Re: Question about Ctrl-C and less

On Wed, Oct 19, 2005 at 04:24:21AM -0700, Kevin Brown wrote:
> It does?  It looked to me like it was setting the signal handler to
> SIG_IGN before doing the popen(), and resetting it to the internal
> signal handler after doing the pclose().

Oops, you're right. It works because less sets it's own signal handler,
as does more. Sure, the status is inherited but I doubt there's a pager
out there that doesn't set its own signal handlers. Still, best do it
properly.

> If I were to take that approach, I'd do the check immediately after
> writing to the pipe for sure, and possibly after reading from the
> database handle.  I'd have to look at the code to see what's going on
> and thus where the best places to stick the checks are.  But in
> general, you want to put a check anywhere within the loop that is
> immediately after a blocking operation such as a read or write.
> However, that said...

Well, the pager is only run after all the data has been collected from
the server so none of this is an issue while the query is being
processed. Send the cancel request and toss the data. Once we start the
pager just check after each line is printed.

> The semantics for handling SIGINT here should be pretty much the same
> as they are when there's no pager at all (which I know works because I

<snip>

You can't do a pclose in a signal handler, it's not one of the
"reeentrant safe" functions and could lead to deadlocks. The signal
manpage documents the ones you can use. Just set a flag. Setting the
descriptor to NULL is worse because then we have check before every
output function. fprintf(NULL, ...) will segfault on most
architechtures I wager.

Also, the signal handler can't free the memory, there's way too many
arrays. A flag allows the output functions to exit sanely.

> If you're hitting SIGQUIT a lot then perhaps you need to assign it to
> a different key.  :-)

I use SIGQUIT to get out of the situation you get currently when you
press Ctrl-C to interrupt less. Somehow all the "q"s go to psql and all
the Ctrl-Ds to less. SIGQUIT just kills them both from where I can
clean the terminal up. So I will press SIGQUIT less once we fix up psql
to not do stupid things on SIGINT. :)

Have a nice day,
--
Martijn van Oosterhout   <kleptog@svana.org>   http://svana.org/kleptog/
> Patent. n. Genius is 5% inspiration and 95% perspiration. A patent is a
> tool for doing 5% of the work and then sitting around waiting for someone
> else to do the other 95% so you can sue them.