Re: pg_replication_origin_xact_reset() and its argument variables

On 2016-06-30 10:14:04 -0400, Tom Lane wrote:
> Fujii Masao <masao(dot)fujii(at)gmail(dot)com> writes:
> > The document explains that pg_replication_origin_xact_reset() doesn't have
> > any argument variables. But it's been actually defined so as to have two
> > argument variables with pg_lsn and timestamptz data types, as follows.
>
> > =# \df pg_replication_origin_xact_reset
> > List of functions
> > Schema | Name | Result data type |
> > Argument data types | Type
> > ------------+----------------------------------+------------------+----------------------------------+--------
> > pg_catalog | pg_replication_origin_xact_reset | void |
> > pg_lsn, timestamp with time zone | normal
>
> > As far as I read the code of the function, those arguments don't seem to
> > be necessary. So I'm afraid that the pg_proc entry for the function might
> > be incorrect and those two arguments should be removed from the definition.
> > Is this analysis right?
>
> Sure looks that way from here. Copy-and-paste from the previous
> line in pg_proc.h, perhaps?

Yes, that's clearly wrong. Damn. Can't fix that for 9.5 anymore. The
function isn't all that important (especially not from SQL), but still,
that's annoying. I'm inclined to just remove the args in 9.6. We could
also add a note to the 9.5 docs, adding that the arguments are there by
error?

Andres