[18852] in Perl-Users-Digest
Perl-Users Digest, Issue: 1020 Volume: 10
daemon@ATHENA.MIT.EDU (Perl-Users Digest)
Wed May 30 14:10:48 2001
Date: Wed, 30 May 2001 11:10:17 -0700 (PDT)
From: Perl-Users Digest <Perl-Users-Request@ruby.OCE.ORST.EDU>
To: Perl-Users@ruby.OCE.ORST.EDU (Perl-Users Digest)
Message-Id: <991246217-v10-i1020@ruby.oce.orst.edu>
Content-Type: text
Perl-Users Digest Wed, 30 May 2001 Volume: 10 Number: 1020
Today's topics:
Re: Problem installing CPAN modules <randy@theoryx5.uwinnipeg.ca>
Re: problem solved.. <somewhere@in.paradise.net>
Re: Problems with Cookie code nobull@mail.com
shell command <Rene.Scheibe@gmx.net>
Re: shell command <kistler@gmx.net>
Re: Silly pipe() question <kistler@gmx.net>
Re: Silly pipe() question nobull@mail.com
Re: Statistics for comp.lang.perl.misc <godzilla@stomp.stomp.tokyo>
Substitution on .htm files (felan)
Substitution on .htm files (felan)
Re: Substitution on .htm files <kistler@gmx.net>
Re: Substitution on .htm files <comdog@panix.com>
Re: Substitution on .htm files <comdog@panix.com>
Re: Substitution on .htm files nobull@mail.com
Re: time (Tweetie Pooh)
Unsure about headers <"relaxedrob@optushome.com.au">
Re: Unsure about headers <kistler@gmx.net>
Re: Unsure about headers <comdog@panix.com>
Re: Unsure about headers nobull@mail.com
Re: use Module (if it exists) nobull@mail.com
Re: What does this perl line mean? <smehrabi@cisco.com>
Digest Administrivia (Last modified: 6 Apr 01) (Perl-Users-Digest Admin)
----------------------------------------------------------------------
Date: 30 May 2001 14:30:59 GMT
From: Randy Kobes <randy@theoryx5.uwinnipeg.ca>
Subject: Re: Problem installing CPAN modules
Message-Id: <9f3073$41b$1@canopus.cc.umanitoba.ca>
In comp.lang.perl.misc Shadow <news@quickdirect.com> wrote:
> I'm a bit new to perl but I was successful in installing some of the
> modules. Here are the commands I used. I ended up getting the errors
> below on some of the modules. It seems that a bunch of the modules are
> corrupt/bad checksum and I find it hard to believe that there are
> problems with so many of these -- including the Bundle::CPAN which I
> was told could be installed and I also got the same checksum/bogus
> file warning. Does anyone know what all this means and how I can
> install the missing modules?
> Commands I used:
> bash% perl -MCPAN -e shell
[ .. ]
> Errors I received from some of the modules:
> cpan> install DBI
> Running make for T/TI/TIMB/DBI-1.15.tar.gz
> Unwrapped into directory yes/build/TIMB000
> CPAN.pm: Going to build T/TI/TIMB/DBI-1.15.tar.gz
> Couldn't chdir yes/build/TIMB000: No such file or directory at (eval
> 26) line 42
It sounds like you have a misconfigured CPAN/Config.pm or perhaps
$HOME/.cpan/CPAN/MyConfig.pm, as it's trying to access the
directory "yes/" (unless this directory really does exist).
In particular, what value does 'build_dir' have? You can
change CPAN.pm's configuration options within the CPAN shell
through the 'o conf' command - see 'perldoc CPAN'.
best regards,
randy kobes
------------------------------
Date: Wed, 30 May 2001 23:11:38 +1000
From: "Tintin" <somewhere@in.paradise.net>
Subject: Re: problem solved..
Message-Id: <tk6R6.27$Vv.1052742@news.interact.net.au>
"Rob" <"relaxedrob@optushome.com.au"> wrote in message
news:n6NQ6.4923$25.17580@news1.eburwd1.vic.optushome.com.au...
> Found the perldocs for it and got it to work!
I couldn't find any doco for "it". Doing a "perldoc it" comes back with:
No documentation found for "it".
------------------------------
Date: 30 May 2001 18:20:58 +0100
From: nobull@mail.com
Subject: Re: Problems with Cookie code
Message-Id: <u93d9mda85.fsf@wcl-l.bham.ac.uk>
Bryan Coon <bcoon@sequenom.com> writes:
> I am having some troubles with some subs that I am using for user login
> and verifications.
No, you are having a fundamental problem with your HTTP/CGI
understanding, nothing to do with Perl.
> The problem specifically is that users have to
> login, then hit reload, because (I think) the cookie is printed in the
> header, and following this header is a check for the existence of a
> cookie. No matter what I seem to tweak, I cannot get the user cookie
> check to work without reloading the page, so I am doing something
> wrong..
>
> Basically all of these are called from one script in one request
HTML is basically tranactional. The client sends an HTTP request
which may contain cookies from the client's cookie jar. The message
passes (via zero or more proxies) to the server which runs a CGI
script passing the request into script is a slightly modified form.
The CGI script looks at the request it was sent and sends a response
back along the reverse path. On arrival at the client cookies in the
response will be placed in the client's cookie jar.
Nothing from the response from a given invocation of a CGI script can
travel back in time and alter the request that triggered the same
invocation of the script.
None of this has anything at all to do with Perl.
--
\\ ( )
. _\\__[oo
.__/ \\ /\@
. l___\\
# ll l\\
###LL LL\\
------------------------------
Date: Wed, 30 May 2001 17:13:32 +0200
From: "Rene Scheibe" <Rene.Scheibe@gmx.net>
Subject: shell command
Message-Id: <9f32k7$20jdq$1@ID-65612.news.dfncis.de>
I have written a CGI-script with perl to
put out some statistics which are read from
a logfile which is generated by a program
once a day.
Now I want realtime statistics so I need to
send my this proggy the signal USR1.
I tried in the following in the perlscript:
`killall -USR1 ipfm`
eval("killall -USR1 ipfm")
system("killall -USR1 ipfm")
`su root killall -USR1 ipfm`
...but nothing worked.
killall needs root priviledges but I think apache
don't let perlscripts run as root.
How about the kill function from perl itself. But
how to get the PID of the proggy.
Please tell me some sugestions...
------------------------------
Date: Wed, 30 May 2001 19:03:00 +0200
From: Per Kistler <kistler@gmx.net>
To: Rene Scheibe <Rene.Scheibe@gmx.net>
Subject: Re: shell command
Message-Id: <3B1527C4.EEE59433@gmx.net>
Hi Rene
SIGUSR1 can only be sent either be root or by the same
user. So you would have to send it as wwwuser, or how
ever the CGI runs. Or if the CGI should send another
program, the same applies.
But signals are only one way for processes to comunicate
with each other. The may use files, or semaphores, or
named pipes etc.
Per Kistler.
Rene Scheibe wrote:
>
> I have written a CGI-script with perl to
> put out some statistics which are read from
> a logfile which is generated by a program
> once a day.
> Now I want realtime statistics so I need to
> send my this proggy the signal USR1.
> I tried in the following in the perlscript:
> `killall -USR1 ipfm`
> eval("killall -USR1 ipfm")
> system("killall -USR1 ipfm")
> `su root killall -USR1 ipfm`
> ...but nothing worked.
>
> killall needs root priviledges but I think apache
> don't let perlscripts run as root.
> How about the kill function from perl itself. But
> how to get the PID of the proggy.
>
> Please tell me some sugestions...
------------------------------
Date: Wed, 30 May 2001 19:10:05 +0200
From: Per Kistler <kistler@gmx.net>
To: Benjamin Goldberg <goldbb2@earthlink.net>
Subject: Re: Silly pipe() question
Message-Id: <3B15296D.5B146E62@gmx.net>
Hi Benjamin
Probably it would polute the global name space...
You could make a small function, which uses the Gensym module.
Per Kistler.
Benjamin Goldberg wrote:
>
> I suppose that this might seem to be a silly question, but...
>
> Why does the builtin function pipe() take two target operands, rather
> than taking no arguments and returning a two-element list?
>
> I would prefer to be able to do these:
> my ($read_fh, $write_fh) = pipe;
> or local (*READER, *WRITER) = pipe;
> or my @p = pipe;
> or pipe my @p;
>
> Than have to do these:
> pipe my($read_fh, $write_fh);
> or pipe local(*READER, *WRITER);
> or my @p; pipe $p[0], $p[1];
>
> It's such a simple, useful thing, and I don't understand why it wasn't
> done. I know where the original syntax comes from... the C function
> pipe takes as an argument a pointer to two ints, and fills them in with
> filedescriptors... but the only reason it was done that way (AFAIKS) is
> that C functions can only return a single value. Perl has no difficulty
> at all returning a list with multiple values. Also, it should have no
> difficulty differentiating the different number of arguments, to do The
> Right Thing.
>
> --
> The longer a man is wrong, the surer he is that he's right.
------------------------------
Date: 30 May 2001 18:32:46 +0100
From: nobull@mail.com
Subject: Re: Silly pipe() question
Message-Id: <u9wv6ybv41.fsf@wcl-l.bham.ac.uk>
Benjamin Goldberg <goldbb2@earthlink.net> writes:
> I suppose that this might seem to be a silly question, but...
>
> Why does the builtin function pipe() take two target operands, rather
> than taking no arguments and returning a two-element list?
History. In the days when most filehandles were written as naked
symbols it made sense to say;
pipe READER, WRITER;
> I would prefer to be able to do these:
> my ($read_fh, $write_fh) = pipe;
> or local (*READER, *WRITER) = pipe;
> or my @p = pipe;
Nobody is preventing you writing an alternative function with those
semantics.
> I know where the original syntax comes from... the C function
> pipe takes as an argument a pointer to two ints, and fills them in with
> filedescriptors... but the only reason it was done that way (AFAIKS) is
> that C functions can only return a single value. Perl has no difficulty
> at all returning a list with multiple values. Also, it should have no
> difficulty differentiating the different number of arguments, to do The
> Right Thing.
Seems to me like there's little gain here.
--
\\ ( )
. _\\__[oo
.__/ \\ /\@
. l___\\
# ll l\\
###LL LL\\
------------------------------
Date: Wed, 30 May 2001 09:04:10 -0700
From: "Godzilla!" <godzilla@stomp.stomp.tokyo>
Subject: Re: Statistics for comp.lang.perl.misc
Message-Id: <3B1519FA.B60D2915@stomp.stomp.tokyo>
Tad McClellan wrote:
> Philip Newton wrote:
> > Greg Bacon wrote:
> >> Top 10 Targets for Crossposts
> >> Articles Newsgroup
> >> 36 alt.perl
> >> 29 comp.lang.perl
> > My goodness. Is this 29 clueless newbies crossposting here, or people
> > not bothering to change their "Newsgroups:" line when following up to an
> > article with bad headers?
> Looks to me like lots of folks helping me automatically reduce
> the number of posts that are visible in clp.misc.
> I appreciate all the help I can get in that endeavor :-)
It is annoyingly clear you view yourself as the sole
proprietary of this newsgroup.
Godzilla!
------------------------------
Date: Wed, 30 May 2001 16:07:45 GMT
From: felan_66@hotmail.com (felan)
Subject: Substitution on .htm files
Message-Id: <3b151878.16116967@news.uio.no>
I am really not a programmer, and I dont really know anything about
perl, but I just need this one program. I have this code that prompts
for an input file and an output file then i substitutes a for b and x
for y in the given file and writes the result in another file:
print "Input file name: ";
chomp($infilename = <STDIN>);
print "Output file name: ";
chomp($outfilename = <STDIN>);
open(IN,$infilename) || die "cannot open $infilename for reading: $!";
undef($/);
{
local $/; # defaults to undef anyway
$text = <IN>;
}
$text =~ s{a}{b}gi;
$text =~ s{x}{y}gi;
open (OUT,">$outfilename") || die "cannot create $outfilename: $!";
print OUT $text;
close(IN);
close(OUT);
now I have a whole lot of files in a my /html directory where I would
like to to this operation on each .htm file and the result should go
to .html files (or better if I could replace the existing files) so
/html/first-file.htm --> /html/first-file.html
/html/second-file.htm --> /html/fsecond-file.html
etc.
or even better replace the existing files with the substitutions done
/html/first-file.htm --> /html/first-file.htm
/html/second-file.htm --> /html/second-file.htm
tnx
felan_66@hotmail.com
------------------------------
Date: 30 May 2001 09:14:59 -0700
From: felan_66@hotmail.com (felan)
Subject: Substitution on .htm files
Message-Id: <a9a034fc.0105300814.2d634302@posting.google.com>
I am really not a programmer, and I dont really know anything about
perl, but I just need this one program. I have this code that prompts
for an input file and an output file then i substitutes a for b and x
for y in the given file and writes the result in another file:
print "Input file name: ";
chomp($infilename = <STDIN>);
print "Output file name: ";
chomp($outfilename = <STDIN>);
open(IN,$infilename) || die "cannot open $infilename for reading: $!";
undef($/);
{
local $/; # defaults to undef anyway
$text = <IN>;
}
$text =~ s{a}{b}gi;
$text =~ s{x}{y}gi;
open (OUT,">$outfilename") || die "cannot create $outfilename: $!";
print OUT $text;
close(IN);
close(OUT);
now I have a whole lot of files in a my /html directory where I would
like to to this operation on each .htm file and the result should go
to .html files (or better if I could replace the existing files) so
/html/first-file.htm --> /html/first-file.html
/html/second-file.htm --> /html/fsecond-file.html
etc.
or even better replace the existing files with the substitutions done
/html/first-file.htm --> /html/first-file.htm
/html/second-file.htm --> /html/second-file.htm
tnx
felan_66@hotmail.com
------------------------------
Date: Wed, 30 May 2001 18:58:00 +0200
From: Per Kistler <kistler@gmx.net>
To: felan_66@hotmail.com
Subject: Re: Substitution on .htm files
Message-Id: <3B152698.E912B53@gmx.net>
Hi Felan
If it has subdirectories than some module like File::Find
may help. If it's on one directory level, than use
opendir and readdir. If it's on unix, then a find and xargs
may do the recursion without any scripting.
Per Kistler.
------------------------------
Date: Wed, 30 May 2001 12:53:51 -0400
From: brian d foy <comdog@panix.com>
Subject: Re: Substitution on .htm files
Message-Id: <comdog-8F14AF.12535130052001@news.panix.com>
In article <3b151878.16116967@news.uio.no>, felan_66@hotmail.com wrote:
> I am really not a programmer, and I dont really know anything about
> perl,
or usenet i take it.
i just answered this question in alt.perl.
--
brian d foy <comdog@panix.com>
CGI Meta FAQ - http://www.perl.org/CGI_MetaFAQ.html
Troubleshooting CGI scripts - http://www.perl.org/troubleshooting_CGI.html
------------------------------
Date: Wed, 30 May 2001 12:54:51 -0400
From: brian d foy <comdog@panix.com>
Subject: Re: Substitution on .htm files
Message-Id: <comdog-244445.12545130052001@news.panix.com>
In article <a9a034fc.0105300814.2d634302@posting.google.com>,
felan_66@hotmail.com (felan) wrote:
> I am really not a programmer, and I dont really know anything about
> perl,
no need to post from multiple sites, or to post multiple
copies at all.
i answered this question in alt.perl already.
--
brian d foy <comdog@panix.com>
CGI Meta FAQ - http://www.perl.org/CGI_MetaFAQ.html
Troubleshooting CGI scripts - http://www.perl.org/troubleshooting_CGI.html
------------------------------
Date: 30 May 2001 18:43:49 +0100
From: nobull@mail.com
Subject: Re: Substitution on .htm files
Message-Id: <u9u222bulm.fsf@wcl-l.bham.ac.uk>
felan_66@hotmail.com (felan) writes:
> I am really not a programmer, and I dont really know anything about
> perl, but I just need this one program.
There are only three basic ways to obtain a program:
1) Learn enough programming to write it yourself.
2) Employ a programmer to write it for you.
3) Find it already existing somewhere.
In this newsgroup you'll find that you are made to feel unwellcome if
you wish to take any path other than (1).
> now I have a whole lot of files in a my /html directory where I would
> like to to this operation on each .htm file and the result should go
> to .html files (or better if I could replace the existing files) so
I have told you already that there is a Perl module to repeat a
operation for each file in a directory. If you have any difficulty
understanding how to use it then please feel free to ask. Be sure to
be precise about what it is you don't understand as you will not be
well received arround here if you give the impression you are simply
asking us to write your program for you so that you don't have to
learn how.
--
\\ ( )
. _\\__[oo
.__/ \\ /\@
. l___\\
# ll l\\
###LL LL\\
------------------------------
Date: Wed, 30 May 2001 15:24:22 GMT
From: tp601553@cia.gov (Tweetie Pooh)
Subject: Re: time
Message-Id: <Xns90B1A76361BCFTweetiePooh@62.253.162.108>
"John W. Krahn" <krahnj@acm.org> honoured comp.lang.perl.misc on Wed 30 May
2001 06:31:32a with news:3B1485C7.4210D1FB@acm.org:
> Rob wrote:
>>
>> Howdy all!
>>
>> Could anyone help me out with my time problem please?
>>
>> I want to get the current time in this format:
>> DD-MM-YYYY HH:MI AM
>
> use POSIX qw(strftime);
>
> print strftime( "%d-%m-%Y %I:%M %p", localtime );
>
>
>
> John
Or you can use Date::Manip from CPAN for non-posix situations.
use Date::Manip;
print UnixDate("today","%d-%m-%Y %I:%M %p"); # check format string
------------------------------
Date: Wed, 30 May 2001 14:13:05 GMT
From: "Rob" <"relaxedrob@optushome.com.au">
Subject: Unsure about headers
Message-Id: <Rd7R6.5216$25.18677@news1.eburwd1.vic.optushome.com.au>
Hello all!
I have a perl script on my web server that produces a form with a submit button
that sends the form attributes back to itself for processing..
myPage.pl
..
<form method="POST" action="myPage.pl">
Elsewhere in this script I am using the DBI package to create a database handle
and statement handle objects to query a database using the form attributes.
The script then re-displays the form so the user can enter the form attributes
again and re-submit - to the *same* script.. myPage.pl
My problem is this: even though I can see that the new form attributes are being
received each time the script calls itself, the sql object seem to be stuck with
old values.
DO I have to do something like refresh header information or something similar
to 'flush' away these old objects?
By sending data back to the same script am I somehow keeping the old objects
alive??
Any help would be most apreciated!
Rob
------------------------------
Date: Wed, 30 May 2001 19:06:17 +0200
From: Per Kistler <kistler@gmx.net>
To: Rob <relaxedrob@optushome.com.au>
Subject: Re: Unsure about headers
Message-Id: <3B152889.881123B3@gmx.net>
Hi Rob
If your WWW server uses mod_perl, than global variables may
remain in the ever running perl interpreter. This could be
prevented by having a local scope or by reseting all relevant
variables at the start of the script.
But maybe your problem is of a totally different nature.
Per Kistler.
Rob wrote:
>
> Hello all!
>
> I have a perl script on my web server that produces a form with a submit button
> that sends the form attributes back to itself for processing..
>
> myPage.pl
> ..
> <form method="POST" action="myPage.pl">
>
> Elsewhere in this script I am using the DBI package to create a database handle
> and statement handle objects to query a database using the form attributes.
>
> The script then re-displays the form so the user can enter the form attributes
> again and re-submit - to the *same* script.. myPage.pl
>
> My problem is this: even though I can see that the new form attributes are being
> received each time the script calls itself, the sql object seem to be stuck with
> old values.
>
> DO I have to do something like refresh header information or something similar
> to 'flush' away these old objects?
>
> By sending data back to the same script am I somehow keeping the old objects
> alive??
>
> Any help would be most apreciated!
>
> Rob
------------------------------
Date: Wed, 30 May 2001 13:03:10 -0400
From: brian d foy <comdog@panix.com>
Subject: Re: Unsure about headers
Message-Id: <comdog-E2BA77.13031030052001@news.panix.com>
In article <3B152889.881123B3@gmx.net>, Per Kistler <kistler@gmx.net>
wrote:
[please post your reply after the original text, as demonstrated here]
> If your WWW server uses mod_perl, than global variables may
> remain in the ever running perl interpreter.
that is, the one in one of the httpd children. ;)
--
brian d foy <comdog@panix.com>
CGI Meta FAQ - http://www.perl.org/CGI_MetaFAQ.html
Troubleshooting CGI scripts - http://www.perl.org/troubleshooting_CGI.html
------------------------------
Date: 30 May 2001 18:09:06 +0100
From: nobull@mail.com
Subject: Re: Unsure about headers
Message-Id: <u9ae3udarx.fsf@wcl-l.bham.ac.uk>
"Rob" <"relaxedrob@optushome.com.au"> writes:
> Subject: Unsure about headers
Unsure what any of this has to do with any concept that anyone usually
describes with the word 'headers'.
> I have a perl script on my web server that produces a form with a submit button
> that sends the form attributes back to itself for processing..
>
> The script then re-displays the form so the user can enter the form attributes
> again and re-submit - to the *same* script.. myPage.pl
>
> My problem is this: even though I can see that the new form attributes are being
> received each time the script calls itself, the sql object seem to be stuck with
> old values.
>
> DO I have to do something like refresh header information or something similar
> to 'flush' away these old objects?
I don't understand what SQL and headers have to do with any of this.
I think you are whitnessing the "sticky" behaviour of the CGI module.
This behaviour can be switched off on a per-field basis using the
'-force' parameter. For further details locate the documentation of
the -foece paramter.
--
\\ ( )
. _\\__[oo
.__/ \\ /\@
. l___\\
# ll l\\
###LL LL\\
------------------------------
Date: 30 May 2001 18:01:59 +0100
From: nobull@mail.com
Subject: Re: use Module (if it exists)
Message-Id: <u9d78qdb3s.fsf@wcl-l.bham.ac.uk>
rgarciasuarez@free.fr (Rafael Garcia-Suarez) writes:
> Todd Smith wrote in comp.lang.perl.misc:
> } Is there a way to say:
> }
> } use String::Approx (if it's installed)
>
> Why not use require() and wrap it in an eval() ?
>
> #!/usr/local/bin/perl -l
> BEGIN {
> eval 'require Foo' or print "No module Foo";
> eval 'require CGI' or print "No module CGI";
> }
Better still, since this is in BEGIN{}, why not keep it as use() and
wrap it in eval?
> } and later:
> }
> } (if String::Approx is on your system) {
> } ...
> } }
if ( exists &String::Approx::amatch ) {
--
\\ ( )
. _\\__[oo
.__/ \\ /\@
. l___\\
# ll l\\
###LL LL\\
------------------------------
Date: Wed, 30 May 2001 10:57:44 -0400
From: "Sultan Mehrabi" <smehrabi@cisco.com>
Subject: Re: What does this perl line mean?
Message-Id: <991234691.899609@sj-nntpcache-3>
Thank you all for your responses :)
Most appreciated.
-Sultan
"Sultan Mehrabi" <smehrabi@cisco.com> wrote in message
news:991167173.333383@sj-nntpcache-3...
> I was reading over a script someone else wrote and saw this line. I can't
> seem to figure out what it means.
> I know the hashes but what's our exactly?
> Is this a list of hashes?
>
> our (%console_svr, %console_port, %checkout);
>
> Thanks in advance for your help.
> -Sultan
>
>
------------------------------
Date: 6 Apr 2001 21:33:47 GMT (Last modified)
From: Perl-Users-Request@ruby.oce.orst.edu (Perl-Users-Digest Admin)
Subject: Digest Administrivia (Last modified: 6 Apr 01)
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.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.
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 V10 Issue 1020
***************************************
