Re: Corrections for 20031122

[Thomas Dickey]

On Fri, Nov 28, 2003 at 08:39:26PM +0100, Philippe Blain wrote:
> >From Philippe Blain, Bordeaux, France.
> My computer: P133 - 8,4 Go - 32 Mo Red Hat Linux 7.0
> 
> Corrections for ncurses-5.3-20031122+
...
> * Recursion problems between napms() and _nc_timed_wait() when not
>   having nanosleep().

in lib_twait.c, currently I have

#if PRECISE_GETTIME && HAVE_NANOSLEEP
    /* 
     * If the timeout hasn't expired, and we've gotten no data, 
     * this is probably a system where 'select()' needs to be left 
     * alone so that it can complete.  Make this process sleep, 
     * then come back for more. 
     */
    if (result == 0 && milliseconds > 100) {
        napms(100);             /* FIXME: this won't be right if I recur! */
        milliseconds -= 100;
        goto retry;
    }
#endif

so there is no recursion problem that I see.  (I agree it would be nice to
redesign the whole thing, but not in the near-term, since I'm interested
mainly in tying off loose ends for 5.4).

-- 
Thomas E. Dickey <address@hidden>
http://invisible-island.net
ftp://invisible-island.net