On Tuesday, September 27, 2022 3:21 PM Alvaro Herrera <alvhe...@alvh.no-ip.org>
wrote:
>
> On 2022-Sep-27, Kyotaro Horiguchi wrote:
>
> > By the way, this is not an issue caused by the proposed patch, I see
> > the following message in the patch.
> >
> > - errdetail("Column list cannot be used
> for a partitioned table when %s is false.",
> > + errdetail("Column list cannot be
> specified for a partitioned
> > +table when %s is false.",
> >
> "publish_via_partition_root")));
> >
> > I think that the purpose of such separation of variable names is to
> > unify multiple messages differing only by the names (to keep
> > translation labor (relatively:p) low). In that sense, that separation
> > here is useless since I see no chance of having the same message with
> > another variable in future.
>
> Well, it also reduces chances for typos and such, so while it's not strictly
> necessary to do it this way, I tend to prefer it on new messages. However, as
> you say it's not very interesting when there's no possibility of duplication,
> so
> changing existing messages to this style when we have no other reason to
> change the message, is not a useful use of time. In this case we're changing
> the message in another way too, so I think it's okay.
Thanks for reviewing!
Just in case I misunderstand, it seems you mean the message style[1] is OK,
right ?
[1]
errdetail("Column list cannot be specified for a partitioned table when %s is
false.",
"publish_via_partition_root")));
Best regards,
Hou zj
