patch 9.0.0070: using utfc_ptr2char_len() when length is negative
Problem: Using utfc_ptr2char_len() when length is negative.
Solution: Check value of length. (closes #10760)
diff --git a/src/screen.c b/src/screen.c
index 6944247..1093b7c 100644
--- a/src/screen.c
+++ b/src/screen.c
@@ -1571,21 +1571,18 @@
// check if this is the first byte of a multibyte
if (has_mbyte)
{
- if (enc_utf8 && len > 0)
- mbyte_blen = utfc_ptr2len_len(ptr, (int)((text + len) - ptr));
- else
- mbyte_blen = (*mb_ptr2len)(ptr);
+ mbyte_blen = enc_utf8 && len > 0
+ ? utfc_ptr2len_len(ptr, (int)((text + len) - ptr))
+ : (*mb_ptr2len)(ptr);
if (enc_dbcs == DBCS_JPNU && c == 0x8e)
mbyte_cells = 1;
else if (enc_dbcs != 0)
mbyte_cells = mbyte_blen;
else // enc_utf8
{
- if (len >= 0)
- u8c = utfc_ptr2char_len(ptr, u8cc,
- (int)((text + len) - ptr));
- else
- u8c = utfc_ptr2char(ptr, u8cc);
+ u8c = len >= 0
+ ? utfc_ptr2char_len(ptr, u8cc, (int)((text + len) - ptr))
+ : utfc_ptr2char(ptr, u8cc);
mbyte_cells = utf_char2cells(u8c);
#ifdef FEAT_ARABIC
if (p_arshape && !p_tbidi && ARABIC_CHAR(u8c))
@@ -1599,8 +1596,10 @@
}
else
{
- nc = utfc_ptr2char_len(ptr + mbyte_blen, pcc,
- (int)((text + len) - ptr - mbyte_blen));
+ nc = len >= 0
+ ? utfc_ptr2char_len(ptr + mbyte_blen, pcc,
+ (int)((text + len) - ptr - mbyte_blen))
+ : utfc_ptr2char(ptr + mbyte_blen, pcc);
nc1 = pcc[0];
}
pc = prev_c;