From e9553a0502fc72fbd047e1976714ffb6b37818f1 Mon Sep 17 00:00:00 2001
From: Tim Kuijsten <info+git@netsend.nl>
Date: Fri, 4 Nov 2022 01:56:44 +0100
Subject: [PATCH] don't trip over unescaped UTF-8 keys

Support UTF-8 in the spirit of jsmn being lean and mean. Don't do any validation
or decoding, just make sure unescaped UTF-8 keys are supported.
---
 jsmn.h | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/jsmn.h b/jsmn.h
index 8ac14c1..8172565 100644
--- a/jsmn.h
+++ b/jsmn.h
@@ -137,11 +137,13 @@ static int jsmn_parse_primitive(jsmn_parser *parser, const char *js,
                                 const size_t num_tokens) {
   jsmntok_t *token;
   int start;
+  unsigned char b;
 
   start = parser->pos;
 
   for (; parser->pos < len && js[parser->pos] != '\0'; parser->pos++) {
-    switch (js[parser->pos]) {
+    b = js[parser->pos];
+    switch (b) {
 #ifndef JSMN_STRICT
     /* In strict mode primitive must be followed by "," or "}" or "]" */
     case ':':
@@ -158,10 +160,20 @@ static int jsmn_parse_primitive(jsmn_parser *parser, const char *js,
                    /* to quiet a warning from gcc*/
       break;
     }
-    if (js[parser->pos] < 32 || js[parser->pos] >= 127) {
+
+    if (b < 32) {
       parser->pos = start;
       return JSMN_ERROR_INVAL;
     }
+
+    if (b >= 127 &&
+      (b & (0xc0 | 0x20)) != 0xc0 &&
+      (b & (0xe0 | 0x10)) != 0xe0 &&
+      (b & (0xf0 | 0x08)) != 0xf0 &&
+      (b & (0x80 | 0x40)) != 0x80) {
+        parser->pos = start;
+        return JSMN_ERROR_INVAL;
+    }
   }
 #ifdef JSMN_STRICT
   /* In strict mode primitive must be followed by a comma/object/array */