diff options
Diffstat (limited to 'Modules')
| -rw-r--r-- | Modules/unicodedata.c | 10 |
1 files changed, 7 insertions, 3 deletions
diff --git a/Modules/unicodedata.c b/Modules/unicodedata.c index 7a9a964a0f..e8788f5036 100644 --- a/Modules/unicodedata.c +++ b/Modules/unicodedata.c @@ -681,15 +681,19 @@ nfc_nfkc(PyObject *self, PyObject *input, int k) if (LBase <= code && code < (LBase+LCount) && i + 1 < len && VBase <= PyUnicode_READ(kind, data, i+1) && - PyUnicode_READ(kind, data, i+1) <= (VBase+VCount)) { + PyUnicode_READ(kind, data, i+1) < (VBase+VCount)) { + /* check L character is a modern leading consonant (0x1100 ~ 0x1112) + and V character is a modern vowel (0x1161 ~ 0x1175). */ int LIndex, VIndex; LIndex = code - LBase; VIndex = PyUnicode_READ(kind, data, i+1) - VBase; code = SBase + (LIndex*VCount+VIndex)*TCount; i+=2; if (i < len && - TBase <= PyUnicode_READ(kind, data, i) && - PyUnicode_READ(kind, data, i) <= (TBase+TCount)) { + TBase < PyUnicode_READ(kind, data, i) && + PyUnicode_READ(kind, data, i) < (TBase+TCount)) { + /* check T character is a modern trailing consonant + (0x11A8 ~ 0x11C2). */ code += PyUnicode_READ(kind, data, i)-TBase; i++; } |