\input texinfo   @c -*-texinfo-*-
@c %**start of header
@setfilename esh.info
@settitle esh, the easy shell.
@c %**end of header

@ifinfo
@dircategory Shells
@direntry
* esh: (esh).                     the easy shell.
@end direntry
@end ifinfo

@titlepage
@sp 10
@center @titlefont{esh, the easy shell.}
@end titlepage

@ifinfo
@node Top, Rationale, (dir), (dir)
@top esh, the easy shell.
     
@code{esh} is a new, lightweight Unix shell designed for maximum simplicity and
generality.

@end ifinfo
     
@menu
* Rationale::               Why write it?
* Overview::                What is it?
* Details::                 In-depth information on programming the shell.
* Differences from Scheme:: How this shell differs from Scheme.
* Differences from sh::     How this shell differs from sh-based shells.
* Command Index::           All the builtin commands.
* Concept Index::           Concepts referred to in this manual.
@end menu

@node     Rationale, Overview, Top, Top
@cindex Rationale
@chapter Rationale

@code{esh} was primarily written out of a need for a simple and lightweight shell
for Unix. As such, it deviates completely from all of the traditional
shells, opting instead for a Lisp-like syntax. This allows exceptionally 
small size, both in terms of lines of code and memory consumption, while
retaining remarkable flexibility and programmability.

@node     Overview, Details, Rationale, Top
@cindex Overview
@cindex Introduction
@cindex Getting started
@chapter Overview

@ifinfo
An overview of the basic information about the shell.
@end ifinfo

@menu
* Starting esh::     How to launch it.
* Interaction::        Simple interaction with the shell.
* Simple Programming:: Writing simple programs.
* Non-interactive::    Running the shell non-interactively.
* Job Control::        How to use job control.
* Basic Usage::        Some essential commands.
@end menu

@node Starting esh, Interaction,     , Overview
@cindex Starting @code{esh}
@cindex Command line parameters
@cindex Launching @code{esh}
@cindex .eshrc
@cindex eshrc
@section Starting @code{esh}

To start the shell, simply type @code{esh}. There are no command-line 
parameters. However, since all the command-line arguments are available
to the shell programmer, it is a simple matter to write your own argument
processing routine and place it in your @code{.eshrc} file.

Both @code{/etc/eshrc} and the @code{.eshrc} in your home directory will
be run by the shell on startup, if they exist.

@node Interaction, Simple Programming, Starting esh, Overview
@cindex Interaction
@cindex Running commands
@cindex Executing files
@cindex Pipes and redirection
@section Interaction

If the shell is running interactively, you will get a prompt, @code{$ } by
default. Since the shell uses the readline library, you can use line-editing
keystrokes, the history buffer, filename completion, and the rest of the 
goodies provided by the readline library.

To run a single executable, simply type it in as you would in any other
shell. Example: @code{/usr/games/fortune -m Unix}.

To run a pipeline, type several commands separated by commas or vertical
lines. Example:
@code{cat /var/log/messages, grep pppd, grep "IP address", xmessage -file -}.

To run a pipeline with file redirection, place the redirection symbols and the
filenames after the pipeline. Example:
@code{sort, uniq < names.raw > names.sorted}. In this case, the contents of
@code{names.raw} are used as the standard input of @code{sort}, and the output 
of the whole pipeline is saved to @code{names.sorted}. The order in which 
the redirection symbols are given does not matter, but a valid filename 
should always follow a redirection symbol.

Note one major pitfall: @code{cd}, @code{bg}, @code{fg}, etc., are all
builtin shell commands! For info on running shell commands, see the
next section. (@pxref{Simple Programming})

@node Simple Programming, Non-interactive, Interaction, Overview
@cindex Simple programming
@cindex Syntax, basic
@cindex Basics of shell programming
@section Simple Programming

All shell commands are specified using a parenthetical syntax
very similar to that of Lisp or Scheme. Example: Use @code{(cd /var)} to change
the current directory. In this case, @code{cd} is the name of the command,
and @code{/var} is the argument. Multiple arguments to the same 
command are separated by white space. Example: @code{(set HOME /home/ivan)}.
Builtin commands never accept options in the
style of compiled programs. Also note that quotes around string values are
not required, though they are certainly allowed at any time if you wish
to escape special characters. (In fact, using quotes is the only way to
do this. Backslash escape sequences are not recognized.) Example: 
@code{(cd "funny,directory name()")}. In this case, the whole string inside
the quotes is used as the argument verbatim. Single quotes and double
quotes are equivalent.

If a command returns anything, the returned data will be printed
by the shell. Example: @code{(get HOME)} @result{} @code{/home/ivan}.

Note that you can certainly use the return values of one command as input
to another. Example: @code{(cd (get HOME))}

If you want to pass a list instead of a string as an argument to a command,
quote the list with a tilde. Example: 
@code{(run-simple ~(ls --color=yes))}. Here, the @code{run} command is given
a list as an argument. 

Note that a sublist of a quoted list is also quoted; that is, 
@code{~(1.1 (2.1 2.2) 1.2)} is a list of three elements, the first and last
are strings and the middle element is a list. 

@xref{Basic Usage}, for an explanation of what @code{cd}, @code{get}, 
@code{set}, and @code{run} do.

@node Non-interactive, Job Control, Simple Programming, Overview
@cindex Non-interactive
@cindex Files, as shell scripts
@section Non-interactive

If the shell is started non-interactively -- for example, when the standard
input to the shell is a file -- the shell will only accept commands.
To run executables and construct pipes, simply use the @code{run} 
command.
(@pxref{Basic Usage})

@node Job Control, Basic Usage, Non-interactive, Overview
@cindex Job control
@cindex Foreground
@cindex Background
@cindex Stopping jobs
@section Job Control

If the shell is running interactively, you will have access to job control
primitives. 

To stop a job in the foreground, simply type @kbd{C-z}, as in other shells.

To bring the last known job into foreground or background, issue the @code{fg}
or @code{bg} command, respectively, without arguments. 
(@pxref{Simple Programming})

To get a listing of all the currently known jobs, issue the @code{jobs} 
command, also without arguments.

If the @code{fg} or @code{bg} command is given a single numeric argument, it
will act on the job number specified by that argument. To find out the job
number for a command, simply issue the @code{jobs} command. 

Note: the job number is @emph{not} the PID!


@node Basic Usage, Syntax, Job Control, Overview
@cindex Command list
@cindex Basic commands
@cindex Builtin command list, basic
@findex jobs
@findex cd
@findex fg
@findex bg
@findex run
@findex run-simple
@findex get
@findex set
@findex env
@findex print
@findex +
@findex -
@findex *
@findex /
@findex if
@findex list
@findex eval
@findex =
@section Basic Usage

@itemize @bullet
@item 
@code{(cd [string])} Change the current directory to the given
argument. If the directory name is a single dash, the previous directory
will be used. (i.e. the value of the environment variable @code{OLDPWD})
If the command is called without any arguments, change to the home directory.

@item
@code{(jobs)} List the current jobs.

@item
@code{(fg [number])} If the optional numeric argument is given, bring
the job number given by the argument into the foreground. If this argument
is omitted, bring the first job into the foreground.

@item
@code{(bg [number])} Same as @code{fg}, but puts a job into the background.

@item
@code{(run <bool> <file> <file> <list>...)} Run a pipeline.
The first argument should be @code{false} if you want the command to be
brought into the foreground, or @code{true} if otherwise.
The next two arguments are the input and output redirection files, 
respectively. You can use @code{(standard)} for standard input/output. 
The rest of the arguments are the commands you want to run,
given as lists.
If the command was run in the background, then the return value
is the PID of the last executable in the pipeline. Otherwise, the return
value is the exit status of the pipeline.

Examples: 
@code{(run (false) (standard) (standard) ~(/usr/games/fortune -m Unix))} 
is equivalent to
@code{/usr/games/fortune -m Unix}.
@code{(run (false) (file-open file "names.raw") (file-open file "names.sorted") ~(sort) ~(uniq))} is equivalent to
@code{sort, uniq < names.raw > names.sorted}.

@item
@code{(run-simple <list>...)} Equivalent to 
@code{(run (false) (standard) (standard) ...)}.
This command returns the exit status of the pipeline.

@item
@code{(get <string>)} Return the environment variable named by the given
argument.

@item
@code{(set <string> <string>)} Set the environment variable named by the first
string to be the second string.

@item
@code{(env)} List all environment variables.

@item
@code{(print ...)} Print the given arguments on the standard output in
a human-readable form. 

@item
@code{(+ <number>...)} Add all the arguments.

@item
@code{(- <number>...)} Subtract the arguments following the first argument
from the first argument.

@item
@code{(* <number>...)} Multiply all the arguments.

@item
@code{(/ <number>...)} Divide the first argument by the rest of the arguments.

@item
@code{(< <number> <number>)} Return @code{true} if the first number is
less than the second; otherwise, return @code{false}.

@item
@code{(> <number> <number>)} Return @code{true} if the first number is
less than the second; otherwise, return @code{false}.

@item
@code{(if <any> <any> <any>)} If the \"eval\" of the
first argument is @code{false}, execute @code{eval} on the second argument. 
Otherwise, execute @code{eval} on the third argument. Note that the 
arguments must be quoted right! (@pxref{Quoting Trickery} for 
more info on that.)

@item
@code{(list ...)} Compose a list made of the given arguments and return it.

@item
@code{(eval ...)} If an argument is a string or a hash table, simply return it.
Otherwise, execute the given list as if it was a command. 
For example, typing in
@code{(eval ~(print Hello World!))} is equivalent to 
@code{(print Hello World!)}. However, you must make sure that the list given
to @code{eval} is quoted properly in order to delay evaluation
until the right moment. (@pxref{Quoting Trickery} for more info.)

@item
@code{(= <string> <string>)} Compare the two strings and return an empty
list if they are not equal. Otherwise, simply return the first string.

@end itemize



@node Details, Differences from Scheme, Overview, Top
@cindex Details
@cindex Programmer's Manual
@cindex Tutorial
@chapter Details

A programmer's manual and a tutorial.

@menu
* Syntax::             Syntax of the shell.
* Semantics::          Writing commands, control structures, data storage.
* Quoting Trickery::   Controlling @code{eval} and using data as code.
* Command List::       More commands.
* Tutorial::           Learn by example.
@end menu


@node Syntax, Semantics, Basic Usage, Details
@cindex Syntax
@section Syntax

Here is a description of the syntax of @code{esh}. Note, however, that 
this grammar is only a descriptive tool, since the actual parser in the 
shell is written by hand.

Also note that using the interactive-command syntax is not allowed when
the shell is not started interactively.

@example
openparen-symbol = '('
closeparen-symbol = ')'
delay-symbol = '~' | '$'
pipe-symbol = ',' | '|'
redirection-symbol = '<' | '>'

script = /* Nothing */ |
         list script

list = openparen-symbol list-elements closeparen-symbol

list-elements = /* Nothing */ |
                string list-elements |
                delay-symbol list list-elements |
                list list-elements

interactive-command = list |
                      interactive-pipeline

interactive-pipeline = /* Nothing */ | 
                      interactive-pipe redirection redirection

interactive-pipe = /* Nothing */ |
                   single-command pipe-symbol interactive-pipe

single-command = /* Nothing */ |
                 string single-command

redirection = /* Nothing */ |
              redirection-symbol string
@end example

@node Semantics, Quoting Trickery, Syntax, Details
@cindex Commands
@cindex Recursion
@cindex Control structures
@cindex Loops
@cindex Variables
@cindex Stack, local variable
@cindex Arguments, to commands
@section Semantics

One major deviation of @code{esh} from other programming languages is
that commands are not functions. Any command can return any number of
values, without explicit unpacking syntax. (e.g. @code{Python})

For example, @code{(rest ~(foo bar baz))} @result{} @code{bar baz}.
These returned elements are not a list! Therefore, this code
will produce an error: @code{(rest (rest ~(foo bar baz)))}. Instead, you
should write: @code{(rest (list (rest ~(foo bar baz))))}.

The first, most important concept is the @code{define} command. This command
allows you to define new commands, which are no different from builtin
commands as far as the programmer is concerned.

The syntax of @code{define} is simple: @code{(define <string> ...)}.
The first argument is simply the name of the command you are defining, and
the rest of the arguments will simply be passed to @code{eval} whenever
your new command is called.

It's important to realize that this code snippet
@code{(define foo ~(print bar) bar) (foo)} is identical to
this one: @code{(eval ~(print bar) bar)}. Therefore, anything that
applies to @code{eval} also applies to commands defined by @code{define}.
(@pxref{Quoting Trickery} for more on that.)

If you are familiar with Scheme or Lisp, you'll notice the lack of
argument passing information in the syntax of @code{define}. The reason for
that is that the argument passing convention is radically different in
@code{esh} from other Lisp-like languages.

Every command in @code{esh} has a stack all to itself for scratch space
and local variables. When a user-defined command is run, all the arguments
are simply placed on the local-variable stack, the first argument at the
top.

To access the local-variable stack, several commands can be used.
@code{(push <any>)} will push a value onto the stack, @code{(pop)} will
remove the top element of the stack and return it, @code{(top)} will return
a copy of the top element, @code{(stack)} will return a copy of the whole
stack, and finally @code{(rot)} will switch the top and the next-to-top
elements of the stack and return a copy of the element that just became
the top one.

The top-level of the interpreter also has a stack; that is, typing the
stack manipulation commands directly into the interpreter will produce the
expected results.

Note that commands called recursively do not inherit the stack of the
calling command.

Also note that @code{(stack)} does not return a list, it returns an 
arbitrary number of elements. If you find typing @code{(list (stack))}
frequently, you can instead use the @code{(l-stack)} command, since they are
identical.

The shell does not support looping constructs in the traditional sense --
like in Scheme, looping is accomplished by using recursion. Here is
a concrete example:

@example
(define print-squish
        ~(if ~(not-null? (top))
             ~(begin 
                   (print (squish (pop)))
                   (print-squish (stack)))
              ()))
@end example

The command @code{begin} simply returns the given arguments; its purpose 
is to allow the use of several commands where one is asked for.
(@pxref{Command List} for the syntax of @code{if})

Since @code{esh} 0.2, there is an explicit @code{true} and @code{false}
value; these values are different from all other possible values. Commands
that operate on predicates (@code{if}, @code{and}, @code{or}, @code{=}, etc.)
use these values extensively. You can use them in your own scripts with
the @code{true} and @code{false} commands. Note that @code{if}, @code{and}, 
and @code{or} do not explicitly check for @code{true}, so to these commands
any value that isn't @code{false} is "true", so to say.

@node Quoting Trickery, Command List, Semantics, Details
@cindex Quoting
@cindex @code{eval}, quoting for
@cindex Code-as-data
@cindex Tilde, as a quote character
@section Quoting Trickery

(Note: Since @code{esh} 0.6, the backslash quote has been removed.
Use the tilde quote instead.)

Note that the meaning of the quote syntax is much different in @code{esh}
than in Lisp -- in Lisp, the single quote is a syntactic trick to
allow inputting lists and symbols as literals. In @code{esh}, the tilde
quote has a very clear semantic meaning -- whenever a list is marked with
a tilde, it is marked as such throughout it's life. Commands such as
@code{eval} can then use this information to see which lists were supposed
to be evaluated, and which ones were meant to be left alone.

In other words, any list marked with a tilde has a numeric flag set on it,
which is preserved on the list until the list is deleted.

Also, nested tildes increase the "strength" of the tilde, so to say. 
@code{eval} will not evaluate lists which have a "strength" greater than
itself, and calling @code{eval} recursively will increase it's strength.

This may sound confusing, but it has a simple intuitive meaning:

@example
(eval ~(foo ~(bar baz)))
@end example

Here, as you would expect, @code{eval} will only evaluate @code{foo}, and
pass @code{(bar baz)} as a list to @code{foo}.

Note that unlike in Lisp, @code{esh} has no "special forms"! A command
that works on lists is indistinguishable from a command that works
on code. That's why there is no @code{lambda} command, and also why
arguments to @code{if} and firends need to be quoted with a tilde.


@node Command List, Tutorial, Quoting Trickery, Details
@cindex Command list
@cindex Builtin command list, detailed
@findex alias
@findex alias-hash
@findex alive?
@findex and
@findex begin
@findex begin-last
@findex car
@findex car-l
@findex cdr
@findex chars
@findex chop!
@findex chop-nl!
@findex clone
@findex copy
@findex define
@findex defined?
@findex exec
@findex exit
@findex false
@findex file-open
@findex file-read
@findex file-type
@findex file-write
@findex filter
@findex first
@findex first-l
@findex gobble
@findex hash-get
@findex hash-keys
@findex hash-make
@findex hash-put
@findex help
@findex interactive?
@findex l-stack
@findex newline
@findex not
@findex not-null?
@findex nl
@findex null
@findex null?
@findex match
@findex or
@findex parse
@findex pop
@findex prompt
@findex push
@findex read
@findex repeat
@findex rest
@findex reverse
@findex rot
@findex script
@findex split
@findex squish
@findex stack
@findex standard
@findex stderr
@findex stderr-handler
@findex substring?
@findex top
@findex true
@findex typecheck
@findex unlist
@findex version
@findex void
@findex while
@section Command List

@itemize @bullet

@item 
@code{(alias <string> <string>...)} 
Create an alias. The first argument is the name of the alias, and the
rest of the arguments is the expansion. For example, @code{(alias rm rm -i)}
will create an alias called @code{rm}, so that any time @code{rm} is run,
@code{rm -i} gets executed in actuality.

@item
@code{(alias <string> <list>...)}
Mimic an executable with a command. Whenever an executable named by
the first string is run, the lists following get evaluated, and 
their output is thrown away. This process will only happen during the
parsing stage, though. That is, this type of alias will only be enabled
when typing commands interactively into te shell.
Example: @code{(alias cd ~(cd (stack)))} to mimic the
traditional syntax of @code{cd}.

@item
@code{(alias-hash)}
Return the alias hash table. Modifying this table also modifies the
actual aliases.

@item
@code{(alive? <PID>)}
Return @code{true} if the given process is alive, and @code{false} otherwise.

@item
@code{(and ...)} Return @code{false} if any of the arguments evaluate to
@code{false}. Warning: the inputs need to be quoted with a tilde, since this
command implements a short-circuited @code{"and"}!
For example, @code{(and ~(under-attack) ~(launch-nukes))} is safe to use.
If the @code{"eval"} of @code{~(under-attack)} evaluates to @code{false}
@code{~(launch-nukes)} will never be evaluated.

@item 
@code{(begin ...)}
Simply copy the inputs to the outputs. This command is useful when you want
to use several commands as if they were a single command. Example:

@example
(if ~(check-me)
    ~(begin (print-message)
            (set-environment-magic))
     ())
@end example

@item
@code{(begin-last ...)} Evaluate the given arguments sequentially, and
return the value of last argument.

@example
(if ~(not-null? (top))
    ~(begin (push foo) 
            (print (stack))
            (pop))
    ())
@end example

@item
@code{(builtin <string> ...)} Execute the builtin command named by the
first argument, even if this name has already been @code{define}d.
For example, here's how to write a custom replacement for @code{cd}:

@example
(define cd
        ~(push (squish (get HOME) "/.cdlog"))
        ~(file-write (file-open file (top))
                     (squish (file-read (file-open file (pop)))
                             (stack) (nl)))
        ~(builtin cd (stack)))
@end example

@item 
@code{(car <list>)} Simply return the first element of the list.

@item
@code{(car-l ...)} Return the first element.

@item 
@code{(cdr <list>)} Return the elements of the list after the first one.

@item
@code{(chars <string>)} Return the charcters in the given string.
@example
(chars "foo bar") => 'f' 'o' 'o' ' ' 'b' 'a' 'r'
@end example

@item
@code{(chop! <string>)} Delete the last character in the given string,
and return the given string.
Note: this function modifies the string in-place!

@item
@code{(chop-nl! <string>)} Like @code{chop!}, except that it only deletes
the last character if it is a newline.

@item
@code{(clone <string> <number>)} Return the first argument X number of times,
where X is the numeric value of the second argument.

@item 
@code{(copy)} Equivalent to @code{begin}.

@item 
@code{(define <string> ...)} Define a new command, named by the
first argument. The rest of the arguments will be passed to @code{eval} 
when the newly defined command is called. Example:

@example
(define print-nl
        ~(print (stack) (nl)))
@end example

@item 
@code{(defined? <string>)} Return @code{false} if the given string has not
been defined as a command.

@item 
@code{(exec <list> ..)} Equivalent to @code{eval}, except that the stack will
be set to the first argument for the duration of evaluation.

@item 
@code{(exit [number])} Exit with the given exit status, or zero if the
command is called with no arguments.

@item
@code{(false ...)} Return @code{false}.

@item
@code{(file-open <string> <string>)} Open a file. The first argument specifies
the type of the file, and the second the name of the file you want to open.
Allowed values for the first argument:

@itemize @bullet
@item @code{"file"} Open a regular file for reading/writing.
@item @code{"truncate"} Open a regular file for reading/writing, but truncate
it first.
@item @code{"append"} Open a regular file for reading/appending.
@item @code{"string"} Simulate a file with a string variable. In this case,
the name is used as the initial contents of the buffer.
(Note: "string" files are implemented as pipes internally.)
@end itemize

This command return the file, or an empty list on error.

@item 
@code{(file-read <file>)} Read the entire file given by the argument
into a single string, and return the string. @emph{Warning:} when reading
the output of a pipeline, you may have to call this command several times.


@item 
@code{(file-type <string>)} Return a string describing what type of file
is named by the given string. If such a file does not exist, return an empty
list. Otherwise, one of these strings is returned:

@itemize @bullet
@item @code{"link"} File is a symbolic link.
@item @code{"regular"} File is a regular file.
@item @code{"directory"} File is a directory.
@item @code{"character"} File is a character device.
@item @code{"block"} File is a block device.
@item @code{"pipe"} File is a FIFO pipe.
@item @code{"socket"} File is a socket.
@end itemize

@item 
@code{(file-write <file> <string>)} Write the second argument into the 
first one.

@item
@code{(filter <string> <list>)} Apply the second argument to every character
of the first argument. This command does not change the original string.

@example
(filter "string with spaces" 
        ~(if ~(= (top) ' ')
             '_'
             ~(top)))
@end example

returns "string_with_spaces".


@item 
@code{(first <list>)} Equivalent to @code{car}.

@item
@code{(first-l ...)} Equivalent to @code{car-l}.

@item 
@code{(gobble <file> <list>...)} Equivalent to @code{run}, except that the
output of the pipeline will be returned, as a string.


@item 
@code{(hash-get <hash> <string>)} Return the element in the given hash table
corresponding to the given key.

@item 
@code{(hash-keys <hash>)} Return all the keys in the given hash table.

@item 
@code{(hash-make)} Return a new, empty hash table.

@item 
@code{(hash-put <hash> <string> ...)} Associate the arguments after the
second one with the given key in the given hash table.

@item 
@code{(help)} Show version number and all builtin commands.

@item
@code{(interactive?)} Return @code{false} if the shell has @emph{not} been
started interactively.

@item
@code{(l-cdr <list>)} Equivalent to @code{(list (cdr ...))}.

@item
@code{(l-rest <list>)} Equivalent to @code{l-cdr}.

@item 
@code{(l-stack)} Equivalent to @code{(list (stack))}.

@item 
@code{(newline)} Return a newline character.

@item 
@code{(nl)} Equivalent to @code{newline}.

@item
@code{(not <bool>)} Return @code{false} if the argument is @code{true}.

@item
@code{(not-null? <any>)} Return @code{false} if the argument is an empty list.

@item 
@code{(null ...)} Return an empty list.

@item
@code{(null? <any>)} Return @code{true} if the argument is an empty list.

@item
@code{(match <string> <string>)} Return @code{true} if the second argument
matches the first. The first argument is a regular expression.

@item
@code{(or ...)} Return @code{false} if all the argument evaluate to 
@code{false}. As with the command @code{and}, this command is also 
short-circuited. See the description of @code{and} for an explanation of why
arguments must be quoted with a tilde.

@item 
@code{(parse <string>)} Parse the given string as if it was typed into the 
shell.

@item 
@code{(pop)} Excise the top element from the stack, and return it.

@item 
@code{(prompt ...)} Run the given arguments through @code{(squish (eval ...))}
to produce the prompt. This command need only be run once.

@item 
@code{(push <any>)} Push the argument onto the stack.

@item 
@code{(read <string>)} Read a line of input from the user, using the given
string as a prompt. Warning: this command only works if the shell is running
interactively.

@item
@code{(repeat <string> ...)} Repeat the arguments after the first argument
@code{n} number of times, where @code{n} is the numeric value of the first
argument.

@item 
@code{(rest <list>)} Equivalent to @code{cdr}.

@item
@code{(reverse ...)} Return the given arguments in reverse order.

@item 
@code{(rot)} Switch the top and next-to-top elements of the stack, and return
the element that just became the top one.

@item 
@code{(script <string>)} Execute the given filename as a shell script.

@item 
@code{(split <string>...)} Separate the given string into words. If there are
more than one arguments given, then use the arguments after the first one as
field separators. Otherwise, assume that fields are separated by whitespace.
Note that splitting is more complex than you'd think at first glance --
@code{(split ":.:.foo:.:bar.:baz." ":" ".")} @result{} @code{foo bar baz}.
In other words, extra field separators are ignored.

@item 
@code{(squish ...)} Concatenate any string values into a single string. List
structures will be ignored. For example: 
@code{(squish ~(foo (bar (baz))))} @result{} @code{foobarbaz}.

@item 
@code{(stack)} Return all the stack elements, top one first.

@item
@code{(standard)} Return the standard input/output.

@item
@code{(stderr)} Return the standard input/error.

@item
@code{(stderr-handler <file>)} Define a standard error handler for
all new subprocesses. This means that from now on, all executables run 
from the shell will send their standard error to the given file.

@item
@code{(substring? <string> <string>)} Return @code{true} if the first argument
is a substring of the second.

@item 
@code{(top)} Return a copy of the top element of the stack.

@item
@code{(true ...)} Return @code{true}.

@item 
@code{(typecheck <string> ...)} Make sure that the given arguments match the
given type specification string. (@pxref{Tutorial} for a description of this
type specification string.)

@item 
@code{(unlist <list>)} Return the elements of the given list.

@item
@code{(version)} Return the version of the shell, as three numbers.

@item
@code{(void ...)} Return nothing at all. Note: This is different from
@code{null}! For example:

@example
(if (null (magic-sequence))
    ~(my-predicate)
    true
    false)
@end example

is an error, since in this case @code{if} is given four arguments.
(@code{null} returns an empty list.)

On the other hand,

@example
(if (void (magic-sequence))
    ~(my-predicate)
    true
    false)
@end example

is perfectly fine since @code{void} returns nothing at all whatsoever!

@item
@code{(while <list> <list> ...)} @code{eval} the second argument as long
as the @code{eval} of the first argument is not @code{false}. The rest of
the arguments are used as the stack for the duration of this command's 
execution. @emph{Warning:} this command is an efficiency hack. It saves the 
waste of creating a huge return value which would be inherent in a recursive 
command. Use @code{while} only if you don't care about the return value of
the given commands.

Example:

@example
(define while-rec
 ~(if ~(condition)
      ~(begin (action)
              (while-rec))
      ()))
@end example

will return a value equivalent to @code{(action) (action) (action) ...}.
This value could be a large memory and time waste. Therefore, when you are
looping over large lists and you don't care about return values, use
@code{while}.


@end itemize

@node Tutorial,     , Command List, Details
@cindex Tutorial
@cindex Examples
@cindex Sample code
@cindex Typechecking
@findex typecheck
@findex prompt
@findex for-each
@findex file-read
@findex file-write
@section Tutorial

For starters, let us write a simple command to iterate through a list:

@example
(define for-each
        ~(if ~(not-null? (rot))
             ~(begin ((rot) (rot))
                     (for-each (cdr (list (stack)))))
             ()))
@end example

Using this command is simple:

@example
(define starrify 
        ~(squish '*' (top) '*'))

(for-each starrify foo bar baz)
@end example

results in @code{*foo* *bar* *baz*}.

Several notes to keep in mind: notice the tricky use of @code{rot} to
shuffle stack elements. If you don't remember, @code{rot} switches the top
and the next-to-top elements of the stack, and returns a copy of the element 
that just became the top one.

Think of it as such: the first call to @code{rot} brings the second argument
to the front and checks that the stack is not empty, at the same time.

The second call to @code{rot} returns the first argument, and the third
call to @code{rot} returns the second element and undoes the previous call
of @code{rot} at the same time. At this point, the second argument is still
on top of the stack.

Finally, when @code{for-each} is recursively called, it is given only
below the top stack element as arguments. Effectively, the second element
was excised from the stack.

Also note that @code{((rot) (rot))} is calling a command, even though the
name of the command is not explicit.

Now for another example. Suppose that you want to remind yourself which
directory serves which purpose, and you'd like a descriptive string to
be listed along with a directory name in the prompt. 
(i.e. @code{[/home/ivan/src/xcf => Backup of GIMP artwork]$ })

@example
(prompt ~(push (get PWD))
        ~(push (hash-get (dir-names) (top)))
        "["
        ~(rot)
        ~(if ~(not-null? (rot))
             ~(squish " => " (top))
             "")
        "]$ "
        ~(null (pop))
        ~(null (pop)))
@end example

The meaning of @code{prompt} is simple -- whatever arguments are given to
it are passed through @code{eval} and @code{squish} to become the string
specifying the prompt. Again, as far as quoting is concerned, @code{prompt}
is identical to @code{eval}.

Also, the command @code{null} simply returns an empty list.

To finish the job, you should insert the following code into your 
@code{.eshrc}:

@example
(define dir-names (hash-make))

(hash-put (dir-names) "/home/ivan/src/xcf" "Backup of GIMP artwork")
@end example


This is a simple script that illustrates simple file I/O:

@example
(define prepend
  ~(push (file-read (file-open file (pop))))
  ~(push (squish (pop)
                 (file-read (file-open file (top)))))
  ~(file-write (file-open truncate (rot)) (rot)))
@end example

A script such as this was used to append a header to every source code file
in the @code{esh} distribution. Note the @code{typecheck} command; it is
very useful for insuring the consistency of arguments to commands.

The only tricky part about using @code{typecheck} is the syntax of the
first argument. Here is an explanation:

@itemize @bullet
@item @code{s} Make sure that the next argument is a single string.
@item @code{l} Make sure that the next argument is a single list.
@item @code{h} Make sure that the next argument is a single hash table.
@item @code{b} Make sure that the next argument is a single boolean.
@item @code{f} Make sure that the next argument is a single file.
@item @code{p} Make sure that the next argument is a PID.
@item @code{S} Match any number of strings.
@item @code{L} Match any number of lists.
@item @code{H} Match any number of hash tables.
@item @code{B} Match any number of booleans.
@item @code{F} Match any number of files.
@item @code{P} Match any number of PID's.
@item @code{?} Match any one element.
@item @code{*} Match any number of any elements.
@item @code{(} Match a list only if the sublist passes typechecking on the
               string after the parentheses.
@item @code{)} Match end-of-list.
@end itemize

For example, @code{"H(ss)s"} will match any list of arguments that begin
with an arbitrary number of hash tables, followed by a list of two strings,
and ends with a single string.


@node     Differences from Scheme, Differences from sh, Details, Top
@cindex Differences from Scheme
@cindex Scheme
@cindex Lisp
@cindex Differences
@chapter Differences from Scheme


Note that in @code{esh} there is no numeric or "symbol" type -- all scalars are
either strings, booleans, files, or process ID's. Most importantly, 
there is no "procedure" type -- hence, no @code{lambda} command. From the point
of view of the interpreter, a command and a list are indistinguishable.

Don't be mislead by the naming of @code{define} -- it is actually equivalent to
a Lisp-like macro definition. 

Also, @code{car} and friends are more complicated because commands in 
@code{esh}
can return any number of arguments. This is why @code{(car (split "foo bar"))}
doesn't work -- @code{car} expects a list, while @code{split} returns 
an arbitrary number of elements. In a case like this, it is more convenient
to write @code{(car-l (split "foo bar"))} instead, which is equivalent
to @code{(car (list (split "foo bar")))}.

@node     Differences from sh, Command Index, Differences from Scheme, Top
@cindex Differences from sh
@cindex Differences
@cindex bash
@cindex csh
@cindex sh
@cindex tcsh
@chapter Differences from sh

The most obvious difference is the syntax -- other shells normally don't
make much of a distinction between commands from a disk executable and
the shell interpreter's builtin commands. Not so in @code{esh}. 
While other shell normally operate by complicated string substitution
and matching rules, @code{esh} mainly uses command definitions and 
recursion. In that sense, it is more similar to a "real" programming 
language. @code{esh} also supports more complicated data structures --
lists and hash tables, for example. However, string operations are
lacking in comparison to other shells. 

@code{esh} is also more verbose and formal, though this could be beneficial 
when you're trying to compose large libraries of useful routines.
(It's possible to write shell "modes", much as in @code{emacs}, when using
@code{esh}.)

@code{esh} has a more generalized support for files. Instead of limiting
file I/O to redirection only, the same commands can be used either on pipes
or on files directly. (e.g. in the future, it may be possible to use a 
network socket as the output of a pipeline.)

Also, traditional shells have "exported" and "non-exported" environment
variables, whereas in @code{esh} all variables are exported. That is,
there's no @code{setenv} command, there's just @code{set}.

@node     Command Index, Concept Index, Differences from sh, Top
@unnumbered Command Index

@printindex fn

@node     Concept Index,     , Command Index, Top
@unnumbered Concept Index

@printindex cp

@contents
@bye