This posting contains answers to general information questions, mostly about programming aids.
The easiest way to do this is to run Perl under its debugger. If you have no program to debug, you can invoke the debugger on an `empty' program like this:
perl -de 0
(The more positive hackers prefer "perl -de 1". :-)
Now you can type in any legal Perl code, and it will be immediately evaluated. You can also examine the symbol table, get stack backtraces, check variable values, and if you want to, set breakpoints and do the other things you can do in a symbolic debugger.
While there isn't one included with the perl source distribution (yet) various folks have written packages that allow you to do at least some sort of profiling. The strategy usually includes modifying the perl debugger to handle profiling. Authors of these packages include
Wayne Thompson <me@anywhere.EBay.Sun.COM> Ray Lischner <lisch@sysserver1.mentor.com> Kresten Krab Thorup <krab@iesd.auc.dk>
The original articles by these folks containing their profilers are available on convex.com in /pub/perl/information/profiling.shar via anon ftp.
Yes!! It's a version of Berkeley yacc that outputs Perl code instead of C code! You can get this from ftp.sterling.com [192.124.9.1] in /local/perl-byacc1.8.1.tar.Z, or send the author mail for details.
That depends on what you mean. If you want something that works like vgrind on Perl programs, then the answer is "yes, nearly". Here's a vgrind entry for perl:
PERL|perl|Perl:\
:pb=^\d?(sub|package)\d\p\d:\
:bb={:be=}:cb=#:ce=$:sb=":se=\e":lb=':\
:le=\e':tl:\
:id=_:\
:kw=\
if for foreach unless until while continue else elsif \
do eval require \
die exit \
defined delete reset \
goto last redo next dump \
local undef return \
write format \
sub package
It doesn't actually do everything right; in particular, things like $#, $', s#/foo##, and $foo'bar all confuse it.
David Levine uses this:
# perl 4.x David Levine <levine@ics.uci.edu> 05 apr 1993
# Derived from Tom Christiansen's perl vgrindef. I'd like to treat all of
# perl's built-ins as keywords, but vgrind fields are limited to 1024
# characters and the built-ins overflow that (surprise :-). So, I didn't
# include the dbm*, end*, get*, msg*, sem*, set*, and shm* functions. I
# couldn't come up with an easy way to distinguish beginnings of literals
# ('...') from package prefixes, so literals are not marked.
# Be sure to:
# 1) include whitespace between a subprogram name and its opening {
# 2) include whitespace before a comment (so that $# doesn't get
# interpreted as one).
perl4:\
:pb=^\d?(sub|package)\d\p\d:\
:id=$%@_:\
:bb=\e{:be=\e}:cb=\d\e#:ce=$:sb=\e":se=\e":\
:kw=accept alarm atan2 bind binmode caller chdir chmod chop \
chown chroot close closedir connect continue cos crypt defined delete \
die do dump each else elsif eof eval exec exit exp fcntl fileno flock \
for foreach fork format getc gmtime goto grep hex if include index int \
ioctl join keys kill last length link listen local localtime log lstat \
m mkdir next oct open opendir ord pack package pipe pop print printf \
push q qq qx rand read readdir readlink recv redo rename require reset \
return reverse rewinddir rindex rmdir s scalar seek seekdir select send \
shift shutdown sin sleep socket socketpair sort splice split sprintf \
sqrt srand stat study sub substr symlink syscall sysread system \
syswrite tell telldir time times tr truncate umask undef unless unlink \
unpack unshift until utime values vec wait waitpid wantarray warn while \
write y:
If what you mean is whether there is a program that will reformat the program much as indent(1) will do for C, then the answer is no. The complex feedback between the scanner and the parser (as in the things that confuse vgrind) make it challenging at best to write a stand-alone Perl parser.
Because the Pascal people would be upset that we stole their name. :-)
The dynamic nature of Perl's do and eval operators (and remember that constructs like s/$mac_donald/$mac_gregor/eieio count as an eval) would make this very difficult. To fully support them, you would have to put the whole Perl interpreter into each compiled version for those scripts using them. This is what undump does right now, if your machine has it. If what you're doing will be faster in C than in Perl, maybe it should have been written in C in the first place. For things that ought to be written in Perl, the interpreter will be just about as fast, because the pattern matching routines won't work any faster linked into a C program. Even in the case of simple Perl programs that don't do any fancy evals, the major gain would be in compiling the control flow tests, with the rest still being a maze of twisty, turny subroutine calls. Since these are not usually the major bottleneck in the program, there's not as much to be gained via compilation as one might think.
However, we're still looking for someone to generate byte-compiled code for Perl, or eventually even C code out of it. These are probably masters and PhD thesis topics respectively, and no one has begun work on it yet.
Since Emacs version 19 patchlevel 22 or so, there has been both a perl-mode.el and support for the perl debugger built in. These should come with the standard Emacs 19 distribution.
In the perl source directory, you'll find a directory called "emacs", which contains several files that should help you.
Not really. Perl is a programming language, not a command interpreter. There is a very simple one called "perlsh" included in the Perl source distribution. It just does this:
$/ = ''; # set paragraph mode
$SHlinesep = "\n";
while ($SHcmd = <>) {
$/ = $SHlinesep;
eval $SHcmd; print $@ || "\n";
$SHlinesep = $/; $/ = '';
}
Not very interesting, eh?
Daniel Smith <dansmith@autodesk.com> is working on an interactive Perl shell called SoftList. It's currently at version 3.0beta. SoftList 3.0 has tcsh-like command line editing, can let you define a file of aliases so that you can run chunks of perl or UNIX commands, and so on. You can send mail to him for further information and availability.
In release 4 of perl, the only way to do this was was to build a curseperl binary by linking in your C curses library as described in the usub subdirectory of the perl sources. This requires a modicum of work, but it will be reasonably fast since it's all in C (assuming you consider curses reasonably fast. :-) Programs written using this method require the modified curseperl, not vanilla perl, to run. While this is something of a disadvantage, experience indicates that it's better to use curseperl than to try to roll your own using termcap directly.
Fortunately, in version 5, Curses is a dynamically loaded extension by William Setzer*. You should be able to pick it up wherever you get Perl 5 from, or at least these places:
ftp://ftp.ncsu.edu/pub/math/wsetzer/cursperl5a6.tar.gz
ftp://ftp.metronet.com/pub/perlinfo/perl5/cursperl5a6.tar.gz
ftp://ftp.cs.ruu.nl/pub/PERL/perl5.0/cursperl5a6.tar.gz
For a good example of using (v4) curseperl, you might want to pick up a copy of Steven L Kunz's* "perl menus" package ("menu.pl") via anonymous FTP from "ftp.iastate.edu". It's in the /pub/perl as
menu.pl.v2.3.shr1 menu.pl.v2.3.shr2 menu.pl.v2.3.tar.Z
menus.pl is a complete menu front-end for curseperl and demonstates a lot of things (plus it is useful to boot if you want full-screen menu selection ability).
Another possibility is to use Henk Penning's cterm package, a curses emulation library written in perl. cterm is actually a separate program with which you communicate via a pipe. It is available from ftp.cs.ruu.nl [131.211.80.17] via anonymous ftp. in the directory pub/PERL. You may also acquire the package via email in compressed, uuencoded form by sending a message to mail-server@cs.ruu.nl containing these lines:
begin send PERL/cterm.shar.Z end
See the question on retrieving perl via mail for more information on how to retrieve other items of interest from the mail server there.
Right now, you have several choices. If you are still using perl4, use the WAFE or STDWIN packages, or try to make your own usub binding.
However, if you've upgraded to version 5, you have several exciting possibilities, with more popping up each day. Right now, Tk and Sx are the best known such extensions.
If you like the tk package, you should get the Tk extension kit, written by Malcolm Beattie*. Here are some places to get it:
Tk (as in tcl/tk, but sans tcl)
ftp://ftp.cis.ufl.edu/pub/perl/src/tkperl/tkperl5a4.tar.gz
ftp://ftp.khoros.unm.edu/pub/perl/extensions/tkperl5a4.tar.gz
ftp://ftp.metronet.com/pub/perlinfo/perl5/tkperl/tkperl5a4.tar.gz
ftp://ftp.cs.ruu.nl/pub/PERL/perl5.0/tkperl5a4.tar.gz
ftp://black.ox.ac.uk/src/ALPHA/tkperl5a4.tar.gz
(try 5a5 everywhere after 2pm UST Thu 20 Oct 1994, as in)
ftp://sable.ox.ac.uk/pub/perl/tkperl5a5.tar.gz
You may also use the old Sx package, (Athena & Xlib), written by originally written by by Dominic Giampaolo*, then and rewritten for Sx by Frederic Chauveau*.fr>. It's available from these sites:
ftp://ftp.pasteur.fr/pub/Perl/Sx.tar.gz
ftp://ftp.khoros.unm.edu/pub/perl/extensions/Sx.tar.gz
ftp://ftp.metronet.com/pub/perlinfo/perl5/Sx.tar.gz
ftp://ftp.cs.ruu.nl/pub/PERL/perl5.0/PerlDoc.ps.gz
STDWIN is a library written by Guido van Rossum* (author of the Python programming language) that is portable between Mac, Dos and X11. One could write a Perl agent to speak to this STDWIN server.
WAFE is a package that implements a symbolic interface to the Athena widgets (X11R5). A typical Wafe application consists in our framework of two parts: the front-end (we call it Wafe for Widget[Athena]front end) and an application program running typically as a separate process. The application program can be implemented in an arbitrary programming language and talks to the front-end via stdio. Since Wafe (the front-end) was developed using the extensible TCL shell (cite John Ousterhout), an application program can dynamically submit requests to the front-end to build up the graphical user interface; the application can even down-load application specific procedures into the front-end. The distribution contains sample application programs in Perl, GAWK, Prolog, TCL, and C talking to the same Wafe binary. Many of the demo applications are implemented in Perl. Wafe 0.9 can be obtained via anonymous ftp from ftp.wu-wien.ac.at[137.208.3.5]:pub/src/X11/wafe-0.9.tar.Z
Alternatively, you could use wish from tcl.
#!/usr/local/bin/perl
#####################################################################
# An example of calling wish as a subshell under Perl and
# interactively communicating with it through sockets.
#
# The script is directly based on Gustaf Neumann's perlwafe script.
#
# Dov Grobgeld dov@menora.weizmann.ac.il
# 1993-05-17
#####################################################################
$wishbin = "/usr/local/bin/wish";
die "socketpair unsuccessful: $!!\n" unless socketpair(W0,WISH,1,1,0);
if ($pid=fork) {
select(WISH); $| = 1;
select(STDOUT);
# Create some TCL procedures
print WISH 'proc echo {s} {puts stdout $s; flush stdout}',"\n";
# Create the widgets print WISH <<TCL; # This is a comment "inside" wish
frame .f -relief raised -border 1 -bg green
pack append . .f {top fill expand}
button .f.button-pressme -text "Press me" -command {
echo "That's nice."
}
button .f.button-quit -text quit -command {
echo "quit"
}
pack append .f .f.button-pressme {top fill expand} \\
.f.button-quit {top expand}
TCL
;
# Here is the main loop which receives and sends commands
# to wish.
while (<WISH>) {
chop;
print "Wish sais: <$_>\n";
if (/^quit/) { print WISH "destroy .\n"; last; }
}
wait;
} elsif (defined $pid) {
open(STDOUT, ">&W0");
open(STDIN, ">&W0");
close(W0);
select(STDOUT); $| = 1;
exec "$wishbin --";
} else {
die "fork error: $!\n";
}
Yes -- dynamic loading comes with the distribution. That means that you no longer need 18 different versions of fooperl floating around. In fact, all of perl can be stuck into a libperl.so library and then your /usr/local/bin/perl binary reduced to just 50k or so. See DynLoader(3pm) for details.
In perl4, the answer is kinda. One package has been released that does this, by Roberto Salama*. He writes:
Here is a version of dylperl, dynamic linker for perl. The code here is based on Oliver Sharp's May 1993 article in Dr. Dobbs Journal (Dynamic Linking under Berkeley UNIX).
dyl.h dyl.c - code extracted from Oliver Sharp's article
hash.h hash.c - Berkeley's hash functions, should use perl's but could not be bothered
dylperl.c - perl usersubs user.c - userinit function
sample.c - sample code to be dyl'ed sample2.c - " test.pl - sample perl script that dyl's sample*.o
The Makefile assumes that uperl.o is in /usr/local/src/perl/... You will probable have to change this to reflect your installation. Other than that, just type 'make'...
The idea behind being able to dynamically link code into perl is that the linked code should become perl functions, i.e. they can be invoked as &foo(...). For this to happen, the incrementally loaded code must use the perl stack, look at sample.c to get a better idea.
The few functions that make up this package are outlined below.
&dyl("file.o"): dynamically link file.o. All functions and non-static
variables become visible from within perl. This
function returns a pointer to an internal hash table
corresponding to the symbol table of the newly loaded
code.
eg: $ht = &dyl("sample.o")
This function can also be called with the -L and -l ld options.
eg: $ht = &dyl(""sample2.o", "-L/usr/lib", "-lm")
will also pick up the math library if sample.o
accesses any symbols there.
&dyl_find("func"): find symbol 'func' and return its symbol table entry
&dyl_functions($ht): print the contents of the internal hash table
&dyl_print_symbols($f): prints the contents of the symbol returned by
dyl_find()
There is very little documentation, maybe something to do for a future release. The files sample.o, and sample2.o contain code to be incrementally loaded, test.pl is the test perl script.
Comments are welcome. I submit this code for public consumption and, basically, am not responsible for it in any way.
The undump program comes from the TeX distribution. If you have TeX, then you may have a working undump. If you don't, and you can't get one, *AND* you have a GNU emacs working on your machine that can clone itself, then you might try taking its unexec() function and compiling Perl with -DUNEXEC, which will make Perl call unexec() instead of abort(). You'll have to add unexec.o to the objects line in the Makefile. If you succeed, post to comp.lang.perl about your experience so others can benefit from it.
If you have a version of undump that works with Perl, please submit its anon-FTP whereabouts to the FAQ maintainer.
John Dallman* has written a program "#!perl.exe" which will do this. It is available through anonymous ftp from ftp.ee.umanitoba.ca in the directory /pub/msdos/perl/hbp_20.zip. This program works by finding the script and perl.exe, building a command line and running perl.exe as a child process. For more information on this, contact John directly.