22505430 need workaround for build failure due to new si_id #define
authorDanek Duvall <danek.duvall@oracle.com>
Wed, 06 Jan 2016 13:07:04 -0800
changeset 5238 ac8017f29145
parent 5236 9f6ec47a1612
child 5239 a4bfde1dd226
22505430 need workaround for build failure due to new si_id #define
components/vim/patches/si_id.patch
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/components/vim/patches/si_id.patch	Wed Jan 06 13:07:04 2016 -0800
@@ -0,0 +1,16 @@
+A change to sys/siginfo.h added a #define for si_id that breaks the vim
+build.  This is a nasty hack to undo those effects, and should not go
+upstream.  The right fix may be to get vim to stop using the "si_" prefix,
+since that appears to be "reserved for the implementation", per the
+OpenGroup specs.
+
+--- src/syntax.c	Sat Aug 10 06:00:24 2013
++++ src/syntax.c	Wed Jan  6 10:34:12 2016
+@@ -279,6 +279,7 @@
+  * When si_m_endpos.lnum is 0, the items other than si_idx are unknown.
+  * (The end positions have the column number of the next char)
+  */
++#undef si_id
+ typedef struct state_item
+ {
+     int		si_idx;			/* index of syntax pattern or