[8777] in Perl-Users-Digest
Perl-Users Digest, Issue: 2394 Volume: 8
daemon@ATHENA.MIT.EDU (Perl-Users Digest)
Wed Apr 22 20:07:19 1998
Date: Wed, 22 Apr 98 17:00:31 -0700
From: Perl-Users Digest <Perl-Users-Request@ruby.OCE.ORST.EDU>
To: Perl-Users@ruby.OCE.ORST.EDU (Perl-Users Digest)
Perl-Users Digest Wed, 22 Apr 1998 Volume: 8 Number: 2394
Today's topics:
Re: 'each' and recursion, do they mix? <jll@skynet.be>
ActiveStates vs. Standard Distribution <mike@borg.com>
Re: Calculating business dates? (Abigail)
Re: Command line instructions don't execute in Win95 <jdf@pobox.com>
Confile & Perl mod ref <snx@snx.com>
Re: Confile & Perl mod ref (Jason Gloudon)
control characters!! (Michael Russo)
Re: control characters!! (Craig Berry)
Re: Count some word in a string. (Bart Lateur)
Cron job and Date problem (Alan Voorhees)
Re: Cron job and Date problem <rootbeer@teleport.com>
Re: Defending Perl <rootbeer@teleport.com>
Re: File Manipulation <quentin@jihad.amd.com>
Re: Getting rid of the DOS console window when running <warwick@webdev.co.uk>
Re: Help: To Kill an Object <rootbeer@teleport.com>
How to save an external program output in a var? <yong@shell.com>
how to use perl behind a firewall? (Yuming Huang)
Re: how to use perl behind a firewall? <Tony.Curtis+usenet@vcpc.univie.ac.at>
ODBC/DBI on WIN32 <dennis.kowalski@daytonoh.ncr.com>
Perl Echo Server <jferris@jdfdesign.com>
Re: Perl for NT <rootbeer@teleport.com>
Re: Perl WIN32 Remote Process Monitoring warren@jen.co.il
Re: POSIX::setuid and changing the UID (Jason Gloudon)
Re: Problem with Perl read on NT (Bbirthisel)
Re: QUESTION: foreach $scalar can this be foreach $hash <rootbeer@teleport.com>
Re: Question? (gaj)
Re: Regular Expression to match a valid IP address (Fritz Knack)
Re: Regular Expression to match a valid IP address <quentin@jihad.amd.com>
Re: Removing spaces from a variable, how? <quentin@jihad.amd.com>
Re: Removing spaces from a variable, how? (Craig Berry)
Re: RMS should be invited to O'Reilly's "Free Software (Eugene O'Neil)
Re: Sendmail Attach <quednauf@nortel.co.uk>
Re: Setuid tips aren't working! <rootbeer@teleport.com>
Re: Testing <quentin@jihad.amd.com>
Re: Testing <quednauf@nortel.co.uk>
Re: Time : Year2000 & 2038 code question lvirden@cas.org
Re: Time : Year2000 & 2038 code question <bland@sprint.ca>
Tk Listbox get("@x,y") confusion. (Wayne C. McCullough)
Re: What is OFFICIAL version of Perl FAQ? (M.J.T. Guy)
Re: where's the faq? <tchrist@mox.perl.com>
Re: WIN32::Process::Create doesn't work in Win 95. Why? <jimbo@soundimages.co.uk>
Digest Administrivia (Last modified: 8 Mar 97) (Perl-Users-Digest Admin)
----------------------------------------------------------------------
Date: Wed, 22 Apr 1998 16:35:58 +0200
From: Jean-Louis Leroy <jll@skynet.be>
Subject: Re: 'each' and recursion, do they mix?
Message-Id: <VA.000000a0.5cf55d7b@jll>
Ok, I've done what I should have began with: run a benchmark. And the
results are interesting. I used a 20-element hash, and tried four
methods: use 'each', do it the obvious way (foreach $k (keys %h)...),
and two variants that iterate over a flattened hash (one with shift,
the other with splice).
Use strict;
use vars qw(%h $n);
%h = 1..40;
print scalar(%h), "\n"; # 16/32
use Benchmark;
$n = 10000;
print 'each: ';
timethis($n, q{ while (my ($k, $v) = each %h) { } } );
print 'obvious: ';
timethis($n, q{ foreach my $k (keys %h) { $h{$k} } } );
print 'flatten_shift: ';
timethis($n, q{ my @pairs = %h;
while (my ($k, $v) = shift @pairs, shift @pairs) { } } );
print 'flatten_splice: ';
timethis($n, q{ my @pairs = %h;
while (my ($k, $v) = splice @pairs, 0, 2) { } } );
__SUSPEND_PERL-SPEAK__
For the fun I also factorized both the flatten-shift and the obvious
methods using prototypes:
__RESUME_PERL-SPEAK__
sub factorized_shift (&%)
{
my $f = shift;
while (@_) { &$f(shift, shift) }
}
sub factorized_obvious (&\%)
{
my ($f, $h) = @_;
foreach my $k (keys %$h) { &$f($k, $h->{$k}) }
}
#factorized_shift { print "$_[0] => $_[1] " } %h; print "\n";
#factorized_obvious { print "$_[0] => $_[1] " } %h; print "\n";
print 'factorized_shift: ';
timethis($n, q{ factorized_shift { $_[0], $_[1] } %h } );
print 'factorized_obvious: ';
timethis($n, q{ factorized_obvious { $_[0], $_[1] } %h } );
The results are, on an Olivetti Modulo P-133-L, 188mb RAM, NT 4.0 SP 3:
each: timethis 10000: 15 secs (15.47 usr 0.00 sys = 15.47 cpu)
obvious: timethis 10000: 7 secs ( 6.49 usr 0.00 sys = 6.49 cpu)
flatten_shift: timethis 10000: 17 secs (16.53 usr 0.00 sys = 16.53
cpu)
flatten_splice: timethis 10000: 21 secs (21.60 usr 0.00 sys = 21.60
cpu)
factorized_shift: timethis 10000: 15 secs (15.71 usr 0.00 sys = 15.71
cpu)
factorized_obvious: timethis 10000: 16 secs (15.33 usr 0.00 sys =
15.33 cpu)
It seems that the most obvious method is also the best...
I tried a 200-pair hash (158/256) and the relative performance of the
four methods remains unchanged.
At this point I'd really like to know why 'each' is so slow...
Jean-Louis Leroy
http://ourworld.compuserve.com/homepages/jl_leroy
------------------------------
Date: Wed, 22 Apr 1998 11:41:52 -0400
From: "Michael Wil. Swiercz" <mike@borg.com>
Subject: ActiveStates vs. Standard Distribution
Message-Id: <353E0FBC.25E7@borg.com>
DISCLAIMER: Before I get zinged for using NT rather than UNIX, let me
state that whenever I have a choice, I use UNIX. But sometimes corporate
decisions are not logical, and other times product integration on a
platform are more complete than others. But that is another discussion
topic.
TO THE QUESTION: Has anyone tried to use ActiveState's ISAPI version of
Perl (PerlIS) with the Standard Distribution? Is it even possible? I
find myself between two extremes. I prefer the Standard Distribution due
to its completeness, however, ActiveState's ISAPI version provides
better response time and performance (yes, I'm forced to use MS IIS).
Many moons ago, I battled with getting Perl installed and usuable on an
NT Server. Thus, I figured to ask now, before I find myself ripping the
remaining hair out of my head. Any feedback would be appreciated.
Best Regards,
Mike :)
------------------------------
Date: 22 Apr 1998 15:14:42 GMT
From: abigail@fnx.com (Abigail)
Subject: Re: Calculating business dates?
Message-Id: <6hl1h2$bs0$1@client3.news.psi.net>
Timothy Reed (treed@cpr.com) wrote on MDCXCV September MCMXCIII in
<URL: news:353D716D.5DB98A7F@cpr.com>:
++ Hi,
++ I need to determine if a given date falls (or fell) on a business date. I als
++ need to tell if a day x days ahead or behind is a business date or note. Has
++ anyone worked out a solution in Perl?
Yes.
It's called Date::Manip.
Abigail.
--
perl5.004 -wMMath::BigInt -e'$^V=new Math::BigInt+qq;$^F$^W783$[$%9889$^F47$|88768$^W596577669$%$^W5$^F3364$[$^W$^F$|838747$[8889739$%$|$^F673$%$^W98$^F76777$=56;;$^U=substr($]=>$|=>5)*(q.25..($^W=@^V))=>do{print+chr$^V%$^U;$^V/=$^U}while$^V!=$^W'
------------------------------
Date: 22 Apr 1998 10:21:00 -0500
From: Jonathan Feinberg <jdf@pobox.com>
To: "Creede Lambard" <fearless@io.com>
Subject: Re: Command line instructions don't execute in Win95
Message-Id: <hg3lanmr.fsf@mailhost.panix.com>
[posted and mailed]
"Creede Lambard" <$_=qq!fearless\@NOSPAMio.com!;y/A-Z//d;print> writes:
> This might be a FAQ but I'll answer it anyway: WinPerl really wants
> double quotes when you enclose parameters for the -e command.
Not so; it's the DOS *shell* that doesn't understand single quotes.
Perl on Win32 has exactly the same syntax as other perls.
--
Jonathan Feinberg jdf@pobox.com Sunny Brooklyn, NY
------------------------------
Date: Wed, 22 Apr 1998 13:41:51 -0400
From: Stacy Hunter <snx@snx.com>
Subject: Confile & Perl mod ref
Message-Id: <353E2BD5.916C0D3E@snx.com>
I have a cronfile which executes fine.
I have a separate script with a perl module and local library reference
that executes fine from the command line.
However, when I try to execute the same script from the cronfile I get a
cron output error message that says,
"...Can't locate Date/Manip.pm in @INC (@INC contains: ..."
The only difference is how the script is getting initiated. The path
from the file getting initiated by cronfile hasn't changed.
Anyone have any ideas as to why this won't execute?
Any help would be most appreciated.
- Stacy
------------------------------
Date: Wed, 22 Apr 1998 23:34:35 GMT
From: jgloudon@hyssop.bbn.com.bbn.com (Jason Gloudon)
Subject: Re: Confile & Perl mod ref
Message-Id: <slrn6jsvk3.2fi.jgloudon@hyssop.bbn.com>
Stacy Hunter <snx@snx.com> wrote:
.
>"...Can't locate Date/Manip.pm in @INC (@INC contains: ..."
>
>The only difference is how the script is getting initiated. The path
>from the file getting initiated by cronfile hasn't changed.
You need to tell perl where your modules are located, as the environment it
runs in when run from cron is not as complete as you interactive shell
environment. You can do this by adding putting
use lib '/path/to/libdir';
where this is the directory containing the Date module directory.
--
Jason Gloudon
------------------------------
Date: 22 Apr 98 16:45:03 GMT
From: russo02@zon.eelab.newpaltz.edu (Michael Russo)
Subject: control characters!!
Message-Id: <353e1e8f.0@motss>
Keywords: characters
I'm trying to write a script to remove '^M' and replace it with control-M,
which looks like '^M', but is only one character. (In vi, you would type
<ctrl-V><ctrl-M> to achieve this.) I can't seem to make any progress.
The search-replace operator sees '^M' as two seperate characters...
------------------------------
Date: 22 Apr 1998 17:49:31 GMT
From: cberry@cinenet.net (Craig Berry)
Subject: Re: control characters!!
Message-Id: <6hlajb$o01$4@marina.cinenet.net>
Michael Russo (russo02@zon.eelab.newpaltz.edu) wrote:
: I'm trying to write a script to remove '^M' and replace it with control-M,
: which looks like '^M', but is only one character. (In vi, you would type
: <ctrl-V><ctrl-M> to achieve this.) I can't seem to make any progress.
: The search-replace operator sees '^M' as two seperate characters...
In a doublequoted string (or other places that do doublequotish
processing, like regex patterns and backtick strings), you can write
ctrl-M as \cM. So your substitution could be written as:
s/\^M/\cM/g;
Do be sure, though, that those '^M' strings are really in your data,
rather than an artifact of the tool you're using to display the data. On
unix systems, the 'od' tool comes in very handy in checking stuff like
this.
---------------------------------------------------------------------
| Craig Berry - cberry@cinenet.net
--*-- Home Page: http://www.cinenet.net/users/cberry/home.html
| Member of The HTML Writers Guild: http://www.hwg.org/
"Every man and every woman is a star."
------------------------------
Date: Wed, 22 Apr 1998 11:58:22 GMT
From: bart.mediamind@tornado.be (Bart Lateur)
Subject: Re: Count some word in a string.
Message-Id: <353dd87e.2145645@news.tornado.be>
Chao-cheng Huang wrote:
>Assume I have a string
>$s = "aa....aa....aa....aa....aa" ;
>Ans I would like to calculate how much times "aa" happened.
>I have these two code:
>
>$cnt = $s =~ s/(aa)/$1/g ; # $cnt = 5
>$cnt = $s =~ /aa/g ; # $cnt = 1
>
>When I read perl man page , the modifier 'g' means global match.
>So, the code should get the same value. Why they are different?
It's a matter of context. See PERLOP.POD. In a scalar context (second
case), the behaviour is intended to be used in code like this:
while ($s =~ /aa/g) {
print "Got one!\n";
}
Try it. It's fun. So every time you go through the loop, you get another
match, until all are souped up.
Anyway, in a list context (aka "array context"), m//g will return an
array of all matches. So if you assign this to an array, you can get at
the count alright:
$cnt = @ary = $s=~/aa/g;
print "$cnt\n"; #prints: 5
In modern Perl ports, you don't even need an actual array:
$cnt = () = $s=~/aa/g;
p.s. If you try this in a line immediately following your second case,
the result will be a puzzling 4, not 5. That is because you just "ate"
one match already.
HTH,
Bart.
------------------------------
Date: Wed, 22 Apr 1998 16:19:10 -0700
From: radio@hotmail.com (Alan Voorhees)
Subject: Cron job and Date problem
Message-Id: <radio-ya02408000R2204981619100001@nntp.best.com>
I'm trying to get a PERL script to run as a cron job. The script is written
to prune items from a database that are 30 days old. If I run the script at
the command line it functions fine, it purunes the list and outputs a file
with the items that were deleted as it should. When it runs as a cron job
it doesn't delete anything and returns an empty list. Any ideas on how to
get PERL to recognize the dates in this process?
I'm running the script on a Sun Sparc w/Solaris.
------------------------------
Date: Wed, 22 Apr 1998 23:44:11 GMT
From: Tom Phoenix <rootbeer@teleport.com>
To: Alan Voorhees <radio@hotmail.com>
Subject: Re: Cron job and Date problem
Message-Id: <Pine.GSO.3.96.980422164324.6132D-100000@user2.teleport.com>
On Wed, 22 Apr 1998, Alan Voorhees wrote:
> I'm trying to get a PERL script to run as a cron job.
Like CGI scripts, cron jobs run in a different environment than you may be
expecting. But if that's not the problem, can you show us some small piece
of code (say, under ten lines) which isn't doing what you want? Hope this
helps!
--
Tom Phoenix Perl Training and Hacking Esperanto
Randal Schwartz Case: http://www.rahul.net/jeffrey/ovs/
------------------------------
Date: Wed, 22 Apr 1998 23:09:13 GMT
From: Tom Phoenix <rootbeer@teleport.com>
To: Marjorie Roswell <roswell@umbc.edu>
Subject: Re: Defending Perl
Message-Id: <Pine.GSO.3.96.980422160220.6132z-100000@user2.teleport.com>
On Wed, 22 Apr 1998, Marjorie Roswell wrote:
> Does anyone else see a need for a "debunking Perl myths" FAQ?
I'm not sure. What questions and answers would it have? Write up the FAQ,
and I can tell you whether I think it's needed or not. :-)
More seriously, I think you're looking for Perl Advocacy. That's good.
Here are some potential resources.
http://language.perl.com/versus/
You may also wish to submit your request to the Perl Wish List, which is
maintained by The Perl Institute.
http://www.perl.org/wishlist.html
http://www.perl.org/
Hope this helps!
--
Tom Phoenix Perl Training and Hacking Esperanto
Randal Schwartz Case: http://www.rahul.net/jeffrey/ovs/
------------------------------
Date: 22 Apr 1998 08:32:37 -0500
From: Quentin Fennessy <quentin@jihad.amd.com>
Subject: Re: File Manipulation
Message-Id: <ximemyqf0cq.fsf@jihad.amd.com>
>>>>> "Roger" == Roger Dooley <dooleyr@ccis12.baylor.edu> writes:
Roger> 1. Can I read a line from a file and extract a certain
Roger> portion of that line without having to read, store the
Roger> text, and then grab the relevant information?
Yes - open a file, slurp it or read it line by line, and when you get
to the correct portion save it to a variable of some sort. THis code
assumes you can match your input line with the regex /pattern-to-match/,
and the code saves that line in $variable, as well as splitting the same
line into @my_list. You never have to explicitly `save the text'
when programming like this -- while (<F>) { } lets you read it and
deal with it line by line.
open(F, "/path/to/file") or die "Could not open file";
while (<F>) {
next unless /pattern-to-match/;
$variable = $_;
@my_list = split;
}
close(F);
Roger> 2. If I have a file with some data in it, can I open the
Roger> file and change some of the data. For example if I have the
Roger> following line:
Not really, but that is not a problem. In perlfaq5(1) (the man page)
there is a section headed:
How do I change one line in a file/delete a line in a
file/insert a line in the middle of a file/append to the
beginning of a file?
Your answer is here, along with suggestions for pseudo-inplace edits.
Have fun with Perl! Learning Perl is an excellent place to start.
You will benefit greatly from reading all of the FAQ man pages
included with Perl.
--
Quentin Fennessy AMD, Austin Texas
Secret hacker rule #11 - hackers read manuals
------------------------------
Date: Wed, 22 Apr 1998 09:25:49 +0100
From: "warwick" <warwick@webdev.co.uk>
Subject: Re: Getting rid of the DOS console window when running a Perl program
Message-Id: <353da81d.0@nnrp1.news.uk.psi.net>
Hi
To get rid of the console window create a VC console project.
Compile the attached file.
When linking goto Project/Setting/Link tab
add these to the project options:
/nodefaultlib
libcmt.lib
and path to perl.lib location
eg c:\perl 5004\bin\perl.lib
FreeConsole(); in the c file loses the console window.
to use:
perlmain flags script option list
perlmain -T script.pl -config config.pl
Warwick
Bill Hess wrote in message ...
>I am running Perl 5.004_04 on WinNT/Win95 and I need to run a Perl Program
>with no DOS Console window popping up - how can I do this???
>
>Bill Hess
>brhess@msn.com
>
>
begin 666 perlmain.c
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7,"D[#0H-"@ER971U<FXH,"D[#0I]#0H`
`
end
------------------------------
Date: Wed, 22 Apr 1998 23:42:37 GMT
From: Tom Phoenix <rootbeer@teleport.com>
To: Jerome Bradenbaugh <jerome.bradenbaugh@firstunion.com>
Subject: Re: Help: To Kill an Object
Message-Id: <Pine.GSO.3.96.980422164135.6132C-100000@user2.teleport.com>
On Wed, 22 Apr 1998, Jerome Bradenbaugh wrote:
> This works fine, but I think the object remains after it has performed
> its dirty work. Is there a way to kill an object after each use, or is
> there perhaps a way to request multiple documents with the same object?
Yes, and yes. Check out the info on destructors in perlobj for the first.
If you're wanting to re-use an object, though, generally... you re-use the
object! :-) Hope this helps!
--
Tom Phoenix Perl Training and Hacking Esperanto
Randal Schwartz Case: http://www.rahul.net/jeffrey/ovs/
------------------------------
Date: Wed, 22 Apr 1998 11:00:39 -0500
From: Yong Huang <yong@shell.com>
Subject: How to save an external program output in a var?
Message-Id: <353E1427.C3AF7B43@shell.com>
Suppose I want to check user login to sqlplus (a program to let you
login to Oracle database):
open SAVEOUT, ">&STDOUT";
open STDOUT, ">mytmpfile" or die "Can't do it"; #Can I not
open a file?
open SQL, " | sqlplus $user/$passwd" or die "Can't run it";
...
close SQL; #followed by restore STDOUT...
...
Then I can check the file mytmpfile to see if it contains "Invalid
username". But I want to optimize the code so the string "Invalid
username" is not saved in an external file (which takes time in I/O) but
in a Perl variable. How can I do it? Thanks for any advice.
Yong Huang
Email:yong@shell.com
------------------------------
Date: Wed, 22 Apr 1998 13:58:10 GMT
From: yh39@research.att.com (Yuming Huang)
Subject: how to use perl behind a firewall?
Message-Id: <ErtIsz.6nr@research.att.com>
I tired to use perl to access a website outside our firewall and failed. As we
know, to use netscape behind a firewall, you should set proxy parameter first.
does anyone know how to config perl to make it work (to access website outside
firewall)?
Thanks,
Yuming
yuming@att.com
630-810-7856
------------------------------
Date: 22 Apr 1998 16:31:11 +0200
From: Tony Curtis <Tony.Curtis+usenet@vcpc.univie.ac.at>
Subject: Re: how to use perl behind a firewall?
Message-Id: <7xvhs22aj4.fsf@beavis.vcpc.univie.ac.at>
Re: how to use perl behind a firewall? , Yuming
<yh39@research.att.com> said:
Yuming> I tired to use perl to access a website outside our
Yuming> firewall and failed. As we know, to use netscape
Yuming> behind a firewall, you should set proxy parameter
Yuming> first. does anyone know how to config perl to make
Yuming> it work (to access website outside firewall)?
The LWP::UserAgent module allows you to set proxies in a
fairly intuitive syntax.
tony
--
Tony Curtis, Systems Manager, VCPC, | Tel +43 1 310 93 96 - 12; Fax - 13
Liechtensteinstrasse 22, A-1090 Wien, AT | http://www.vcpc.univie.ac.at/
"You see? You see? Your stupid minds! Stupid! Stupid!" ~ Eros, Plan9 fOS.
------------------------------
Date: Wed, 22 Apr 1998 10:26:51 -0400
From: Dennis Kowalski <dennis.kowalski@daytonoh.ncr.com>
Subject: ODBC/DBI on WIN32
Message-Id: <353DFE2A.425E@daytonoh.ncr.com>
I am running on a NT 4.0 host and using the build 315 port from
ActiveWare.
Can anyone tell me how to get the modules needed to access an ORACLE
database in this WIN32 environment ??
I do not have a C compiler on my NT box.
I have read about the ODBC and DBI interfaces but I do not seem to have
either one in my Perl installation.
Thanks
------------------------------
Date: Wed, 22 Apr 1998 16:11:53 -0700
From: "Joseph M. Ferris" <jferris@jdfdesign.com>
Subject: Perl Echo Server
Message-Id: <353E7939.792021DB@jdfdesign.com>
Hi Everyone,
I am writing a VB app that is TCP/IP capable. What I need to know is
that if knows where I could find the Perl source for an echo server that
allows multiple connections. I run on a Unix webserver, so I can't use
a Windows server for this.
Please respond by email,
Joseph M. Ferris
jferris@jdfdesign.com
------------------------------
Date: Wed, 22 Apr 1998 23:01:32 GMT
From: Tom Phoenix <rootbeer@teleport.com>
To: Eric Finlayson <Buzzbane@iname.com>
Subject: Re: Perl for NT
Message-Id: <Pine.GSO.3.96.980422155607.6132y-100000@user2.teleport.com>
On 21 Apr 1998, Eric Finlayson wrote:
> I have been hired to build a shopping car for a site.
Wow, a shopping car. Gas or electric? :-)
> But it is on a NT box and I have only had experience programming CGI on
> Unix boxes. I was told there is a major difference.
The differences between Unix Perl and NT Perl are merely the differences
between the underlying systems. For example, if you're working with
filenames, those are different. (But, of course, if you use
File::Basename, your programs will be portable.)
There may be differences between Unix CGI scripting and NT CGI scripting,
but I'm sure that, if you wanted to know about those, you would have asked
in a newsgroup about CGI scripting. :-)
A detailed look at the differences in Windows Perl is part of the
Perl-for-Win32 FAQ.
http://cpan.perl.org/doc/FAQs/win32/Perl_for_Win32_FAQ.html
Hope this helps!
--
Tom Phoenix Perl Training and Hacking Esperanto
Randal Schwartz Case: http://www.rahul.net/jeffrey/ovs/
------------------------------
Date: Wed, 22 Apr 1998 16:36:34 -0600
From: warren@jen.co.il
Subject: Re: Perl WIN32 Remote Process Monitoring
Message-Id: <6hlnt2$jsb$1@nnrp1.dejanews.com>
In article <353D2F78.8400AB93@tir.com>,
reboot@tir.com wrote:
>
> Does anyone know of a way to monitor processes running on a remote/network
> server? Thanx in advance for any info.
I don't know, but I've been wondering for a while how to monitor processes
running locally on Win32 (all processes, not just processes I created), and
I'm hoping anyone answering your question can answer mine, too. On UNIX I
open a pipe from ps to do this.
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/ Now offering spam-free web-based newsreading
------------------------------
Date: Wed, 22 Apr 1998 23:15:57 GMT
From: jgloudon@hyssop.bbn.com.bbn.com (Jason Gloudon)
Subject: Re: POSIX::setuid and changing the UID
Message-Id: <slrn6jsuh5.2fi.jgloudon@hyssop.bbn.com>
Gary Holt <holt@kelp.usc.edu> wrote:
>o Assigning to $EUID. This works:
>
># perl -e 'use English; $EUID = 500; open(X, ">x");'
># ls -l x
>-rw-r--r-- 1 holt root 0 Apr 22 09:18 x
>
> which changed the user name, as I wanted. However, there seems to be a
> security problem with this: I can change the $EUID back. For example,
>
># perl -e 'use English; $EUID = 500; $EUID = 0; open(X, ">x");'
This is meant to be this way so that you can juggle privileges in a setuid
program.
> When my daemon process forks off other processes running under other
> UIDs, I don't want them to be able to change UID back to root. (This
.
>o Trying to change both $EUID and $UID doesn't work any differently from
> changing $EUID alone.
This is not true across an exec. If you only change EUID or UID then it
is possible for the process you exec to regain a UID 0 privileges.
If you change BOTH UID and EUID to a non-root uid, then once you exec a
process, that process is no longer able to regain UID 0 privileges.
On linux 2.0.30, and Solaris 2.5.1 this illustrates what i'm saying:
---- parent
#!/usr/local/bin/perl
use English;
print "Parent:$UID $EUID\n";
$UID = 1001;
$EUID = 1001;
print "Parent:$UID $EUID\n";
unless (fork){
exec './child';
}
---- child
#!/usr/local/bin/perl
use English;
print "Child:$UID, $EUID\n";
$UID = 0;
$EUID = 0;
print "Child:$UID, $EUID\n";
-----
Additionally you should also set the GID's -and call initgroups (3) as well-
to fix the group memberships of the child processes.
--
Jason Gloudon
------------------------------
Date: 22 Apr 1998 13:41:00 GMT
From: bbirthisel@aol.com (Bbirthisel)
Subject: Re: Problem with Perl read on NT
Message-Id: <1998042213410000.JAA19303@ladder03.news.aol.com>
Hi Virgina:
>try to read from a file that we have opened successfully using
> while (<INPUT_FILE>)
>we can't manage to enter the while loop eventhough stat
>shows that the file exist and the file size is not 0.
> and only fails on NT.
Does the file contain non-text characters? NT cares (you
need "binmode" for those).
-bill
------------------------------
Date: Wed, 22 Apr 1998 22:06:28 GMT
From: Tom Phoenix <rootbeer@teleport.com>
To: Terrence Brannon <brannon@lnc.usc.edu>
Subject: Re: QUESTION: foreach $scalar can this be foreach $hash{ref}
Message-Id: <Pine.GSO.3.96.980422144941.6132u-100000@user2.teleport.com>
On 22 Apr 1998, Terrence Brannon wrote:
> As you can see, below I had to use a scalar as the argument for my
> foreach loops because apparently perl cant assign to the keys of a
> hash within the foreach stmt.
Who says "perl cant"? :-) But you never really "assign to the keys of a
hash". You assign to a hash element. And you can do that within a loop, no
problem.
Maybe you mean that the control variable of a foreach loop can't be
anything but a simple scalar. It can't be an element of a hash or array.
You'd very rarely need that, but that very rare case is easy to work
around.
> foreach $v_factor_40 (@factor) {
> foreach $v_factor_51 (@factor) {
> foreach $v_factor_32 (@factor) {
> foreach $v_factor_25 (@factor) {
> $v{factor_40}=$v_factor_40;
> $v{factor_51}=$v_factor_51;
> $v{factor_32}=$v_factor_32;
> $v{factor_25}=$v_factor_25;
>
> run_jobs_{$branch}'br';
Huh? What's that? Is that valid syntax?
It would be more efficient, though, to not re-assign each element each
time through the innermost loop.
foreach $v_factor_40 (@factor) {
$v{factor_40}=$v_factor_40;
foreach $v_factor_51 (@factor) {
$v{factor_51}=$v_factor_51;
foreach $v_factor_32 (@factor) {
$v{factor_32}=$v_factor_32;
foreach $v_factor_25 (@factor) {
$v{factor_25}=$v_factor_25;
If, of course, nothing else in the loops will damage those elements of %v.
Hope this helps!
--
Tom Phoenix Perl Training and Hacking Esperanto
Randal Schwartz Case: http://www.rahul.net/jeffrey/ovs/
------------------------------
Date: Wed, 22 Apr 1998 18:14:07 GMT
From: gjandl@rf.mnschoolbiz.com (gaj)
Subject: Re: Question?
Message-Id: <353e328d.90071669@news.spacestar.net>
On 18 Apr 1998 14:06:41 GMT, mrjeff@blithe.ircbar.com (Jeff Workman)
wrote:
>Don't you ever have anything constructive to say? Or are you here merely to
>belittle those who aren't as 3733+ as you?
Um...reading the fscking manual *would* be a rather constructive use
of this person's time. In addition, contrary to the (false)
accusations that are posted to clpm, Abigail often does answer
questions. She did this time. How would *you* do a 'getline' in Perl,
if not with the diamond operator? Besides, I don't see *you* posting
an answer to the question.
------------------------------
Date: Wed, 22 Apr 1998 12:16:46 GMT
From: fritz.knack@nospam.POPULUS.net (Fritz Knack)
Subject: Re: Regular Expression to match a valid IP address
Message-Id: <353ddefc.3286858@cnews.newsguy.com>
On 22 Apr 1998 02:56:59 GMT, Tom Christiansen <tchrist@mox.perl.com>
wrote:
>In comp.lang.perl.misc, "Rob Smith" <robsmith@writeme.com> writes:
>:I'm definitely new to Perl, and have been working on a project for work.
>:I'm at a point where I want to validate an IP entered.
>
>Skip the regex approach. Seize the wheel, don't rebuild it. What you
>should do is this:
>
> #!/usr/bin/perl
> use Socket;
> for $ip ( @ARGV ) {
> printf "%s is %s a valid ip address\n", $ip,
> defined(inet_aton($ip)) ? "indeed" : "NOT";
> }
>
>Anything else will be very complicated, wrong, or most likely both.
>
>--tom
That seems to work nicely on the Sun box to which I have access, but I
gotta tell you it stinks on my Win95 machine. No formal benchmarks,
but it took approximately 4 seconds (on a Dell P200) to verify that
127.0.0.1 is indeed valid. Yuck.
Fritz
-------------------------
Sorry 'bout the nospam in the From field. You know how those 'bots are.
------------------------------
Date: 22 Apr 1998 09:38:24 -0500
From: Quentin Fennessy <quentin@jihad.amd.com>
Subject: Re: Regular Expression to match a valid IP address
Message-Id: <xim67k1gbvj.fsf@jihad.amd.com>
>>>>> "John" == John Porter <jdporter@min.net> writes:
John> It is "equivalent to", it does not "expand to". 10.1 is a
John> valid IP address, and should be recognized so by any test
John> one writes. For that matter, 10.16777215 is a valid IP
John> address, and is equivalent to 10.255.255.255.
John, you may be technically correct, but the intent of the original
post (I suppose) is to match via regex the common way of representing
IP addresses. An IP address is a 32 bit quantity. Some 32 bit values
are unlikely to be valid IP addresses. By using the common
dotted-quad representation we can visually (or via regex) check for
basic errors.
The valid IP address of 192.100.12.1 might be represented in several
ways: base 16, base 27, Roman numerals, or whatever.
192.100.12.1 = CXCII.C.XII.I
= 3,190,129,561
= ...
You suggest that some uncommon representations be matched by `any test
one writes' (and I know you did not suggest Roman numnerals!). I
don't think that is useful for most folks. Most folks want what most
folks recognize, dotted quad addresses.
--
Quentin Fennessy AMD, Austin Texas
Secret hacker rule #11 - hackers read manuals
------------------------------
Date: 22 Apr 1998 17:08:46 -0500
From: Quentin Fennessy <quentin@jihad.amd.com>
Subject: Re: Removing spaces from a variable, how?
Message-Id: <ximvhs1ecgh.fsf@jihad.amd.com>
>>>>> "Dan" == Dan Boorstein <danboo@negia.net> writes:
Dan> i'm not sure this is doing what you expect. after the first
Dan> s/ //g $spaces no longer has spaces. if you move those
Dan> assignments inside the sub declaration you will see some
Dan> different results:
Dan> draw your own conclusions, but it looks to me like 'tr' is
Dan> usually a safe bet for these sort of operations.
Oops. Thanks for catching my mistake, Dan. I idly considered the fact
that you repeated the 'my $var =...' line and thought it a mistake.
I guess I missed the point. tr/// remains the champ.
--
Quentin Fennessy AMD, Austin Texas
Secret hacker rule #11 - hackers read manuals
------------------------------
Date: 22 Apr 1998 17:37:19 GMT
From: cberry@cinenet.net (Craig Berry)
Subject: Re: Removing spaces from a variable, how?
Message-Id: <6hl9sf$o01$3@marina.cinenet.net>
Sleep (mhchau@cse.cuhk.edu.hk) wrote:
: Craig Berry <cberry@cinenet.net> wrote:
: > bjf (ben@wallroyds.demon.co.uk) wrote:
: > : How do you remove all spaces ... ie ... " " .... from a variable using
: > : pattern-matching?
:
: > This is actually a much better job for tr:
: > $var =~ tr/ //d;
:
: what's the difference if we use :
: $var =~ s/ //g;
: ?
No difference semantically, but it's slower.
---------------------------------------------------------------------
| Craig Berry - cberry@cinenet.net
--*-- Home Page: http://www.cinenet.net/users/cberry/home.html
| Member of The HTML Writers Guild: http://www.hwg.org/
"Every man and every woman is a star."
------------------------------
Date: Wed, 22 Apr 98 15:38:27 GMT
From: eugene@cs.umb.edu (Eugene O'Neil)
Subject: Re: RMS should be invited to O'Reilly's "Free Software Summit"
Message-Id: <6hl2tb$euf@dfw-ixnews8.ix.netcom.com>
In article <m3iuoam7w5.fsf@stoli.spirits.org.au>, nathanh@chirp.com.au wrote:
>tzs@halcyon.com (Tim Smith) writes:
>> (2) If by having something free of mine in the proprietary code, the
>> proprietary product is improved, I have helped the users of that
>> proprietary product.
>
>You have allowed commercial software to continue to exist, when in
>all probability the free software would have eventually caught and
>surpassed the commercial equivalents. You have allowed an inferior
>parasitic form of software to survive when it should have died.
But commercial software is only a small part of a much larger problem: We must
destroy commercialism itself, to be truly free of commercial software! Once
the workers control the means of production, we will have a glorious
Revolution in software productivity! Kill the capitalist lacky running dogs!
All hail the Revolution, comrade!
>Supposedly "free" licenses that allow commercial parasites to take
>without giving are unfairly slanting the playing field away from a
>free software society towards a commercial software society. It is
>not conducive to the goal of software being free.
The BSD license allows people to enhance software with new features, and
actually CHARGE PEOPLE MONEY to use their improved version. This fiendish
practice of expecting money in exchange for work must be stopped! We should
put all free software under the GPL, and if those BSD-loving bastards don't
like it, they don't have to improve anything at all. That will make the world
a better place, dammit!
>You are right that it doesn't (directly) affect the freedom of the
>individual package, but non-GPL alike licenses hurt the freedom of
>software itself. You owe it to yourself to contribute code under a
>license that promotes further contributions.
Yes, software that people can do anything they want with isn't really free.
The GPL is better because it FORCES people who make improvements to contribute
them to free software: they have no choice. I mean, think about it! If people
get to CHOOSE whether or not they want to contribute to free software, in what
sense is it really FREE? That's just plain illogical!
>BSD licenses don't do this, thus they are inferior to the GPL.
Don't worry, anyone who belives in inferior ideologies like that will be
hunted down and killed like the vermin they are as soon as we have our
glorious Revolution, which is coming any day now!
Any day... NOW!
NOW NOW NOW NOW!
Oh, well... I'll try again tomorrow.
-Eugene
------------------------------
Date: Wed, 22 Apr 1998 15:03:07 +0100
From: "Frank L. Quednau" <quednauf@nortel.co.uk>
Subject: Re: Sendmail Attach
Message-Id: <353DF89B.B364EFD9@nortel.co.uk>
InterRed wrote:
> How can I send email with attach!...
>
> thanks!
>
Or, to come back to the subject visit my webpage:
http://www.surrey.ac.uk/~me51fq
There, in the perl section (the wrench) you will find a perl
subroutine which uses MIME::Base64 to send an attachment through
a Perl script.
Hope this helps
>
--
____________________________________________________________
Frank Quednau
Phone: +44 (0)1279 402447
http://www.surrey.ac.uk/~me51fq MailTo:F.L.Quednau@bnr.co.uk
____________________________________________________________
------------------------------
Date: Wed, 22 Apr 1998 22:15:23 GMT
From: Tom Phoenix <rootbeer@teleport.com>
To: Nate Grover <natedogg@webbnet.com>
Subject: Re: Setuid tips aren't working!
Message-Id: <Pine.GSO.3.96.980422150952.6132w-100000@user2.teleport.com>
On Wed, 22 Apr 1998, Nate Grover wrote:
> ($<,$>) = ($>,$<);
> # thought that might help, but it doesn't seem to do anything
Your system might not allow you to swap uid and euid. In any case, since
errors here can open security holes, it's always good to check that the
change "took".
> If I am the owner, and I run this, it
> works, but calling it over the web fails.
On some systems, the webserver runs as user nobody. If user nobody is uid
-1 (or 65535), some systems (such as Linux) won't allow that user to
change uids, as a security precaution. On those systems, setid scripts
still run as nobody. For this reason, some webmasters make a user 'noone'
which is id -2 (or 65534), and which can properly run setid scripts.
> The exact same program translated into C works fine.
If C can do this, but Perl can't, your perl binary may be miscompiled. For
example, was it compiled to do setid emulation? My guess is that it
wasn't, so you don't have an sperl or suidperl binary. Recompile, and this
may work for you.
Hope this helps!
--
Tom Phoenix Perl Training and Hacking Esperanto
Randal Schwartz Case: http://www.rahul.net/jeffrey/ovs/
------------------------------
Date: 22 Apr 1998 09:39:20 -0500
From: Quentin Fennessy <quentin@jihad.amd.com>
Subject: Re: Testing
Message-Id: <xim4szlgbtz.fsf@jihad.amd.com>
>>>>> "Frank" == Frank L Quednau <quednauf@nortel.co.uk> writes:
>> mindmore@mindless.com
Frank> WAY COOL EMAIL ADDRESS! IS IT REAL???
Don't we have a regex around to check this address for validity?
(ducking)
--
Quentin Fennessy AMD, Austin Texas
Secret hacker rule #11 - hackers read manuals
------------------------------
Date: Wed, 22 Apr 1998 15:07:46 +0100
From: "Frank L. Quednau" <quednauf@nortel.co.uk>
Subject: Re: Testing
Message-Id: <353DF9B2.DE1B0AA5@nortel.co.uk>
Fain wrote:
> Charlene Yanke wrote:
>
> > This is only a test.
>
> Your test worked.
> Is this something you do often or is it spontaneous (like you wake up in
> the morning having the urge to run to your computer and test your
> USENET program to make sure that it is working OK (you would do this for
> the same reason the rest of us do it; because USENET is the worlds most
> valuable resource))?
I agree and I find it very difficult to resist the urge to live on USENET.
> mindmore@mindless.com
WAY COOL EMAIL ADDRESS! IS IT REAL???
--
____________________________________________________________
Frank Quednau
Phone: +44 (0)1279 402447
http://www.surrey.ac.uk/~me51fq MailTo:F.L.Quednau@bnr.co.uk
____________________________________________________________
------------------------------
Date: 22 Apr 1998 12:47:24 GMT
From: lvirden@cas.org
Subject: Re: Time : Year2000 & 2038 code question
Message-Id: <6hkoss$j7g$2@srv38s4u.cas.org>
According to Mark-Jason Dominus <mjd@op.net>:
:Yes, but not as a localtime. Unless, as you note, the clock is
:mis-set. So the question becomes: Is it more important to operate
:correctly when the clock is mis-set, or in the presence of a broken
:localtime implementation?
Ah but 'mis-set' can mean more than one thing. For instance, what
does your system do if the battery that keeps the clock going during
power outs do when the battery is dead? I suspect different software
and hardware systems are going to do different things.
There are at least 3 scenarios that I can see:
1. User specifically sets system clock to old date; perhaps to
fool some kind of software licensing, or just to have fun, or to see
how good his / her software really is.
2. User accidentally sets the system clock wrong.
3. In some way, the system software and hardware malfunction so as
to perceive an ancient time.
Each new scenario someone thinks up as a reason the clock might be
set to a date pre 1970 is another area that someone wanting truely
harded software needs to consider. I _do_ hope that folk writing
mission critical software (you know, running life support systems,
guiding missles, creating the ultimate cheesecakes, that sort of stuff)
consider all these types of things...
--
<URL:mailto:lvirden@cas.org> Quote: In heaven, there is no panic,
<*> O- <URL:http://www.teraform.com/%7Elvirden/> only planning.
Unless explicitly stated to the contrary, nothing in this posting
should be construed as representing my employer's opinions.
------------------------------
Date: Wed, 22 Apr 1998 09:42:15 -0400
From: Bojan Landekic <bland@sprint.ca>
Subject: Re: Time : Year2000 & 2038 code question
Message-Id: <353DF3B6.1ACA0439@sprint.ca>
Brad Baxter wrote:
> If I'm reading this right, the author knows that the year will not be less
> than 70 until the year 2000, when the year will be '00'. Therefore, he is
> adding 2000 instead of 1900.
That's right I think.. but the same problem we will experience again in 2100,
2200....xx00 years will have this.. A solution for "all" of them needs to be
found, not just for the 2k problem, which I think is what most companies are
concerned, just watch in 2100, the same problem will come again (-:..
sheeshh...
Adious,
Bojan Landekic (bland@sprint.ca)
------------------------------
Date: 22 Apr 1998 22:15:39 GMT
From: wayne@Glue.umd.edu (Wayne C. McCullough)
Subject: Tk Listbox get("@x,y") confusion.
Message-Id: <6hlq6b$h56$1@hecate.umd.edu>
I am having a very frustrating time useing the @x,y index with
the listbox widget. I have been unable to find a faq that deals with
this.
I am trying to attach a ballon help message that is dependent upon
what the mouse is over in the listbox. So I figured to use the get(@x,y)
command on the listbox. My problem is finding the right combination
of quotes or what have you that works so perl will accept it. I have
tried ''s, qw//, and in desperation nothing, ``s and ""s.
At best I get the error:
bad listbox index "@x,y": must be active, anchor, end, @x,y, or a number
at D:\PERL\lib\site/Tk/Derived.pm line 464.
[above line split for readability]
This is using perl on a win95 machine. Version 5.004_02. My Tk version,
as near as I can tell is "4.2", as reported by:
perl -e "use Tk;print $Tk::version;"
Thank you for your help.
W
------------------------------
Date: 22 Apr 1998 14:38:54 GMT
From: mjtg@cus.cam.ac.uk (M.J.T. Guy)
Subject: Re: What is OFFICIAL version of Perl FAQ?
Message-Id: <6hkvdu$i74$1@lyra.csx.cam.ac.uk>
jc22a70a0-Beyerl <db21@ih4gp756.ih.att.com> wrote:
>
>My question to the authors/maintainers is "What is OFFICIAL version
>of Perl FAQ?" If your answer is Ver. 4, then why can I not get this
>from the CPAN sites?
You can get it from the CPAN sites - it's included in the current
version of Perl.
But it's rather unfortunate that
CPAN/doc/FAQs/FAQ
doesn't contain the latest version.
Mike Guy
------------------------------
Date: 22 Apr 1998 12:07:21 GMT
From: Tom Christiansen <tchrist@mox.perl.com>
Subject: Re: where's the faq?
Message-Id: <6hkmhp$n5e$1@csnews.cs.colorado.edu>
[courtesy cc of this posting sent to cited author via email]
In comp.lang.perl.misc, Jay Eckles <eckje@rhodes.edu> writes:
:Where do I find the faq (and archive if there is one) for this newsgroup?
$ man perlfaq
--
Tom Christiansen tchrist@jhereg.perl.com
What about WRITING it first and rationalizing it afterwords? :-)
--Larry Wall in <8162@jpl-devvax.JPL.NASA.GOV>
------------------------------
Date: 22 Apr 1998 13:17:41 +0100
From: Jim Brewer <jimbo@soundimages.co.uk>
Subject: Re: WIN32::Process::Create doesn't work in Win 95. Why?
Message-Id: <uwwcivymy.fsf@soundimages.co.uk>
I have been able to get Win32::Process to run on Win 95, under perl 5.004_04 standard edition, out of libwin32-0.08. However, Win32::Process from versions 0.09 and 0.10 refuse to load. All three modules have been compiled using MSVC 5.0 using the same build configuration as perl.
Apparently there have been some changes to Win32::Process that make it compatible with Win32::IPC, does anyone know what effect these changes would have to the module when run under Win 95. Bear in mind, it builds and loads and executes flawlessly under Win NT 4.
Any and all suggestions greatly appreciated.
Sincereley,
Jim Brewer
------------------------------
Date: 8 Mar 97 21:33:47 GMT (Last modified)
From: Perl-Request@ruby.oce.orst.edu (Perl-Users-Digest Admin)
Subject: Digest Administrivia (Last modified: 8 Mar 97)
Message-Id: <null>
Administrivia:
The Perl-Users Digest is a retransmission of the USENET newsgroup
comp.lang.perl.misc. For subscription or unsubscription requests, send
the single line:
subscribe perl-users
or:
unsubscribe perl-users
to almanac@ruby.oce.orst.edu.
To submit articles to comp.lang.perl.misc (and this Digest), send your
article to perl-users@ruby.oce.orst.edu.
To submit articles to comp.lang.perl.announce, send your article to
clpa@perl.com.
To request back copies (available for a week or so), send your request
to almanac@ruby.oce.orst.edu with the command "send perl-users x.y",
where x is the volume number and y is the issue number.
The Meta-FAQ, an article containing information about the FAQ, is
available by requesting "send perl-users meta-faq". The real FAQ, as it
appeared last in the newsgroup, can be retrieved with the request "send
perl-users FAQ". Due to their sizes, neither the Meta-FAQ nor the FAQ
are included in the digest.
The "mini-FAQ", which is an updated version of the Meta-FAQ, is
available by requesting "send perl-users mini-faq". It appears twice
weekly in the group, but is not distributed in the digest.
For other requests pertaining to the digest, send mail to
perl-users-request@ruby.oce.orst.edu. Do not waste your time or mine
sending perl questions to the -request address, I don't have time to
answer them even if I did know the answer.
------------------------------
End of Perl-Users Digest V8 Issue 2394
**************************************
