Re: SELECT ... FOR UPDATE OF clause documentation implies use of table_names rather than aliases

2023-11-20 Thread David G. Johnston
On Mon, Nov 20, 2023 at 8:16 PM Bruce Momjian wrote: > On Mon, Nov 20, 2023 at 07:19:39PM -0700, David G. Johnston wrote: > > The placement in the numbered listing section feels wrong, I am OK with > > the wording. It should be down in the clause details. > > > > FOR lock_strength [ OF from_refe

Re: SELECT ... FOR UPDATE OF clause documentation implies use of table_names rather than aliases

2023-11-20 Thread Bruce Momjian
On Mon, Nov 20, 2023 at 07:19:39PM -0700, David G. Johnston wrote: > The placement in the numbered listing section feels wrong, I am OK with > the wording.  It should be down in the clause details. > > FOR lock_strength [ OF from_reference [, ...] ] [ NOWAIT | SKIP LOCKED ]  -- > need to change th

Re: SELECT ... FOR UPDATE OF clause documentation implies use of table_names rather than aliases

2023-11-20 Thread David G. Johnston
On Mon, Nov 20, 2023 at 7:04 PM Bruce Momjian wrote: > On Fri, Nov 17, 2023 at 03:44:04PM -0700, David G. Johnston wrote: > > I don't like this particular solution to the stated complaint. When a > FROM > > entry has an alias it must be referenced via that alias anywhere it is > > referenced in

Re: SELECT ... FOR UPDATE OF clause documentation implies use of table_names rather than aliases

2023-11-20 Thread Bruce Momjian
On Fri, Nov 17, 2023 at 03:44:04PM -0700, David G. Johnston wrote: > I don't like this particular solution to the stated complaint.  When a FROM > entry has an alias it must be referenced via that alias anywhere it is > referenced in the query - and indeed it is an error to not write the alias in >

Re: T is a mandatory date time separator in RFC3339 but documentation states differently

2023-11-20 Thread Erik Wienhold
On 2023-11-20 08:14 +0100, Peter Eisentraut wrote: > On 19.11.23 21:34, Erik Wienhold wrote: > > * Making explicit references to ISO 8601:2004 where section numbers are > >referenced. Mostly in source comments but also a couple of places in > >the docs. This is about avoiding confusion as