Handle too-large code points in generated code

This is simpler than doing so by hand.
QubesOS · Jan 31, 2023 · 039273c · 039273c
1 parent f6e93a8
commit 039273c
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 1 deletion.
diff --git a/gentbl.py b/gentbl.py
@@ -23,5 +23,6 @@ def print_interval(interval, last_cat):
                 print_interval(interval, last_cat)
             interval = [i, i]
     print_interval(interval, last_cat)
+    print('    case 0x10FFFF ... UINT32_MAX: // too large')
 if __name__ == '__main__':
     main()
diff --git a/qrexec-lib/unpack.c b/qrexec-lib/unpack.c
@@ -276,10 +276,11 @@ static int validate_utf8_char(const unsigned char *untrusted_c) {
     case 0xE0020 ... 0xE007F: // category Cf
     case 0xF0000 ... 0xFFFFD: // category Co
     case 0x100000 ... 0x10FFFD: // category Co
+    case 0x110000 ... UINT32_MAX: // too large
         return 0; // Invalid UTF-8 or forbidden codepoint
 
     default:
-        return code_point > 0x10FFFF ? 0 : total_size;
+        return total_size;
     }
 }