Re: [vile] vile strange behavior in 9.5n

[Thomas Dickey]

On Sat, 13 Jan 2007, Gary Jennejohn wrote:

> OK, without digraph.rc 9.5n's behavior is back to what it was prior to 9.5l.
>
> The behavior was exactly as you describe it - vile seemed to be waiting for
> the next character before deciding what to do with it.
>
> I'd say that not too many people use digraph.rc.

It's an interesting bug.  Here's what happened:

        a) in eval.c, I changed two places that had

                while (isBlank(*src))
                        ++src;

           to

                src = skip_blanks(src);

        b) but skip_blanks() uses isSpace() rather than isBlank().

        c) ^K is treated as a "space" by vile (because of the locale
           code - see the items marked "sp" when doing show-printable).
           A "blank" is exactly space or tab (built-in).

        d) ^K is the first character of the map! command.

        e) so the code skipped over the ^K, and mapped the more-intrusive
           equivalent, e.g., "ss" rather than "^Kss".

On reflection, I don't think I wanted skip_blanks() to skip over ^K.
It's probably using isSpace() in case I ran into a ^M.

Anyway, as you note, not many people seem to use digraph.rc (and not
many would have seen this bug because it's mostly limited to the 
map/map! commands).

--
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net