Re: [HACKERS] New buildin function

2003-01-28 Thread Stephan Szabo
On Fri, 24 Jan 2003, Olleg Samoylov wrote: > On 23 Jan 2003, Rod Taylor wrote: > Nope. CURRENT_USER and SESSION_USER return username. Sometimes need uid, > it's key usesysid in table pg_shadow, for instance, for row based > permissions. Explain in example: > > create table role { >role smalli

[HACKERS] New buildin function

2003-01-28 Thread Olleg Samoylov
On 23 Jan 2003, Rod Taylor wrote: RT>On Wed, 2003-01-22 at 08:09, Olleg Samoylov wrote: RT>> Hi! RT>> RT>> What about adding new function: RT>> pg_uid() RT>> pg_session_uid() RT>> RT>> as reference to internal function GetUserId() and GetSessionUserId(). RT>> RT>> These can help useful, for instan

Re: [HACKERS] New buildin function

2003-01-23 Thread Rod Taylor
On Wed, 2003-01-22 at 08:09, Olleg Samoylov wrote: > Hi! > > What about adding new function: > pg_uid() > pg_session_uid() > > as reference to internal function GetUserId() and GetSessionUserId(). > > These can help useful, for instance in row based securety. Do CURRENT_USER and SESSION_USER no

Re: [HACKERS] New buildin function

2003-01-23 Thread Peter Eisentraut
Olleg Samoylov writes: > What about adding new function: > pg_uid() > pg_session_uid() > > as reference to internal function GetUserId() and GetSessionUserId(). CURRENT_USER, SESSION_USER -- Peter Eisentraut [EMAIL PROTECTED] ---(end of broadcast)

[HACKERS] New buildin function

2003-01-23 Thread Olleg Samoylov
Hi! What about adding new function: pg_uid() pg_session_uid() as reference to internal function GetUserId() and GetSessionUserId(). These can help useful, for instance in row based securety. -- Olleg Samoylov ---(end of broadcast)--- TIP 3: if