On Sep 5, 2014, at 11:07 AM, Alan M. Carroll <a...@network-geographics.com> wrote:
> Hello All, > > Sorry for the delays but we have a release candidate for 5.1.0 ready now. > > Changes since 5.0.0: > > https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12310963&version=12324894 > > A summary of the new features are here: I think we might want to respin this, because we broke the builds for OmniOS between 5.0 and 5.1 :-/. I’m still trying to isolate exactly what we did, but the short story is that we should most likely do something like the patch below. I’m trying to git bisect what exactly we did to make us lose this weird typedef, but it’s probably a moot point in that we ought to use the appropriate type instead. I have not committed this yet, but I filed a Jira on it (TS-3066). So, I’m -1 on this release; I’m not an OmniOS user, but I think we should continue supporting it. Cheers, — Leif diff --git a/lib/ts/ink_resolver.h b/lib/ts/ink_resolver.h index 1532c40..b0373ba 100644 --- a/lib/ts/ink_resolver.h +++ b/lib/ts/ink_resolver.h @@ -200,8 +200,8 @@ parse_host_res_preference( #ifndef NS_GET16 #define NS_GET16(s, cp) do { \ const u_char *t_cp = (const u_char *)(cp); \ - (s) = ((u_int16_t)t_cp[0] << 8) \ - | ((u_int16_t)t_cp[1]) \ + (s) = ((uint16_t)t_cp[0] << 8) \ + | ((uint16_t)t_cp[1]) \ ; \ (cp) += NS_INT16SZ; \ } while (0) @@ -221,7 +221,7 @@ parse_host_res_preference( #ifndef NS_PUT16 #define NS_PUT16(s, cp) do { \ - u_int16_t t_s = (u_int16_t)(s); \ + uint16_t t_s = (uint16_t)(s); \ u_char *t_cp = (u_char *)(cp); \ *t_cp++ = t_s >> 8; \ *t_cp = t_s; \ @@ -269,7 +269,7 @@ struct ts_imp_res_state { int _vcsock; /*%< PRIVATE: for res_send VC i/o */ u_int _flags; /*%< PRIVATE: see below */ u_int _pad; /*%< make _u 64 bit aligned */ - u_int16_t _nstimes[INK_MAXNS]; /*%< ms. */ + uint16_t _nstimes[INK_MAXNS]; /*%< ms. */ }; typedef ts_imp_res_state *ink_res_state;
