Re: Facility for detecting insecure object naming - Mailing list pgsql-hackers

From Noah Misch
Subject Re: Facility for detecting insecure object naming
Date
Msg-id 20180816055616.GB3626005@rfd.leadboat.com
Whole thread Raw
In response to Re: Facility for detecting insecure object naming  (Nico Williams <nico@cryptonector.com>)
List pgsql-hackers
On Wed, Aug 15, 2018 at 10:40:55AM -0500, Nico Williams wrote:
> On Tue, Aug 14, 2018 at 11:50:24PM +0000, Nasby, Jim wrote:
> > On Aug 14, 2018, at 4:01 PM, Nico Williams <nico@cryptonector.com> wrote:
> > > 
> > > On Tue, Aug 14, 2018 at 03:00:55PM +0000, Robert Haas wrote:
> > >> The more I think about it, the more I think having a way to set a
> > >> lexically-scoped search path is probably the answer.  [...]
> > > 
> > > Yes please!
> > > 
> > > This is what I want.  Evaluate the search_path at function definition
> > > time, and record code with fully-qualified symbols in the catalog.
> > 
> > Unfortunately, that falls apart for relocatable extensions.
> 
> It would only require recomputing the bindings at relocation time.  IMO
> PG should not allow extension relocation after installation either, but
> sadly it does.

Agreed.  I wouldn't mind deprecating relocatable=true, if it helps.


pgsql-hackers by date:

Previous
From: Andrew Gierth
Date:
Subject: Re: Fix quadratic performance of regexp match/split functions
Next
From: Thomas Munro
Date:
Subject: Re: Index Skip Scan