Re: Unnecessary pointer-NULL checks in pgp-pgsql.c
От | Robert Haas |
---|---|
Тема | Re: Unnecessary pointer-NULL checks in pgp-pgsql.c |
Дата | |
Msg-id | CA+Tgmoa4a2R1O1CWZ--1mLK2VdL1+2r-7SCikXTwDwR6gy9VUw@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: Unnecessary pointer-NULL checks in pgp-pgsql.c (Michael Paquier <michael.paquier@gmail.com>) |
Ответы |
Re: Unnecessary pointer-NULL checks in pgp-pgsql.c
|
Список | pgsql-hackers |
On Wed, Feb 4, 2015 at 7:00 AM, Michael Paquier <michael.paquier@gmail.com> wrote: > Robert Haas wrote: >> I wrote: >>> src, dst and ctx are created respectively from mbuf_create_from_data, >>> mbuf_create and pgp_init which never return NULL and they are palloc'd >>> all the time. I think that we could simplify things with the patch >>> attached, note that I added an assertion for correctness but I don't >>> really think that it is much necessary. >> >> Yeah, I'd drop the assertion. Also, how about changing things around >> slightly so that we lose the goto-label construct? There's only one >> goto, and its only about 6 lines before the label, so we could just >> flip the sense of the if-test and put the code that gets skipped >> inside the if-block. > Good idea. This gives the patch attached then. Committed. (BTW, why do you not leave a blank line between quoted text and your responses?) -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise PostgreSQL Company
В списке pgsql-hackers по дате отправления: