Reindexing corrupts my database

Steve Albin steve at steve-albin.com
Fri Feb 6 07:49:32 CST 2009


On Friday, February 6, 2009 sunshine at public.kherson.ua (Ruslan Zasukhin) wrote:

>> Do you think that if I removed the object pointers from the table that it
>> would help?  I implemented them, but I have rdb keys also that I'm using.  I
>> realize that object pointers are faster, but with my small database, I don't
>> think it matters.
>
>Why ObjectPtr ?
>Diagnose complains on ObjectPtr?
>I do not see this in first letter.
>
>What field name and type cause
>
>> PAGE (276, 275) Corrupted Header (Values Count). Corrupted Header (Values
>> count < RecID count). At least one value counter is wrong.

Just guessing on ObjectPtr because the error message is complaining about the Recid count and an objectptr is just a recid placed in another table.

Forget it.  I will upgrade and get it over with.  But this really bugs me because reindexing should not be something that has a bug like this.  If I was using triggers or client/server, I could understand upgrading to fix a bug, but reindexing has been around since version 1.  My database is standalone and only uses features introduced in version 2, so upgrading to fix a bug in indexing is a bit irritating.  

The money is the least of the problem.   I now have to change my app to handle the new version and also regression test to find the bugs introduced in version 4.  If I were getting a new feature, I'd be happy to do it.  But to make reindexing work?  Woo-hoo!


Steve




-- 
Steve Albin - Montclair, NJ
http://www.jazzdiscography.com/
http://www.steve-albin.com/


More information about the Valentina mailing list