Re: [Babeldoc-devel] Database Tables

[<ek...@ba...>]

Let me attempt to understand what this implies.

1. Does this mean that in order for me to do an XSL transformation, I
will need 2 stages?  One for SQLEnrich to read the transformation script
from the database and another to perform the XslTransform?

2. Where would I store the XSL Transformation text in the database?  Not
in CONFIG, right?  Yet, it would have to be "aligned" with the specific
Pipeline Name and Pipeline Stage for which it is intended.  So, this
other table would still need to have the same "key", namely CFG_NAME.
In that case, database normalization rules tell me that they should be
in the same table, right?  Otherwise, I would have two tables with the
same key.

Tends to lead me toward including it in CONFIG.  Our "problem" is based
on database implementations (varchar vs. text) ... which we should
really try to abstract ourselves from when writing our Java code.  I am
*very* confident that JDBC 2.0 supports BLOBs and CLOBs (text) just like
every other SQL data type in that you can issue getBlob(), setBlob()
just as you would setInt(), getInt() on a prepared statement (i.e. no
"stupid stuff" like the Oracle BLOB implementation ... which I am sure
is only JDBC 1.0 dependent).  In addition, depending on the database
implementation, even using getString(), setString() is possible as long
as the data length of the underlying driver is not exceeded (which can
be determined by using Statement.getMaxFieldSize()) ... I don't
recommend this, just thought you might find it interesting.

Here's a link for more info:
http://java.sun.com/j2se/1.3/docs/guide/jdbc/spec/jdbc-spec.frame7.html#4149

Erik

<<
Does this meet want you need?



On Friday 28 March 2003 09:56 am, bruce wrote:
> On Friday 28 March 2003 09:49 am, Dejan Krsmanovic wrote:
> > You could store script as document atribute. There is pipeline stage
> > called sqlEnrich wich you can use to get data from database using
any sql
> > query and store the result as document attribute. That way you can
use it
> > as inline script in XslTransform witout changing original babeldoc
db
> > schema
>
> Yes - thats a great idea.  But even so, we will still have to make the
> change to the XslTransformPipelineStage to load the script inline.  I
will
> make this change.
>>