*******************************************************************************
2.0.11 public release
*******************************************************************************

Thu Sep 23 00:10:52 1993  Kevin Layer  (layer@sole)

	* fi-lemacs.el: fix more menu specs

	* fi-indent.el: search's should never error

	* fi-modes.el: look for -*- only in first line of file

Thu Sep 16 22:48:49 1993  Kevin Layer  (layer@sole)

	* Makefile: fi-dist hacking

	* UserGuide.n: just include major version number

	* fi-leep-lemacs.el: backout in progress fix 

	* fi-lep.el: fi::error-if-request-in-progress is outta here, so
	don't call it.

*******************************************************************************
2.0.10 public release
*******************************************************************************

Mon Sep 13 13:49:07 1993  Kevin Layer  (layer@sole)

	* fi-composer.el: add fi:show-calls-to-and-from

	* fi-lemacs.el: finish menus (some undefined functions were being
	called for xref stuff)

Fri Sep 10 10:34:14 1993  Kevin Layer  (layer@sole)

	* fi-lemacs.el: missed fixing one of the menus

*******************************************************************************
2.0.8 public release
*******************************************************************************

Fri Sep 10 08:43:07 1993  Kevin Layer  (layer@sole)

	* fi-site-init.el: lemacs 19.8 needs cl-compat, too.

Thu Sep  9 11:07:53 1993  Kevin Layer  (layer@sole)

	* fi-lemacs.el: 19.8 active-p menu spec change

Tue Sep  7 17:17:52 1993  Kevin Layer  (layer at sole)

	* fi-lemacs.el: more fiddling with the menus; fix bugs in active-p
	menu functions

	* fi-lep.el: only call lep::kill-list-all-completions-session when
	it exists

	* fi-modes.el: remove put of 'screen-name since we'll do it
	differently now

	* fi-subproc.el: fix calling of fi:display-buffer-function tcp
	connection code, too

	* fi-utils.el: fi::switch-to-buffer-new-screen doesn't care about
	which mode anymore.

Thu Sep  2 11:32:57 1993  Kevin Layer  (layer at sole)

	* fi-utils.el:
	* fi-lep.el:
	* fi-dmode.el:
	* fi-changes.el:
	* fi-basic-lep.el:
	  change all `package' local variables to `xpackage', because
	  `package' is often local (due to the file mode line) and
	  set-buffer undoes dynamic bindings of local variables when
	  switching from one buffer to another.  Foo!

	* fi-lemacs.el: just found out that the active-p menu predicates
	run with the locals from the buffer in the screen where the
	menubar resides.  If fi:package is something, like :pclos, that
	doesn't exist in the lisp world, then running the active-p preds
	that do fi:eval-in-lisp causes problems.  So, bind fi:package to
	nil.

Wed Sep  1 16:08:59 1993  Kevin Layer  (layer at sole)

	* fi-lemacs.el: lots of fixes, improvements

	* fi-lze.el: remove fi::error-if-request-in-progress, since it was
	turned off anyway; remove fi:reset-background-request since it was
>>>	never documented or needed; fi:lisp-evals-always-compile initial
	value is `t'; fix docstrings.

	* fi-modes.el: use fi:common-lisp-mode-hook instead of
	fi:lisp-mode-hook to hack mode-line-process; put readtable in
	mode-line-process; lemacs hacks

>>>	* fi-subproc.el: new fi:display-buffer-function for lemacs; new
	variable fi:new-screen-for-common-lisp-buffer.

	* fi-utils.el: fi::switch-to-buffer-new-screen and support

	* examples/emacs.el: fix for new fi-site-init

Tue Aug 31 20:00:00 1993  Kevin Layer  (layer@sole)

>>>	* fi-subproc.el: add fi:readtable defvar

	* fi-basic-lep.el: pass :buffer-readtable to make-session

	* fi-modes.el: note readtable in mode line formatting; revamp -*-
	mode line parsing to be more general (to handle "readtable: ").

Tue Aug 31 16:23:49 1993  Kevin Layer  (layer@sole)

	* fi-site-init.el: fi::emacs-type, which tells which type of emacs
	we are running; install fix for fi::build-time.

	* Makefile: load fi-version.el at byte-compile time

	* fi-keys.el: fi::emacs-type
	* fi-leep.el: fi::emacs-type
	* fi-utils.el: fi::emacs-type

*******************************************************************************
2.0.7, limited test release
*******************************************************************************

Tue Aug 31 12:01:14 1993  Kevin Layer  (layer@sole)

	* fi-changes.el: fix remaining problems with numbered backups.
	The problem was that find-backup-file-name for numbered backups
	returns the "next" backup filename; fix docstrings.

	* fi-gnu.el: new file

	* fi-site-init.el: add load of fi-gnu.elc.

	* fi-utils.el: fsf emacs 19 fix

Thu Aug 12 16:42:02 1993  Kevin Layer  (layer@sole)

	* UserGuide.n: remove fi:restart-common-lisp, since it is complete
	redundant.

>>>	* fi-changes.el: new variable fi:change-definitions-since-default.

	* fi-lemacs.el: remove ref to fi:restart-common-lisp.

	* fi-modes.el: fix for "-*- package: -*-" mode line

	* fi-subproc.el: detect when rsh fails and print something
	intelligent; remove fi:restart-common-lisp

*******************************************************************************
2.0.6, limited test release
*******************************************************************************

Wed Aug 11 10:05:19 1993  Kevin Layer  (layer@sole)

	* fi-keys.el: fix for fsf emacs 19

	* fi-lep.el: fix doc string

Tue Jul 27 13:07:02 1993  Kevin Layer  (layer at sole)

	* clmanaux.c: exit status <0 means `failed'
	* clman.c: exit status 0 means `failed'
	* makeman.c: exit status -1 means `failed'
	* clman.h: extern for `failed'

	* fi-basic-lep.el: open-network-stream -> fi::open-network-stream;
	prin1-to-string -> fi::prin1-to-string

	* fi-clman.el: use minibuffer in `failed' case

	* fi-leep-lemacs.el: start of fix for bug2757

	* fi-subproc.el: open-network-stream -> fi::open-network-stream

	* fi-utils.el: fi::open-network-stream, so we can catch certain
	errors (like /etc/hosts not being readable); fi::prin1-to-string,
	so we can make sure print-escape-newlines is nil.

	* Doc.el: prin1-to-string -> fi::prin1-to-string

Fri Jul 23 00:01:17 1993  Kevin Layer  (layer@sole)

	* Makefile: fix fi-dist rule, add clman-dist; move contents of
	Makefile.fi here

Thu Jul 22 20:45:32 1993  Kevin Layer  (layer at sole)

	* Makefile: new file fi-leep-lemacs.el

	* fi-clman.el: look in exec-path for `clman'

	* fi-leep-lemacs.el: presenting listeners work for lemacs!

	* fi-leep.el: push more stuff on fi::initialization-forms

	* fi-site-init.el: load fi-leep-lemacs.elc, when in lemacs

	* fi-subproc.el: generalize and hooks to fi::subprocess-filter so
	that the fi-leep-lemacs.el can use it.

	* fi-utils.el: add fi::insert-string (used by epoch and lemacs
	leeps)

Thu Jul 22 16:55:18 1993  Kevin Layer  (layer at sole)

	* fi-lemacs.el: more menu hacking

	* fi-lep.el: fix fi::compile-or-load-file to give better
	notification of what it is doing.

Thu Jul 22 16:04:14 1993  Kevin Layer  (layer at sole)

	* Makefile.fi: local rdist rules; new rules for makeman, et al

	* clman.c, clman.h, clmanaux.c, makeman.c

	* fi-clman.el: use new clman.data, if it exists

	* fi-indent.el: &env -> &environment

	* fi-lemacs.el: redesign menus, looks better but still not done

	* fi-utils.el: fi::get-default-symbol looks at new variable,
	fi::use-symbol-at-point, to determine whether the user is asked
	for a symbol--mode-specific menus bind this to `t'.

	* fi-leep.el: push forms onto fi::initialization-forms that need
	to be eval'd at interface startup time

	* fi-site-init.el: new variable, fi::initialization-forms, used to
	store forms during loading that need to be evaluated at interface
	startup time.

Wed Jul 14 16:58:12 1993  Kevin Layer  (layer at sole)

	* Makefile: add fi-emacs19.elc

	* fi-emacs19.el: stub file, for now

>>>	* fi-keys.el: fix require of 'tags; add fi:uncomment-region

	* fi-lemacs.el: lots of additions, still work in-progress

	* fi-lep.el: when m-. is done, tell what is being searched for

	* fi-site-init.el: load only one of fi-lemacs or fi-emacs19

	* fi-subproc.el: set fi::common-lisp-first-time even when
	non-interactive.

	* fi-utils.el: the "get default symbol" function now behaves
	differently when m-. called from a menu.

Tue Jul 13 11:46:44 1993  Kevin Layer  (layer@sole)

	* Makefile: add fi-lemacs.el

>>>	* UserGuide.n: add fi:package, fi:default-package,
	fi:restart-common-lisp

	* fi-basic-lep.el: fix fi::show-some-text-1 per cer (comment added
	to describe the problem)

	* fi-indent.el: install fix for fi:lisp-semicolon;
	fi::make-request indent spec; fix to 4/28/92 fix for aborting the
	application of an indent spec--the old fix made some things indent
	badly, the new fix seems better.

	* fi-lep.el: fi:bug-report is more verbose (include info on eli
	version, etc); when quit is typed and we are doing a completion,
	kill the process we started on the lisp side.

>>>	* fi-modes.el: use fi:default-package instead of hardwired string
	"user".

	* fi-site-init.el: load fi-lemacs conditionally.

	* fi-subproc.el: fix fi:common-lisp-image-name doc string;
>>>	fi:restart-common-lisp; expand-file-name the thing the user gives
	as the image only if there is a $ or ~ in the name; minor bug in
	handling of fi:common-lisp-buffer-name.

	* fi-utils.el: fi::display-pop-up-window-other now doesn't split
	the only window if (current-buffer) is buffer.

	* doc/announce2_0_4.n: amend for accuracy.

	* fi-lemacs.el: work in progress

Thu Jul  1 13:15:57 1993  Kevin Layer  (layer at sole)

	* Makefile.fi: more fixes to the rdist rule

	* fi-clman.el: fix fi::manual-dir init form

Tue Jun 29 16:04:48 1993  Kevin Layer  (layer@sole)

	* Makefile.fi: release rules
	* UserGuide.n: change version number.

	* Doc.el: *.el -> fi-*.el
	* Makefile: *.el -> fi-*.el, can't use lemacs to compile .el
	files, because they can't be loaded into an 18.x emacs.
	* fi-site-init.el: *.el -> fi-*.el
	* fi-site-load.el: *.el -> fi-*.el
	* fi-test.el: *.el -> fi-*.el
	* fi-utils.el: fix test for lemacs

*******************************************************************************
renamed files (*.el -> fi-*.el)
*******************************************************************************

Tue Jun 29 15:00:03 1993  Kevin Layer  (layer at sole)

	* Makefile: use lemacs instead of emacs

	* keys.el: put a condition-case around the require of 'tags,
	fi::mark

	* utils.el, ring.el, lze.el: fi::mark

Fri Jun 18 13:00:20 1993  Kevin Layer  (layer@sole)

	* modes.el: make mode line checking of "package: " more robust

	* keys.el: deal with comment-start being a regexp

	* subproc.el: print process separators on a line by themselves

	* site-init.el:
	* site-load.le: work in progress (to make eli loadable at build
	time)

===============================================================================
public release 2.0.5 (with acl 4.2.beta2.0 on sun4)
===============================================================================

Thu Apr 29 11:52:57 1993  Kevin Layer  (layer@sole)

>>>	* lep.el: new externals: fi:trace-definition and fi:gdb

	* clman.el: fix display of *clman* buffer

Thu Apr 15 15:32:12 1993  Kevin Layer  (layer@sole)

	* clman.el: fix defaulting of symbol in fi:clman

Mon Apr 12 09:36:15 1993  Kevin Layer  (layer@sole)

	* indent.el: add clim 2.x indent methods

===============================================================================
public release 2.0.4
===============================================================================

Fri Apr  9 08:09:23 1993  Kevin Layer  (layer@sole)

	* subproc.el: properly string-match $ for env variabie substitution

Tue Mar 23 10:08:35 1993  Kevin Layer  (layer@ice)

	* Makefile: testit rule

	* UserGuide.n: update

	* modes.el: move fi:in-package-regexp and
	fi:default-in-package-regexp to this file; set comment-column
	only when unbound;
>>>	fi:define-emacs-lisp-mode, default is `t'.

	* indent.el: change defclass indent spec

	* keys.el: comment

	* utils.el: fix fi::explode

	* subproc.el: fix directory tracking for relative pathnames

	* site-init.el: new variable fi::load-subprocess-files (default
	`t').

	* clman.el: totally revamped (uses OBLIST.el now)

===============================================================================
public release 2.0.3 (with ACL 4.1 on the HP700)
===============================================================================

Mon Dec 14 17:30:28 1992  Kevin Layer  (layer@ice)

>>>	* shell.el: put fi:shell-mode-bang under control of variable,
>>>	turned off by default: fi:shell-mode-use-history

	* subproc.el: for the hp700: don't assume `rsh' is there, check
	for `remsh' first.  Also, `localhost' doesn't exist so default to
	(system-name).

	* utils.el: new function fi::command-exists-p

Thu Dec 10 06:01:08 1992  Kevin Layer  (layer@ice)

	* rlogin.el: new function: fi:rlogin-new-user

===============================================================================
public release 2.0.2 (with ACL 4.2 beta under CLIM 2.0 beta)
===============================================================================

Mon Oct  5 08:43:27 1992  Kevin Layer  (layer@ice)

	* clman.el: allow symbols from different packages to be retrieved

Fri Oct  2 09:08:06 1992  Kevin Layer  (layer@ice)

	* leep.el: hack for epoch 4.2

Thu Sep 24 17:58:38 1992  Kevin Layer  (layer@ice)

	* indent.el: handle named-function

Thu Sep 17 10:46:30 1992  Kevin Layer  (layer@ice)

	* indent.el: make auto-file-mode in #||# comments work like
	indented text mode.

	* site-init.el: version 2.0.2

	* keys.el: use only the first 128 entries in keymap (for Epoch
	4.2)

	* changes.el: pass fi:emacs-to-lisp-transaction-directory to
	scm::list-changed-definitions

	* lze.el: pass fi:emacs-to-lisp-transaction-directory to
	evaluation request

Wed Aug 19 07:14:19 1992  Kevin Layer  (layer@ice)

	* changes.el: use find-backup-file-name instead of
	make-backup-file-name.

>>>	* shell.el: fi:shell-mode-bang that expands !$

	* ring.el: added fi:pop-input-last-word (for new shell-mode !
	command)

Tue Aug  4 15:49:10 1992  Kevin Layer  (layer@ice)

	* Doc.el: member-equal -> fi:member-equal

	* utils.el: added fi:member-equal

	* keys.el: use (current-global-map) instead of variable global-map

	* filec.el: replace-match for file completion only if search successful

Thu Jul 23 15:06:32 1992  Kevin Layer  (layer@ice)

	* lze.el: don't echo results in minibuffer if
	fi:echo-evals-from-buffer-in-listener-p is non-nil

Wed Jul 22 11:34:56 1992  Kevin Layer  (layer@ice)

>>>	* filec.el: implement filename abbreviations, where something like
	/foo/bar/baz could expand to /foobar/bartab/bazmaster.

===============================================================================
public release 2.0.1
===============================================================================

Mon Jul 20 10:09:06 1992  Kevin Layer  (layer@ice)

	* subproc.el: make default of
	fi:pop-to-sublisp-buffer-after-lisp-eval be nil

Fri Jul 17 12:18:09 1992  Kevin Layer  (layer@ice)

	* indent.el: without-interrupts indent method fixed

	* clman.el lep.el db.el keys.el: use (goto-char (point-max))
	instead of (end-of-buffer) because the latter causes "Mark set" to
	appear in the minibuffer.

	* lze.el: fi:pop-to-sublisp-buffer-after-lisp-eval (bug2683) was
	being ignored.

Thu May 28 15:29:15 1992  Kevin Layer  (layer@ice)

	* utils.el: move fi::explode here from subproc.el; new
	fi::shell-command-output-to-string

	* subproc.el: use csh's cdpath, not CDPATH env variable; move
	fi::explode to utils.el

Tue May 19 08:31:52 1992  Kevin Layer  (layer@ice)

	* subproc.el: fix bug2601 (fi:common-lisp-buffer-name not handled
	in non-interactive fi:common-lisp); make sure DISPLAY is set
	properly

Tue Apr 28 23:20:59 1992  Kevin Layer  (layer@ice)

	* lze.el: don't lock buffer during eval/compile

Tue Apr 28 13:49:18 1992  Kevin Layer  (layer@ice)

	* indent.el: make fi:lisp-semicolon take a numerical prefix arg so
	it is easy to insert multiple semis; fix egregious bug in
	indenting forms like (cond ((let (x) (foo x)) #|RET HERE|# bar));
	fix indent spec for without-scheduling (0 instead of 1)

>>>	* basic-lep.el: add fi:send-reply

>>>	* subproc.el: handle cdpath variable

	* site-init.el: set version to 2.0.1

	* lze.el: fix bug with where eval/compile background status gets
	wedged

	* stream.el: add `name' keyword to lep::create-listener-stream
	(called from ACL)

	* leep.el: conditionalize for Epoch 4.0

===============================================================================
release 2.0 with ACL 4.1 for Sun.
===============================================================================

Fri Feb 21 13:50:54 1992  Kevin Layer  (layer@ice)

	* subproc.el, basic-lep.el: fix obscure race condition with
	starting interface.

Thu Feb 20 10:19:31 1992  Kevin Layer  (layer@ice)

	* utils.el: fix fi:verify-emacs-support (accept-process-output
	mods no longer needed with 18.58).

Fri Jan 24 18:49:48 1992  Kevin Layer  (layer at ice)

	* utils.el: split fi:error into fi:note and fi:error--fi:note does
	everything but signal the error that the old fi:error used to do.

	* lep.el: scm::signal-transaction-file-error used when :edit, for
	example, can't find the file it expects.

Wed Jan 22 16:39:45 1992  Kevin Layer  (layer at ice)

	* utils.el: fix fi::listify-string so that spaces can be escaped
	(and included in single strings)

Thu Jan 16 14:28:54 1992  Kevin Layer  (layer at ice)

	* clman.el: make fi::setup-default-clman-package-info just grab
	the dirs in fi/manual instead of hardwiring them in... 

Thu Jan 16 11:08:34 1992  Kevin Layer  (layer at ice)

	* clman.el: split fi:clman-apropos into two functions (so some
	consistency checking code can use the internal version)

	* keys.el: fix docstring

	* lze.el: unused var

Wed Jan 15 13:28:51 1992  Kevin Layer  (layer at ice)

	* db.el: fix help text

Mon Jan 13 22:29:55 1992  Kevin Layer  (layer at ice)

	* keys.el: added fi:lisp-mode-auto-indent

Sun Jan 12 11:35:46 1992  Kevin Layer  (layer at ice)

	* su.el: make sure root password doesn't hang around

	* clman.el: add defsys, prof and stream.  misc cleanup.

	* subproc.el: fix some docstrings.

	* keys.el: [rfe1658] add new binding: C-cl (fi:toggle-to-lisp)

	* keys.el, lze.el: [rfe1659] [rfe1657] change fi:lisp-eval-* to
	fi:lisp-eval-or-compile-*.  New functions
	fi:lisp-{compile,eval}-{defun,region,current-buffer,last-sexp}.
	New variable fi:lisp-evals-always-compile.

	* lep.el: [bug2444] fix bug in offering matches

Thu Dec 19 19:00:34 1991  Kevin Layer  (layer at ice)

	* utils.el: if fi::abbrev-to-symbol doesn't find a match, then
	return the original pattern.

	* lep.el: hack completion to better work with upper case

Thu Dec 12 14:21:35 1991  Kevin Layer  (layer at ice)

	* clman.el: [bug2380] and [spr4707] (misc weirdness)

Tue Dec 10 18:34:38 1991  Kevin Layer  (layer at ice)

	* db.el: fix help text

Mon Dec  9 15:53:15 1991  Kevin Layer  (layer at ice)

	* lze.el: [bug2376] fix docstrings

	* db.el: [rfe1638] and [bug2377]: use same mechanism for saving
	window conf as C-c SPC in other modes

Thu Dec  5 16:51:56 1991  Kevin Layer  (layer at ice)

	* leep.el: fix presentation bug whereby &'s get into buffer and
	presentations are turned off thereafter

Wed Nov 27 17:32:44 1991  Kevin Layer  (layer at ice)

	* subproc.el: package qualify s-e-l-i; expand-file-name of
	image-file.

	* lep.el: fix undefine form

Mon Nov 25 10:58:56 1991  Kevin Layer  (layer at ice)

	* clman.el: cltr1 -> cltl1, numbers in packages

Thu Nov 21 13:15:21 1991  Kevin Layer  (layer at ice)

	* subproc.el: [rfe1622] fix defaulting for image-name

	* indent.el: fix loop method (again!)

	* utils.el: get-symbol-at-point must do save-excursion

>>>	* keys.el: put ; binding on lisp-mode maps

Mon Nov 11 14:41:49 1991  Kevin Layer  (layer at ice)

	* basic-lep.el: beef up error handling in *background-interaction*
	buffers (in case user does a :reset).

	* lze.el: fix buffer handling in filter

	* leep.el: fix C-cC-c in presentation buffers

Fri Nov  1 00:11:29 1991  Kevin Layer  (layer at ice)

	* utils.el: fix fi::get-default-symbol

	* lep.el: "Complete, but not unique." message was bogus

Tue Oct 29 14:56:34 1991  Kevin Layer  (layer at ice)

	* sublisp.el: fi::sublisp-select: check
	fi::common-lisp-backdoor-main-process-name before signalling an
	error.

	* utils.el: fi::setup-shell-buffer-for-common-lisp-interaction
	used by make-dist.

	* subproc.el: support for
	fi::setup-shell-buffer-for-common-lisp-interaction

>>>	* db.el: next-line binding (d) is now fi:ss-next-frame;
	previous-line binding (u) is now fi:ss-previous-frame.

	* dmode.el: fix `q' for both modes

Mon Oct 21 20:01:37 1991  Kevin Layer  (layer at ice)

	* indent.el: fix `do' method

	* subproc.el: don't start interface for anything but the first cl

	* composer.el: fix docstrings

	* dmode.el: use one, buffer local variable to hold the window
	configuration because `q' in inverse and normal mode expect it

Tue Oct 15 11:31:39 1991  Kevin Layer  (layer at ice)

	* indent.el: indent methods were being ignored in some cases

	* basic-lep.el: abstract the fi:pop-up-temp-window-behavior and
	move it to utils.el, so it can be used by definition modes

	* utils.el: fi::display-pop-up-window, and friends, now used by
	things that pop up *CL-temp* and definition mode buffers.

	* Makefile: make tags depends on TAGS and TAGS depend on elcs

	* dmode.el: don't bind M-. in definition mode (this was resolved a
	long time ago in `tech' discussions, but was never implemented);
	fix some doc strings; use new fi::display-pop-up-window

>>>	* lep.el: new fi:list-unused-functions

Thu Oct 10 11:35:10 1991  Kevin Layer  (layer at ice)

	* subproc.el, utils.el: if process-environment isn't boundp, then
	use setenv instead

	* lze.el: make compiling/evaluating message in mode line stand out
	more.

>>>	* basic-lep.el: new function: fi:eval-in-lisp-asynchronous

	* keys.el: have fi::tcp-simulate-special-char use
	fi:eval-in-lisp-asynchronous so things like C-cC-c in a lisp
	listener are not synchronous.

Thu Oct  3 12:41:28 1991  Kevin Layer  (layer at ice)

	* lep.el: fix to lep::completing-read-complete from cer

	* basic-lep.el: bind fi:lisp-evalserver-timeout to 1000 in
	fi::wait-for-reply-to-come-back.  The user can type ^G if they get
	tired of waiting.

	* utils.el: fi::get-default-symbol takes another optional:
	ignore-keywords.

	* subproc.el: fix fi:common-lisp to use
	fi:common-lisp-buffer-name; also, always read the buffer name, if
	reading other things.

	* lep.el: specify explicit third argument to
	fi::get-default-symbol; ignore-keywords hack.

Tue Oct  1 13:36:43 1991  Kevin Layer  (layer at ice)

	* keys.el: move fi::get-default-symbol and friends to utils.el.

	* utils.el: hack fi::get-default-symbol to "do the right thing".
	Check for funcall, apply and handle #' and (function ...) first
	args to those functions.

	* lep.el: fix [bug2197] "scm::metadot-session editor session N"
	processes don't go away when they can; list and edit who calls get
	default at point.

Sun Sep 29 17:52:37 1991  Kevin Layer  (layer at ice)

	* subproc.el: fix command line arg starting of emacs-lisp interface
	for localhost

Fri Sep 27 00:54:49 1991  Kevin Layer  (layer at ice)

	* modes.el: make the syntax of dot in lisp-mode be same as other
	symbol chars.

	* basic-lep.el: defconst's -> defvar's

	* utils.el: fi:map-lines

	* subproc.el: start-emacs-lisp-interface is external now;
	deinstall the common-lisp filter when the rendezvous is made with
	lisp.

	* keys.el: doc string hacking

	* db.el: make fi::ss-help look at the keymap

	* dmode.el: make sure definitions don't span lines

>>>	* lze.el: annotate mode-line with "Compiling" or "Evaluating"

Tue Sep 17 12:39:13 1991  Kevin Layer  (layer at ice)

	* keys.el: fix fi:find-unbalanced-parenthesis yet again

Mon Sep 16 14:42:13 1991  Kevin Layer  (layer at ice)

	* basic-lep.el: fix show-some-text so that text isn't lost;
	fi::note-background-reply in fi:show-some-text.

>>>	* subproc.el: fi:start-lisp-interface-function is history.
>>>	fi:start-lisp-interface-arguments is `in'.  New variable
>>>	fi:use-background-streams.
>>>	fi:start-backdoor-interface is history.
>>>
>>>	emacs-lisp interface started via command line args (-e) instead of
>>>	stuffing forms into buffer.
>>>
>>>	fi:common-lisp doesn't change fi:common-lisp- variables unless run
>>>	interactively.

>>>	* keys.el: C-ck is fi:kill-definition in lisp modes.

>>>	* lze.el: background eval/compile updates mode line with
>>>	"Compiling" or "Evaling" (uses minor-mode-alist).

Thu Sep 12 11:13:13 1991  Kevin Layer  (layer at ice)

	* Makefile: add new files
  from cxh:
	* clman.el: directory fixes
	* clman-generate.el: new file
	* clman-mode.el: new file
	* clman-preview.el: new file

Wed Sep 11 15:18:11 1991  Kevin Layer  (layer at ice)

	* su.el: fix su-filter messing with the current buffer

	* basic-lep.el: fix error message when lep can't start.  Don't
	create *LEP...* buffers by default.

	* utils.el: make fi:error signal an error

	* subproc.el: fi:goto-common-lisp removed (was never documented).
	Hacks in cl filter so errors starting the interface can be
	reported.

	* site-init.el: fi::required-ipc-version

	* keys.el: make fi:lisp-delete-pop-up-window work in epoch, too

	* examples/emacs.el: example binding of fi:common-lisp

Tue Sep 10 10:44:56 1991  Kevin Layer  (layer at ice)

	* indent.el: also indent ^def forms (like defun).  Fix
	define-condition (the indent for this seems to have changed
	recently, but the indent method in this file hasn't).

>>>	* basic-lep.el: new variable: fi:pop-up-temp-window-behavior
	move fi:show-some-text el al here from lep.el.

	* UserGuide.n: add fi:pop-up-temp-window-behavior, fi:raw-mode*

	* subproc.el: abstract the commands used to start remote lisps so
	things other than `rsh' can be used.

	* site-init.el: inc versions number

>>>	* keys.el: support for raw mode.  Two new variables: fi:raw-mode
	and fi:raw-mode-echo.  Fix fi:find-unbalanced-parenthesis.

	* rlogin.el: fix where args are consed.

	* lep.el: code moved to basic-lep.el.  Fix calls to
	fi::get-default-symbol to specify how default is parsed.
	fi::reset-metadot-session, and fix problems with new metadot
	stack.

	* examples/emacs.el: new file

Fri Aug 23 08:58:17 1991  Kevin Layer  (layer at ice)

	* utils.el: beef up fi:verify-emacs-support
	* subproc.el: run fi:verify-emacs-support when the lep is started

Thu Aug 22 20:18:11 1991  Kevin Layer  (layer at ice)

	* indent.el: fix problems with ;'s.  Do automagic indent for with-
	and do- forms.

	* basic-lep.el: bury the *lep...* buffer.  Give error for some
	commands if connection not open.  Make fi:eval-in-lisp wait for
	the interface to startup.

	* clman.el: fix skip-chars-* calls (arg is not a regexp).

	* subproc.el: hacks to make interrupting of listeners work better
	(we don't depend on the process name in lisp anymore...)

	* keys.el: fix functions that send pseudo chars to listener.
>>>	fi:tcp-lisp-listener-interrupt-process renamed
	fi:interrupt-listener.

>>>	Also, made C-cC-c behave the same in all common lisp listener
	buffers (subprocess and otherwise): interrupts the single process
	tied to the buffer.  Give C-cC-c a prefix argument and it does the
	"select process" thang, that used to happen only in *common-lisp*,
	but not in *listener*.

	Call fi:indent-sexp instead of indent-sexp, when appropriate.  Fix
	for fi:find-unbalanced-parenthesis.

	* changes.el: fi'ify
	* composer.el: fi'ify
	* db.el: fi'ify
	* dmode.el: fi'ify

>>>	* lep.el: new variable: fi:maintain-definition-stack (see doc
	string for more info)

	* lze.el: for eval-buffer, don't send package over, since it may
	not exist.

	* stream.el: fix for interrupting correct process.

	* leep.el: fix for interrupting correct process.

Tue Jul 30 20:49:18 1991  Kevin Layer  (layer at ice)

	* subproc.el: fi:common-lisp was not using given values of
	interactive arguments, if they were `nil'.

	* indent.el: fix fi:lisp-indent-line for ;'s.

>>>	* keys.el: fi:superkey-shadow-universal-argument (initial value
	`t') control whether C-u superkey is made.

Wed Jul 24 14:00:44 1991  Kevin Layer  (layer at ice)

	* sublisp.el, clman.el, keys.el, db.el, lep.el: raise minibuffer
	before doing completing-read

	* subproc.el: fix tracking of in-package in subprocess buffers

===============================================================================
2.0beta release made.
===============================================================================

Wed Jul 10 22:25:38 1991  Kevin Layer  (layer at ice)

	* Makefile, site-init.el: new files leep0.el and leep.el

	* leep.el, leep0.el: the epoch side of presentations in a
	lisp-listener window

Sun Jul  7 21:16:09 1991  Kevin Layer  (layer at ice)

	* indent.el: indent property for define-compiler-macro

Tue Jul  2 10:06:51 1991  Kevin Layer  (layer at ice)

	* subproc.el: try and work around a bug whereby the form sent to
	cl is doesn't make it there

	* keys.el: tweak fi:find-unbalanced-parenthesis 

Fri Jun 28 09:57:33 1991  Kevin Layer  (layer at ice)

	* subproc.el: fix message about image-args

Thu Jun 27 16:06:44 1991  Kevin Layer  (layer at ice)

	* indent.el: if fi:lisp-comment-indent-specification specifies
	`t', then really indent like an sexp.

Thu Jun 27 15:20:15 1991  Kevin Layer  (layer at ice)

	* indent.el: make fi:lisp-comment-indent-specification
	initialization really come from the defvar in this file.  Handle
	comment-spec of a negative integer.  Don't setf car of
	fi:lisp-comment-indent-specification when it is not a number.

	* modes.el: don't set fi:lisp-comment-indent-specification at mode
	init time, because it wipes out the initial value given by the
	user

	* clman.el: don't include "graph/" in initial list

	* subproc.el: fix cl process startup message; fix out buffer is
	found from the process name.

	* Makefile: make a tags file for this directory.

>>>	* keys.el: New binding in lisp editing modes

		C-c^	       fi:center-defun

	fix fi:find-unbalanced-parenthesis.

	* dmode.el: do fi:lisp-push-window-configuration before poping up
	buffers

	* lep.el: fix typo

	* stream.el: *compilation-output* -> *background-interaction*

Thu Jun 20 15:22:03 1991  Kevin Layer  (layer at ice)

>>>	* subproc.el: added new function: fi:goto-common-lisp.

	* lep.el: fix scm::make-and-initialize-metadot-session wrt to
	messages

===============================================================================
======== changes given to doc group and folded into beta documentation ========
===============================================================================

Thu Jun 20 10:26:14 1991  Kevin Layer  (layer at ice)

	* modes.el: move write-file-hooks hacking to keys.el

>>>	* keys.el: new variable fi:check-unbalanced-parentheses-when-saving

Wed Jun 19 21:59:06 1991  Kevin Layer  (layer at ice)

	* indent.el: make sure fi:lisp-comment-indent-specification is
	updated when comment-column is changed.

	* rlogin.el, shell.el, su.el, telnet.el: fi::mkae-subprocess
	addition argument (nil, in this case).

>>>	* modes.el: write-file-hooks runs fi:find-unbalanced-parenthesis

	* basic-lep.el: remove lep::create-session-for-lisp

>>>	* utils.el: fi:verify-emacs-support

>>>	* subproc.el: fi:source-info-not-found-hook is gone (and so is
	fi:find-tag-somewhere which was the initial value).  Get the
	expansion of directories right for fi:common-lisp.  Added startup
	message in fi::make-subprocess so users can be informed as to what
	is being run.

>>>	* keys.el: fi:lisp-find-tag -> fi:lisp-find-definition and
	fi:lisp-tags-loop-continue -> fi:lisp-find-next-definition.
	Furthermore, fi:lisp-find-definition is on ``C-c .'' and
	fi:lisp-find-next-definition is on ``C-c ,''.  ``C-c 4 .'' is
	bound to fi:lisp-find-definition-other-window.

	Implement abbreviations (for example, f-d-c-p expands to
	fast-digit-char-p).

	Previously ``C-c .'' was bound to
	fi:lisp-sync-current-working-directory in *common-lisp* buffers.
	It is now ``C-c =''.

	``C-c SPC'' restores the window configuration as it was before the
	temporary window was made visible.

	M-. is locked for first-time use, displaying a message about the
	new changes.  New variable fi:find-tag-lock.

	Make the second argument to interrupt-process, kill-process and
	quit-process depend on the mode--in rlogin, shell and telnet modes
	it is `t', otherwise `nil'.

	* lep.el: fi:lisp-find-definition support.  Make sure
	fi:lisp-find-definition and fi:lisp-find-next-definition print
	correct message in minibuffer.  Support for abbreviation
	expansion.

	* dmode.el, composer.el: fi:lisp-find-tag ->
	fi:lisp-find-definition.

	* lze.el: remove debugging code

	* stream.el: rename buffer in which compilation warnings appear

Tue May 28 16:10:41 1991  Kevin Layer  (layer at ice)

	* Doc.el: set fi:lisp-do-indentation to `t' before loading
	site-init

	* indent.el: add with-package-iterator and
	with-hash-table-iterator methods

>>>	* modes.el: define lisp-mode and common-lisp-mode, so that we
	don't need to redefine set-auto-mode

	* basic-lep.el: fi'ify.  Commented out
	lep::create-session-for-lisp since it couldn't work

	* utils.el: move functions to this file that are of general
	internal use

	* clman.el: hacking with cxh

	* subproc.el, changes.el, dmode.el, lep.el, lze.el, stream.el: fi'ify

	* Makefile: load utils.el during compile, now

>>>	* db.el: `export' fi:scan-stack-mode-display-help, and fi'ify;
>>>	lep:scan-stack-mode-hook renamed fi:scan-stack-mode-hook

Fri Apr 26 13:01:09 1991  Kevin Layer  (layer at ice)

	* lep.el: fix compile-or-load-file make-request

===============================================================================
==================== Final ALPHA DOC GIVEN TO DOCUMENTATION GROUP =============
===============================================================================

===============================================================================
==================== 2.0alpha release =========================================
===============================================================================

Fri Apr 26 09:32:56 1991  Kevin Layer  (layer at ice)

	* subproc.el: grok inspector prompts

Wed Apr 24 10:04:48 1991  Kevin Layer  (layer at ice)

>>>	* composer.el: browse -> inspect; doc strings

	* lep.el: fix lep::prompt-for-values

Tue Apr 23 21:05:09 1991  Kevin Layer  (layer at ice)

	* clman.el: cxh/layer hacking

===============================================================================
==================== Updated ALPHA DOC GIVEN TO DOCUMENTATION GROUP ===========
===============================================================================

Tue Apr 23 15:45:42 1991  Kevin Layer  (layer at ice)

>>>	* subproc.el: fi:emacs-to-lisp-transaction-directory is back!

	* db.el: lep: -> fi:

	* dmode.el: lep::dmode-undo -> fi:definition-mode-undo; doc string
	additions

	* lep.el: doc string hacking

Mon Apr 22 23:24:00 1991  Kevin Layer  (layer at ice)

	* db.el: give unique names to scan stack buffers

	* lep.el: make bug-report prompt for process on which to zoom

Mon Apr 22 14:22:00 1991  Kevin Layer  (layer at ice)

	* dmode.el: print message when finding a definition

	* clman.el: new fi::setup-default-clman-package-info from cxh

	* lep.el, dmode.el: finish support for fi:list-undefined-functions
	(inverse-definitions)

	* subproc.el: fi'ify find-tag-somewhere

	* Makefile: `lep' directory is outta here (files moved to .)

Mon Apr 22 13:37:33 1991  Kevin Layer  (layer at ice)

	* Doc.el: fi:use-lep is outta here

	* utils.el: fi::frob-case-*-lisp moved here from tcplisp.el (which
	is outta here)

	* subproc.el: remove non-lep conditionalization

	* Makefile: nonlep/ is outta here

	* site-init.el: nonlep code is outta here

	* keys.el: backdoor-eval replaced with eval-in-lisp

	* lep/basic-lep.el: fi:eval-in-lisp works again

	* lep/lep.el: lep::eval-from-lisp used by the Lisp function
	lep::eval-in-emacs.

	* lep/db.el: minor

Wed Apr 17 16:09:44 1991  Kevin Layer  (layer at ice)

>>>	* subproc.el: changed fi:source-info-not-found-hook to be a
	function of two arguments: tag and other-window-p.  Fix prompt
	to compatible with default prompts in all lisps.

	* Makefile: lep/stream.el is back!

	* site-init.el: fi:use-lep is on by default

>>>	* keys.el: changed fi::get-symbol-at-point to support symbol
	completion.

>>>	* lep/basic-lep.el: new function fi:reset-lep-connection.
	Support for completing read.

	* lep/composer.el: epoch support; fi:'ed the functions
	show-subclasses & show-superclasses

>>>	* lep/dmode.el: `t' in definition-mode toggles tracing for that
	definition; "C-n" binding is outta here.  Added epoch mouse
	support and undo to revert buffer back to what it was
	previously. Made it check for dmode being the only window and
	to not shrink it - epoch death.

>>>	inverse-definition-mode stuff for displaying output of
...	fi:list-undefined-functions (see below)

	* lep/lep-init.el: load stream.elc again

>>>	* lep/lep.el: hacked the completion code into separate
	functions, and added lep::completing-read that is an interface
	to completing-read.  Deleted lep::create-listener-stream
	duplicate.  Added lep::make-request-in-session-and-wait does
	what it says.

>>>	New function fi:compile-file.

>>>	New function fi:list-undefined-functions.

	* lep/stream.el: modified lep::create-listener-streams to take
	args that specify the xparent etc.

>>>	* lep/changes.el: Made a prefix arg of 4 do a compute changes
...	relative to a ",0" file

Wed Apr  3 08:53:50 1991  Kevin Layer  (layer at vapor)

	* subproc.el: special case interace startup when require'ing lep
	and scm

Fri Mar 29 11:41:02 1991  Kevin Layer  (layer at vapor)

	* clman.el: new one from teddy.  clman.data is outta here!

Thu Mar 28 12:50:49 1991  Kevin Layer  (layer at vapor)

   *** reorganize code in many files, remove dead code, move nonlep code
   *** to nonlep files.

>>>	* sublisp.el: the following features have been removed for the lep
	version of the interface, since they are superfluous:

	    fi:emacs-to-lisp-transaction-directory (variable)
	    fi:remove-all-temporary-lisp-transaction-files (function)

	* keys.el: fi:inferior-lisp-send-input moved to this file

	* lep/lze.el: when evaluating, print appropriate messages

Sat Mar 16 13:50:48 1991  Kevin Layer  (layer at vapor)

	* lep/basic-lep.el: [bug1730] bind inhibit-quit to t in connection
	filter

	* subproc.el: give the `process' created by open-network-stream a
	sentinel, so that pesky message on exit doesn't appear

	* Makefile: `print' rule

===============================================================================
==================== ALPHA DOC GIVEN TO DOCUMENTATION GROUP ===================
===============================================================================

Fri Mar 15 12:37:35 1991  Kevin Layer  (layer at vapor)

	* ug.n: done, except for clman!

	* Makefile: spec.n is automatically generated from ug.n

	* doc.el: differentiate between commands and functions

	* nonlep/query.el, keys.el: lisp-who-calls -> list-who-calls

	* lep/composer.el: fi'ify

	* lep/dmode.el: lisp-who-calls -> list-who-calls

	* lep/lep.el: fix fi:list- and fi:edit- functions up a little

	* lep/changes.el: fix a few bugs, fi'ify

*******************************************************************************

    global change: make all keymaps non-sparse (all were previously
    sparse).  There seems to be a bug in emacs which doesn't deal with
    sparse keymaps correctly.  For example, I often get "error wrong
    type argument listp ..." when doing a describe-bindings.  This has
    gone away since I've started using vector keymaps.  Furthermore,
    the doc strings for the various modes have the keys in the keymap
    descriptions ordered randomly with sparse keymaps--I want them
    ordered. 

    Many, many doc strings were fixed or written.  Basically, all have
    been regularized to conform to the Emacs standard doc string
    format, giving the argument names in upper case and mentioning
    them in the order in which they are found in the formal arglist.
    When prefix arguments are taken, this has been explained.

*******************************************************************************

Wed Mar 13 22:00:00 1991  Kevin Layer  (layer at vapor)

	* doc.el: new look to .out files 

	* lep/dmode.el: fix it up a little; don't find first definition,
	by default

	* lep/lep.el: fi'ify some code

	* lep/changes.el: fix typo

Wed Mar 13 15:30:00 1991  Kevin Layer  (layer at vapor)

	* subproc.el: read host second, in
	fi::get-lisp-interactive-arguments, and make sure file names exist
	if `local'; make fi:common-lisp and fi:franz-lisp program callable

Wed Mar 13 15:00:00 1991  Kevin Layer  (layer at vapor)

	* Makefile: new file: lep/changes.el

	* site-init.el: reload keys.el when toggling lep

	* spec.n: add fi:scan-stack

	* lep/basic-lep.el: make session ids negative numbers; debugging aids

	* lep/composer.el: edit-generic-function-methods, edit-callers,
	and edit-callees moved elsewhere; fi'ify stuff; misc

	* lep/dmode.el: misc

	* lep/lep-init.el: new file lep/changes.el

	* lep/lze.el: message with done with evaluations

	* lep/lep.el: fi:edit-* moved here; misc

	* lep/changes.el: changed-definition support!

Wed Mar 13 10:38:06 1991  Kevin Layer  (layer at vapor)

	* db.el: doc string hacking, rename some interactive functions

Tue Mar 12 20:56:00 1991  Kevin Layer  (layer at vapor)

	* utils.el: new function to canonicalize filenames

	* subproc.el: better handle reading interactive arguments for
	fi:common-lisp and fi:franz-lisp (share common code, too)

Tue Mar 12 18:23:52 1991  Kevin Layer  (layer at vapor)

	* Makefile: new file ug.n

	* ug.n: user guide chapter draft

	* spec.n: significantly update 

	* subproc.el, sublisp.el: massive reorganization of function to
	start lisp subprocesses, fully documented in fi:common-lisp doc
	string, doc string hacking.

	* indent.el, ring.el: doc string hacking

	* rlogin.el, shell.el, telnet.el: doc string hacking, new
	fi::make-subprocess

	* su.el: new fi::make-subprocess

	* doc.el: fix a few bugs

	* modes.el, keys.el, tcplisp.el: `tcp' generally replaced with
	`lisp-listener', doc string hacking.

	* lep/basic-lep.el: variable name hacking

	* lep/composer.el: name hacking

	* lep/lep.el: doc string hacking

	* lep/stream.el: name hacking

	* nonlep/query.el: provide correct args to fi::get-default-symbol

Thu Mar  7 14:52:54 1991  Kevin Layer  (layer at vapor)

	* subproc.el: don't look for ^A's in the subprocess filter after
	the lisp-emacs connection has been established.

	* subproc.el: rework regexp to avoid \255, which phreaks out nemacs

	* indent.el: fix defgeneric--'(like defmethod) doesn't work for
	some reason

	* keys.el: don't silently fixup syntax upon newline in subprocess
	buffer.

	* lep/lep.el: make find-tag-other-window do it in the other
	window!

	* lep/db.el: fix help text, and make reset, continue and pop not
	frob with the current frame

Thu Feb 28 23:00:22 1991  Kevin Layer  (layer at vapor)

	* lep/lep.el: when *CL-TEMP* is used, print `package' on a
	separate line; make sure there is a message printed before
	macroexpansion starts

	* indent.el: add a bunch of methods for ACL and generic CL.

	* subproc.el: fix filter frammis to send back stuff before and
	after ^A

	* spec.n, keys.el: new function: fi:lisp-apropos; changed the name of
	fi:lisp-walk to fi:lisp-macroexpand-recursively

	* lep/basic-lep.el: lep::lep-open-connection-p to check for open
	lep connection

	* lep/lep.el: add hook in fi::show-some-text-1; fi:lisp-apropos;
	don't start lisp to do M-.; make M-, behave like the standard one
	does.

	* lep/db.el: add binding to disassemble

	* nonlep/query.el: M-, hacking, fi:lisp-walk renamed
	fi:lisp-macroexpand-recursively

Wed Feb 27 13:28:37 1991  Kevin Layer  (layer at vapor)

>>>	* spec.n, indent.el: rename fi:lisp-case-sensitive to
	  fi:indent-methods-case-sensitive.

	* su.el: go to the correct buffer before reading the passwd

	* nonlep/ltags.el, utils.el: fi:filename-frobber-hook and
	  fi::discombobulate-automounter-lint moved from
	  nonlep/ltags.el to utils.el.

	* lep/db.el: don't shadow standard bindings in scan stack mode

>>>	* spec.n, tcplisp.el, subproc.el: fi:unix-domain and
	fi:unix-domain-socket have been removed from the interface.

	* site-init.el: variable fi:use-lep controls whether or not lep is
	loaded by default (nil, for now)

	* lep/lep.el: lep::who-calls may return `nil', so handle this case

Fri Feb 22 00:42:38 1991  Kevin Layer  (layer at vapor)

	* lep/composer.el, lep/lze.el, lep/lep.el: regularize
	  show-some-text*

	* lep/db.el: use window configuration to restore windows

Thu Feb 21 21:55:38 1991  Kevin Layer  (layer at vapor)

	* spec.n, subproc.el: fi:subprocess-map-nl-to-cr and
	  fi:subprocess-write-quantum variables are history.
	  fi::send-region-split and fi::send-string-split functions are
	  history.

	* telnet.el, su.el, sublisp.el, keys.el, rlogin.el: use
	send-string and send-region instead of our own versions of them.

Fri Feb 15 23:34:00 1991  Kevin Layer  (layer at vapor)

	* site-init.el, Makefile: query and ltags moved to nonlep
	directory (soon to be depricated)

	* keys.el, nonlep/query.el: fi:lisp-eval-* functions moved to
	nonlep/query.el

	* lep/lze.el: finish fi:lisp-eval-* suite of functions

Fri Feb 15 23:18:24 1991  Kevin Layer  (layer at vapor)

	* keys.el: fi:scan-stack entry point; beginning-of-defun fixes

	* lep/lep-init.el: load db.elc

	* subproc.el: send listener-protocol to correct process

	* Makefile: added lep/db

	* lep/db.el: implement stack scan mode, an interface to the
	allegro debugger.

	* lep/lze.el: use fi:beginning-of-defun

	* lep/lep.el: more show-some-text fixes

Wed Feb 13 10:19:08 1991  Kevin Layer  (layer at vapor)

	* lep/basic-lep.el: print `done' message after starting lep
	  connection
	* lep/lep.el: fix show-some-text; remove debugging print's
	* lep/lze.el: cosmetic

Tue Feb 12 17:00:00 1991  Kevin Layer  (layer at vapor)

	* lep/lep.el: fix show-some-text to be more friendly;
	  *always-in-a-window* -> fi:always-in-a-window

  from cer and layer:

	* Makefile: added lep/lze.el
	* site-init.el: message about loading LEP and non-LEP versions.
	* sublisp.el: fix #+ version's.  Don't use compile-file-if-needed,
	  since compiling is always the right thing.
	* subproc.el: new variable: fi::frammis-hook, used to startup the
	  lep connection between lisp and emacs.
	* lep/composer.el: fix typo
	* lep/dmode.el: definition mode hacking
	* lep/lep-init.el: load lze.elc
	* lep/lze.el: new file + cosmetic changes from cer's version.
	* lep/lep.el: show output from lisp in a window (needs work,
	  however).  Lots of other misc changes.
	* lep/basic-lep.el: auto-ensure-lep-connection, which uses
	  fi::frammis-hook

Tue Feb 12 14:52:18 1991  Kevin Layer  (layer at vapor)

	* dot.emacs: remove setting of fi:common-lisp-image-arguments
	* indent.el: indent for ANSI `loop'
	* sublisp.el: tell from what buffer forms are loaded
	* subproc.el: give require 2nd arg for :emacs, since people may
	  have a directory called `emacs'.  Frob defaulting in common-lisp
	  functions.
	* rlogin.el: set TERM to dumb in remote shell

Thu Jan 31 22:03:23 1991  Kevin Layer  (layer at akbar)

	* tcplisp.el: make fi:eval-in-lisp work in all buffer.

	* keys.el: for compatibility with standard shell-mode, make C-cC-y
	  to fi:pop-input; lep::toggle-trace-definition is fi:toggle-...;
	  fi:lisp-find-tag-other-window on "C-x4."; make sure too many
	  parens don't hang on newline in subprocess buffer.

Thu Jan 31 14:00:00 1991  Kevin Layer  (layer at akbar)

>>>	* modes.el: define an initial fi:lisp-mode-hook which puts the
	  package in the mode line.  fi::check-for-package-info renamed
	  fi:parse-mode-line-and-package.

Thu Jan 31 09:23:43 1991  Kevin Layer  (layer at akbar)

	* su.el: run fi:su-mode-hook instead of fi:shell-mode-hook
	* spec.n, doc.el: revamp both
	* telnet.el, modes.el, shell.el, rlogin.el, keys.el: doc string
	  hacking
	* query.el, lep/lep.el: lep::toggle-trace-definition ->
	  fi:toggle-trace-definition

>>>	* subproc.el: new variable, fi:subprocess-env-vars, contains the
	  environment variables to pass to subprocesses (even remote ones
	  started with `rsh').

	* site-init.el: fix fi:toggle-lep

>>>	* keys.el: new lisp-mode bindings:
		C-c %		fi:extract-list
		C-c ;		fi:comment-region
	make "C-c%", "C-c;", "C-cC-e" and "C-c]" available in subprocess
	modes, too.  New function, fi:beginning-of-defun, which works in
	subprocess modes.

Wed Jan 30 10:35:10 1991  Kevin Layer  (layer at akbar)

>>>	* keys.el: new functions and bindings, for editing modes:

		C-c C-e		fi:end-of-defun
		C-c ]		fi:super-paren
		ESC T		lep::toggle-trace-definition

	new functions:

		fi:find-unbalanced-parenthesis
		fi:fill-paragraph

	the latter fills and justifies comments.

	* query.el: non-lep stub version of lep::toggle-trace-definition.

	* lep/lep.el: fix interactive arguments to a bunch of functions.
	Fix lep::toggle-trace-definition. 

Tue Jan 29 14:19:49 1991  Kevin Layer  (layer at akbar)

	* sublisp.el: 
	* tcplisp.el: 
	* utils.el: 
	* ltags.el: 
	* subproc.el: 
	* lep/basic-lep.el: new predicate fi:process-running-p

>>>	* site-init.el: fi:toggle-lep goes between lep and non-lep
	versions of emacs-lisp interface.

	* Makefile: lep/lep-utils.el is outta here (in comes `cl')

	* sublisp.el: make fi::send-string-load setup things properly for
	ACL 4.1 features.

	* tcplisp.el: ipc version control

>>>	* subproc.el: ipc version control; make fi:tcp-common-lisp take an
	optional buffer-name

	* new files in lep:
		composer.el
		lep-init.el
		lep.el
		basic-lep.el
		dmode.el
		lep-utils.el
		stream.el

	* keys.el: move functions effected by the lisp-editor protocol
	portions into query.el.

	* query.el: new file, containing the function effected by the
	lisp-editor protocol.

	* site-init.el: load query.elc

=============== 1.5.2 release (included with ACL 4.0 on the Sun4) =============

Fri Jan 11 10:25:22 1991  Kevin Layer  (layer at akbar)

	* ltags.el: when the package specified in the file's mode
	specification does not exist in the Lisp, make sure the backdoor
	doesn't get an error.

Fri Jan 11 10:10:39 1991  Kevin Layer  (layer at akbar)

	* clman.el: `abouts' is called `about' in manual/.

Fri Dec 28 14:26:53 1990  Kevin Layer  (layer at akbar)

	* sublisp.el: make sure fi::sublisp-name is really a name and not
	a process object, since caching the object is bad in the face of
	someone restarting the process!

Thu Dec 27 13:04:07 1990  Kevin Layer  (layer at akbar)

	* tcplisp.el: split handling of case-mode into `to' and `from'
	lisp functions.

	* ltags.el: new hook, fi:filename-frobber-hook, used to strip off
	/tmp_mnt/net at the beginning of filenames.

	* subproc.el: case-mode hacking
	* keys.el: case-mode hacking

	* spec.n: add fi:filename-frobber-hook

Mon Dec 17 17:47:32 1990  Kevin Layer  (layer at akbar)

	* subproc.el: frob the case of the host name coming in from Lisp,
	when starting the backdoor; fi::is-running-p hack is outta here.

Mon Dec 17 14:50:48 1990  Kevin Layer  (layer at akbar)

	* tcplisp.el: new functions: fi:backdoor-is-running-p and
	fi:common-lisp-is-running-p.

	* keys.el: augment M-x behaviour: don't startup backdoor, unless a
	Lisp is running.  If can't then use default tags mechanism.

	* clman.data: new one for 4.0, to go with the new manual/

	* manual/: new one for 4.0

Thu Dec 13 17:16:03 1990  Kevin Layer  (layer at akbar)

	* subproc.el: don't set fi::remote-host if process was already
	running

Fri Dec  7 00:32:20 1990  Kevin Layer  (layer at akbar)

	* keys.el: make sure M-. and M-, in CL buffer starts the backdoor.

Tue Dec  4 23:04:58 1990  Kevin Layer  (layer at akbar)

	* indent.el: add indent methods for a couple of new x3j13 things:
	with-standard-io-syntax, print-unreadable-object

	* modes.el: move local variable hackin in fi::lisp-edit-mode-setup
	to another place.

	* sublisp.el: setup some lisp-mode specific variables; don't use
	"noname" as the default transaction buffer name, use the buffer
	name.

	* tcplisp.el: fi:unix-domain-socket is not a read-only variable
	(ie, no one but the user looks at it); fi:inet-host-name is outta
	here (no one was setting it); generally fix unix domain.

	* utils.el: fi:test-tcp-interface is outta here since it didn't
	work in the first place.

	* subproc.el: unix domain hacking to make it work again.

Thu Nov 29 14:07:47 1990  Kevin Layer  (layer at akbar)

	* indent.el: indent method for defpackage

	* sublisp.el: prevent sublisp-name from pointing to a dead process

	* ltags.el: require 'tags

	* subproc.el: make sure value of
	fi:remote-lisp-track-image-name-directory has a directory.

	* Makefile: use emacs instead of /usr/local/emacs

	* keys.el: qualify chdir in excl package

	* dot.emacsX11: give fi:lisp-find-tag an argument

=============== 1.5.1 release (ftp'able from uunet) ===========================

Fri Oct 19 13:48:25 1990  Kevin Layer  (layer at akbar)

	* spec.n: version is now 1.5.1

	* site-init.el: added version number variable,
	fi:emacs-lisp-interface-version, the value is a string which
	represents the version number.

=============== 1.5 release (included with Sun4 ACL 4.0.beta.0) ===============

Wed Oct 17 16:43:38 1990  Kevin Layer  (layer at akbar)

	* subproc.el: if tracking directory for remote lisps, make sure
	the rsh can access the image via an unqualified (relative)
	pathname.

Tue Oct 16 22:00:08 1990  Kevin Layer  (layer at akbar)

	* subproc.el: put *'s before docstrings in some recent external
	variables additions; new variable

	   fi:default-remote-common-lisp-directory
		If non-nil, used by fi:remote-common-lisp and
		fi:explicit-remote-common-lisp as the directory in which
		the Lisp is started.  Note that the pathname of the Lisp
		image must be give relative to this directory, or the home
		directory if this variable is nil.

	fix startup of listener daemon so it works when ipc isn't loaded.

Sat Oct 13 19:27:56 1990  Kevin Layer  (layer at akbar)

	* sublisp.el: fix fi::sublisp-select so it knows about
	tcp-common-lisp-mode.

	* tcplisp.el: [bug1631] handle null string in fi::backdoor-filter

	* keys.el: for remote lisps, defined C-d to use the backdoor and
	do a debug-pop (fi:remote-ilsp-send-eof)

	* subproc.el: tcp lisps shouldn't set
	fi::freshest-common-sublisp-name and new variables:
	  fi:remote-lisp-track-image-name-directory
	  	If non-nil, then fi:remote-common-lisp and
		fi:explicit-remote-common-lisp will cause both Emacs and
		Lisp to cd into the directory where the Common Lisp image
		resides.  Otherwise, the remote Common Lisp will have
		$HOME as its current working directory and Emacs will have
		whatever directory you were in when you evaluated either
		function.
	  fi:default-explicit-common-lisp-image-name
	  	If non-nil, then the value of this variable is used as the
		name of the image fi:explicit-common-lisp and
		fi:explicit-remote-common-lisp uses to run Lisp, instead
		of reading the name from the minibuffer.
	  fi:default-explicit-common-lisp-image-arguments
	  	If non-nil, then the value of this variable (a string)
		contains the arguments given to Lisp by
		fi:explicit-common-lisp and fi:explicit-remote-common-lisp
		when Lisp is invoked, instead of reading the name from the
		minibuffer.
	  fi:default-remote-common-lisp-host
	  	If non-nil, then the value of this variable (a string) is
		names the host on which fi:remote-common-lisp and
		fi:explicit-remote-common-lisp executes Lisp, instead of
		reading the name from the minibuffer.

	* telnet.el
	* su.el
	* modes.el
	* rlogin.el
	* shell.el: main modes take optional mode-hook, which is funcall'd
	after local variables are killed.

Tue Oct  9 10:42:42 1990  Kevin Layer  (layer at akbar)

	* dot.emacs: add fi:subprocess-mode-hook and example of
	calculating the process name

	* subproc.el: make starting interface incur as little output as
	possible in lisp subprocess buffer

	* README: bring it up-to-date

Thu Sep 13 23:28:51 1990  Kevin Layer  (layer at akbar)

	* subproc.el: put package qualifiers on REQUIRE, since it is
	movin' to the excl package.

Tue Sep 11 16:48:28 1990  Kevin Layer  (layer at akbar)

	* utils.el: new function: fi::fast-string-search, two args CHAR
	and STRING.  Return t if CHAR is in STRING.

	* subproc.el: use fi::fast-string-search in
	fi::subprocess-control-a-frammis, to speed up processing of
	subprocess output.

Mon Sep 10 15:01:56 1990  Kevin Layer  (layer at akbar)

	* subproc.el: fi:display-buffer-function back to
	'switch-to-buffer.  Subprocess filter must bind inhibit-quit to t
	so that output from the subprocess can be process fully
	(independent of the user typing ^G's).

	* tcplisp.el: inhibit-quit during filter operation.

Sat Sep  8 01:00:22 1990  Kevin Layer  (layer at akbar)

	* sublisp.el: use display-buffer & select-window instead of
	switch-to-buffer-other-window, since the former will use a window
	already showing the buffer.  fi::select-sublisp fixed so that when
	it starts a lisp because a command needs it, it doesn't bomb.
	Third arg to fi::eval-string-send was useless.

	* keys.el: fi::eval-string-send takes two args, now.

	* subproc.el: use display-buffer & select-window instead of
	switch-to-buffer.

Fri Sep  7 18:04:48 1990  Kevin Layer  (layer at akbar)

	* sublisp.el: fix fi::send-string-load to not create a "visible"
	buffer (put a space before the name).

	* tcplisp.el: move constant fi::lisp-macroexpand-command to
	keys.el (where it is used).

	* keys.el: frob fi::lisp-macroexpand-command so that it binds
	*print-length*, *print-level*, and *print-circle* to something
	reasonable (nil, nil and t, for example).

	* subproc.el: fi:remote-common-lisp and
	fi:explicit-remote-common-lisp: give the remotely executing
	command an environment of three variables: EMACS (so that programs
	can discriminate whether they are running in an emacs buffer),
	TERM and TERMCAP (where the screen-width is set properly).

Thu Sep  6 23:11:22 1990  Kevin Layer  (layer at akbar)

	* sublisp.el: create better name for
	fi::emacs-to-lisp-transaction-file and use set-buffer on
	fi::emacs-to-lisp-transaction-buf instead of pop-to-buffer so that
	it doesn't appear on buffer alist.

Wed Sep  5 16:57:57 1990  Kevin Layer  (layer at akbar)

	* keys.el: define fi:shell-sync-current-working-directory and bind
	to C-c. in shell modes 

	* ipc.cl: use defpackage for ACL 4.0.

Tue Sep  4 15:16:00 1990  Kevin Layer  (layer at akbar)

>>>	* subproc.el: new variable, fi:start-lisp-interface-function,
	which controls whether the lisp interface is started or not.

Tue Sep  4 10:39:01 1990  Kevin Layer  (layer at akbar)

	* subproc.el: funcall image-file in fi::make-subprocess only when
	the process is created.  Also, funcall
	fi::start-tcp-lisp-interface in fi::make-subprocess when it is
	passed as the optional argument initial-func.

Mon Sep  3 11:40:00 1990  Kevin Layer  (layer at akbar)

>>>	* keys.el: C-c. in lisp subprocess modes bound to
	fi:sync-current-working-directory.

	* ltags.el: don't startup backdoor on "M-.".

	* ipc.cl: cstruct-len -> cstruct-length (for 4.0)

Mon Sep  3 00:21:03 1990  Kevin Layer  (layer at akbar)

	* ipc.cl: merge 3.1 master and 4.0 hacks into our master.

	* emacs.cl: merge 4.0 hacks into our master.

Sun Sep  2 20:38:00 1990  Kevin Layer  (layer at akbar)

>>>	* subproc.el: no long have to put anything in .clinit.cl to have
	the lisp interface started!!!  Now that's progress.

Sun Sep  2 20:00:00 1990  Kevin Layer  (layer at akbar)

	* ipc.cl: [bug1303] don't print ";;; starting listener..." since
	it screws up lisp symbol completion.

	* tcplisp.el: fi::make-connection was bogusly implemented earlier
	today.  Give it two args, name and buffer, and fix the callers.

	* keys.el: case frobing in lisp symbol completion.

Sun Sep  2 19:00:00 1990  Kevin Layer  (layer at akbar)

	* Makefile: doc.el and spec.n are in . not ../somethingorother.

	* spec.n: fi:excl-service-name and fi:local-host-name are outta
	here.  Add fi:lisp-case-sensitive.

	* doc.el: spec.n is in . not ../doc.

Sun Sep  2 18:00:26 1990  Kevin Layer  (layer at akbar)

	* ipc.cl: [rfe972] parameterize for ACL 3.1 and 4.0.
>>>	Symbol ipc:*inet-port* is history.

>>>	New symbols: *inet-port-min*, *inet-port-max*, *inet-port-used*,
	and *inet-listener-password*.

	Revamp the way internet port selection occurs.  Also, make the
	internet domain the default!  The first pass as design and
	implementation courtesy of Louis Steinberg, Rutgers University
	(lou@aramis.rutgers.edu).  Thanks!  The scheme is this:

		When ACL start the Lisp-side of the emacs-lisp interface,
		it prints the a string to standard-output of the form:
		^Aport password case-mode^A.  PORT is the internet
		port which emacs uses to connect to ACL.  PASSWORD is the
		number (randomly chosen by ACL) which identifies
		connections as valid.  CASE-MODE is the case that ACL is
		currently using.  When emacs sees this string, it sets the
		state of various global and buffer local variables.

>>>	* tcplisp.el: fi:unix-domain defaults to nil!  fi:local-host-name
	changed to fi:inet-host-name.  fi:excl-service-name is outta here!

	* modes.el: better handle symbolic-style package: mode line
	format.

>>>	* keys.el: in Lisp editing mode keymaps, define C-c-, which
	annotates a function with a log entry of what was just done to it
	(see fi:log-functional-change).  Also, frobing of symbols for
	correctness of case.

	* dot.emacs: fixup fi:common-lisp-image-arguments for above hacks. 
	* dot.clinit: fixup for above hacks. 

	* subproc.el: New variables: fi::remote-host (buffer local),
	fi::remote-port (buffer local), fi::remote-password (buffer
	local), and fi::lisp-case-mode.

	All the fi: functions that startup subprocesses or tcp connections
	now note their host.

	fi::subprocess-filter watches for ^A in Lisp subprocess buffers.

	New functions: fi::get-buffer-password, fi::get-buffer-port,
	fi::get-buffer-host, and fi::set-buffer-host.

	The case of symbols sent to Lisp is now correct, due to the above
	^A hack.

	* ltags.el: frob symbol case.

	* utils.el: fix fi:test-tcp-interface due to above changes (even
	though this function doesn't do what it was designed to do (yet)).

Sat Sep  1 20:12:28 1990  Kevin Layer  (layer at akbar)

	* sublisp.el: make sure transaction filenames don't have a length
	of more than 14 characters

	* keys.el: fix heuristic for determining when the collected input
	should be send to the sublisp process.

	* subproc.el: fix doc strings re: buffer naming.  Fix
	fi::make-subprocess to expand image-file's with tildes.

Fri Aug 31 23:30:13 1990  Kevin Layer  (layer at akbar)

	* sublisp.el: qualify symbols in excl package; when loading lisp
	code, do ":print t"; use sys::*fasl-default-type* for loading
	.fasls.

	* filec.el: ignore errors while searching for files

	* indent.el: nasty bug whereby certain indent methods caused an
	error (defmethod, for one).  New intenter for define-condition and
	defclass.

	* ipc.cl: cstruct-len -> cstruct-length; hacks for sony, sgi4d and
	SunOS 4.1 in open-network-stream; use double-buffered-terminal
	streams.

	* rlogin.el: use fi::subprocess-mode-super-keys to setup keymap;
	fi:rlogin-send funcs moved elsewhere.

	* Makefile: added su.el and telnet.el; new directory structure in
	/usr/emacs.  Remove `tags' rule (in ../../Makefile).  Added
	`install' rule used by ../../install_emacs.sh.  Set SHELL (for
	some system V machines) to /bi/sh.

	* su.el: new file
	* telnet.el: new file

	* modes.el: fi::lisp-subprocess-mode-variables sets buffer local
	fi:subprocess-mode to t so that we know when we are in a
	subprocess mode (as opposed to just an editing mode).

	* keys.el: handle rlogin mode, too.  Move fi:rlogin-send funcs to
	this file from elsewhere.  fi:lisp-arglist: allow macros and
	special forms, since ACL 4.0.alpha saves arglists for these frobs.

	* dot.clinit: make sure lookup of
	ipc:start-lisp-listener-daemon is not case sensitive.  Make
	*directory-stack* a defparameter (from a defvar).

	* subproc.el: fix fi:remote-common-lisp so that it evaluates
	fi:common-lisp-image-name before passing it to fi::make-subprocess
	as the image arguments.

	* README: bring it up-to-date.

	* site-init.el: load telnet and su.

	* shell.el: fix doc string.

	* utils.el: fi::substitute-chars-in-string: make sure
	CHAR-ASSOC-LIST is hunky dory.  New function: fi::read-password.

	* docompile.sh: make sure $3 looks like a directory.  Require
	:defctype instead of :cstructs (the wave of the future).

============================== release 1.4 ==============================
Also distributed with NeXT 1.0
=========================================================================

Thu Aug 17 20:47:22 1989  Kevin Layer  (layer at snooze)

	* subproc.el: save fi::input-ring-yank-pointer across processes,
	just like fi::input-ring.

Wed Aug 16 13:51:37 1989  Kevin Layer  (layer at snooze)

	* ipc.cl: `hostent' structure changed for SunOS 4.0.  Put hacks in
	open-network-stream which handle that structure differently if on
	SunOS 4.0 (if /lib/ld.so exists, then we know we are on 4.0).

Mon Aug  7 16:06:38 1989  Kevin Layer  (layer at snooze)

	* ipc.cl: don't use excl::machine-case, because some targets (or
	hosts) are not known on some ports of acl.  fi:unix-domain-socket
	can now be a function, which is funcalled to get the string which
	represents the socket pathname.  The form of the default socket
	file is /tmp/GnuToAclXXX, where XXX is the uid of the person
	running emacs and acl.  The length of this is more friendly to
	system V users.

	* tcplisp.el: fi:unix-domain-socket is `evaluated' at interface
	startup-time, instead of interface load-time.

	* Makefile: new way to compile .fasl files (output goes into the
	library *or* to the ${cl_library}.  See docompile.sh.

	* docompile.sh: used to compile the .cl files.  Mostly it just
	created the output of compile-file in the lisp library.

	* keys.el: during macroexpansion, bind
	fi:echo-evals-from-buffer-in-listener-p to nil.

	* subproc.el: wrap the funcall of `mode' (in make-subprocess) with
	a condition-case.

	* clman.el, clman.data: final touches to new scheme.

	* dot.clinit, dot.emacs: remove a bunch of junk

Wed Jul 19 14:01:32 1989  Kevin Layer  (layer at akbar)

	* Makefile: change `depend' to `elcs' ("make depend" means
	something very different to many people).

	* ipc.cl: *unix-domain* defaults to nil on sgi4d.

	* tcplisp.el: typo

	* clinit.cl: test for setting socket-file fixed

	* subproc.el: read interactive arguments for fi:explicit-*
	functions better.  Specifically, don't read a list of strings for
	the image arguments, ie ("a" "b" "c"), read "a b c" and do the
	conversion for the user.

	* ltags.el: fully qualify symbols not in the lisp package (someone
	may not use the `excl' package).

	* indent.el: move fi::lisp-find-char to utils.el

	* utils.el: fi::lisp-find-char moved to this file (from
	indent.el); new function fi::listify-string, which takes one arg,
	a string, and converts it to a list of strings (space is the
	separator in the string).  Also, first pass at a diagnosic tool
	find failures in the interface.

	* clman.el and clman.data: new (incompatible with previous
	versions) format for manual files.  Names in the manual directory
	are now *all* less than 14 characters!  Viva la system5!  Pardon
	me, whilest I become sick...

Tue Jul 11 18:10:08 1989  Kevin Layer  (layer at snooze)

	* ipc.cl: better error messages.

	* rlogin.el: get `env' from exec-directory.

	* tcplisp.el: fully qualify `errorset' in the `excl' package.

	* Makefile: clman-oblist.el renamed clman.data, and it is not
	byte-compiled anymore.

	* clinit.cl: lisp-listener-daemon-ff-loaded is now
	.lisp-listener-daemon-ff-loaded.

	* subproc.el: make-subprocess: substitute-in-file-name on image-file;
	make-tcp-connection only does expand-file-name when unix-domain
	non-nilA.

	* indent.el: added indent property for `defclass'.

	* clman.el: setup fi:clman-package-info first time manual entry
	looked up, not at load time.

============================== release 1.4beta ==============================

Thu Jun  1 22:27:46 1989  Kevin Layer  (layer at snooze)

	* modes.el: added some vars made local in mode setup.

	* indent.el: these vars
	    fi::calculate-lisp-indent-state-temp
	    fi::lisp-most-recent-parse-result
	    fi::lisp-indent-state-temp
	  setup in mode initialization.  Also, fix indent hook for
	  handler-bind and -case.A

	* utils.el: make third arg to fi::{slow,fast}-parse-partial-sexp
	optional (just like parse-partial-sexp).

	* clman.el: use fi::clman-downcase instead of downcase.

Wed May 24 14:36:02 1989  Kevin Layer  (layer at snooze)

	* indent.el: defvar'd value of
	fi:lisp-comment-indent-specification is bogus, because
	comment-column given value in mode setup.

	* modes.el: fi:lisp-do-indentation defaults to `t'.  Also, set
	fi:lisp-comment-indent-specification in mode setup.

	* subproc.el: make buffer names created by fi::make-subprocess
	follow the standard buffer naming convention.  A second *shell*
	buffer is now called *shell*<2>.

Mon May 22 15:29:21 1989  Kevin Layer  (layer at snooze)

	* clinit.cl: don't use handler-case, since it is not in all
	versions of ACL in which this interface runs, but use errorset.Q

	* ipc.cl: new and improved: 1) correctly restart daemon after
	dumplisp [NOTE: this requires at least version 3.1.beta.19 to
	work correctly], and 2) put #+'s around use of
	process-property-list, since it doesn't exist in all versions in
	which this interface runs.

	* sublisp.el: forms sometimes sent to lisp in the wrong package.

Mon May 15 22:22:12 1989  Kevin Layer  (layer at akbar)

	* modes.el: Many changes:

	  + define lisp mode syntax tables with fi:lisp-mode-syntax-table
	    and emacs lisp mode with fi:emacs-lisp-mode-syntax-table.
	  + added Emacs Lisp mode: variable: fi:emacs-lisp-mode-map.
	    Function fi:emacs-lisp-mode.
	  + new variable, fi:lisp-do-indentation, which controls whether
	    the FI indentation is used.  This should be set in .emacs.
	    The only reason not to use it is that M-C-q is slow for large
	    functions.  This will be fixed in the future, but I don't have
	    the time right.   Btw, the reason to use the FI indentation is
	    that it indents everything *just right*!
	  + fi:indent-setup-hook no longer run from all the lisp modes.
	  + all the modes set fi:lisp-indent-hook-property according to
	  the style of indentation for that mode (ie, either
	  fi:common-lisp-indent-hook, fi:franz-lisp-indent-hook, etc).

	* subproc.el: fi::make-subprocess: the last argument is optional,
	a filter to install on the subprocess.

	* site-init.el: only load indent.elc if fi:lisp-do-indentation is
	non-nil.

>>>	* indent.el: new file.   The meta language for describing how to
	indent forms has been improved.  Among the features are: it
	properly indents `flet', `labels', `macrolet', all the pitman
	condition system macros (this is no small thing!!), and lambda
	list keywords are indented specially (finally!).  Here are the
	variables which control indentation:

	  + fi:lisp-electric-semicolon: If `t', semicolons that begin
	    comments are indented as they are typed.
	  + fi:lisp-comment-indent-specification: Specification list for
	    indentations of semicolon comments. The nth element of the
	    list specifies the indentation for a comment beginning with n
	    semicolons (e.g. the first element of the list is the
	    indentation for comments beginning with one semicolon).  Each
	    element of the list may be one of `t' (indent comment just
	    like an s-expression), `nil' (don't change the indentation of
	    the comment, i.e. leave the semicolon where it is), a non-
	    negative integer (specifying the absolute column to which the
	    comment is to be indented), or a negative integer (specifying
	    a negative offset for the comment relative to the current
	    column)
	  + fi:lisp-body-indent: The indentation of a list continued on
	    another line
	  + fi:lisp-indent-offset: If non-nil, then indent by a constant
	    amount of the column in which the sexp starts
	  + fi:lisp-tag-indentation: Indentation of tags relative to
	    containing list. This variable is used by the function
	    `fi:lisp-indent-tagbody' to indent tags that occur within
	    special forms whose symbols have a 'fi:lisp-indent-hook
	    property of 'tag or 'tagbody.  The indentation is relative to
	    the indentation of the parenthesis enclosing the special form
	  + fi:lisp-tag-body-indentation: Indentation of non-tagged lines
	    relative to containing list. This variable is used by the
	    function `fi:lisp-indent-tagbody' to indent normal lines
	    (lines without tags) that occur within special forms whose
	    symbols have a 'fi:lisp-indent-hook property of 'tag or
	    'tagbody.  The indentation is relative to the indentation of
	    the parenthesis enclosing the special form. If the value is T,
	    the body of tags will be indented as a block at the same
	    indentation as the first s-expression following the tag.  In
	    this case, the s-expressions before the first tag are indented
	    as an undistinguished form
	  + fi:lisp-tag-indentation-hook: Name of function to apply to
	    return indentation of tag. This variable may be bound to the
	    name of a function to be applied (to three arguments: the
	    character position of the beginning of the tag, the last parse
	    state, and the indent point) to return the appropriate
	    indentation for tags occurring within special forms whose
	    symbols have a 'fi:lisp-indent-hook property of 'tag or
	    'tagbody.  The indentation returned is absolute
	  + fi:lisp-keyword-indentation: Indentation of keywords relative
	    to containing list. This variable is used by the function
	    `fi:lisp-indent-keyword-list' to indent keywords that occur
	    within special forms whose symbols have a 'fi:lisp-indent-hook
	    property of 'keyword or 'keyword-list.  The indentation is
	    relative to the indentation of the parenthesis enclosing the
	    special form
	  + fi:lisp-keyword-argument-indentation: Indentation of keyword
	    argument lines relative to containing list. This variable is
	    used by the function `fi:lisp-indent-keyword-list' to indent
	    keyword-argument lines that occur within special forms whose
	    symbols have a 'fi:lisp-indent-hook property of 'keyword or
	    'keyword-list.  The indentation is relative to the indentation
	    of the parenthesis enclosing the special form. If the value is
	    T, the argument(s) of keywords will be indented as a block at
	    the same indentation as the first s-expression following the
	    tag.  See the documentation for the function
	    `fi:lisp-indent-keyword-list'
	  + fi:lisp-keyword-indentation-hook: Name of function to apply to
	    return indentation of a keyword. This variable may be bound to
	    the name of a function to be applied (to three arguments: the
	    character position of the beginning of the keyword, the last
	    parse state, and the indent point) to return the appropriate
	    indentation for keywords occurring within special forms whose
	    symbols have a 'fi:lisp-indent-hook property of 'keyword or
	    'keyword-list.  The indentation returned is absolute
	  + fi:lisp-maximum-indent-struct-depth: Maximum depth to
	    backtrack out from a sublist for structured indentation. If
	    this variable is NIL, no backtracking will occur and lists
	    whose `car' is a symbol with a 'fi:lisp-indent-hook property
	    of 'label, 'labels, 'flet, 'macrolet, 'defun, or a list may
	    not be indented properly.  In addition, quoted lists will not
	    be treated specially.  If this variable is T, there is no
	    limit placed on backtracking.  A numeric value specifies the
	    maximum depth to backtrack.  A reasonable value is 3.
	  + fi:lisp-case-sensitive: If non-NIL, the code that is being
	    edited is for a case-sensitive dialect of Lisp.  This variable
	    is buffer-local.  If a Lisp is case-insensitive, indentation
	    specifications should be placed on the Emacs Lisp symbol that
	    corresponds to the lowercase name of the function, macro, or
	    special form
	  + fi:lisp-package: This variable may be NIL, T, or a symbol or
	    string. If the value is NIL, a package qualifier is ignored
	    when getting the indentation specification for a symbol.  If
	    the value is T, the package qualifier is not ignored.  If this
	    variable is any other symbol or a string, it names the package
	    to be used for all unqualified symbols.  When this variable is
	    not NIL, the qualified symbol is first checked for an
	    indentation specification, then the unqualified symbol is
	    checked.  This variable is buffer-local.

	You may safely ignore these variables: the defaults are very
	reasonable.

	* utils.el: added an abstraction of parse-partial-sexp which has a
	sixth argument (I tried to add a sixth argument in the C coded
	function parse-partial-sexp, but the limit of 5 args is so
	hardwire in Emacs that I didn't want to make the change) which is
	used as the place to put the result.  Otherwise, this function
	does lots of consing.
 
	* tcplisp.el: for the backdoor loop use package qualifier on
	`errorset' in case the package has changed and excl not inherited.

	* keys.el: for new indentation (see fi:lisp-do-indentation).
	fi:lisp-reindent-newline-indent changed to fi:lisp-mode-newline,
	and the default is now to just do a newline, whereas before it
	used to do "reindent, newline, reindent".

	* filec.el: don't bind completion-ignored-extensions to nil.

	* rlogin.el: made the filter an argument to fi::make-subprocess.

============================== release 1.3 ==============================

Fri Feb 17 18:49:01 1989  Kevin Layer  (layer at snooze)

	* moved ChangeLog to parent directory of fi/.

	* include patch43.{68k,sparc} in distribution.  This is so that
	the following feature will work (otherwise the output has
	duplicate characters, since unpatched Allegro CL echo streams
	stutter).

	* a new variable
	      fi:echo-evals-from-buffer-in-listener-p
	which defaults to nil (since the necessary patch to Allegro CL is
	not yet widely distributed) and controls whether or not lisp echos
	what is evaluated from emacs lisp source buffers.  Setting this to
	nil (default is t) will cause the feature to be disabled.
	Files: sublisp.el, patch43.*

Tue Feb 14 17:13:54 1989  Kevin Layer  (layer at snooze)

>>>	* added two new files: clman.el and clman-oblist.el.  These are
	the emacs/on-line manual interface and are now part of this more
	general interface.  Also, there is a directory called `manual'
	which contains the database for the on-line manual.

	The entry points are fi:clman and fi:clman-apropos.  Nothing other
	than a binding for these two functions is needed for their use
	(they are `interactive').

	* added autoloads for clman stuff
	File: site-init.el

	* new function, fi::find-path, takes a string and looks through
	load-path for that file or directory and returns the pathname
	(absolute), if found.
	File: utils.el

Sat Nov 26 00:37:01 1988  Kevin Layer  (layer at snooze)

	* new variable, fi:emacs-to-lisp-transaction-directory, which
	controls where the transaction files are kept.  This is especially
	useful for running lisp on remote machines (i.e., using
	fi:remote-common-lisp).
	Files: keys.el and subproc.el

Tue Nov 22 20:17:56 1988  Kevin Layer  (layer at snooze)

	* ipc.cl: don't load from the C library unless absolutely
	necessary!  SunOS 4.0 on the Sun4 takes a bogus amount of time to
	dynamically load under certain conditions.

	* fix the way fi:XXX-image-name is handled (again, from 11/18/88).
	Now it is `reduced' before displaying the buffer but it is NOT
	done when the process already exists.
	File: subproc.el

Mon Nov 21 13:54:17 1988  Kevin Layer  (layer at snooze)

	* fi::subprocess-hack-directory renamed to
	fi::subprocess-watch-for-special-commands, to which smarts were
	added about tracking package changes in inferior lisp modes.  The
	new variable fi:common-lisp-package-regexp is used for this
	purpose.
	File: modes.el, keys.el and subproc.el

	* fix fi:lisp-complete-symbol in the case of an explicit package
	reference.
	File: keys.el

	* made the lisp evalserver more bullet-proof (output of more than
	one process read wasn't doing the right thing).
	File: tcplisp.el

Fri Nov 18 20:34:39 1988  Kevin Layer  (layer at snooze)

	* in fi::make-subprocess, funcall mode after starting process;
	when fi:XXX-image-name is a cons, it used to be funcall'd after
	the buffer was displayed.  It is now done before the process it
	created.
	File: subproc.el

	* changed default service number from 1123 (which is taken on some
	machines around here) to 6789.
	Files: tcplisp.el, spec.n, and ipc.cl

Thu Nov 17 12:15:57 1988  Kevin Layer  (layer at snooze)

	* two new files in this directory: ipc.cl and emacs.cl.  To ipc.cl
	added open-network-stream; to emacs.cl added list-all-completions
	which is used by M-TAB in lisp modes to complete a Common Lisp
	symbol!

	* Implemented fi:eval-in-lisp with a lisp eval server over a tcp
	socket and a new binding for M-TAB (fi:lisp-complete-symbol).
	Files: tcplisp.el and keys.el

	* new file clinit.cl, which contains the preferred way to startup
	the emacs/lisp interface.  It requires that EMACSLIBRARY be set to
	the root of the emacs tree (/usr/local/lib/emacs on almost all
	machines).  The following should be put in .login:

		setenv EMACSLIBRARY /usr/local/lib/emacs

	where /usr/local/lib/emacs is replaced by the appropriate
	pathname.

Thu Nov  3 17:08:21 1988  Kevin Layer  (layer at snooze)

	* filename completion as first input to a buffer now works.
	File: filec.el

	* new variable `fi:common-lisp-file-types', which is a list of the
	file extentions for files automatically put in
	fi:common-lisp-mode.  This variable should be set before
	fi/site-init is loaded.
	File: modes.el

>>>	* new variable `fi:rlogin-initial-input', which is the initial
	input to an rlogin process (after the first prompt is seen)
	File: rlogin.el

Thu Aug 11 21:24:19 1988  Kevin Layer  (layer at akbar)

	* fixed typo in fi:explicit-rlogin; TERM set to `dumb' before
	calling rlogin.
	File: rlogin.el

Sat Jul 23 12:09:56 1988  Kevin Layer  (layer at akbar)

	* fix another bug in directory tracking code (wasn't handling
	"pushd ...; ..." correctly.
	File: subproc.el

>>>	* added command completion to the file name completion code.  The
	entry point is now fi:shell-do-completion, and the following added
	to fi:subprocess-mode-hook will do the trick

	   (define-key (current-local-map) "\C-i" 'fi:shell-do-completion)

	File: filec.el

Thu Jul 21 23:47:31 1988  Kevin Layer  (layer at akbar)

	* directory tracking now handles "cd ... ; ..." and ignore
	(rightly so) "cd ... &" since the latter is executed in a
	subshell.
	File: subproc.el

	* M-. package hacking: do source-file on proper symbol by doing a
	find-symbol with `fi:package' as the package argument.
	File: ltags.el

Mon Jul 18 11:31:28 1988  Kevin Layer  (layer at akbar)

	* fixed fi:explicit-remote-common-lisp, which was calling
	fi::make-subprocess with too many arguments.
	File: subproc.el

Sat Jul 16 16:22:02 1988  Kevin Layer  (layer at akbar)

	* fix fi::get-default-symbol to send the proper package info to
	lisp.
	File: keys.el

============================== release 1.2 ==============================

  This release put in ~ftp/pub/fi/gnudist-1.2-tar.Z on ucbvax and ucbarpa

=========================================================================

Mon Jul 11 23:33:15 1988  Kevin Layer  (layer at akbar)

	* if there is a space in the tag name then don't query lisp.  In
	the future we may handle function specs, but for now only symbols
	are done correctly.
	File: ltags.el

	* fixed up fi::subprocess-hack-directory to handle `pushd +n' and
	`popd +n'.
	File: subproc.el

	* have filename completion do environment variable substitution
	File: filec.el

============================== release 1.1 ==============================

 This release on the Allegro CL 3.0 tapes for sun3 and sun4.

=========================================================================

Tue Jun  7 14:49:17 1988  Kevin Layer  (layer at akbar)

	* reading form for M-W from minibuffer caused an error.
	File: keys.el

Wed May 25 13:48:02 1988  Kevin Layer  (layer at akbar)

	* fix bug (that's been buzzing jdi) in
	fi:inferior-lisp-newline--RET in the middle of the buffer was
	copying the wrong input to the end.
	File: keys.el

	* fix bug in fi:set-associated-sublisp (" not backquoted in
	docstring)
	File: sublisp.el

============================== release 1.0 ==============================

 This release given to selected customers, and really constitutes a beta
 release of this software.

 The compressed tar file is in arpa:~ftp/pub/layer/dist-1.0.tar.Z

=========================================================================

Wed May 25 10:54:27 1988  Kevin Layer  (layer at akbar)

	* have M-sh-{M,F,D,A,W} commands read a form or symbol from the
	minibuffer if parsing the current buffer around the point fails.
	File: keys.el

Thu May 19 15:08:00 1988  Kevin Layer  (layer at akbar)

	* new function: fi:remove-all-temporary-lisp-transaction-files

	* minor bugs

Tue May 17 17:22:59 1988  Kevin Layer  (layer at akbar)

	* change in bindings:

	  in all subprocess modes:
	     C-m bound to fi:subprocess-input-region

	  in inferior-lisp mode:
	     RET is bound to a new version of fi:inferior-lisp-newline
	     which, when not at the end of the buffer, sends `input'
	     surrounding the point to the lisp process (also copying it to
	     the end of the buffer).

Mon May 16 15:10:02 1988  Kevin Layer  (layer at akbar)

	* fi::make-process and fi::make-tcp-connection use the default
	directory of the buffer from which they were invoked.

Fri May 13 10:27:46 1988  Kevin Layer  (layer at akbar)

	* frontdoor equivalents to  M-A, M-D, M-F, M-M commands.

Thu May 12 20:01:11 1988  Kevin Layer  (layer at akbar)

	* if fi:MUMBLE-image-name is not a string, then it is funcall'd to
	yield the real image name.

	* fi:pop-input and fi:push-input arguments are now optional.

	* new variable `fi:pop-to-sublisp-buffer-after-lisp-eval'
	controls whether or not `eval' commands cause one to be put in the
	sublisp buffer.

	* `common-lisp-mode' and `lisp-mode' are outta here.  To do this,
	though, we redefine `set-auto-mode' so we can capture the mode in
	the -*- line.

	* we now downcase the package name in the -*- line and the package
	can be a list, in which case we just take the `car' of it.

	* lisp tag stuff revamped so that package info doesn't hang
	around.

	* New function: fi:lisp-find-tag-other-window.

	* TAB, in subprocess modes, is no longer bound to
	fi:shell-file-name-completion, but rather is bound to
	lisp-indent-line.

	* fi:inferior-lisp-newline does a fi:inferior-lisp-send-sexp-input
	instead of a fi:subprocess-send-input when in the middle of a
	sublisp buffer.

Wed May 11 14:07:51 1988  Kevin Layer  (layer at akbar)

	* fi:lisp-find-tag, if given a prefix argument, will do a
	tags-loop-continue.

	* fi::get-cl-symbol renamed to fi::get-default-symbol, greatly
	repaired in the process.

	* fi:lisp-macroexpand and fi:lisp-walk now parse the sexp in the
	current buffer (at the point) like fi:inferior-lisp-send-sexp-input.

	* fi:set-associated-sublisp-buffer now sets the sublisp buffer
	for eval commands for all buffers which are in the same mode as
	the current buffer.

	* fi:subprocess-kill-output no longer insert `input' of
	"[output flushed]" into subprocess buffer.

	* `fi:inferior-lisp-send-sexp-input' and
	`fi:inferior-lisp-send-list-input' again have bindings in
	fi:inferior-common-lisp and fi:tcp-common-lisp modes (ESC RET and
	C-x RET, respectively).

	* The fi:another-MUMBLE functions have been removed.  In
	their place fi:MUMBLE has been enhanced and fi:explicit-MUMBLE has
	been added.

	* fi:explicit-MUMBLE-file-name and
	fi:explicit-MUMBLE-image-arguments are outta here.

	* the fi:default-MUMBLE-file-name and
	fi:default-MUMBLE-image-arguments are now fi:MUMBLE-image-name
	(Note: `file-name' -> `image-name') and fi:MUMBLE-image-arguments.

	* fi:remote-MUMBLE and fi:explicit-remote-MUMBLE functions added.

	* the TCP lisp functionality of fi:common-lisp was moved to
	fi:tcp-common and fi:explicit-tcp-common.

	* As you have undoubtedly noticed, the new form of the interface
	is:
	   fi:common-lisp
	   fi:remote-common-lisp
	   fi:tcp-common-lisp

	   fi:explicit-common-lisp
	   fi:explicit-remote-common-lisp
	   fi:explicit-tcp-common-lisp

	all of the above functions take a prefix argument, a number, which
	is the buffer number.  If the buffer number is > 0, then the buffer
	named `*MUMBLE-N*' is used.  The default is 1, in which case
	the `-1' is dropped.  If the buffer number is < 0, then N is
	chosen to be as small as possible while not pointing to an
	existing buffer with a running process.

	* the functions which start subprocesses and connections do not
	reset the input ring.

	* internal name change: tcp-lisp changed to tcp-common-lisp.
	There was mention of fi:tcp-lisp-mode in the document, and that is
	now fi:tcp-common-lisp-mode.

Mon May  9 17:43:22 1988  Kevin Layer  (layer at akbar)

	* remove all remaining indenting stuff from this interface
	(actually, move it to ../local/indent.el).
	File: modes.el

	* add binding for who-calls (M-C, fi:lisp-who-calls)
	File: modes.el
	Note: this change will not work with allegro CL 3.0 (who calls is
	      not in that lisp)

Fri Apr 29 15:37:18 1988  Kevin Layer  (layer at akbar)

	* removed from fi::make-process the "image name from the
	environment" hack.  Not even Harry, the original author of this
	feature, uses it.
	File: subproc.el

	* cl-indent is no longer loaded, since This interface has nothing
	to do with indenting.
	File: site-init.el

Mon Apr 25 15:14:59 1988  Kevin Layer  (layer at akbar)

	* fi::unix-domain-socket renamed to fi:unix-domain-socket
	(exported)
	File: subproc.el

	* fi:emacs-lisp-mode moved from file modes.el to elisp.el

	* The functions fi:inferior-lisp-send-sexp-input and 
	fi:inferior-lisp-send-list-input are no longer bound in inferior
	lisp modes, since they didn't work (at least I couldn't figure
	them out).  They were previously bound to the keys M-RET and
	C-x RET.

	* fi::package renamed fi:package (exported)

	* generic `lisp' mode removed (since nothing referred to it

	* lisp-mode and common-lisp-mode are the same as
	fi:common-lisp-mode (this is for -*- mode defs)

