Just to make the review easier.
The point I was missing and that made it hard for me to understand how
endword() works:
If your cursor is already sitting at the end of a word, pressing 'e'
must jump to the end of the next word.
Let's take a word that ends with a UTF-8 ntilde (\xc3\xb1):
++ncursor moves to \xb1:
while (++ncursor < es->linelen) {
uc = es->cbuf[ncursor];
isspace(\xb1) returns false:
if (isspace(uc)) {
if (skip_space)
continue;
else
break;
}
Tracking state is cleared:
skip_space = 0;
The next round of the loop (uc == ' ') breaks here:
if (isspace(uc)) {
if (skip_space)
continue;
else
break;
}
Finally domove() counts the continuation bytes backward, dragging the
cursor to the beginning of ntilde '\xc3':
case 'e':
case 'E':
if (!sub && es->cursor + 1 >= es->linelen)
return -1;
ncursor = (*cmd == 'e' ? endword : Endword)(argcnt);
if (!sub)
while (isu8cont((unsigned char)es->cbuf[--ncursor]))
continue;
break;
You're again where you started. Next time you type 'e' your cursor will
end on that same space. :-)
--
Walter
