Source-Changes-HG archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
[src/trunk]: src/external/mit/lua/dist/src lua: apply upstream bugfix for "'u...
details: https://anonhg.NetBSD.org/src/rev/b577658937c4
branches: trunk
changeset: 374301:b577658937c4
user: nikita <nikita%NetBSD.org@localhost>
date: Mon Apr 17 20:27:40 2023 +0000
description:
lua: apply upstream bugfix for "'utf8.codes' does not raise an error on spurious continuation bytes."
diffstat:
external/mit/lua/dist/src/lutf8lib.c | 28 ++++++++++++++++------------
1 files changed, 16 insertions(+), 12 deletions(-)
diffs (104 lines):
diff -r 789f2fbe8dbe -r b577658937c4 external/mit/lua/dist/src/lutf8lib.c
--- a/external/mit/lua/dist/src/lutf8lib.c Mon Apr 17 20:17:58 2023 +0000
+++ b/external/mit/lua/dist/src/lutf8lib.c Mon Apr 17 20:27:40 2023 +0000
@@ -1,4 +1,4 @@
-/* $NetBSD: lutf8lib.c,v 1.9 2023/04/16 20:46:17 nikita Exp $ */
+/* $NetBSD: lutf8lib.c,v 1.10 2023/04/17 20:27:40 nikita Exp $ */
/*
** Id: lutf8lib.c
@@ -29,6 +29,8 @@
#define MAXUTF 0x7FFFFFFFu
+#define MSGInvalid "invalid UTF-8 code"
+
/*
** Integer type for decoded UTF-8 values; MAXUTF needs 31 bits.
*/
@@ -39,7 +41,8 @@ typedef unsigned long utfint;
#endif
-#define iscont(p) ((*(p) & 0xC0) == 0x80)
+#define iscont(c) (((c) & 0xC0) == 0x80)
+#define iscontp(p) iscont(*(p))
/* from strlib */
@@ -69,7 +72,7 @@ static const char *utf8_decode (const ch
int count = 0; /* to count number of continuation bytes */
for (; c & 0x40; c <<= 1) { /* while it needs continuation bytes... */
unsigned int cc = (unsigned char)s[++count]; /* read next byte */
- if ((cc & 0xC0) != 0x80) /* not a continuation byte? */
+ if (!iscont(cc)) /* not a continuation byte? */
return NULL; /* invalid byte sequence */
res = (res << 6) | (cc & 0x3F); /* add lower 6 bits from cont. byte */
}
@@ -144,7 +147,7 @@ static int codepoint (lua_State *L) {
utfint code;
s = utf8_decode(s, &code, !lax);
if (s == NULL)
- return luaL_error(L, "invalid UTF-8 code");
+ return luaL_error(L, MSGInvalid);
lua_pushinteger(L, code);
n++;
}
@@ -194,16 +197,16 @@ static int byteoffset (lua_State *L) {
"position out of bounds");
if (n == 0) {
/* find beginning of current byte sequence */
- while (posi > 0 && iscont(s + posi)) posi--;
+ while (posi > 0 && iscontp(s + posi)) posi--;
}
else {
- if (iscont(s + posi))
+ if (iscontp(s + posi))
return luaL_error(L, "initial position is a continuation byte");
if (n < 0) {
while (n < 0 && posi > 0) { /* move back */
do { /* find beginning of previous character */
posi--;
- } while (posi > 0 && iscont(s + posi));
+ } while (posi > 0 && iscontp(s + posi));
n++;
}
}
@@ -212,7 +215,7 @@ static int byteoffset (lua_State *L) {
while (n > 0 && posi < (lua_Integer)len) {
do { /* find beginning of next character */
posi++;
- } while (iscont(s + posi)); /* (cannot pass final '\0') */
+ } while (iscontp(s + posi)); /* (cannot pass final '\0') */
n--;
}
}
@@ -230,15 +233,15 @@ static int iter_aux (lua_State *L, int s
const char *s = luaL_checklstring(L, 1, &len);
lua_Unsigned n = (lua_Unsigned)lua_tointeger(L, 2);
if (n < len) {
- while (iscont(s + n)) n++; /* skip continuation bytes */
+ while (iscontp(s + n)) n++; /* skip continuation bytes */
}
if (n >= len) /* (also handles original 'n' being negative) */
return 0; /* no more codepoints */
else {
utfint code;
const char *next = utf8_decode(s + n, &code, strict);
- if (next == NULL)
- return luaL_error(L, "invalid UTF-8 code");
+ if (next == NULL || iscontp(next))
+ return luaL_error(L, MSGInvalid);
lua_pushinteger(L, n + 1);
lua_pushinteger(L, code);
return 2;
@@ -257,7 +260,8 @@ static int iter_auxlax (lua_State *L) {
static int iter_codes (lua_State *L) {
int lax = lua_toboolean(L, 2);
- luaL_checkstring(L, 1);
+ const char *s = luaL_checkstring(L, 1);
+ luaL_argcheck(L, !iscontp(s), 1, MSGInvalid);
lua_pushcfunction(L, lax ? iter_auxlax : iter_auxstrict);
lua_pushvalue(L, 1);
lua_pushinteger(L, 0);
Home |
Main Index |
Thread Index |
Old Index