Fix decompression of unicode values above 2047

jepler · jepler · commit 0eee93729a30 · 2020-09-08T20:54:47.000-05:00
Two problems: The lead byte for 3-byte sequences was wrong, and one
mid-byte was not even filled in due to a missing "++"!

Apparently this was broken ever since the first "Compress as unicode,
not bytes" commit, but I believed I'd "tested" it by running on the
Pinyin translation.

This rendered at least the Korean and Japanese translations completely
illegible, affecting 5.0 and all later releases.
diff --git a/supervisor/shared/translate.c b/supervisor/shared/translate.c
@@ -59,8 +59,8 @@ STATIC int put_utf8(char *buf, int u) {
         *buf   = 0b10000000 | (u & 0b00111111);
         return 2;
     } else { // u <= 0xffff
-        *buf++ = 0b11000000 | (u >> 12);
-        *buf   = 0b10000000 | ((u >> 6) & 0b00111111);
+        *buf++ = 0b11100000 | (u >> 12);
+        *buf++ = 0b10000000 | ((u >> 6) & 0b00111111);
         *buf   = 0b10000000 | (u & 0b00111111);
         return 3;
     }
