Re: UU: Single-db-call philosophy

["Alexey Parshin" <alexeyp@gmail.com>]

2009/1/8 Ilya A. Volynets-Evenbakh <ilya@total-knowledge.com>

> Err.. Group_find_for_object already exists and is fine as-is, I think.
> And yes, I'm fine with bugging you about creating an extra two-line
> procedure every now and then, if you think it's the right thing to do :)
>
> I guess what I really need is umo_list_acl(umo_version_id, access_type)
> returning recordset of PERSON.


Ok. Do you want any person that has *exactly* or *at least* that access?
Also, since ACL isn't in parameters or results, may be it's
umo_list_persons?
I'm not sure which name sucks more :)

Goes to my TODO list.

>
>
> Alexey Parshin wrote:
> > Remember what I was trying to do in Slonic project - replace all the
> > sequences of SQL stmts with a single call?
> > I'd really prefer to join 'em together.
> >
> > Speaking of these functions. Group_list_people() is understood. But what
> > should group_find_for_object() take as parameters and return?
> >
> > 2009/1/8 Ilya A. Volynets-Evenbakh <ilya@total-knowledge.com>
> >
> >
> >> I've run into another function pair that I need to call:
> >> group_find_for_object and group_list_people (seems like we
> >> don't have that one yet).
> >>
> >> Do we really want to create two-call wrapper functions
> >> for all such pairs, or do we want to delegate combining these
> >> things to C++?
> >>
> >> --
> >> Ilya A. Volynets-Evenbakh
> >> http://www.total-knowledge.com
> >>
> >>
> >>
> >
> >
> >
>
>
> --
> Ilya A. Volynets-Evenbakh
> http://www.total-knowledge.com
>
>


-- 
Alexey Parshin,
http://www.sptk.net