view tests/sigpipe.c @ 5803:b79e1e02bf01

Preserve extent information in the command builder code. src/ChangeLog addition: 2014-07-14 Aidan Kehoe <kehoea@parhasard.net> * event-stream.c: * event-stream.c (mark_command_builder): * event-stream.c (finalize_command_builder): Removed. * event-stream.c (allocate_command_builder): * event-stream.c (free_command_builder): Removed. Use free_normal_lisp_object() instead. * event-stream.c (echo_key_event): * event-stream.c (regenerate_echo_keys_from_this_command_keys): Detach all extents here. * event-stream.c (maybe_echo_keys): * event-stream.c (reset_key_echo): * event-stream.c (execute_help_form): * event-stream.c (Fnext_event): * event-stream.c (command_builder_find_leaf_no_jit_binding): * event-stream.c (command_builder_find_leaf): * event-stream.c (lookup_command_event): * events.h (struct command_builder): Move the command builder's echo_buf to being a Lisp string rather than a malloced Ibyte array. This allows passing through extent information, which was previously dropped. It also simplifies the allocation and release code for the command builder. Rename echo_buf_index to echo_buf_fill_pointer, better reflecting its function. Don't rely on zero-termination (something not particularly compatible with Lisp-level code) when showing a substring of echo_buf that differs from that designated by echo_buf_fill_pointer, keep a separate counter instead and use that. * minibuf.c: * minibuf.c (echo_area_append): Use the new START and END keyword arguments to #'append-message, rather than consing a new string for basically every #'next-event prompt displayed. test/ChangeLog addition: 2014-07-14 Aidan Kehoe <kehoea@parhasard.net> * automated/extent-tests.el: Check that extent information is passed through to the echo area correctly with #'next-event's PROMPT argument. lisp/ChangeLog addition: 2014-07-14 Aidan Kehoe <kehoea@parhasard.net> * simple.el (raw-append-message): Use #'write-sequence in this, take its START and END keyword arguments, so our callers don't have to cons as much. * simple.el (append-message): Pass through START and END here.
author Aidan Kehoe <kehoea@parhasard.net>
date Mon, 14 Jul 2014 13:42:42 +0100
parents 308d34e9f07d
children
line wrap: on
line source

/* code is all from loser.c and loser.el by Mly

Copyright (C) 2002 Richard Mlynarik <mly@pobox.com>

This file is part of XEmacs.

XEmacs is free software: you can redistribute it and/or modify it
under the terms of the GNU General Public License as published by the
Free Software Foundation, either version 3 of the License, or (at your
option) any later version.

XEmacs is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
for more details.

You should have received a copy of the GNU General Public License
along with XEmacs.  If not, see <http://www.gnu.org/licenses/>.
Commentary:

Compile this file.  Run it in the background giving it a command line
argument PORT which is a positive integer 1024 < PORT < 32768 (avoid the
numbers assigned in /etc/services).

Then start up a fresh (you're going to crash) XEmacs.  Execute the following

(defun lose (port)
  (interactive "nUrk: ")
  (require 'comint)
  (while t
    (condition-case e
        (let* ((name "*lose*")
	       (b (get-buffer-create name)))
          (switch-to-buffer b)
          (comint-mode)
          (comint-exec b name (cons "127.0.0.1" port) nil '())
          (process-send-string (get-buffer-process b) "\377\373\001")
          (process-send-string (get-buffer-process b) "\377\373\001"))
      (error (message "URK: %s" e)) (sit-for 1))))

Then M-x lose RET PORT RET and you lose big (in XEmacs 21.1, anyway).
Note: the error messages are proper functioning.  What should eventually
happen after a number of SIGPIPEs is that you get a SIGSEGV and life is
bad and XEmacs is dead.
*/

#include <arpa/inet.h>

int
main (int argc, char **argv)
{
  struct sockaddr_in junk;
  int s;

  memset (&junk, 0, sizeof (junk));

  junk.sin_family = AF_INET;
  junk.sin_addr.s_addr = htonl (INADDR_ANY); /* un*x sucks */
  junk.sin_port = htons (atoi (argv[1])); /* un*x blows */

  s = socket (PF_INET, SOCK_STREAM, 0);

  bind (s, (struct sockaddr *)&junk, sizeof (junk));

  listen (s, 1);
  
  for (;;)
  {
    int loser = accept (s, NULL, 0);
    close (loser);
  }
}