Re: mvwprintw

[Thomas Dickey]

On Thu, 30 Mar 2006, Bob Rossi wrote:

> On Thu, Mar 30, 2006 at 01:19:43PM -0500, Bob Rossi wrote:
> > On Thu, Mar 30, 2006 at 01:01:13PM -0500, Thomas Dickey wrote:
> > > On Thu, 30 Mar 2006, Bob Rossi wrote:
> > >
> > > > Can a char string like this:
> > > >
> > > > $7 = 0x810e700 "(tgdb) b main.c:7                   ..."  where the
> > > > string is 81 characters long, and from the '7' up until char 80 they
> > > > are all spaces, and char 81 is the NUL char.
> > > >
> > > > be used with the function like this?
> > > > mvwprintw(win, 13, 0, "%s", buffer);
> > > > where buffer is the string above.

In the original report I didn't understand what "curropt" is.  Can you
clarify that remark?

> > > It looks ok to me - how big is the window?
> >
> > It's 60 rows x 80 columns. This is a frustrating bug to find. I've been
> > looking at it for an hour and have haven't got anywhere.

So the window is 80 columns and presumable also the terminal.  Offhand
I don't know of wrapping problems (except for people who set TERM to 
"xterm" while using PuTTY).  What is TERM and the actual terminal 
emulator?

(of course this is bug-ncurses list - if I had a short test program
for the feature, I could test that and see if I got the same result).

> > I did notice that my application was sending ^G (bell character) on
> > stderr. Would this screw something up?

no (unless the ^G is being echoed as "^" and "G").

> > I can't figure out why the terminal is printing currupt data when it
> > appears that I am passing it a normal string. I ran it through valgrind,
> > and it didn't detect any problems. :(
>
> OK, I downloaded the latest ncurses. The problem still exists. However,
> when I build ncurses with trace and have NCURSES_TRACE set or not,
> everything works fine. This really bothers me.

That sounds like the "still exists" case is picking up the original
shared library (which you can verify with ldd).  It's possible that
the traces make some side-effect, but it's been quite a while since
anyone reported _that_ problem.

> So, I have ncurses with out trace where it fails to write the screen
> properly and I have ncurses_g where it writes to the screen properly.
>
> I don't know if this is my fault or ncurses. Any help?
>
> Thanks,
> Bob Rossi

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