annotate README.multi-tty @ 53341:d4e6a050c9b1

Bugfix festival. lib-src/emacsclient.c (main_argc, main_argv): New variables. (main): Initialize them. (fail): Use them. (window_change, copy_from_to): Don't kill if emacs_pid is zero. (pty_conversation): Watch the command socket, too. Read emacs_pid here. Emacs and emacsclient could deadlock if Emacs tried to do a reset_sys_modes before sending its pid. lisp/server.el: Automatically delete the client frame when done editing. (server-frames): New variable. (server-process-filter, server-sentinel, server-buffer-done): Use it. (server-process-filter): Do a redisplay before evaluating other parameters. (Prevents "emacsclient -h -e '(delete-frame)'" from messing up the system. src/dispextern.h: Update prototypes. src/dispnew.c (window_change_signal): Do nothing if !term_initted. (init_display): Set the frame size from the tty data after term_init. src/emacs.c (main): Make sure things that init_sys_modes needs are initialized before init_display (which calls init_sys_modes now). (sort_args): Use xfree, not free. (shut_down_emacs) [!EMACS_HAVE_TTY_PGRP]: Use reset_all_sys_modes instead of reset_sys_modes. src/frame.c (make_terminal_frame): Sigh. Move terminal initialization back to the middle of frame setup. Handle errors by making sure that the delete_tty() called from term_init() will see and delete this frame. (Fdelete_frame): Kill the frame before calling delete_tty(). Fix condition for tty deletion. src/keyboard.c (Fset_input_mode): Use reset_sys_modes on the current terminal only. src/lisp.h: Remove duplicate prototypes. src/msdos.c (croak): use reset_all_sys_modes(). src/sysdeps.c (init_baud_rate): Added tty parameter, use it instead of CURTTY. (child_setup_tty): Reset sigio on stdin, not CURTTY(). (reset_sigio): Added fd parameter, put explicit fcntl there. (request_sigio, unrequest_sigio)[FASYNC]: Simply block/unblock the SIGIO signal, don't touch the file params. There are multiple ttys now, and we can't disable the SIGIO from emacsclient. (get_tty_size)[VMS]: Use tty_out instead of CURTTY(). (reset_sys_modes): Don't call cursor_to, clear_end_of_line; call cmgoto and tty_clear_end_of_line instead. The frame may already be dead. Updated reset_sigio call. src/term.c (clear_and_of_line): Separate tty-dependent stuff to tty_clear_end_of_line() for reset_sys_modes. (tty_clear_end_of_line): New function. (term_init): Added frame parameter, don't use selected_frame. Set the frame's output_data.tty value (in case there is an error later). Set the frame size in Wcm, not in the frame. Only free the termcap buffer if there is a termcap-related error. Call init_sys_modes last, not first. (deleting_tty): New variable. (delete_tty): Use it for handling recursive calls. Free deleted tty, except its Wcm (there is still a dangling reference somewhere). git-archimport-id: [e-mail]--2004/emacs--multi-tty--0--patch-19

			-*- coding: utf-8; -*-
GOAL
----

The ultimate goal of this branch is to implement support for opening
multiple, different tty devices and simultaneous X and tty frames from
a single Emacs session.

WHO IS DOING IT
---------------

I'm Károly Lőrentey.  My address: [e-mail].

Patches or suggestions are welcome!

Retrieving the latest version of the branch:

	tla register-archive [e-mail]--2004 http://lorentey.web.elte.hu/arch/2004/
	tla get [e-mail]--2004/emacs--multi-tty--0 <directory>

(I use tla 1.1.)


STATUS
------

Basic multi-tty support is there; there are some rough edges, but it
already seems to be usable.  Emacsclient has been extended to support
opening a new terminal frame.

To try it out, compile the multi-tty branch with the following
commands:

   	mkdir +build
	cd +build
	../configure --with-x-toolkit=no --without-x
	make bootstrap

then start up the emacs server (src/emacs, M-x server-start), and then
(from a shell prompt on another terminal) start emacsclient with

	lib-src/emacsclient -h  /optional/file/names...

You'll hopefully have two fully working, independent frames on
separate terminals.  (This seems to be very useful, emacsclient starts
up even faster than vi!) :-) You can close the newly opened frame and
return to the shell without exiting Emacs by pressing C-x 5 0, i.e.,
delete-frame.  (Note that this does not seem to work yet on the
original terminal.)  Creating new frames on the same tty with C-x 5 2
works exactly as before.  Suspending Emacs is disabled at the moment.
If you exit emacs, all terminals should be restored to their previous
states.

X, Mac, Windows and DOS support is broken, probably doesn't even
compile -- this will be solved later.  Please see the attached README
for other issues, implementation notes and sub-TODO items.

Only tested on my GNU/Linux box.


NEWS
----

For the NEWS file:

** Support for multiple terminal devices has been added.  You can
   specify a terminal device (`tty' parameter) and a terminal type
   (`tty-type' parameter) to `make-terminal-frame'. `tty' must be a
   terminal device created by the updated emacsclient, or there will
   be problems with terminal input and window resizes.  

   You can test for the presence of multiple terminal support by
   testing for the `multi-tty' feature.

** A make-frame-on-tty function has been added to make it easier to
   create frames on new terminals.

** Emacsclient has been extended to support opening a new terminal
   frame.

CHANGELOG
---------

See arch logs.


DIARY OF CHANGES
----------------

(ex-TODO items with explanations.)

-- Introduce a new abstraction for terminal devices.  

   (Done, see struct tty_output.  The abstraction is not yet
   complete.)

-- Change the bootstrap procedure to initialize tty_list.

   (Done, but needs review.)

-- Change make-terminal-frame to support specifying another tty.

   (Done, new frame parameters: `tty' and `tty-type'.)

-- Implement support for reading from multiple terminals.

   (Done, read_avail_input tries to read from each terminal, until one
   succeeds.  MULTIKBOARD is not used.  Secondary terminals don't send
   SIGIO!)

-- other-frame should cycle through the frames on the `current'
   terminal only.  

   (Done, by trivially modifiying next_frame and prev_frame.)

-- Support different terminal sizes.
   
   (Done, no problem.)

-- Make sure terminal resizes are handled gracefully.  (Could be
   problematic.)

   (Done.  We don't get automatic SIGWINCH for additional ttys,
   though.)

-- Extend emacsclient to automatically open a new tty when it connects
   to Emacs.
   
   (Done.  It's an ugly hack, needs more work.)

-- Redisplay must refresh the topmost frame on *all* terminals, not
   just the initial terminal.
   
   (Done, but introduced an ugly redisplay problems.  Ugh.)

-- Fix redisplay problems.

   (Done; it turned out that the entire Wcm structure must be moved
   inside tty_output.  Why didn't I catch this earlier?)

-- Provide a way for emacsclient to tell Emacs that the tty has been
   resized.

   (Done, simply forward the SIGWINCH signal.)

-- Each keypress should automatically select the frame corresponding
   to the terminal that it was coming from.  This means that Emacs
   must know from which terminal the last keyboard event came from.

   (Done, it was quite simple, the input event system already
   supported multiple frames.)

-- Fix SIGIO issue with secondary terminals.

   (Done, emacsclient signals Emacs after writing to the proxy pseudo
   terminal.  Note that this means that multi-tty does not work with
   raw ttys!)

   (Update: This is bullshit.  There is a read_input_waiting function,
   extend that somehow.)

-- Make make-terminal-frame look up the `tty' and `tty-type' frame
   parameters from the currently selected terminal before the global
   default.

   (Done.)

-- Put all cached terminal escape sequences into struct tty_output.
   Currently, they are still stored in global variables, so we don't
   really support multiple terminal types.

   (Done.  It was not fun.)

-- Implement sane error handling after initialization.  (Currently
   emacs exits if you specify a bad terminal type.)  The helpful error
   messages must still be provided when Emacs starts.

   (Done.)

-- Implement terminal deletion, i.e., deleting local frames, closing
   the tty device and restoring its previous state without exiting
   Emacs.

   (Done, but at the moment only called when an error happens during
   initialization.  There is a memory corruption error around this
   somewhere.)

-- Implement automatic deletion of terminals when the last frame on
   that terminal is closed.

   (Done.)

-- Restore tty screen after closing the terminal.

   (Done, we do the same as Emacs 21.2 for all terminals.)

-- 'TERM=dumb src/emacs' does not restore the terminal state.
   
   (Done.)

THINGS TO DO
------------

** Fix mysterious memory corruption error with tty deletion.  To
   trigger it, try the following shell command:

	while true; do TERM=no-such-terminal-definition emacsclient -h; done

   Emacs usually dumps core after a few dozen iterations.  (The bug
   seems to be related to the xfree()ing or bzero()ing of
   tty_output.Wcm or some other tty_output part.  Maybe there are
   outside references to struct Wcm?  Why were these vars collected
   into a struct before multi-tty support?)

   The bug does not seem to happen if the error occurs before terminal
   initialization or if I comment out all xfree()s in delete_frame.
   Update: yes it does, although it is much rarer.  Or maybe it's
   another bug.

** Change emacsclient/server.el to support the -h argument better,
   i.e. automatically close the socket when the frame is closed.

** Export delete_tty to the Lisp environment, for emacsclient.

** C-g should work on secondary terminals.

** Make parts of struct tty_output accessible from Lisp.  The device
   name and the type is sufficient.

** Find out why does Emacs abort when it wants to close its
   controlling tty.  Hint: chan_process[] array.  Hey, maybe
   noninterrupt-IO would work, too?

** Implement support for starting an interactive Emacs session without
   an initial frame.  (The user would connect to it and open frames
   later, with emacsclient.)  Not necessarily a good idea.

** Support raw secondary terminals.  (Note that SIGIO works only on
   the controlling terminal.) Hint: extend read_input_waiting() for
   multiple ttys and hopefully this will be fixed.

** What does interrupt_input do?  I tried to disable it for raw
   secondary tty support, but it does not seem to do anything useful.

** Issue with SIGIO: it needs to be disabled during redisplay.  See if
   fcntl() kernel behaviour could be emulated by emacsclient.

** Move optimalization parameters (costs) from union output_data to
   a backend-neutral per-device structure.

** Find out the best way to support suspending Emacs with multiple
   ttys.

** Deal with SIGHUP in Emacs and in emacsclient.  (After this, the
   server-frames may be removed from server.el.)

** Do tty output through term_hooks, like all other display backends.

** Fix X support.

** Allow simultaneous X and tty frames.

** Fix Mac support (I can't do this myself).

** Fix W32 support (I can't do this myself).

** Fix DOS support (I can't do this myself).



;;; arch-tag: 8da1619e-2e79-41a8-9ac9-a0485daad17d