Added the beginnings of proper Eshell documentation.

[bpt/emacs.git] / man / eshell.texi

\input texinfo  @c -*-texinfo-*-

@c "@(#)$Name:  $:$Id: $"

@c Documentation for Eshell: The Emacs Shell.
@c Copyright (C) 1999-2000  Free Software Foundation, Inc.

@c This file is part of GNU Emacs

@c GNU Emacs is free software; you can redistribute it and/or modify it
@c under the terms of the GNU General Public License as published by the
@c Free Software Foundation; either version 2 of the License, or (at
@c your option) any later version.

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

@c You should have received a copy of the GNU General Public License
@c along with Eshell; see the file COPYING.  If not, write to the Free
@c Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.

@c %**start of header
@setfilename ../info/eshell
@settitle Eshell: The Emacs Shell
@c %**end of header

@c @dircategory Emacs
@direntry
* Eshell: (eshell).     A command shell implemented in Emacs Lisp.
@end direntry
@setchapternewpage on
     
@ifinfo
Copyright @copyright{} 1999-2000  Free Software Foundation, Inc.

Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.

@ignore
Permission is granted to process this file through TeX and print the
results, provided the printed document carries copying permission notice
identical to this one except for the removal of this paragraph (this
paragraph not being relevant to the printed manual).
@end ignore

Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided also that the
section entitled ``GNU General Public License'' is included exactly as
in the original, and provided that the entire resulting derived work is
distributed under the terms of a permission notice identical to this
one.

Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that the section entitled ``GNU General Public License'' and this
permission notice may be included in translations approved by the Free
Software Foundation instead of in the original English.
@end ifinfo

@synindex vr fn
@c The titlepage section does not appear in the Info file.
@titlepage
@sp 4
@c The title is printed in a large font.
@center @titlefont{User's Guide}
@sp
@center @titlefont{to}
@sp
@center @titlefont{Eshell: The Emacs Shell}
@ignore
@sp 2
@center release 2.3.2
@c -release-
@end ignore
@sp 3
@center John Wiegley
@c -date-

@c  The following two commands start the copyright page for the printed
@c  manual.  This will not appear in the Info file.
@page
@vskip 0pt plus 1filll
Copyright @copyright{} 1999-2000  Free Software Foundation, Inc.

Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.

Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided also that the
section entitled ``GNU General Public License'' is included exactly as
in the original, and provided that the entire resulting derived work is
distributed under the terms of a permission notice identical to this
one.

Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that the section entitled ``GNU General Public License'' and this
permission notice may be included in translations approved by the Free
Software Foundation instead of in the original English.
@end titlepage

@c ================================================================
@c                   The real text starts here
@c ================================================================

@node Top, What is Eshell?, (dir), (dir)
@ifinfo
@top Eshell

This manual documents Eshell, a shell-like command interpretor
implemented entirely in Emacs Lisp.  It invokes no external processes
beyond those requested by the user.  It is intended to be a functional
replacement for command shells such as @command{bash}, @command{zsh},
@command{rc}, @command{4dos}; since Emacs itself is capable of handling
most of the tasks accomplished by such tools.
@c This manual is updated to release 2.3.2 of Eshell.
@end ifinfo

@menu
* What is Eshell?::             A brief introduction to the Emacs Shell.
* Bugs and ideas::              
@end menu

@node What is Eshell?, Bugs and ideas, Top, Top
@chapter What is Eshell?
@cindex What is Eshell?

Eshell is a command shell written using Emacs Lisp.  All of what it does
it uses Emacs' facilities to do.  This means Eshell is as portable as
Emacs itself.  It also means that cooperation with other Lisp code is
natural and seamless.

So what is a command shell?  To properly understand the role of a shell,
it's necessary to visualize what a computer does for you.  Basically, a
computer is a tool; in order to use that tool, you must tell it what to
do---or give it ``commands''.  These commands take many forms, such as
clicking with a mouse on certain parts of the screen.  But that is only
one form of command input.

By far the most versatile way to express what you want the computer to
do is using an abbreviated language, called script.  In script, instead
of telling the computer, ``list my files, please'', we write just
``list''.  In fact, this command is so commonly used that we abbreviate
it to ``ls''.  Typing @code{ls} in a command shell is a script way of
telling the computer to list your files.  This is comparable to viewing
the contents of a folder using a graphical display.

The real flexibility is apparent only when you realize that there are
many, many ways to list your files.  Perhaps you want them sorted by
name, or sorted by date, or in reverse order, or grouped by type.  Most
graphical browsers have simple ways to express this.  But what about
showing only a few files, or only files that meet a certain criteria?
In very complex and specific situations, the request becomes too
difficult to express with a mouse.  It is just these kinds of requests
that are solvable using a command shell.

For example, what if you want to list every Word file on your hard
drive, larger than 100 kilobytes in size, and which hasn't been looked
at in over six months?  That is a good candidate list for deletion, when
you go to clean up your hard drive.  But have you ever tried asking your
computer for such a list?  There is no way to do it!  At least, not
without using a command shell.

So the role of a command shell is to give you more control over what
your computer does for you.  Not everyone needs this amount of control,
and it does come at a cost: Learning the necessary script commands to
express what you want done.  A complicated query, such as the example
above, takes time to learn.  But if you find yourself using your
computer frequently enough, it is more than worthwhile in the long run.
Any tool you use often deserves your time in learning to master it.
@footnote{For the understandably curious, here is what that command
looks like: But don't let it fool you; once you know what's going on,
it's easier than it looks: @code{ls -lt **/*.doc(Lk+50aM+5)}.}

As of Emacs 21, Eshell is part of the standard Emacs distribution.

@menu
* Contributors to Eshell::      
* Installation::                
@end menu

@node Contributors to Eshell, Installation, What is Eshell?, What is Eshell?
@section Contributors to Eshell
@cindex Contributors
@cindex Authors

Contributions to Eshell are welcome.  I have limited time to work on
this project, but I will gladly add any code you contribute to me to
this package.

The following persons have made contributions to Eshell.

@itemize @bullet
@item
Eli Zaretskii made it possible for Eshell to run without requiring
asynchronous subprocess support.  This is important for MS-DOS, which
does not have such support.@refill

@item
Miles Bader contributed many fixes during the port to Emacs 21.@refill

@item
Stefan Monnier fixed the things which bothered him, which of course made
things better for all.@refill

@item
Gerd Moellmann also helped to contribute bug fixes during the initial
integration with Emacs 21.@refill

@item
Alex Schroeder contributed code for interactively querying the user
before overwriting files.@refill

@item
Sudish Joseph helped with some XEmacs compatibility issues.@refill

@end itemize

Apart from these, a lot of people have sent suggestions, ideas,
requests, bug reports and encouragement.  Thanks a lot!  Without you
there would be no new releases of Eshell.

@node Installation,  , Contributors to Eshell, What is Eshell?
@section Installation
@cindex Installation

As mentioned above, Eshell comes preinstalled since Emacs 21.  If you're
using Emacs 20.4 or later, or XEmacs 21, you can download the most
recent version of Eshell from
@url{http://www.emacs.org/~johnw/Emacs/eshell.tar.gz}.

If you are using Emacs 21, please skip this section.

@subsection Short Form

Here's exactly what to do, with no explanation why:

@enumerate
@item @samp{M-x load-file RET eshell-auto.el RET}
@item @samp{ESC : (add-to-list 'load-path "<path where Eshell resides>") RET}
@item @samp{ESC : (add-to-list 'load-path "<path where Pcomplete resides>") RET}
@item @samp{M-x eshell RET}

You should see a version banner displayed.

@item @samp{ls RET}

Confirm that you see a file listing.

@item @samp{eshell-test RET}

Confirm that everything runs correctly.  Use `M-x eshell-report-bug' if
not.

@item @samp{cd $@{dirname (locate-library "eshell-auto")@} RET}
@item @samp{find-file Makefile RET}
@item Edit the Makefile to reflect your site.
@item @samp{M-x eshell RET}
@item @samp{make install RET}
@item @samp{find-file $user-init-file RET}
@item Add the following lines to your @file{.emacs} file:

@example
(add-to-list 'load-path "<directory where you install Eshell>")
(load "eshell-auto")
@end example

@item @samp{M-x eshell RET}
@item @samp{customize-option #'eshell-modules-list RET}
@item Select the extension modules you prefer.
@item Restart Emacs!
@item @samp{M-x info RET m Eshell RET}

Read the manual and enjoy!
@end enumerate

@subsection Long Form

@enumerate
@item
Before building and installing Eshell, it is important to test that it
will work properly on your system.  To do this, first load
@file{eshell-auto}, which will define certain autoloads required to run
Eshell.  This can be done using the command @kbd{M-x load-file}, and
then selecting the file @file{eshell-auto.el}.

@item
In order for Emacs to find Eshell's files, the Eshell directory must be
added to the @code{load-path} variable.  This can be done within Emacs by
typing:

@example
ESC : (add-to-list 'load-path "<path where Eshell resides>") RET
ESC : (add-to-list 'load-path "<path where Pcomplete resides>") RET
@end example

@item
Start Eshell from the distributed sources, using default settings, by
typing @kbd{M-x eshell}.

@item
Verify that Eshell is functional by typing @command{ls} followed by
@kbd{RET}.  You should have already seen a version banner announcing the
version number of this release, followed by a prompt.

@item
Run the test suite by typing @command{eshell-test} followed by @kbd{RET}
in the Eshell buffer.  It is important that Emacs be left alone while
the tests are running, since extraneous command input may cause some of
the tests to fail (they were never intended to run in the background).
If all of the tests pass, Eshell should work just fine on your system.
If any of the tests fail, please send e-mail to the Eshell maintainer
using the command @kbd{M-x eshell-report-bug}.

@item
Edit the file @file{Makefile} in the directory containing the Eshell
sources to reflect the location of certain Emacs dircetories at your
site.  The only things you really have to change are the definitions of
@code{lispdir} and @code{infodir}.  The elisp files will be copied to
@code{lispdir}, and the info file to @code{infodir}.

@item
Type @code{make install} in the directory containing the Eshell sources.
This will byte-compile all of the @file{.el} files and copy both the
source and compiled versions to the directories specified in the
previous step.  It will also copy the info file, and add a corresponding
entry to your @file{dir} file----if @file{install-info} can be found.

If you only want to create the compiled elisp files, but don't want to
install them, you can type just @command{make} instead.

@item
Add the directory into which Eshell was installed to your
@code{load-path} variable.  This can be done by adding the following
line to your @file{.emacs} file:

@example
(add-to-list 'load-path "/usr/local/share/emacs/site-lisp/eshell")
@end example

The actual directory on your system may differ.

@item
To install Eshell privately, edit your @file{.emacs} file; to install
Eshell site-wide, edit the file @file{site-start.el} in your
@file{site-lisp} directory (usually
@file{/usr/local/share/emacs/site-lisp} or something similar).  In
either case enter the following line into the appropriate file:

@example
(load "eshell-auto")
@end example

@item
Restart Emacs.  After restarting, customize the variable
@code{eshell-modules-list}.  This variable selects which Eshell
extension modules you want to use.  You will find documentation on each
of those modules in the Info manual.

@end enumerate

If you have @TeX{} installed at your site, you can make a typeset manual
from @file{eshell.texi}.

@enumerate
@item
Run @TeX{} by typing @samp{texi2dvi eshell.texi}.
@item
Convert the resulting device independent file @file{eshell.dvi} to a
form which your printer can output and print it.  If you have a
postscript printer there is a program, @code{dvi2ps}, which does.  There
is also a program which comes together with @TeX{}, @code{dvips}, which
you can use.
@end enumerate

@c @node Forming commands, Known problems, What is Eshell?, Top
@c @chapter Forming commands

@c A command shell is nothing more than a place to enter commands.

@c What is a command?

@c A command is piece of ``script''---or special shorthand
@c language---that the computer can understand.

@c What does script look like?

@c Script is an extremely simplified language.  Oddly enough, this
@c actually makes it look more complicated than it is.  Whereas normal
@c languages can use many different embellishments, the form of a script
@c command is always: a command verb, following by its arguments.

@c A verb?  Arguments?

@c The verb is the thing you want your computer to do.  There are a set
@c number of verbs, although this number is quite large.  On my
@c computer, it reaches almost 1400 in number!  But of course, only a
@c handful of these are necessary most of the time.

@c Sometimes, the verb is all that's necessary.  A verb is always a
@c single word, usually related to the task it will perform.
@c @command{reboot} is a good example.  Entering that will cause your
@c computer to reboot, assuming you have sufficient privileges.

@c Other verbs need more information.  These are usually very capable of
@c verbs, but they must be told more specifically what to do.  This
@c extra information is given in the form of arguments.  Arguments are
@c also words, that appear after the verb.  For example, @command{echo}
@c is a command verb that will print back to you whatever you say.
@c @command{echo} requires a set of arguments, to know what you want it
@c to echo!  So a proper use of echo might look like:

@c @example
@c echo This is an example of using echo!
@c @end example

@c This command would result in the computer printing back to you,
@c ``This is an example of using echo!''.  Pretty easy, no?

@c Although commands are always simple words, arguments can take
@c different forms.  There are textual arguments, numeric arguments,
@c even Lisp arguments.  Distinguishing among these different types of
@c arguments requires some special typing, because the computer needs
@c very specific directions to understand what you mean.

@node Bugs and ideas,  , What is Eshell?, Top
@chapter Bugs and ideas
@cindex Reporting bugs and ideas
@cindex Bugs, how to report them
@cindex Author, how to reach
@cindex Email to the author
@cindex Known bugs
@cindex Bugs, known
@cindex FAQ
@cindex Problems, list of common

If you find a bug or misfeature, don't hesitate to let me know!  Send
email to @samp{johnw@@gnu.org}.  Feature requests should also be sent
there.  I prefer discussing one thing at a time.  If you find several
unrelated bugs, please report them separately.

If you have ideas for improvements, or if you have written some
extensions to this package, I would like to hear from you.  I hope you
find this package useful!

@menu
* Known problems::              
@end menu

@node Known problems,  , Bugs and ideas, Bugs and ideas
@section Known problems

Below is a partial list of currently known problems with Eshell version
2.3.2, which is the version distribution with Emacs 21.1.

@table @asis
@item @samp{for i in 1 2 3 @{ grep -q a b && *echo has it @} | wc -l} fails

In fact, piping to a process from a looping construct doesn't work in
general.  If I change the call to @code{eshell-copy-handles} in
@code{eshell-rewrite-for-command} to use @code{eshell-protect}, it seems
to work, but the output occurs after the prompt is displayed.  The whole
structured command thing is too complicated at present.

@item Error with @command{bc} in @code{eshell-test}

On some XEmacs system, the subprocess interaction test fails
inexplicably, since @command{bc} works fine at the command prompt.

@item @command{ls} in remote directories sometimes fails

For XEmacs users, using @command{ls} in a remote directory sometimes
fails.  The reason why has not yet been found.

@item Eshell does not delete @file{*Help*} buffers in XEmacs 21.1.8+

In XEmacs 21.1.8, the @file{*Help*} buffer has been renamed such that
multiple instances of the @file{*Help*} buffer can exist.

@item Pcomplete sometimes gets stuck

When @kbd{TAB}, no completions appear, even though the directory has
them.  This behavior is rare.

@item @samp{grep python $<rpm -qa>} doesn't work, but using @samp{*grep} does

This happens because the @code{grep} Lisp function returns immediately,
and then the asynchronous @command{grep} process expects to examine the
temporary file, which has since been deleted.

@item Problem with C-r repeating text

If the text @emph{before point} reads "./run", and you type @kbd{C-r r u
n}, it will repeat the line for every character typed.

@item Backspace doesn't scroll back after continuing (in smart mode)

Hitting space during a process invocation, such as @command{make}, will
cause it to track the bottom of the output; but backspace no longer
scrolls back.

@item It's not possible to fully @code{unload-feature} Eshell

@item Menu support was removed, but never put back

@item Using C-p and C-n with rebind gets into a locked state

This happened a few times in Emacs 21, but has been unreproducable
since.

@item If an interactive process is currently running, @kbd{M-!} doesn't work

@item Use a timer instead of @code{sleep-for} when killing child processes

@item Piping to a Lisp function is not supported

Make it so that the Lisp command on the right of the pipe is repeatedly
called with the input strings as arguments.  This will require changing
eshell-do-pipeline to handle non-process targets.

@item Input redirection is not supported

See the entry above.

@c @item problem running "less" without argument on Windows
@c Before running telnet, I noticed that 'less' (for example) was already
@c configured as a visual command. So I invoked it from eshell to see what
@c would happen.
@c 
@c Here's the result in the eshell buffer:
@c 
@c      Spawning child process: invalid argument
@c 
@c Also a new 'less' buffer was created with nothing in it .. (presumably this
@c holds the output of less)
@c 
@c If I run 'less.exe' from the eshell command line, I get the output I expect
@c simply written to the buffer.
@c 
@c Note that I'm using FSF NT-Emacs 20.6.1 on Win2000. The term.el package and
@c the supplied shell both seem to use the 'cmdproxy' program to run things
@c like shells.
@c @item implement -r, -n and -s switches for cp
@c @item Make M-5 eshell -> switch to *eshell<5>*, creating it if need be
@c @item mv DIR FILE.tar does not remove directories
@c This is because the tar option --remove-files doesn't do so.  Should
@c it be Eshell's job?
@c @item Write an article about Eshell for the LinuxWorld journal.
@c @item bind standard-output and standard-error, so that if a Lisp function
@c   calls `print', everything will happen as it should (albeit slowly)
@c @item when the extension modules fail to load, cd / gives a Lisp error
@c @item if a globbing patterns returns only one match, should it still be a
@c   list?
@c @item make sure that the syntax table correctly in eshell mode
@c So that M-DEL acts in a predictable manner, etc.
@c @item allow all Eshell buffers to share the same history and list-dir
@c @item error with script commands and outputting to /dev/null
@c If a script file, somewhere in the middle, does a "> /dev/null",
@c output from all subsequent commands will be swallowed
@c @item split up parsing of the text after a $ in eshell-var
@c Similar to way that eshell-arg is structured.  Then add parsing of
@c $[?\n]
@c @item after pressing M-RET, redisplay before running the next command
@c @item argument predicates and modifiers should work anywhere in a path
@c   /usr/local/src/editors/vim $ vi **/CVS(/)/Root(.)
@c   Invalid regexp: "Unmatched ( or \\("
@c 
@c with zsh, the glob above expands to all files named Root in
@c directories named CVS.
@c @item typing "echo ${locate locate}/bin<tab>" results in a Lisp error
@c Perhaps it should interpolate all permutations, and make that the
@c globbing result, since otherwise hitting return here will result in
@c "(list of filenames)/bin", which is never very valuable.  Thus, one
@c could cat only c backup files by using "ls ${identity *.c}~".  In that
@c case, having an alias command name `glob' for `identity' would be
@c useful
@c @item for XEmacs on Win32, fix `file-name-all-completions'
@c Make sure it returns directory names terminated by
@c `directory-sep-char' (which is initialized to be ?/), rather than
@c backslash
@c @item once symbolic mode is supported for umask, implement chmod in Lisp
@c @item create `eshell-expand-file-name'
@c Which uses a data table to transform things like "~+", "...", etc
@c @item abstract `eshell-smart.el' into `smart-scroll.el'
@c It only really needs: to be hooked onto the output filter and the
@c pre-command hook, and to have the input-end and input-start markers.
@c And to know whether the last output group was "successful".
@c @item allow for fully persisting the state of Eshell
@c vars, history, buffer, input, dir stack, etc.
@c @item implement D in the predicate list
@c It means that files beginning with a dot should be included in the
@c glob match
@c @item a comma in a predicate list means OR
@c @item error if a glob doesn't expand due to a predicate
@c An error should be generated only if `eshell-error-if-no-glob' is
@c non-nil
@c @item the following doesn't cause an indent-according-to-mode to occur
@c (+ RET SPC TAB
@c @item create `eshell-auto-accumulate-list'
@c It is a list of commands for which, if the user presses RET, the text
@c gets staged as the next Eshell command, rather than being sent to the
@c current interactive
@c @item display file and line number if an error occurs in a script
@c @item wait doesn't work with process ids at the moment
@c @item enable the direct-to-process input code in eshell-term.el
@c @item problem with repeating "echo ${find /tmp}"
@c With smart display active, if I hold down RET, after a while it can't
@c keep up anymore and starts outputting blank lines.  It only happens if
@c an asynchronous process is involved...
@c 
@c I think the problem is that `eshell-send-input' is resetting the input
@c target location, so that if the asynchronous process is not done by
@c the time the next RET is received, the input processor thinks that the
@c input is meant for the process; which, because smart display is
@c enabled, will be the text of the last command line!  That is a bug in
@c itself.
@c 
@c In holding down RET while an asynchronous process is running, there
@c will be a point in between termination of the process, and the running
@c of eshell-post-command-hook, which would cause `eshell-send-input' to
@c call `eshell-copy-old-input', and then process that text as a command
@c to be run after the process.  Perhaps there should be a way of killing
@c pending input between the death of the process, and the
@c post-command-hook.
@c @item allow for a more aggressive smart display mode
@c Perhaps toggled by a command, that makes each output block a smart
@c display block
@c @item create more meta variables
@c $! -- the reason for the failure of the last disk command, or the text
@c       of the last Lisp error
@c 
@c $= -- a special associate array, which can take references of the form
@c       $=[REGEXP].  It also indexes into the directory ring
@c @item eshell scripts can't execute in the background
@c @item support zsh's "Parameter Expansion" syntax, i.e. ${NAME:-VAL}
@c @item write an `info' alias that can take arguments
@c So that the user can enter "info chmod"
@c @item split off more generic code from Eshell
@c parse-args.el --- parse a list of arguments
@c interpolate.el --- interpolate $variable $(lisp)... references
@c interp.el --- find which interpretor to run a script with
@c sh-ring.el --- extend ring.el for persistant, searchable history
@c zsh-glob.el --- zsh-style globbing and predicate/modifiers
@c smartdisp.el --- smart scrolling in input buffers
@c egetopt.el --- `eshell-eval-using-options'
@c prompt.el --- code for outputting and navigating prompts
@c cmd-rebind.el --- rebind certain keys in the input text
@c unix.el --- provides Lispish UNIX command, such as unix-rm, etc.
@c emacs-ls.el --- implementation of ls in Emacs Lisp
@c texidoc.el
@c pushd.el --- implementation of pushd/popd in Lisp
@c interface.el -- a mode for reading command-line input from the user
@c @item create a mode `eshell-browse'
@c It would treat the Eshell buffer as a outline.  Collapsing the outline
@c hides all of the output text.  Collapsing again would show only the
@c first command run in each directory
@c @item look through the Korn Shell book for feature ideas
@c @item allow other version of a file to be referenced by "file{rev}"
@c This would be expanded by `eshell-expand-file-name'
@c @item print "You have new mail" when the "Mail" icon gets turned on
@c @item implement M-|
@c @item implement input redirection
@c If it's a lisp function, input redirection implies "xargs" (in a
@c way..).  And if input redirection is added, don't forget to update the
@c file-name-quote-list, and the delimiter list.
@c @item allow #<WORD ARG> to be a generic syntax
@c With the handling of "word" specified by an `eshell-special-alist'.
@c @item in `eval-using-options', have a :complete tag
@c It would be used to provide completion rules for that command.  Then
@c the macro will automagically define the completion function
@c @item for `eshell-command-on-region', redirections apply to the result
@c So that "+ > 'blah" will cause the result of the `+' (using input from
@c the current region) to be inserting in the symbol `blah'.
@c 
@c If a disk command is being invoked, the input is sent as standard
@c input, as if a "cat <region> |" were invoked.
@c 
@c If a lisp command, or an alias, is invoked, then: if the line has no
@c ^J characters, it is divided by whitespace and passed as arguments to
@c the lisp function.  Otherwise, it is divided at the ^J characters.
@c Thus, invoking `+' on a series of numbers will add them; `min' would
@c display the smallest figure.
@c @item write `eshell-script-mode' as a minor mode
@c It would provide syntax, abbrev, highlighting and indenting support
@c like emacs-lisp-mode + shell-mode.
@c @item in the history mechanism, finish bash-style support
@c For !n, !#, !:%, and !:1- as separate from !:1*
@c @item support the -n command line option for "history"
@c @item implement `fc'
@c @item specifying a frame as a redirection target implies point's buffer
@c @item implement ">FUNC-OR-FUNC-LIST"
@c This would allow for an "output translator", that takes a function to
@c modify output with, and the target.  Devise a syntax that words well
@c with pipes, and can accomodate multiple functions (i.e.,">'(upcase
@c regexp-quote)" or ">'upcase").
@c @item allow Eshell to read/write to/from standard input and output
@c This would be optional, rather than always using the Eshell buffer.
@c This would allow it to be run from the command line.
@c @item write a "help" command
@c It could even call subcommands with "--help" (or "-h" or "/?").
@c @item implement stty
@c @item support rc's matching operator, "~ (list) regexp"
@c @item implement "bg" and "fg" to edit `eshell-process-list'
@c Using "bg" on a process that is already in the background does
@c nothing.  Specifying redirection targets replaces (or adds) to the
@c list current being used.
@c @item have "jobs" print only the processes for the current eshell
@c @item how do I discover that a background process has requested input?
@c @item support 2>&1 and >& and 2> and |&
@c The syntax table for parsing these should be customizable, such that
@c the user could change it to use rc syntax: >[2=1].
@c @item allow $_[-1], which reads the last element of the array, etc.
@c @item make $x[*] equal to listing out the full contents of x
@c Return them as a list, so that $_[*] is all the arguments of the last
@c command.
@c @item move ANSI code handling from `term' into `eshell-term'
@c And make it possible for the user to send char-by-char to the
@c underlying process.  Ultimately, I should be able to move away from
@c using term.el altogether, since everything but the ANSI code handling
@c is already part of Eshell.  Then, things would work correctly on Win32
@c as well (which doesn't have "/bin/sh", though term tries to use it)
@c @item have other shell spawning commands be visual
@c Make (su, bash, telnet, rlogin, rsh, etc.) be part of
@c `eshell-visual-commands'.  The only exception is if rsh/su/bash are
@c simply being used to invoke a single command.  Then, it should be
@c based on what that command is.
@c @item create an alias "open"
@c This will search for some way to open its argument (similar to opening
@c a file in the Windows Explorer).  Perhaps using ffap...
@c @item alias "read" to be the same as "open", except read-only
@c @item write a "tail -f" alias which does a view-file
@c I.e., it moves point to the end of the buffer, and then turns on
@c auto-revert mode in that buffer at frequent intervals -- and a head
@c alias which assums an upper limit of `eshell-maximum-line-length'
@c characters per line.
@c @item make dgrep load dired, mark everything, then execute the A binding
@c @item write emsh.c
@c It just runs Emacs with the appropriate arguments to invoke eshell.
@c That way, it could be listed as a login shell.
@c @item use an intangible PS2 string for multi-line input prompts
@c @item auto-detect when a command is visual, by checking TERMCAP usage
@c @item First keypress after M-x watson triggers `eshell-send-input'
@c @item Emacs 20.3: Figure out why pcomplete won't make
@c @item Make / electric
@c So that it automatically expands and corrects pathnames.  Or make
@c pathname completion for pcomplete auto-expand "/u/i/std<TAB>" to
@c "/usr/include/std<TAB>".
@c @item Write pushd/popd out to disk along with last-dir-ring
@c @item add options to eshell/cat which would cause it to sort and uniq
@c @item implement in Lisp: wc.  Also count sentences, paragraphs, pages.
@c @item once piping is added, implement sort and uniq
@c @item implement touch
@c @item implement epatch
@c Calls ediff-patch-file, or ediff-patch-buffer, depending on its
@c argument.
@c @item have an option for bringing up ls -l result in a dired buffer
@c @item write a version of xargs that's based on command rewriting
@c find X | xargs Y == Y ${find X}.  Maybe I could change
@c eshell-do-pipelines to perform this on-thy-fly rewriting.
@c @item implement head and tail in Lisp
@c @item write an alias for less and more that brings up a view buffer
@c Such that they can press SPC and DEL, and then q to return to eshell.
@c The more command would be equivalent to: X > #<buffer Y>; view-buffer
@c #<buffer Y>
@c @item differentiate between aliases and functions
@c Allow for a bash-compatible syntax, such as:
@c 
@c   alias arg=blah
@c   function arg () { blah $* }
@c @item find the various references to shell-mode within Emacs
@c And add support for Eshell there, since now Eshell is going to be part
@c of Emacs.
@c @item permit umask to be set on a cp target during the cp command
@c @item if the first thing that I do after I enter Emacs
@c is to run eshell-command and invoke ls, and then I use M-x eshell, it
@c doesn't show me anything.
@c @item M-RET during a long command doesn't quite work
@c Since it keeps the cursor up where the command was invoked.
@end table

@unnumbered Function and Variable Index

@printindex fn

@unnumbered Concept Index

@printindex cp

@unnumbered Key Index

@printindex ky

@setchapternewpage odd
@summarycontents
@contents
@bye