LookupDict2: Fix searching entries longer than 128

This is a fix for https://github.com/nvaccess/nvda/issues/7740.

With the addition of emoji support, dictionary entries can now be
longer than 128 bytes. This fix makes sure the character is
interpreted as an unsigned byte so it does not treat long entries
as having a negative offset.

Treating the offset as a signed byte (like in the previous code)
could cause the hash chain search to loop indefinitely when
processing certain input, like the Tamil characters in the NVDA
issue noted above that is added as a test case to translate.test.

src/libespeak-ng/dictionary.c

@@ -2607,7 +2607,7 @@ static const char *LookupDict2(Translator *tr, const char *word, const char *wor
 	// This corresponds to the last matching entry in the *_list file.
 
 	while (*p != 0) {
-		next = p + p[0];
+		next = p + (p[0] & 0xff);
 
 		if (((p[1] & 0x7f) != wlen) || (memcmp(word, &p[2], wlen & 0x3f) != 0)) {
 			// bit 6 of wlen indicates whether the word has been compressed; so we need to match on this also.

tests/translate.test

@@ -18,6 +18,9 @@ test_phonemes en " h@l'oU" "hello"
 # correct word stress
 test_phonemes en " s'VmTIN Imp'o@t@nt" "something important"
 
+# bug: https://github.com/nvaccess/nvda/issues/7740
+test_phonemes ta " 'il." "ள்"
+
 # bug: https://github.com/nvaccess/nvda/issues/7805
 test_phonemes hi " r'UcI" "रुचि"
 test_phonemes hi " dUk'a:n" "दुकान"