Hi Sylvain,

> Contribution? Why not!!
> Could you explain me what are the tricks about reverse-db and the
> improvements in reverse-db2?

The tricks are mostly about "performance improvements" by using a wildcard
in certain queries, e.g. the list of primary key fields of a table is (if I
remember correctly) not retrieved per table (with the table name in the
parameters of the method call) but for the complete database with a
wildcard. Then a method goes through the result set and sets a "primary key"
flag for the fields. While this works for some databases, it doesn't for
others. The reason I did it this way was that it takes hours to read the
metadata of an oracle database if each of the metadata queries was done with
separately.

reverse-db2 uses a different approach, it is multithreaded and this fixes
most of the performance problems while having better interoperability.

> In reverse-db2, what are the tasks to finish it?

Currently there are two major open items:

 - Complete repository.xml editing
 - Complete reverse engineering so all metadata is read from the database

> Could you easely modify reverse-db to adapt it with MSSQL?
> Where are the sources of these tools?

I am not sure how reverse-db could be adapted to work with MSSQL while
maintaining compatibility with databases it currently works with. There were
patches to make it work with MySQL, I am not sure whether they are
integrated in the repository or not.

You can find all sources in CVS in src/java/apache/org/ojb/tools/mapping. If
you send patches to the list I'll be more than happy to integrate them.

best regards,
Florian

>
> Thank you
> Sylvain
>



--
To unsubscribe, e-mail:   <mailto:ojb-user-unsubscribe@;jakarta.apache.org>
For additional commands, e-mail: <mailto:ojb-user-help@;jakarta.apache.org>

Reply via email to