Re: Add support for specifying tables in pg_createsubscriber.
| От | Peter Smith |
|---|---|
| Тема | Re: Add support for specifying tables in pg_createsubscriber. |
| Дата | |
| Msg-id | CAHut+PuBrust-bZD2dYutDBu7B+-L_Z8u5piMKxMNzEEiFea7Q@mail.gmail.com обсуждение исходный текст |
| Ответ на | Re: Add support for specifying tables in pg_createsubscriber. (Shubham Khanna <khannashubham1197@gmail.com>) |
| Список | pgsql-hackers |
Hi Shubham.
A comment about the v17-0001.
======
1.
+ if (check_publication_exists(conn, dbinfo[i].pubname, dbinfo[i].dbname))
+ {
+ /* Reuse existing publication on publisher. */
+ pg_log_info("dry-run: would use existing publication \"%s\" in
database \"%s\"",
+ dbinfo[i].pubname, dbinfo[i].dbname);
+ dbinfo[i].made_publication = false;
+ }
Is that correct? Won't this code now unconditionally log with the
"dry-run:" prefix, even when the tool is *not* doing a dry-run?
I thought code would be something like:
SUGGESTION #1 (if/else)
/* Reuse existing publication on publisher. */
if (dry_run)
pg_log_info("dry-run: would use existing publication ...);
else
pg_log_info("use existing publication ...);
~~~
OTOH, (since here is just an info message with no destructive
operation) perhaps it would be harmless also to keep the original log
message for both dry-run and normal mode.
SUGGESTION #2 (do nothing)
pg_log_info("use existing publication ...);
======
Kind Regards,
Peter Smith.
Fujitsu Australia
В списке pgsql-hackers по дате отправления: