@Richard: I've thought about having one DB user for each APP user. However, a coworker told me that it would infeasible to do that on the web enviroment, specifically for J2EE where a DB connection pool is used, so I gave up on that.
@Jorge: Is this "connection id" you say equivalent to the "applicationid" mentioned in the ibm db2 article? If so, how could I get this data through my application?
On 4/24/07, Marcelo de Moraes Serpa <celoserpa@xxxxxxxxx> wrote:
Thank you for the replies.
@Richard: I've thought about having one DB user for each APP user. However, a coworker told me that it would infeasible to do that on the web enviroment, specifically for J2EE where a DB connection pool is used, so I gave up on that.
@Jorge: Is this "connection id" you say equivalent to the "applicationid" mentioned in the ibm db2 article? If so, how could I get this data through my application?
Marcelo.On 4/24/07, Jorge Godoy <jgodoy@xxxxxxxxx> wrote:"Marcelo de Moraes Serpa" <celoserpa@xxxxxxxxx> writes:
> I forgot to add the link to the article I've mentioned:
>
> http://www-128.ibm.com/developerworks/db2/library/techarticle/0302stolze/0302stolze.html#section2b
>
> This is what I'd like to do on PostgreSQL,
So, translating it to a simpler example:
You want that your function gets the connection ID it is using and
ties it to your current user ID at your application and then have
all your tables use a trigger to retrieve the user name from the
auxiliar table that maps "connection ID -> user", right?
That's what's in that page: a UDF (user defined function) named
getapplicationid() that will return the user login / name / whatever and
triggers.
What is preventing you from writing that? What is your doubt with
regards to how create that feature on your database?
--
Jorge Godoy <jgodoy@xxxxxxxxx >
