db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "David Van Couvering" <da...@vancouvering.com>
Subject Re: Roles for Derby - draft spec uploaded
Date Thu, 20 Sep 2007 16:06:10 GMT
It doesn't seem like a lot of work to add this support:

- Add a column to the user metadata (I'm assuming we have such a
thing) indicating default role

- At connection time, set the role to the default role (can be
overriden later by SET ROLE), or null if there is no default role.

I agree, let's get stuff done incrementally -- let's get the basic
role functionality in first.   Could we make this a "stretch goal?"
If things are getting tight as we approach release date, and you need
some help getting this in, let me know, and I'll see what I can do.

David

On 9/18/07, Knut Anders Hatlen <Knut.Hatlen@sun.com> wrote:
> Dag.Wanvik@Sun.COM (Dag H. Wanvik) writes:
>
> > Daniel John Debrunner <djd@apache.org> writes:
> >
> >> On 9/17/07, Dag H. Wanvik <Dag.Wanvik@sun.com> wrote:
> >>>> As for having a default role on connect, that could be added later.
> >>>> This is implementation defined behavior, according to SQL std.
> >>
> >> Where does the standard say that? I see in section 4.37.2 the sentence:
> >>
> >>   "An SQL-session initially has no SQL-session role name."
> >
> > Ah, yes I knew I had seen that somewhere. I was referring to Annex B
> > "Implementation-defined elements", clause 69 b:
> >
> > 69) Subclause 17.1, <connect statement>:
> >   :
> >   b) The initial value of the current role name is the null value.
>
> So, for those of us who haven't studied the details of those chapters,
> does null/no current role name mean "no current role" or "unknown
> current role"?
>
> --
> Knut Anders
>

Mime
View raw message