letters to explain you. You get the activity on
the view, not on the underlying tables.
The bottom line is that the behavior is different than IB6.
C.
--
Claudio Valderrama C. - http://www.cvalde.com - http://www.firebirdSql.org
Independent developer
Owner of the Interbase® WebRing
news:002501c17e7f$0b0507f0$0401a8c0@picard...
> hi folks,We are like a living room as seen in a TV series... everyone can watch for
>
> I have seen some disturbing opinions/observations on the atkin group
> sourceforge.firebird-devel regarding views.
free, at any time the show is running.
> it seems that some people onViews are VERY BUGGY. No joke. And we inherited those bugs from the unnamed
> that group have the opinion that views are far too buggy and should not be
> used. Can someone please comment on this? I have have using views with
> triggers for quite a while and have not found any issues yet.
company. Just search our bug tracker and you will find at least 10 serious
issues. As it's not enough, Dmitry Kuzmenko logged two more today. The only
serious one I remember I fixed (shown among closed bugs) is that one that
came "only" from IB4: joins of views using indices in the underlying tables
kill the engine at prepare time, no need to run the query. Several of those
bugs come from 1994 and never were fixed.
I use views in my applications. If I happen to run in one of the bugs, I
will find a way to restructure my SQL view so it doesn't fail. It's like
running near the nest of very aggressive bees.
> I ran into this wehen I was searching for the rowsaffected problem onviews
> with triggers. I wanted to see if this bug was going to be addressed inIt was fixed according to Ann suggestions. If someone's still not happy with
> RC2, but it does not appear to be the case, which leaves me at v0.94 i
> guess.
the new "virtual behavior" for views, please take the sources and branch a
new fork of the original product. Questions about the philosophical
foundation for the new behavior can be sent to Jason Wharton. <g> I think he
will copy/paste one of his letters to explain you. You get the activity on
the view, not on the underlying tables.
The bottom line is that the behavior is different than IB6.
C.
--
Claudio Valderrama C. - http://www.cvalde.com - http://www.firebirdSql.org
Independent developer
Owner of the Interbase® WebRing