[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: notimeout() function

From: Thomas Dickey
Subject: Re: notimeout() function
Date: Sun, 05 Mar 2006 15:04:39 -0500
User-agent: Mutt/1.5.11+cvs20060126

On Thu, Feb 23, 2006 at 02:33:07PM -0800, Dan Gookin wrote:
> I'm running some tests on the notimeout() function and haven't noticed
> any real difference between the TRUE and FALSE states. I've turned on
> keypad() for the window, and with keypad() on, the delay is very
> noticeable. But when notimeout(win,TRUE) is set, the same delay
> happens. I believe, according to the documentation, that the TRUE
> state should disable the delay after the Esc key is pressed.

       While  interpreting an input escape sequence, wgetch sets a timer while
       waiting for the next character.  If  notimeout(win,  TRUE)  is  called,
       then  wgetch  does  not  set a timer.  The purpose of the timeout isto
       differentiate between sequences received from a function key and  those
       typed by a user.

It sounds as if it should, but reading the code (this dates back quite a
while), I see that notimeout is only checked near the beginning of
wgetch(), when it is checking if there is data - the first character in
an input escape sequence - it's not going to stop it in the middle of an
escape sequence.

If there is no data in the input buffer (fifo), ncurses sets a timeout
unless notimeout is set.

Furthermore, there are two other variables that it checks before
deciding to do the timeout, set by these calls:

        nodelay(win, TRUE);

Either of those calls will make that initial timeout in wgetch() become
very small, so it returns an ERR.

There are actually 3 places that I see which are doing timeouts in the
code (via a function misleadingly named check_mouse_activity()):

        a) the timeout controlled by notimeout/nodelay/halfdelay.

        b) a timeout within the mouse-click code

        c) the timeout used to terminate an escape sequence (which is
           actually set by ESCDELAY).
> Does the ESCDELAY environment variable affect the delay? Or is the

Yes, that does affect it.

> delay set internally by NCurses? Or is there something else I'm
> missing? Here is the code I'm using to test:
>     keypad(stdscr,TRUE);
>     notimeout(stdscr,FALSE);
>     addstr("Press the Esc key and note the delay:\n");
>     refresh();
>     wgetnstr(stdscr,input,79);
>     notimeout(stdscr,TRUE);
>     addstr("\nNow with notimeout TRUE, press Esc:\n");
>     refresh();
>     wgetnstr(stdscr,input,79);
> This is NCurses 5.4 on OS X 10.4.5

I checked back to 5.0, and it's the same behavior.

Thomas E. Dickey <address@hidden>

Attachment: signature.asc
Description: Digital signature

reply via email to

[Prev in Thread] Current Thread [Next in Thread]