Add support for building on SunOS with gcc #1154
Conversation
| #if SOCI_OS != SOCI_OS_SUN | ||
| // As in SunOS: typedef char int8_t; |
There was a problem hiding this comment.
I don't think it's a property of Solaris per se, we should almost certainly check whether char and int8_t are the same in CMake and define some SOCI_HAVE_INT8_T_TYPEDEF or something like this and test for it here.
| @@ -6,7 +6,7 @@ | |||
| #if defined(__GNUC__) || defined(__clang__) | |||
| #if defined(__LP64__) | |||
| #define SOCI_LONG_IS_64_BIT 1 | |||
| #if SOCI_OS == SOCI_OS_LINUX || SOCI_OS == SOCI_OS_FREE_BSD | |||
| #if SOCI_OS == SOCI_OS_LINUX || SOCI_OS == SOCI_OS_FREE_BSD || SOCI_OS == SOCI_OS_SUN | |||
There was a problem hiding this comment.
Here too we should just check for the condition directly instead of testing for the platform... In fact, I really don't know why did we do it like this in the first place, @zann1x do you remember if there was some good reason for it?
There was a problem hiding this comment.
I don't think there was any other reason for it than to just make it work both locally and in the CI pipeline. In the end, it's only important that we know which data type definitions the underlying OS uses.
I've tested it on SunOS 5.11.3 with gcc (GCC) 13.1.0