Re: Duplicated assignment of slot_name in walsender.c
От | José Luis Tallón |
---|---|
Тема | Re: Duplicated assignment of slot_name in walsender.c |
Дата | |
Msg-id | 5628DB10.1000603@adv-solutions.net обсуждение исходный текст |
Ответ на | Re: Duplicated assignment of slot_name in walsender.c (Alvaro Herrera <alvherre@2ndquadrant.com>) |
Список | pgsql-hackers |
On 10/22/2015 12:36 AM, Alvaro Herrera wrote: > Andres Freund wrote: > >> That seems fairly insignificant. For one this is a rather infrequent and >> expensive operation, for another every decent compiler can optimize >> those away. Note that those duplicate strlen() calls are there in a lot >> of places in walsender.c > diff --git a/src/backend/replication/walsender.c b/src/backend/replication/walsender.c > index c6043cd..5487cc0 100644 > --- a/src/backend/replication/walsender.c > +++ b/src/backend/replication/walsender.c > @@ -762,10 +762,10 @@ logical_read_xlog_page(XLogReaderState *state, XLogRecPtr targetPagePtr, int req > static void > CreateReplicationSlot(CreateReplicationSlotCmd *cmd) > { > - const char *slot_name; > const char *snapshot_name = NULL; > char xpos[MAXFNAMELEN]; > StringInfoData buf; > + int len; Surely "size_t len" ? Or am I missing some platform where size_t is not defined ? Minor nitpicking, of course. But once we are cleaning the code up, might as well change this too.... Thanks! / J.L.
В списке pgsql-hackers по дате отправления: