expectperl-discuss — Discussions about Perl modules Expect.pm, IO::Tty and IO::Stty

Re: [Expectperl-discuss] Understanding interact

[Roland G. <r.g...@xs...>]

Simon Taylor wrote:
> I cannot start to get the equivalent functionality in the perl interact, ie:
> 
>     print "$prog: Handing control back to the user on $host...\n";
>     $Expect::Log_Stdout=1;
>     $telnet->interact(\*STDIN, "~~r");
> 
> Am I doing something wrong?

No, this is a bug in the interact routine which will be fixed in the
next release, which I hopefully will find time for after my
easter vacation next week.

I'm attaching a patch from Jorgen Wahlsten <jo...@ti...>
for those who need it in the meantime.

Happy Easter!

Roland
--
RGi...@cp...


*** Expect.pm   Wed Mar 21 17:01:43 2001
--- Expect.pm.DIST      Mon Feb 19 05:21:22 2001
***************
*** 1055,1063 ****
        foreach $escape_sequence (keys(%{${*$read_handle}{exp_Function}})) {
          print STDERR "Tested escape sequence $escape_sequence from ${*$read_handle}{exp_Pty_Handle}"if ${*$read_handle}{"exp_Debug"} > 1;
          # Make sure it doesn't grow out of bounds.
!         if ((defined(${*$read_handle}{"exp_Max_Accum"}) &&
!             (${*$read_handle}{"exp_Max_Accum"}) &&
!             (length($escape_character_buffer)) > ${*$read_handle}{"exp_Max_Accum"})) {
            $offset = length($escape_character_buffer) - ${*$read_handle}{"exp_Max_Accum"};
            $escape_character_buffer = substr($escape_character_buffer,$offset,${*$read_handle}{"exp_Max_Accum"});
          }
--- 1055,1061 ----
        foreach $escape_sequence (keys(%{${*$read_handle}{exp_Function}})) {
          print STDERR "Tested escape sequence $escape_sequence from ${*$read_handle}{exp_Pty_Handle}"if ${*$read_handle}{"exp_Debug"} > 1;
          # Make sure it doesn't grow out of bounds.
!         if ((defined(${*$read_handle}{"exp_Max_Accum"}) &&(length($escape_character_buffer)) > ${*$read_handle}{"exp_Max_Accum"})) {
            $offset = length($escape_character_buffer) - ${*$read_handle}{"exp_Max_Accum"};
            $escape_character_buffer = substr($escape_character_buffer,$offset,${*$read_handle}{"exp_Max_Accum"});
          }

[Expectperl-discuss] Understanding interact

[Simon T. <si...@un...>]

Folks,

I am trying to understand the Expect.pm version of interact().

I have a telnet automation program written in tcl/expect that I am converting
to perl so that I can learn Expect.pm. The tcl version can telnet to a variety
of remote hosts, log in, handle any host-specific settings and then hand
control back to the user via interact:  

    send_user "$Prog: Handing control back to user on ${host}...\n";
    interact {
        "~~r" { send "tput rmso; clear\r" }
    }

This fragment does two nice things, 

  1) it looks for the "~~r" string from the user and responds to it and then 
      continues interacting, and 

  2) as soon as the spawned process exits, the interact loop ends.

I cannot start to get the equivalent functionality in the perl interact, ie:

    print "$prog: Handing control back to the user on $host...\n";
    $Expect::Log_Stdout=1;
    $telnet->interact(\*STDIN, "~~r");  

Am I doing something wrong?

Regards,

Simon Taylor
--
(+613 9568 2005)
Unisolve Pty Ltd 
Melbourne, Australia

[Expectperl-discuss] Re: Expect.pm

[Roland G. <r.g...@xs...>]

Dmitry Mottl wrote:
> When I use Expect.pm to interact with program (mysqldump) which asks
> for a password and dumps data to stdout, all lines are ended with "\r\n=
".
> How can I change this behavior? I need lines ended with "\n"

This behaviour could be changed by adjusting the terminal
settings via exp_stty(), which takes a string with commands similar to
=B4stty=B4.  I'm no specialist for terminal settings, but "igncr" and/or=20
"-onlcr" could do what you want.

OTOH, why do you need lines with "\n"?  How about post-processing the
output and removing the "\r" with =B4perl -e 's/\r+//g'=B4?  If you
have problems matching multiple lines, adjust your regexp by
adding "\r*" before "\n".

An even better way would be to use the Mysql perl module to directly=20
access the database without spawning an intermediate program. =20
Alternatively you could use the DBI module with its DBD::mysql driver=20
to keep the database exchangable.

Hope this helps,

Roland
--
RGi...@cp...

[Expectperl-discuss] Re: Bug in Expect::interconnect v1.11

[Roland G. <r.g...@xs...>]

Jorgen Wahlsten wrote:
> We are using Expect v1.11, and had some problems getting
> Expect::interact() to return control after the escape_sequence was
> sent on STDIN. Eventually we tracked it down to exp_Max_Accum not
> being checked for zero (unlimited) in Expect::interconnect().

Thanks for telling, I'm aware of this problem and will fix
it in the next release, as soon as I can find the time to 
package everything up.

Roland
--
RGi...@cp...

[Expectperl-discuss] Re: Expect.pm

[Roland G. <r.g...@xs...>]

[I'm CCing this to the discussion list to generate some traffic there :-)]

Dennis Hammer wrote:
> I've got a problem with Expect.pm:
> 
> $router->expect(2, ['Password:', sub {$router->send('password\r'); }]);
>
> the above line seems to match as the password string is sent, but
> seemingly the string is only printed to the tty without sending it to
> the application referenced by $router. Output looks like this:
>
> User Access Verification
> 
> Password: password\r
> 
> which is not quite the thing I wanted to achieve.
> 
> Please tell me what the problem is.

Make this '$router->send("password\r")' and it will work.  
q'' doesn't interpret '\r', so you are missing the end-of-line here.  
Don't be ashamed, it happens to the best of us... :-)

Roland
--
RGi...@cp...

RE: [Expectperl-discuss] exiting and a spawned process

[Rogaski, M. A. <ro...@at...>]

: From: James E. Pace [mailto:je...@do...]
: 
: Thank you for the reply.  I don't see the {soft|hard}_close in the
: pulic functions list from the man page.  Where can I find out about
: them?
: 

Yeah, I don't think they've ever been officially documented, but they are
demonstrated in the tutorial (hard close in rlogin, debugging, modem-init;
and soft close in smtp-verify).

Mark

--
[] Mark Rogaski                    "Computers save time like 
[] AT&T Frame Relay Service           kudzu prevents soil erosion"
[] email: ro...@at...                          -- Al Castanoli
[] voice: 732-885-7534 fax: 732-885-7699
[] pager: 888-858-7243 pin 200853 or 20...@pa... 

RE: [Expectperl-discuss] exiting and a spawned process

[Rogaski, M. A. <ro...@at...>]

: From: Rogaski, Mark, ALNTK 
: 
: Expect.pm doesn't have a destructor that does any cleanup 
: when exit() is
: called.  So, there's nothing to reap the little children.  
: However, it does
: provide soft_close() and hard_close() which will do the job.  
: Since this is
: a case where a problem has occurred, you're probably going to 
: want to add a
: $lrb->hard_close() before that exit.
: 

Oh, I forgot to add that if you still have a problem, you can always use
$lrb->exp_pid() to get the process id and use kill() explicitly.

Mark

--
[] Mark Rogaski                    "Computers save time like 
[] AT&T Frame Relay Service           kudzu prevents soil erosion"
[] email: ro...@at...                          -- Al Castanoli
[] voice: 732-885-7534 fax: 732-885-7699
[] pager: 888-858-7243 pin 200853 or 20...@pa... 

RE: [Expectperl-discuss] exiting and a spawned process

[James E. P. <je...@do...>]

On Fri, 16 Mar 2001, Rogaski, Mark, ALNTK wrote:

> Expect.pm doesn't have a destructor that does any cleanup when exit() is
> called.  So, there's nothing to reap the little children.  However, it does
> provide soft_close() and hard_close() which will do the job.  Since this is
> a case where a problem has occurred, you're probably going to want to add a
> $lrb->hard_close() before that exit.

Thank you for the reply.  I don't see the {soft|hard}_close in the
pulic functions list from the man page.  Where can I find out about
them?

Thanks,

James
__ 
                           __
James E. Pace              \o\_____         Manager of Software Quality
je...@do...     ##[==_____>        "Put a Rocket
(408) 371-8977 x106        /_/ dotRocket     on your Socket."

RE: [Expectperl-discuss] exiting and a spawned process

[Rogaski, M. A. <ro...@at...>]

: -----Original Message-----
: From: James E. Pace [mailto:je...@do...]
: 
: unless ($lrb->expect($Settings::timeout, $Settings::boot_prompt))
: {
:             print STDERR "reboot: No boot prompt found.\n";
:             #exit $Settings::UNDECIDED;
:             exit 2;
: }
:  

Expect.pm doesn't have a destructor that does any cleanup when exit() is
called.  So, there's nothing to reap the little children.  However, it does
provide soft_close() and hard_close() which will do the job.  Since this is
a case where a problem has occurred, you're probably going to want to add a
$lrb->hard_close() before that exit.

Mark

--
[] Mark Rogaski                    "Computers save time like 
[] AT&T Frame Relay Service           kudzu prevents soil erosion"
[] email: ro...@at...                          -- Al Castanoli
[] voice: 732-885-7534 fax: 732-885-7699
[] pager: 888-858-7243 pin 200853 or 20...@pa... 

RE: [Expectperl-discuss] telnet client

[Rogaski, M. A. <ro...@at...>]

: It is for one automatic login, I need give user name and password with
: the apropiate
: timeout or response of server and then give some commands and 
: exit, for
: example each day.
: 
: Is Expectperl a good alternative for it? or I need expect?
: Coul you give me any suggestions?
: 

As you describe it, Expect.pm will fit your needs quite well.  Look at
tutorial/2.B.rlogin in the distribution, but use telnet instead of rlogin.

Mark

--
[] Mark Rogaski                    "Computers save time like 
[] AT&T Frame Relay Service           kudzu prevents soil erosion"
[] email: ro...@at...                          -- Al Castanoli
[] voice: 732-885-7534 fax: 732-885-7699
[] pager: 888-858-7243 pin 200853 or 20...@pa... 


 

[Expectperl-discuss] telnet client

[Tomas C. <to...@tc...>]

Hi list

I'm newbie in expect and expectperl.

I have some questions:

Can I have a telnet client with expectperl?

It is for one automatic login, I need give user name and password with
the apropiate
timeout or response of server and then give some commands and exit, for
example each day.

Is Expectperl a good alternative for it? or I need expect?
Coul you give me any suggestions?

Thanks in advance

---
Tomas Carmen
TCC S.A.
+51-1-4428842
Lima-Peru

[Expectperl-discuss] exiting and a spawned process

[James E. P. <je...@do...>]

Hi.

How do I kill my Expect object?  I've got lingering processes that
were $obj->spawn()'ed which don't get killed when my perl script
finishes.  I don't want to interact, I want it dead.


(OK, this next question isn't really Expect specific, but...)

This chunk of code is in a library of routines that I've written to
help write test cases ..

unless ($lrb->expect($Settings::timeout, $Settings::boot_prompt))
{
            print STDERR "reboot: No boot prompt found.\n";
            #exit $Settings::UNDECIDED;
            exit 2;
}
 
This is being run from a shell script:

#!/bin/sh

        $testcase > ./$testname.out 2> ./$testname.err
        ret=$?
        case $ret in
            0) echo "$testname: PASS";;
            1) echo "$testname: FAIL";;
            2) echo "$testname: UNDECIDED";;
            *) echo "$testname: INSANITY";;
        esac
    

I'm getting the printout from that error case above, but the return
value I'm getting in the shell is 0.

What am I doing wrong?


Thanks,

James
__ 
                           __
James E. Pace              \o\_____         Manager of Software Quality
je...@do...     ##[==_____>        "Put a Rocket
(408) 371-8977 x106        /_/ dotRocket     on your Socket."

Re: [Expectperl-discuss] Expect 1.11 exp_internal

[Roland G. <r.g...@xs...>]

> $object -> exp_internal(1) does not have the same effect as
> $Expect::Exp_Internal = 1 (the patterns are not printed out in the former).

Yeah, the debug code is a mess, thanks for telling.  I'll do some cleanup
in the next release (patches always welcome).

Roland
--
RGi...@cp...

[Expectperl-discuss] Expect 1.11 exp_internal

[Ian M. <iw...@do...>]

$object -> exp_internal(1) does not have the same effect as
$Expect::Exp_Internal = 1 (the patterns are not printed out in the former).

In _multi_expect we have 
      # What are we expecting? What do you expect? :-)
      if ($Expect::Exp_Internal) {
with no test of ${*$self}{exp_Exp_Internal}  

--
Ian W Moor                     Department of Computing,     
  i....@do...          Imperial College.            
  Tel +44 20 7594 8250         180 Queensgate               
                               London SW7 2BZ UK.               
                               

[Expectperl-discuss] NOTE: Expect v1.11

[Roland G. <r.g...@xs...>]

Expect users, please note:

The testsuite in Expect v1.11 will produce a false negative
under perl 5.6: test # 7 will fail.  This is OK, it's a bug
in the testsuite, not Expect, so you can still install it.

Sorry for the inconvenience!

Roland Giersig (Expect.pm maintainer)
RGi...@cp...

2001-02-20

[Expectperl-discuss] ANNOUNCE: Expect v1.11

[Roland G. <r.g...@xs...>]

Expect v1.11 has been uploaded to CPAN and is also
available from http://sourceforge.net/projects/expectperl/

From the README:

        Expect.pm v1.11

Expect requires IO::Tty and IO::Stty, also available from CPAN.
There is a Bundle::Expect available that installs everything for
you.

If you prefer manual installation, the usual

  perl Makefile.PL
  make
  make install

should work.  I finally started a simple testsuite for Expect, but
it doesn't really test anything deeper right now.

It would probably be wise to read the docs on IO::Tty before
doing a make install with it.

The Expect for Perl module was inspired more by the functionality the
Tcl tool provides than any previous Expect-like tool such as Comm.pl
or chat2.pl.

The Tcl version of expect is a creation of Don Libes (li...@ni...).
The Tcl Expect home page is http://expect.nist.gov/.  Don has written
an excellent in-depth tutorial of Tcl/Expect, which is _Exploring
Expect_.  It is the O'reilly book with the monkey on the front.  Don has
several references to other articles on the Expect web page.

I try to stay as close to Tcl/Expect in interface and semantics
as possible (so I can refer questions to the Tcl/Expect docu ;-).
Suggestions for improvement are always welcome.

I started a new FAQ (Expect_FAQ.pod), so please let me know if there's
something you'd like to see answered there that isn't.


From the Changelog:
===================

        Bugs fixed:

! max_accum (match_max) finally really restricts the match-buffer-size.

! expect() didn't return the correct error state upon EOF

! soft_close() was re-reading the filehandle even when an EOF
  had already been detected, resulting in a very long delay.

! exp_continue() returned a value that got truncated with certain
  perl versions (5.6 :-( ) on certain systems.


        Features added:

+ Lee Eakin <le...@ja...>contributed a perl version of the
  kibitz script which might be of common interest.  See the
  examples/kibitz subdir.  Thanks Lee!

+ Expect got it's own print() so we can show what's getting
  sent to the spawned program when exp_internal is set.
  (suggested by ho...@ea...)

+ a session can now be logged to a file by setting log_file().  Thanks to
  Marcel Widjaja <mar...@su...> for suggesting this.

+ I added some aliases for certain methods to help Tcl/Expect users
  to a WLIE experience (Work Like I Expect).  'exp_pid', 'match_max',
  'log_file', 'log_user', even 'send' (which is an alias for 'print')
  are all there now.  But I won't make 'send_user' an alias for
  'print STDOUT', you have to draw a line somewhere.

+ some diagnosis messages now use cluck to print a stacktrace
  (suggested by ho...@ea...)

+ there is a new option 'restart_timeout_upon_receive', that, when
  set to 1, will restart the timeout within the expect call.  This is
  useful when supervising an application that produces periodic, but
  not well-defined output and you still want to react to certain
  patterns.  Just say

    $exp->restart_timeout_upon_receive(1);
    $exp->expect($timeout,
                 [ timeout => \&report_timeout ],
                 [ qr/pattern/ => \&handle_pattern ]);

  (suggested by ho...@ea...)
  

Thanks to everybody who wrote to me, either with bug reports or
enhancement suggestions!

Roland Giersig (Expect.pm maintainer)
RGi...@cp...

2001-02-20

[Expectperl-discuss] Prerelease Expect v1.11

[Roland G. <r.g...@xs...>]

Hi Friends and Netizens!

I intend to release Expect v1.11 next week.  As there is no
testsuite yet, I thought I'd post the version here and ask
you to try it out as beta-testers so I can clean up remaining
problems.  Find the prerelease version attached.

Also, if you have small scripts or code snippets that may
be suitable for a testsuite, please send them to me so I
can finally get some tests written.
 
Thanks for participating!

Roland


Bugs fixed (hopefully):

- expect() didn't return the correct error state upon EOF

- soft_close() was re-reading the filehandle even when an EOF
  had already been detected, resulting in a very long delay.

- exp_continue() returned a value that got truncated with certain
  perl versions on certain systems

- max_accum now finally really restricts the match-buffer-size

Features added:

- some diagnosis messages now use cluck to print a stacktrace

- Expect now has it's own print() so we can show what's getting
  sent to the spawned program when exp_internal is set.

- SIGWINCH is now propagated to child processes; unfortunately 
  this is only the first part to make resizing a window work
  for the underlying application

--
RGi...@cp...

Re: [Expectperl-discuss] Return Values from spawned process

[Roland G. <r.g...@xs...>]

Ryan Heatwole schrieb:
>   I have been trying to find a way to determine the return value (that
> is: success/fail) of a spawned process (specifically scp).

You mean the exit status of the spawned process?  Or do you just
want to check if the process got spawned OK?

For the first, there is $exp->exp_exitstatus(), but I haven't tested
if this gives you the right thing.  Another way to do this manually
is to add "; echo EXIT $?" to the end of your command and expect for
qr/^EXIT (\d)/. =20

In the second case, you would see the error as first line from the=20
forked process, saying something like "Cannot exec `command=B4: file not
found".

Hope this helps,

Roland
--
RGi...@cp...

RE: [Expectperl-discuss] Return Values from spawned process

[Don S. <don...@te...>]

I've been spawning processes via eval and then checking the return code.
For example: 
        eval {$MIMS = Expect->spawn($cmd); };
        if ($@) {
                print STDOUT "$ALARM ::: Failed to spawn '$cmd': $! \n";
                exit(1);
        }
I've found this to work well in capturing problems with the spawn process
itself.

Don

-----Original Message-----
From: Ryan Heatwole [mailto:hea...@re...]
Sent: Thursday, January 25, 2001 5:21 AM
To: exp...@li...
Subject: [Expectperl-discuss] Return Values from spawned process


Greetings,

  I have been trying to find a way to determine the return value (that
is: success/fail) of a spawned process (specifically scp).

Currently, I am using an Expect match, along with the values the match
returns to determine if the spawned command finished or failed.

For Example:

# Spawn our command
$cmd = Expect->spawn($scp);
$cmd->expect(60, -re => 'Enter passphrase| password:');
print $cmd "$password\r";

# Check for any returned info from scp command
@array = $cmd->expect(undef, -re => '\r\n.+');
$stuff = $cmd->exp_match();
$cmd->soft_close();

if ($array[1] eq '2:EOF')     # Finished OK
{ ... }
else                          # Command Failed
{ print "We had an error: $stuff\n"; }


Is this a proper way to check the status of the spawned process?  It
seems to work ok, but it also feels a little bit contrived.

Thank you in advance
Ryan Heatwole


_______________________________________________
Expectperl-discuss mailing list
Exp...@li...
http://lists.sourceforge.net/lists/listinfo/expectperl-discuss

[Expectperl-discuss] Return Values from spawned process

[Ryan H. <hea...@re...>]

Greetings,

  I have been trying to find a way to determine the return value (that
is: success/fail) of a spawned process (specifically scp).

Currently, I am using an Expect match, along with the values the match
returns to determine if the spawned command finished or failed.

For Example:

# Spawn our command
$cmd = Expect->spawn($scp);
$cmd->expect(60, -re => 'Enter passphrase| password:');
print $cmd "$password\r";

# Check for any returned info from scp command
@array = $cmd->expect(undef, -re => '\r\n.+');
$stuff = $cmd->exp_match();
$cmd->soft_close();

if ($array[1] eq '2:EOF')     # Finished OK
{ ... }
else                          # Command Failed
{ print "We had an error: $stuff\n"; }


Is this a proper way to check the status of the spawned process?  It
seems to work ok, but it also feels a little bit contrived.

Thank you in advance
Ryan Heatwole

Re: [Expectperl-discuss] does Expect.pm 1.10 work with perl 5.6.0?

[Roland G. <Rol...@a1...>]

"Brian R. Jones" wrote:
> I've been trying to get the example remote login script from the FAQ to work,
> and I have a different problem than the one that John Stumbles was reporting.
> For me, it looks like expect is ignoring the exp_continue and just exiting.
> I've tried it on both SunOS 5.6 and RedHat 7.0, with the same result, so it's
> either me being stupid or 1.10 doesn't play nice with perl 5.6.0.

Hmm, I haven't tried Perl 5.6 myself, so I cannot say.  I'll try it at
home
once I can find a spare moment (I'm between jobs right now and I guess
it will take another month until I have time to devote to Expect). 
Seems
some kind of regression test suite would be good (but we know that
already).

Anybody has got some spare time to write up the basics for a start?
Right now
we only need one simple testcase (the automated login script comes to
mind,
telnet to localhost and use some random user and password and catch the
error response from telnet), but this will make adding tests much
easier.

Any takers?

Roland
--
RGi...@cp...

[Expectperl-discuss] does Expect.pm 1.10 work with perl 5.6.0?

[Brian R. J. <scu...@ea...>]

I've been trying to get the example remote login script from the FAQ to work,
and I have a different problem than the one that John Stumbles was reporting. 
For me, it looks like expect is ignoring the exp_continue and just exiting. 
I've tried it on both SunOS 5.6 and RedHat 7.0, with the same result, so it's
either me being stupid or 1.10 doesn't play nice with perl 5.6.0.  

Here's the debug output
--------------------------------------------------------------------------
...
Waiting for new data (4 seconds)...
spawn id(3): new data.
spawn id(3): read 76 byte(s).

Red Hat Linux release 7.0 (Guinness)
Kernel 2.2.16-22 on an i686
login: 
spawn id(3): Does `Trying 10.1.1.42...\r\r\nConnected to
deepthought.alacritech.com.\r\r\012Escape character is
\'^]\'.\r\r\012\r\012Red Hat Linux release 7.0 (Guinness)\r\012Kernel
2.2.16-22 on an i686\r\012login: '
match:
  pattern #1: -re `login: $'? YES!!
    Before match string: `Trying 10.1.1.42...\r\r\nConnected to
deepthought.alacritech.com.\r\r\012Escape character is
\'^]\'.\r\r\012\r\012Red Hat Linux release 7.0 (Guinness)\r\012Kernel
2.2.16-22 on an i686\r\012'
    Match string: `login: '
    After match string: `'
    Matchlist: ()
Calling hook CODE(0x8338ed8)...
Returning from expect successfully.
  spawn id(3): accumulator: `'
Closing spawn id(3).
test
Password: Timed out waiting for an EOF from spawn id(3).
spawn id(3) closed.
Pid 2753 of spawn id(3) exited, Status: 1
--------------------------------------------------------------------------

Everything up to 'Password: ' happens more or less instantaneously.  The rest
comes after the timeout.  Now I haven't done any Expect for several years, and
I never touched Expect.pm before today, but I would expect that it's supposed
to wait for more data rather than close id(3) isn't it?

Any enlightenment would be appreciated. 
-- 
Brian R. Jones   scu...@ea...   Slackin when I should be workin

[Expectperl-discuss] Maintainance Downtime

[Roland G. <Rol...@al...>]

Hi all,

I'm sorry for not responding to any inquiries lately, but I am
between jobs right now and have other more pressing things to
do.  I hope to be up and running in the second week of january,
so please be patient.

Merry Christmas and a Happy New Year to all of you!

Roland
--
RGi...@cp...
Maintainer [Expect, IO::Tty, IO::Stty, Tie:Persistent]

[Expectperl-discuss] question about intertact sub

[K. H. <k4...@ya...>]

Hi Roland, just wanted to know if the following change
to the interact method is possible:

here is the existing code:

  $self->set_group(@old_group);
  # If old_group was undef, make sure that occurs.
This is a slight hack since
  # it modifies the value directly.
  # Normally an undef passed to set_group will return
the current groups.
  # It is possible that it may be of worth to make it
possible to undef
  # The current group without doing this.
  unless (defined (@old_group)) {
    @{${*$self}{exp_Listen_Group}} = ();
  }

Can this be replaced by:

if(not defined(@old_group))
{
   @old_group = ();
}
$self->set_group(@old_group);

i.e, I want to avoid the direct access as in the
following statement:

 unless (defined (@old_group)) {
    @{${*$self}{exp_Listen_Group}} = ();
  }

Please let me know.
Thanks

__________________________________________________
Do You Yahoo!?
Yahoo! Shopping - Thousands of Stores. Millions of Products.
http://shopping.yahoo.com/

[Expectperl-discuss] Here comes the FAQ...

[Roland G. <Rol...@al...>]

Hi folks,

I'm in between jobs right now and have more pressing things 
to attend to, so I cannot be of much help in the next few days.
I planned to release another bugfix version but this will have
to wait.

Still, I'm monitoring the questions that come up here and take note
of them, so please be patient.

So for now, find the new FAQ-list attached below.

Roland
--
RGi...@cp...