From maintainers-request at octave dot org Sun Jan 30 16:24:08 2005
Subject: Re: Reprise: external pager quitting kills standard error
From: Daniel J Sebald <daniel dot sebald at ieee dot org>
To: maintainers at octave dot org
Date: Sun, 30 Jan 2005 16:30:33 -0600

Daniel J Sebald wrote:

> Furthermore, the external pager is a stream.  And it wouldn't surprise 
> me that when large amounts of data are put into the stream and the 
> other end quits while there is still stuff to be pulled out, that the 
> stream might think the external pager has died.  (Does "less" pull 
> everything out of the stream at the beginning?  Or only as the cursor 
> is advanced?)


And below is the portion of code that print's out the "warning: broken 
pipe" message.  The "FIXME" message  is a bit suspicious as it has 
something to do with the octave interrupt state, which is what that 
chunk of code I sent last time from pager.cc seems to mess with.

Now, the code below is beyond my skill.  Is my compilation in fact 
including "SIGPIPE"?  I don't know.  Is that chunk of code just not 
being run for some reason?  I don't know.  MAYBE_ACK...?  Don't know.  
MAYBE_REINSTALL...?  Don't know.  Do I know how sig pipes work?  Not really.

Dan






#ifdef SIGPIPE
static RETSIGTYPE
sigpipe_handler (int /* sig */)
{
  MAYBE_ACK_SIGNAL (SIGPIPE);

  MAYBE_REINSTALL_SIGHANDLER (SIGPIPE, sigpipe_handler);

  if (pipe_handler_error_count++ == 0)
    std::cerr << "warning: broken pipe\n";

  // Don't loop forever on account of this.

  // XXX FIXME XXX -- will setting octave_interrupt_state really help
  // here?

  if (pipe_handler_error_count  > 100 && octave_interrupt_state >= 0)
    octave_interrupt_state++;

  SIGHANDLER_RETURN (0);
}
#endif /* defined(SIGPIPE) */