Re: BUG #4731: compile with -fast fails on ld: duplicate symbol _PGP_S2K_TYPE in pgp.o and pgp-mpi-openssl.o

Поиск
Список
Период
Сортировка
От Tom Lane
Тема Re: BUG #4731: compile with -fast fails on ld: duplicate symbol _PGP_S2K_TYPE in pgp.o and pgp-mpi-openssl.o
Дата
Msg-id 23058.1237989504@sss.pgh.pa.us
обсуждение исходный текст
Ответ на BUG #4731: compile with -fast fails on ld: duplicate symbol _PGP_S2K_TYPE in pgp.o and pgp-mpi-openssl.o  ("Ceriel Jacobs" <cerieljacobs@gmail.com>)
Ответы Re: BUG #4731: compile with -fast fails on ld: duplicate symbol _PGP_S2K_TYPE in pgp.o and pgp-mpi-openssl.o
Список pgsql-bugs
"Ceriel Jacobs" <cerieljacobs@gmail.com> writes:
> ld: duplicate symbol _PGP_S2K_TYPE in pgp.o and pgp-mpi-openssl.o

I think at bottom that's a compiler/linker bug which you should mention
to Apple.  However, I see what is causing it: pgp.h contains

enum
{
    PGP_S2K_SIMPLE = 0,
    PGP_S2K_SALTED = 1,
    PGP_S2K_ISALTED = 3
} PGP_S2K_TYPE;

I think what was meant was

enum PGP_S2K_TYPE
{
    PGP_S2K_SIMPLE = 0,
    PGP_S2K_SALTED = 1,
    PGP_S2K_ISALTED = 3
};

because what the first one is doing is defining a global variable
that is of an anonymous enum type.

Marko, am I right that this is just a syntax mistake, and not
intentional creation of a variable?  There are several occurrences
of the pattern in pgp.h.

            regards, tom lane

В списке pgsql-bugs по дате отправления:

Предыдущее
От: Guillaume Smet
Дата:
Сообщение: Re: BUG #4728: segfault with window function partition involving subquery
Следующее
От: "Kevin Grittner"
Дата:
Сообщение: Re: BUG #4730: Vacuum full verbose analyze "deadlock"