db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Kristian Waagan <Kristian.Waa...@Sun.COM>
Subject Derby on-disk file format "decoder"
Date Fri, 10 Jul 2009 16:00:30 GMT
Hello,

Does anyone know if a tool to decode Derby conglomerate files exist?
I'm looking for something that will produce output like the header and 
page classes, i.e. showing me the contents of the pages, the number of 
columns in a table, if it is a overflow page, format ids etc.

I'm investigating a bug where Derby apparently tries to read the last 
page in an overflow chain (for a BLOB) as a separate field value. This 
fails, as the stream class is trying to interpret the BLOB content as 
the stream length header. This gets way too big (this is dependent on 
the content of course), and it then tries to read off the page.
I don't know why readExternalFromArray is being used instead of 
readExternal, but my first hunch was to make sure the data is written 
correctly.

Any general debugging tips or pointers for such problems in general are 
also welcome!


Thanks,
-- 
Kristian

Mime
View raw message