This is a discussion on Re: Proposed Patchs within the pgsql Interfaces Pgadmin Support forums, part of the PostgreSQL category; --> > -----Original Message----- > From: Thomas Sondag [mailto:thomas.sondag@gmail.com] > Sent: 29 May 2006 10:02 > To: Dave Page > ...
| |||||||
| FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
| ||||
| > -----Original Message----- > From: Thomas Sondag [mailto:thomas.sondag@gmail.com] > Sent: 29 May 2006 10:02 > To: Dave Page > Cc: pgadmin-support@postgresql.org > Subject: Re: [pgadmin-support] Proposed Patchs > > Hum, I miss this option ... sorry, but the main difference with the > current behaviour is for object owning. The main idea was to set > object owner to a group like that : > database foo -> group foo > schema bar -> group bar > schema bar read user -> user toto > > I don't know if that's a good policy, but this case may exist, we may > add an option like "Show Group for object owning" ? Yes, that seems like a much better option - though I wonder if it should default to just groups and then use the existing setting as per the permissions. That would seem more consistent. > The last bug I have is for database creation, I don't know how to get > the current login. Well dlgDatabase will automatically use the parent node's database (ie. The maintenance DB) when appropriate so you should just be able to get the username from that. From dlgDatabase.cpp dlgProperty *pgDatabaseFactory::CreateDialog(frmMain *frame, pgObject *node, pgObject *parent) { dlgDatabase *dlg=new dlgDatabase(this, frame, (pgDatabase*)node); if (dlg && !node) { // use the server's connection to avoid "template1 in use" dlg->connection=parent->GetConnection(); } return dlg; } Regards, Dave. ---------------------------(end of broadcast)--------------------------- TIP 2: Don't 'kill -9' the postmaster |
| ||||
| 2006/5/29, Dave Page <dpage@vale-housing.co.uk>: > > > > -----Original Message----- > > From: Thomas Sondag [mailto:thomas.sondag@gmail.com] > > Sent: 29 May 2006 10:02 > > To: Dave Page > > Cc: pgadmin-support@postgresql.org > > Subject: Re: [pgadmin-support] Proposed Patchs > > > > Hum, I miss this option ... sorry, but the main difference with the > > current behaviour is for object owning. The main idea was to set > > object owner to a group like that : > > database foo -> group foo > > schema bar -> group bar > > schema bar read user -> user toto > > > > I don't know if that's a good policy, but this case may exist, we may > > add an option like "Show Group for object owning" ? > > Yes, that seems like a much better option - though I wonder if it should > default to just groups and then use the existing setting as per the > permissions. That would seem more consistent. > I'm not sure this is the better way, because most of the time objects are owned by the current user, so showing only group by default can be a bit confusing. If you want to see the result with the "Show group ..." option try the folowing patch . the main idea for AddRoles function was to remove one sql query. I test this patch only with postgresql 8.1 and it stop segfaulting now Thomas ---------------------------(end of broadcast)--------------------------- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match |