Re: [HACKERS][PATCH] adding simple sock check for windows
От | CharSyam |
---|---|
Тема | Re: [HACKERS][PATCH] adding simple sock check for windows |
Дата | |
Msg-id | CAMrLSE7CC1=LHH5r+mHooS44ny=B69=qKMEVONVJt7NM41CzDg@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: [HACKERS][PATCH] adding simple sock check for windows (Amit Kapila <amit.kapila16@gmail.com>) |
Ответы |
Re: [HACKERS][PATCH] adding simple sock check for windows
|
Список | pgsql-hackers |
Hi, Amit, It's good question. I also thought about it. But, I want to leave original code intention. Actually I think there is no case ( slot->sock is not PGINVALID_SOCKET and slot->sock < 0) but if original code want to check (sock < -1) I think it is better to leave that condition. but I think slot->sock == PGINVALID_SOCKET is enough 2018-03-31 14:38 GMT+09:00 Amit Kapila <amit.kapila16@gmail.com>: > On Fri, Mar 30, 2018 at 8:10 PM, CharSyam <charsyam@gmail.com> wrote: >> Hi, I found some missing check for windows int init_slot function in vacuumdb.c >> >> in windows >> SOCKET is unsigned type. so >> >> slot->sock < 0 never can be under 0. >> >> so this patch just check using slot->sock == PGINVALID_SOCKET >> > > - if (slot->sock < 0) > + if (slot->sock == PGINVALID_SOCKET || slot->sock < 0) > > If you are checking for PGINVALID_SOCKET, why do you need the second > part of check (slot->sock < 0)? > > -- > With Regards, > Amit Kapila. > EnterpriseDB: http://www.enterprisedb.com
В списке pgsql-hackers по дате отправления: