autoexpect - generate an Expect script from watching a session
autoexpect [ args ] [ program args... ]
autoexpect watches you interacting with another program and creates an Expect
script that reproduces your interactions. For straightline scripts, autoexpect
saves substantial time over writing scripts by hand. Even if you are an Expect
expert, you will find it convenient to use autoexpect to automate the more
mindless parts of interactions. It is much easier to cut/paste hunks of
autoexpect scripts together than to write them from scratch. And if you are a
beginner, you may be able to get away with learning nothing more about Expect
than how to call autoexpect.
The simplest way to use autoexpect is to call it from the command
line with no arguments. For example:
% autoexpect
By default, autoexpect spawns a shell for you. Given a program
name and arguments, autoexpect spawns that program. For example:
% autoexpect ftp ftp.cme.nist.gov
Once your spawned program is running, interact normally. When you
have exited the shell (or program that you specified), autoexpect will
create a new script for you. By default, autoexpect writes the new script to
"script.exp". You can override this with the -f flag followed by a
new script name.
The following example runs "ftp ftp.cme.nist.gov" and
stores the resulting Expect script in the file "nist".
% autoexpect -f nist ftp ftp.cme.nist.gov
It is important to understand that autoexpect does not guarantee a working
script because it necessarily has to guess about certain things - and
occasionally it guesses wrong. However, it is usually very easy to identify
and fix these problems. The typical problems are:
- •
- Timing. A surprisingly large number of programs (rn, ksh, zsh, telnet,
etc.) and devices (e.g., modems) ignore keystrokes that arrive "too
quickly" after prompts. If you find your new script hanging up at one
spot, try adding a short sleep just before the previous send.
You can force this behavior throughout by overriding the
variable "force_conservative" near the beginning of the
generated script. This "conservative" mode makes autoexpect
automatically pause briefly (one tenth of a second) before sending each
character. This pacifies every program I know of.
This conservative mode is useful if you just want to quickly
reassure yourself that the problem is a timing one (or if you really
don't care about how fast the script runs). This same mode can be forced
before script generation by using the -c flag.
Fortunately, these timing spots are rare. For example, telnet
ignores characters only after entering its escape sequence. Modems only
ignore characters immediately after connecting to them for the first
time. A few programs exhibit this behavior all the time but typically
have a switch to disable it. For example, rn's -T flag disables this
behavior.
The following example starts autoexpect in conservative
mode.
autoexpect -c
The -C flag defines a key to toggle conservative mode. The following example
starts autoexpect (in non-conservative mode) with ^L as the toggle. (Note
that the ^L is entered literally - i.e., enter a real control-L).
autoexpect -C ^L
The following example starts autoexpect in conservative mode with ^L as the
toggle.
autoexpect -c -C ^L
- •
- Echoing. Many program echo characters. For example, if you type
"more" to a shell, what autoexpect actually sees is:
you typed 'm',
computer typed 'm',
you typed 'o',
computer typed 'o',
you typed 'r',
computer typed 'r',
...
Without specific knowledge of the program, it is impossible to
know if you are waiting to see each character echoed before typing the
next. If autoexpect sees characters being echoed, it assumes that it can
send them all as a group rather than interleaving them the way they
originally appeared. This makes the script more pleasant to read.
However, it could conceivably be incorrect if you really had to wait to
see each character echoed.
- •
- Change. Autoexpect records every character from the interaction in the
script. This is desirable because it gives you the ability to make
judgements about what is important and what can be replaced with a pattern
match.
On the other hand, if you use commands whose output differs
from run to run, the generated scripts are not going to be correct. For
example, the "date" command always produces different output.
So using the date command while running autoexpect is a sure way to
produce a script that will require editing in order for it to work.
The -p flag puts autoexpect into "prompt mode". In
this mode, autoexpect will only look for the the last line of program
output - which is usually the prompt. This handles the date problem (see
above) and most others.
The following example starts autoexpect in prompt mode.
autoexpect -p
The -P flag defines a key to toggle prompt mode. The following example
starts autoexpect (in non-prompt mode) with ^P as the toggle. Note that
the ^P is entered literally - i.e., enter a real control-P.
autoexpect -P ^P
The following example starts autoexpect in prompt mode with ^P as the
toggle.
autoexpect -p -P ^P
The -quiet flag disables informational messages produced by autoexpect.
The -Q flag names a quote character which can be used to
enter characters that autoexpect would otherwise consume because they are
used as toggles.
The following example shows a number of flags with quote used to
provide a way of entering the toggles literally.
autoexpect -P ^P -C ^L -Q ^Q
I don't know if there is a "style" for Expect programs but autoexpect
should definitely not be held up as any model of style. For example,
autoexpect uses features of Expect that are intended specifically for
computer-generated scripting. So don't try to faithfully write scripts that
appear as if they were generated by autoexpect. This is not useful.
On the other hand, autoexpect scripts do show some worthwhile
things. For example, you can see how any string must be quoted in order to
use it in a Tcl script simply by running the strings through autoexpect.
"Exploring Expect: A Tcl-Based Toolkit for Automating Interactive
Programs" by Don Libes, O'Reilly and Associates, January 1995.
Don Libes, National Institute of Standards and Technology
expect and autoexpect are in the public domain. NIST
and I would appreciate credit if these programs or parts of them are
used.