[10448] in Perl-Users-Digest
Perl-Users Digest, Issue: 4040 Volume: 8
daemon@ATHENA.MIT.EDU (Perl-Users Digest)
Wed Oct 21 22:03:35 1998
Date: Wed, 21 Oct 98 19:00:16 -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, 21 Oct 1998 Volume: 8 Number: 4040
Today's topics:
Re: automatic garbage collection and memory leak (Ilya Zakharevich)
Re: automatic garbage collection and memory leak <rootbeer@teleport.com>
Re: CPAN and Netscape 4.5/Linux (Alan Barclay)
Re: dbmopen and file contention problems (Martien Verbruggen)
how to redirect POST parameters ? <cusf1@pen.eiu.edu>
opening a remote web page and saving the content (Douglas Melniker)
Re: opening a remote web page and saving the content <rootbeer@teleport.com>
Re: opening a remote web page and saving the content (Martien Verbruggen)
Re: Perl & Java - differences and uses <kick@cig.mot.com>
Re: Perl & Y2K - booby trap code finsol@ts.co.nz
Re: Perl & Y2K - booby trap code <rootbeer@teleport.com>
Perl and prolog <cja1@doc.ic.ac.uk>
Re: Perl Cookbook - is this the best perl book? <lmui@oreilly.com>
perl debugger crash <shenge@ece.ucdavis.edu>
perl locale problem. (Steve .)
Re: Perl on NT (Martien Verbruggen)
Re: Perl on NT (Larry Rosler)
Re: Perl on NT (Martien Verbruggen)
Re: Pink? Blue? What color _is_ it??? (Tad McClellan)
Re: Raleigh.pm (Raleigh, NC, USA perl mongers) has regi (Michael J Gebis)
Re: strange qwirk with READ (Martien Verbruggen)
Sybase DBD Solaris Probs standj@earthlink.net
Re: The story of WinNT & the almost-written data... <rootbeer@teleport.com>
Re: automatic garbage collection and memory leak <dgris@rand.dimensional.com>
Re: automatic garbage collection and memory leak (Larry Wall)
Re: Forks and other Utensils <cynical@iname.com>
Help with perlcc <sy1004@email.com>
Re: Is there an XEmacs wrapper for perldoc? indhiraa@hotmail.com
MD% installation hermeet11@my-dejanews.com
MD5 module installation problems hermeet11@my-dejanews.com
Re: RegExp Hell! (Tom Bates)
Re: Sybase DBD Solaris Probs <mpeppler@mbay.net>
What's the "best" way to call one Perl script from anot (Alfred von Campe)
Special: Digest Administrivia (Last modified: 12 Mar 98 (Perl-Users-Digest Admin)
----------------------------------------------------------------------
Date: 21 Oct 1998 23:06:31 GMT
From: ilya@math.ohio-state.edu (Ilya Zakharevich)
Subject: Re: automatic garbage collection and memory leak
Message-Id: <70lpdn$7qf$1@mathserv.mps.ohio-state.edu>
[A complimentary Cc of this posting was sent to Tom Phoenix
<rootbeer@teleport.com>],
who wrote in article <Pine.GSO.4.02A.9810211332290.5534-100000@user2.teleport.com>:
> On Wed, 21 Oct 1998, Gulriz Aytekin Kurban wrote:
>
> > One of the programs I wrote uses up 1Mb memory at every cycle of a
> > while loop. But, since all my variables are lexically scoped within
> > the innermost block possible, Perl is supposed to reclaim the
> > space(garbage collection), then reallocate it.
>
> Yes, so long as you aren't leaving circular references.
No, Perl will never reclaim the space used by any lexical variable or
a temporary variable.
env PERL_DEBUG_MSTATS=2 perl -wle \
'{my $x = "a" x 1000000; $y = "a"}; $y x= 1000000'
shows 3M used: one for the temporary having the result of "a" x 1000000,
another for the storage left in $x, and the third for $y.
This is considered to be a "feature" by Perl Performance Police.
Ilya
------------------------------
Date: Wed, 21 Oct 1998 23:24:57 GMT
From: Tom Phoenix <rootbeer@teleport.com>
Subject: Re: automatic garbage collection and memory leak
Message-Id: <Pine.GSO.4.02A.9810211623320.5534-100000@user2.teleport.com>
On 21 Oct 1998, Ilya Zakharevich wrote:
> This is considered to be a "feature" by Perl Performance Police.
In that case, even though I'm not sure why it's so, I'm fairly confident
that it's the Right Thing to Do, and I'll not worry about it. :-)
--
Tom Phoenix Perl Training and Hacking Esperanto
Randal Schwartz Case: http://www.rahul.net/jeffrey/ovs/
------------------------------
Date: 21 Oct 1998 22:48:53 GMT
From: gorilla@elaine.drink.com (Alan Barclay)
Subject: Re: CPAN and Netscape 4.5/Linux
Message-Id: <909010130.879995@elaine.drink.com>
In article <slrn72ogsp.ath.PerSteinar.Iversen@elbonia.p52.hioslo.no>,
Per Steinar Iversen <PerSteinar.Iversen@adm.hioslo.no> wrote:
>so nice. So, in fact a browser can do whatever it want with
>.html files found on an FTP server, displaying them as text
>is a reasonable default. I wish CPAN could be bothered to
>use HTTP, much nicer for caching too.
You can access CPAN via HTTP, in fact that's the way I usually do
it, as getting through the firewall at work is easier with HTTP than with
FTP. http://www.perl.com/CPAN would be a good starting point.
------------------------------
Date: Wed, 21 Oct 1998 23:09:51 GMT
From: mgjv@comdyn.com.au (Martien Verbruggen)
Subject: Re: dbmopen and file contention problems
Message-Id: <3RtX1.40$vq4.114712@nsw.nnrp.telstra.net>
In article <4LsX1.18300$WE.956@news.cwix.com>,
"Norman Bunn" <norman.bunn@mci.com> writes:
> dbmopen(%groupBase, $groupFile, 0666) || die "Can't open $groupFile\n";
>
> and the program is dying regularly. I assume this is because the file is
> still open from another user. I have tried changing the mode, but it's
> still happening. Any suggestions?
In the die message, include $!, so you at least know why it dies.
If you can have concurrent access, ensure that you have some locking
mechanism in place, and that all applications use the same locking
mechanism. Make sure you handle it gracefully.
Martien
--
Martien Verbruggen |
Webmaster www.tradingpost.com.au | The world is complex; sendmail.cf
Commercial Dynamics Pty. Ltd. | reflects this.
NSW, Australia |
------------------------------
Date: Wed, 21 Oct 1998 19:53:41 -0500
From: flash <cusf1@pen.eiu.edu>
Subject: how to redirect POST parameters ?
Message-Id: <362E8215.663A8D34@pen.eiu.edu>
hi all,
do someone know how to redirect an URL with POST parameters ?
I have 2 web pages (so 2 forms) for 2 search engines. I want to POST
the parameters to the same CGI perl.
Depending on these parameters, I would like to rePOST to another CGI.
Sometimes to people.pl, sometimes to contents.pl .
I can do the redirection but I loose the POST !
I think redirect() ( CGI.pm ) can help me, but I still loose the POST
!!!
I'm using Perl 5.00404, and CGI.pm 2.36.
Do u have an idea ?
Thanks, flash
------------------------------
Date: Wed, 21 Oct 98 23:02:17 GMT
From: melniker@eecs.tufts.edu (Douglas Melniker)
Subject: opening a remote web page and saving the content
Message-Id: <70lp8v$94i$1@news3.tufts.edu>
Hi. How can i have a perl program (not a cgi script) open up a web page (such
as "http://www.yahoo.com/index.html") and either save the content of it to a
file, or better yet, print it to stout (this part would be trivial...)
i'm trying to do this with a c++ program, but i figured it would be easier to
do in perl, so i'll just call this perl program from inside the c++ prog.
thanks a lot! please cc by mail
- doug
---------------------------------------------------------
Doug Melniker
Center for Connected Learning and Computer-Based Modeling
Tufts University
www.ccl.tufts.edu
------------------------------
Date: Wed, 21 Oct 1998 23:26:22 GMT
From: Tom Phoenix <rootbeer@teleport.com>
Subject: Re: opening a remote web page and saving the content
Message-Id: <Pine.GSO.4.02A.9810211625500.5534-100000@user2.teleport.com>
On Wed, 21 Oct 1998, Douglas Melniker wrote:
> How can i have a perl program (not a cgi script) open up a web page
Have you seen the FAQ? It should be with your other Perl docs. Hope this
helps!
--
Tom Phoenix Perl Training and Hacking Esperanto
Randal Schwartz Case: http://www.rahul.net/jeffrey/ovs/
------------------------------
Date: Wed, 21 Oct 1998 23:31:33 GMT
From: mgjv@comdyn.com.au (Martien Verbruggen)
Subject: Re: opening a remote web page and saving the content
Message-Id: <p9uX1.44$vq4.144223@nsw.nnrp.telstra.net>
In article <70lp8v$94i$1@news3.tufts.edu>,
melniker@eecs.tufts.edu (Douglas Melniker) writes:
>
> Hi. How can i have a perl program (not a cgi script) open up a web page (such
> as "http://www.yahoo.com/index.html") and either save the content of it to a
> file, or better yet, print it to stout (this part would be trivial...)
Use the LWP modules. Get it from CPAN: http://www.perl.com/CPAN/
Martien
--
Martien Verbruggen |
Webmaster www.tradingpost.com.au | If it isn't broken, it doesn't have
Commercial Dynamics Pty. Ltd. | enough features yet.
NSW, Australia |
------------------------------
Date: 21 Oct 1998 18:47:30 -0500
From: Damien Kick <kick@cig.mot.com>
Subject: Re: Perl & Java - differences and uses
Message-Id: <so8u30xjx4d.fsf@cig.mot.com>
Patricia Shanahan <pats@acm.org> writes:
> The main thing is to learn plenty of languages, and think about their
> uses. Learning programming languages is a skill in its own right, and
> one that improves with practice. You never know when you will hit a
> problem that can best be solved in some language you don't know (yet).
<snip>
Like TAL <grin> <pause> <shudder>.
--
Damien Kick
------------------------------
Date: Wed, 21 Oct 1998 22:46:40 GMT
From: finsol@ts.co.nz
Subject: Re: Perl & Y2K - booby trap code
Message-Id: <70lo8h$32s$1@nnrp1.dejanews.com>
In article <8c1zo2t321.fsf@gadget.cscaper.com>,
Randal Schwartz <merlyn@stonehenge.com> wrote:
> >>>>> "finsol" == finsol <finsol@ts.co.nz> writes:
>
> finsol> I posted a link in an earlier thread on this topic -
> finsol> unfortunately the link was changed & the article referenced
> finsol> had nothing to do with Perl - it did spark off an lively
> finsol> discussion however!
>
> finsol> The correct link for this article (unless they change it again!) is:
> finsol> http://www.idg.co.nz/WWWfeat/Y2000/amon1010.htm
>
> Bogus bogus BOGUS.
>
> Geez. I get quite irate when I see sloppy writing like the following:
>
> quote> Perl, MacPerl, CGI using localtime
> quote> [...]
> quote> One of the booby-trap languages listed is MacPerl, a language
> quote> used on Apple Macintosh computers. While it is true that Apple
> quote> hardware does not have the same internal clock problems as the
> quote> industry standard PCs, any software is just as likely to have
> quote> year 2000 software problems. MacPerl applications are just as
> quote> likely to be booby-trapped as applications written in the
> quote> other affected languages.
>
> There is nothing WRONG with MacPerl, or any other flavor of Perl.
> localtime() and gmtime() have *always* returned year - 1900 starting
> with perl version 0. And was documented as such.
>
> If a sloppy programmer wrote "19$year" instead of ($year + 1900),
> that's *their* fault. STOP BLAMING PERL.
>
> We can give yoy a pocket chainsaw. We cannot keep you from
> sawing your arm off from it.
>
> STOP BLAMING PERL.
>
> STOP PROPOGATING NONSENSE.
>
Randal, I checked out your web page and obviously you are *the* expert when it
comes to Perl.
In your expert opinion:
Would you recommend that Perl applications be checked for Year 2000 problems?
Or only the code of 'sloppy' programmers - in which case how do you know who
is a sloppy programmer and which applications thet have written?
Do all these sloppy programmers now know how to correctly use year values and
therefore there is no longer any need to mention this unfortunate
misunderstanding ever again?
Just what do you thing the Y2K problem is? My understanding is that it relates
to software which works OK now but will encounter problems when using dates
beyond 1999. You obviously have a very narrow view of the issue.
Y2K is not about blame - its about understanding the problem, fixing it and
ensuring as best we can, that it doesn't happen again. This can only be
achieved by raising awareness on the issue. Burying our heads in the sand
does not help.
I agree that the term 'booby trap' does get an emotive response but it aptly
describes the nature of the problem in Perl and the other affected languages.
Any response the topic receives is a good response because each time it is
discussed, it increases the chances of others realising
a) They may have misunderstood how year values should be used
b) There is a lot of confusion amongst programmers on this subject and perhaps
Perl code should be checked for Y2K problems.
I would recommend for anyone responsible for Perl applications, to do a
search on "localtime" in the code and ensure that it is handled correctly.
Better to find out now that you have a problem than when your application
encounters 2000 - for some applications that could be as early as January
1999.
The localtime issue is not the only Y2K problem that can occur in Perl but it
is good example of why you can't be complacent about your programming
language and Y2K.
Jocelyn Amon
--
Financial Solutions Limited
http://www.ts.co.nz/~finsol/
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
Date: Wed, 21 Oct 1998 23:20:27 GMT
From: Tom Phoenix <rootbeer@teleport.com>
Subject: Re: Perl & Y2K - booby trap code
Message-Id: <Pine.GSO.4.02A.9810211611111.5534-100000@user2.teleport.com>
On Wed, 21 Oct 1998 finsol@ts.co.nz wrote:
> Randal, I checked out your web page and obviously you are *the* expert
> when it comes to Perl.
>
> In your expert opinion:
> Would you recommend that Perl applications be checked for Year 2000
> problems?
That doesn't take Randal's expert opinion: ALL applications should be
checked for ALL bugs, in a perfect world.
> Y2K is not about blame - its about understanding the problem, fixing it and
> ensuring as best we can, that it doesn't happen again. This can only be
> achieved by raising awareness on the issue. Burying our heads in the sand
> does not help.
Who advocated burying our heads? Don't put words into people's mouths.
But I challenge your implicit claim that you're merely "raising awareness
on the issue". You're stirring up discord over trivia, but you have
nothing new to say about Perl and Y2K.
> I would recommend for anyone responsible for Perl applications, to do a
> search on "localtime" in the code and ensure that it is handled correctly.
I would recommend for anyone responsible for any applications, to do a
search of all tokens in the code and ensure that they are handled
correctly.
> Better to find out now that you have a problem than when your application
> encounters 2000 - for some applications that could be as early as January
> 1999.
For some applications, which calculated the due dates of 30-year bonds, it
was January 1970. So what?
In summary "all programs should be checked for bugs". But there's nothing
Perl-specific about this; please take it to a more appropriate forum.
Thank you.
--
Tom Phoenix Perl Training and Hacking Esperanto
Randal Schwartz Case: http://www.rahul.net/jeffrey/ovs/
------------------------------
Date: Tue, 20 Oct 1998 15:18:58 +0100
From: cja1 <cja1@doc.ic.ac.uk>
Subject: Perl and prolog
Message-Id: <362C9BD2.9EE7EE73@doc.ic.ac.uk>
Does anybody know of a perl/prolog bridge?
If not, any ideas on using perl as a medium to find shortest paths
between nodes?
Thanks,
Billy
-------------------------------------------------------------
Chris/Billy Abbott <c.abbott@doc.ic.ac.uk>
-------------------------------------------------------------
3rd year computer science student Imperial College
------------------------------
Date: Wed, 21 Oct 1998 15:33:54 -0400
From: Linda Mui <lmui@oreilly.com>
Subject: Re: Perl Cookbook - is this the best perl book?
Message-Id: <362E3722.31C7@oreilly.com>
Larry Rosler wrote:
>
> The authors set up a very capable system to handle feedback from a
> considerable community of reviewers. More could not be expected of them
> (the authors, that is).
>
> The final editing seemed rather rushed to me, but was undoubtedly
> determined by the publisher's intense desire to have the book available
> at The Perl Conference, where it rapidly sold out all the copies they
> had available. If any apologies are necessary (which I don't think!),
> they should come from the publishers, not the authors.
I'm the editor of the book. Thanks for coming to the
authors' defense; they have been tremendous in
writing the book, working with our production staff
on a tight schedule, and dealing with the errata.
If you're going to complain about typos, please
direct it at O'Reilly and leave them out of it.
Yes, we did put the Perl Cookbook on the fast track
to make the Perl Conference. Still, the book was reviewed
by dozens of programmers, and carefully read by myself, one
insanely obsessive copyeditor, three proofreaders, one
creator of Perl, and the authors themselves. We all
missed some things. It's a big book. But please don't
think that we just published this book blindly. Many
people slaved over this book in addition to the authors,
and while the end result isn't perfect, few things are.
At some point with every book, we need to call it a book
and just hope that readers appreciate it enough for what
it does well that they take the time to tell us how to
improve it.
I hope Mr. (Ms.?) Jenkins has been sending us errata
so we can incorporate them into the next printing.
-Linda Mui
------------------------------
Date: 21 Oct 1998 23:44:53 GMT
From: Eric Sheng <shenge@ece.ucdavis.edu>
Subject: perl debugger crash
Message-Id: <70lrll$s0p$1@mark.ucdavis.edu>
Hi all,
I am trying to debug a script that I wrote with the perl -d debugger.
The problem is that it crashes at certain point where I tell it to "c" or
continue after setting a break point. The problem got worse and worse and
eventually, it crashed after the first break point I set.
The following is what happened:
%perl5 -d ./my_script.pl
Loading DB routines from perl5db.pl version 1.01
Emacs support available.
Enter h or `h h' for help.
main::(./my_script.pl:22): $pattern_file = "ppsim";
DB<1> b 204
DB<2> c
DB::fake::(/opt/perl5.004_04/lib/PA-RISC1.1/5.00404/perl5db.pl:2083):
2083: "Debugged program terminated. Use `q' to quit or `R' to restart.";
DB<2>
after setting b 204, I did a c for continue. Debugged program then terminated
for no apparent reason, and I couldn't go on anymore. Any ideas?
--Eric
------------------------------
Date: Thu, 22 Oct 1998 00:39:11 GMT
From: syarbrou@nospam.ais.net (Steve .)
Subject: perl locale problem.
Message-Id: <362f7ea2.8216512@news.ais.net>
I have redhat 5.1 installed. I try to install pRPC-modules-0.1005.
Get the following error when doing a perl Makefile.PL
perl: warning: Setting locale failed.
perl: warning: Please check that your locale settings:
LC_ALL = (unset),
LANG = "en_RN"
are supported and installed on your system.
perl: warning: Falling back to the standard locale ("C").
Checking for Storable ... ok
Checking for Sys::Syslog ...
You must have installed the 'Sys::Syslog' module.
You get it at the CPAN. See 'man CPAN' or 'perldoc CPAN'for more info.
[root@olsoracle pRPC-modules-0.1005]#
I keep getting the locale error, and also now the Sylog error. I've
done the h2ph and there are man pages available for Sys::Syslog.
What's up? Thanks.
Steve
------------------------------
Date: Wed, 21 Oct 1998 23:04:45 GMT
From: mgjv@comdyn.com.au (Martien Verbruggen)
Subject: Re: Perl on NT
Message-Id: <hMtX1.38$vq4.114712@nsw.nnrp.telstra.net>
In article <Pine.GSO.3.95.981021135155.22418B-100000@hollywood.cinenet.net>,
Craig Berry <cberry@cinenet.net> writes:
> Hm, interesting point. I worked for so long on CGI servers that did it
> the "right way" (i.e., the way I got used to) that I never quite realized
> that the spec doesn't specify cwd at script runtime. I suppose we should
> all be doing FindBin in a begin block to get to the app's dir.
That might be handy, but certainly not preferable in all cases.
Especially scripts that need to run very often would waste enormous
amount of server resources just to determine where they are.
If you absolutely don't want to hardcode directories in your script,
use a configuration file which is hardcoded. If you don't even want to
do that, have the script dump a file in a well known, always available
place (/tmp, c:\), which contains some config info. Next time the
script fires up, all it has to do is read the file. Since that
will be cached if the script is heavily utilised, that should be a
minimum load. To tune even further, you could consider loading the
FindBin module at runtime, instead of compile time.
Martien
--
Martien Verbruggen |
Webmaster www.tradingpost.com.au | 75% of the people make up 3/4 of the
Commercial Dynamics Pty. Ltd. | population.
NSW, Australia |
------------------------------
Date: Wed, 21 Oct 1998 17:20:54 -0700
From: lr@hpl.hp.com (Larry Rosler)
Subject: Re: Perl on NT
Message-Id: <MPG.10981a43ed768b0989832@nntp.hpl.hp.com>
[Posted to comp.lang.perl.misc and a copy mailed.]
In article <hMtX1.38$vq4.114712@nsw.nnrp.telstra.net> on Wed, 21 Oct
1998 23:04:45 GMT, Martien Verbruggen <mgjv@comdyn.com.au> says...
> In article <Pine.GSO.3.95.981021135155.22418B-100000@hollywood.cinenet.net>,
> Craig Berry <cberry@cinenet.net> writes:
>
> > Hm, interesting point. I worked for so long on CGI servers that did it
> > the "right way" (i.e., the way I got used to) that I never quite realized
> > that the spec doesn't specify cwd at script runtime. I suppose we should
> > all be doing FindBin in a begin block to get to the app's dir.
>
> That might be handy, but certainly not preferable in all cases.
> Especially scripts that need to run very often would waste enormous
> amount of server resources just to determine where they are.
Why is this? I can find out where the program is via $0 and where the
web root is from $ENV{PATH_TRANSLATED} -- hardly any resources needed.
--
(Just Another Larry) Rosler
Hewlett-Packard Laboratories
http://www.hpl.hp.com/personal/Larry_Rosler/
lr@hpl.hp.com
------------------------------
Date: Thu, 22 Oct 1998 00:48:13 GMT
From: mgjv@comdyn.com.au (Martien Verbruggen)
Subject: Re: Perl on NT
Message-Id: <hhvX1.69$vq4.194802@nsw.nnrp.telstra.net>
In article <MPG.10981a43ed768b0989832@nntp.hpl.hp.com>,
lr@hpl.hp.com (Larry Rosler) writes:
> [Posted to comp.lang.perl.misc and a copy mailed.]
>
> In article <hMtX1.38$vq4.114712@nsw.nnrp.telstra.net> on Wed, 21 Oct
> 1998 23:04:45 GMT, Martien Verbruggen <mgjv@comdyn.com.au> says...
>> In article <Pine.GSO.3.95.981021135155.22418B-100000@hollywood.cinenet.net>,
>> Craig Berry <cberry@cinenet.net> writes:
>>
>> > Hm, interesting point. I worked for so long on CGI servers that did it
>> > the "right way" (i.e., the way I got used to) that I never quite realized
>> > that the spec doesn't specify cwd at script runtime. I suppose we should
>> > all be doing FindBin in a begin block to get to the app's dir.
>>
>> That might be handy, but certainly not preferable in all cases.
>> Especially scripts that need to run very often would waste enormous
>> amount of server resources just to determine where they are.
>
> Why is this? I can find out where the program is via $0 and where the
> web root is from $ENV{PATH_TRANSLATED} -- hardly any resources needed.
Notice the mention of FindBin? That is where my comment originated.
FindBin is expensive.
Martien
--
Martien Verbruggen |
Webmaster www.tradingpost.com.au | I'm just very selective about what I
Commercial Dynamics Pty. Ltd. | accept as reality - Calvin
NSW, Australia |
------------------------------
Date: Wed, 21 Oct 1998 18:09:58 -0500
From: tadmc@flash.net (Tad McClellan)
Subject: Re: Pink? Blue? What color _is_ it???
Message-Id: <6kpl07.jch.ln@flash.net>
Randal Schwartz (merlyn@stonehenge.com) wrote:
: >>>>> "John" == John G Dobnick <jgd@alpha3.csd.uwm.edu> writes:
: John> Now, we all know what the "pink" camel is.
: John> But what's this "blue" thing? Since when did "teal" become "blue"?
: John> Have I missed something
: Of the colors: red, orange, yellow, green, blue, purple, white, and
: black, the second edition camel is closest to "blue". Therefore, I
: call it "blue". I'm sorry if you find this confusing. :)
Perl FAQ, part 99:
"How do you round colors in Perl?" ;-)
--
Tad McClellan SGML Consulting
tadmc@metronet.com Perl programming
Fort Worth, Texas
------------------------------
Date: 21 Oct 1998 22:55:46 GMT
From: gebis@fee.ecn.purdue.edu (Michael J Gebis)
Subject: Re: Raleigh.pm (Raleigh, NC, USA perl mongers) has registered
Message-Id: <70lopi$nlg@mozo.cc.purdue.edu>
Elaine -HappyFunBall- Ashton <eashton@bbnplanet.com> writes:
}John Porter wrote:
I'll pay both of you a dollar to take this thread to e-mail.
--
Mike Gebis gebis@ecn.purdue.edu mgebis@eternal.net
------------------------------
Date: Wed, 21 Oct 1998 23:07:03 GMT
From: mgjv@comdyn.com.au (Martien Verbruggen)
Subject: Re: strange qwirk with READ
Message-Id: <rOtX1.39$vq4.114712@nsw.nnrp.telstra.net>
In article <70la60$prn2@valley.rtpnc.epa.gov>,
"William Smith" <smith.will@epa.gov> writes:
> I am using perl to read dbase files --see code below.
> When run on NT I find
^^
> open(DB,$dbname) || die "Unable to open $dbname: $!";
# binary file, so:
binmode(DB);
> # read dbase header using "little-endians"
> read(DB,$buf,12);
> ($a,$nrecs,$hdrlen,$reclen) = unpack("A4 V v v",$buf);
# perldoc -f binmode
Martien
--
Martien Verbruggen |
Webmaster www.tradingpost.com.au | If it isn't broken, it doesn't have
Commercial Dynamics Pty. Ltd. | enough features yet.
NSW, Australia |
------------------------------
Date: Wed, 21 Oct 1998 23:45:51 GMT
From: standj@earthlink.net
Subject: Sybase DBD Solaris Probs
Message-Id: <70lrnf$83j$1@nnrp1.dejanews.com>
I'm back, with _more_problems!
Now I'm trying to get the sybase oc library installed correctly,
and get DBI/DBD up and running.
We have an existing Sybase server, Solaris 2.6, on a Enterprise4k.
Since the client libs weren't installed, and we didn't
want to stomp on the existing instalation, we
installed them into a subdir of $SYBASE.
/sybase
/sybase/OpenClient
Is this an acceptable way to do things, or can we install them under the
base $SYBASE dir without stomping on anything?
So, DBI compiled fine, and then in order to _compile_ DBD::Sybase, I had
to add -I/sybase/OpenClient/include to all the gcc calls in the Makefile.
Is this acceptable, or did I do something wrong?
It compiled fine, but when I tried to make test, the following errors
were produced, it looks like it can't find the client header files again,
but I can't find a place to add it.
install_driver(Sybase) failed: Cant load
'./blib/arch/auto/DBD/Sybase/Sybase.so for module DBD::Sybase ld.so.1 failed:
/bin/perl: fatal: relocation error: file
/blib/arch/auto/DBD/Sybase/Sybase.sop: symbol ct_con_props: referenced symbol
not found at /usr/local/lib/perl5/sun4-solaris/5.00404/DynaLoader.pm line 166
Typos are mine since I can't email from work (don't ask :) ).
So, is this a DBD induced problem, or is it our instalation of the CT lib?
Any help via email, news, telegraph, smoke signals, appreciated.
Sean
--
-----------------------------------------------
Sean M. Tompkins
standj@earthlink.net
http://pookanator.dyn.ml.org
-----------------------------------------------
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
Date: Wed, 21 Oct 1998 23:04:24 GMT
From: Tom Phoenix <rootbeer@teleport.com>
Subject: Re: The story of WinNT & the almost-written data...
Message-Id: <Pine.GSO.4.02A.9810211603550.5534-100000@user2.teleport.com>
On Wed, 21 Oct 1998, R. King wrote:
> When I run the script below with ActivePerl (5.002 w/ perl 5.005_2) on
> NT4.0, STDOUT gets the whole thing, but the script only writes about
> 3/4 of the web page to the file...
Could it be binmode? See perlfunc. Hope this helps!
--
Tom Phoenix Perl Training and Hacking Esperanto
Randal Schwartz Case: http://www.rahul.net/jeffrey/ovs/
------------------------------
Date: Thu, 22 Oct 1998 01:15:16 GMT
From: Daniel Grisinger <dgris@rand.dimensional.com>
Subject: Re: automatic garbage collection and memory leak
Message-Id: <m3lnm92ymf.fsf@rand.dimensional.com>
Tom Phoenix <rootbeer@teleport.com> writes:
> On 21 Oct 1998, Ilya Zakharevich wrote:
>
> > This is considered to be a "feature" by Perl Performance Police.
>
> In that case, even though I'm not sure why it's so, I'm fairly confident
> that it's the Right Thing to Do, and I'll not worry about it. :-)
Yes, but I really like the idea of some sort of simple memory management
in perl. It just matches my personal opinion of TIMTOWTDI.
dgris
--
Daniel Grisinger dgris@perrin.dimensional.com
Supporter of grumpiness where grumpiness is due on clpm.
perl -Mre=eval -e'$_=shift;;@[=split//;;$,=qq;\n;;;print m;(.{$-}(?{$-++}));,q;;while$-<=@[;;' 'Just Another Perl Hacker'
------------------------------
Date: 21 Oct 1998 18:19:37 -0700
From: larry@kiev.wall.org (Larry Wall)
Subject: Re: automatic garbage collection and memory leak
Message-Id: <70m179$j1i@kiev.wall.org>
In article <70lpdn$7qf$1@mathserv.mps.ohio-state.edu>,
Ilya Zakharevich <ilya@math.ohio-state.edu> wrote:
>[A complimentary Cc of this posting was sent to Tom Phoenix
><rootbeer@teleport.com>],
>who wrote in article <Pine.GSO.4.02A.9810211332290.5534-100000@user2.teleport.com>:
>> On Wed, 21 Oct 1998, Gulriz Aytekin Kurban wrote:
>>
>> > One of the programs I wrote uses up 1Mb memory at every cycle of a
>> > while loop. But, since all my variables are lexically scoped within
>> > the innermost block possible, Perl is supposed to reclaim the
>> > space(garbage collection), then reallocate it.
>>
>> Yes, so long as you aren't leaving circular references.
>
>No, Perl will never reclaim the space used by any lexical variable or
>a temporary variable.
>
> env PERL_DEBUG_MSTATS=2 perl -wle \
> '{my $x = "a" x 1000000; $y = "a"}; $y x= 1000000'
>
>shows 3M used: one for the temporary having the result of "a" x 1000000,
>another for the storage left in $x, and the third for $y.
>
>This is considered to be a "feature" by Perl Performance Police.
If his program is indeed reusing the same megabyte each time through
the loop, then he's been saved the overhead of reallocating it.
However, he said "used up...at every cycle", indicating a continuous
leak, and hence, while your statement is correct with respect to
Tom's statement, with respect to the original problem, your answer
is "unresponsive", as they say in court.
If it's leaking a megabyte of memory every time through the loop, I'd
say it's a bad bug.
Larry
------------------------------
Date: Thu, 22 Oct 1998 00:55:02 +0000
From: Jordan Conley <cynical@iname.com>
Subject: Re: Forks and other Utensils
Message-Id: <362E8266.663038BB@iname.com>
Luckily enough I am the sys admin...But what do you mean that zero forks
are safe...I know I have to check the return, but why do you say zero
forks are safe??
Jordan
Tom Phoenix wrote:
>
> On Wed, 21 Oct 1998, Jordan Conley wrote:
>
> > How many forks are safe??
>
> Zero. But if you check the return value from fork(), you'll know when it
> fails. Of course, the number of processes you can run depends upon your
> system, not on Perl; check with your system administrator before you start
> too many processes and bog the system down. Cheers!
>
> --
> Tom Phoenix Perl Training and Hacking Esperanto
> Randal Schwartz Case: http://www.rahul.net/jeffrey/ovs/
--
"In the beginning the Universe was created. This has made
many people very angry and has been widely regarded as a
bad move."
-Douglas Adams, Hitchhikers Guide to the Galaxy
------------------------------
Date: Thu, 22 Oct 1998 10:31:53 +0900
From: "Novice" <sy1004@email.com>
Subject: Help with perlcc
Message-Id: <70m218$9p3$1@usenet.kreonet.re.kr>
Hi, there!!!
I'm trying to compile perl script with perlcc. (5.00502)
I have trouble to compile because of using socket module in my perl script.
(log messges are : No definition for sub Socket::SO_DONTROUTE (unable to
autoload) ...)
Is it possible to compile that kind of script?
Regards,
Novice (mailto:sy1004@email.com)
------------------------------
Date: Thu, 22 Oct 1998 01:00:27 GMT
From: indhiraa@hotmail.com
To: lynn_newton@mcg.mot.com
Subject: Re: Is there an XEmacs wrapper for perldoc?
Message-Id: <70m03c$des$1@nnrp1.dejanews.com>
In article <cn1n26pr3q9.fsf@miles.phx.mcd.mot.com>,
lynn_newton@mcg.mot.com (Lynn D. Newton) wrote:
>
> Has anyone written an XEmacs front end for perldoc?
You can get the info files from :
http://www.perl.com/CPAN-local/authors/Krishna_Shamu_Sethuraman/
Hope it helps,
Indira.
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
Date: Thu, 22 Oct 1998 01:07:58 GMT
From: hermeet11@my-dejanews.com
Subject: MD% installation
Message-Id: <70m0he$e7o$1@nnrp1.dejanews.com>
Hi Guys !
i am trying to install the MD501.7 module for perl , the version of perl i
have is 5.004_04
I am not able to install it , is gives me errors while doing a make
************* ---- start of debug output---------************
cc -c -I/usr/local/include -O -DVERSION=\"1.7\" -DXS_VERSION=\"1.7\"
-fpic -I/usr/local/lib/perl5/sun4-solaris/5.00404/CORE
-DPERL_BYTEORDER=4321 MD5.c
In file included from MD5.xs:20:
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:163: sys/types.h:
No
such file or directory
In file included from
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perlio.h:23,
from
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:181,
from MD5.xs:20:
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perlsdio.h:5: stdio.h: No
such file or directory
In file included from MD5.xs:20:
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:192: ctype.h: No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:200: locale.h: No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:217: setjmp.h: No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:223: sys/param.h:
No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:229: stdlib.h: No
such file or directory
In file included from MD5.xs:20:
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:271: string.h: No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:361: netinet/in.h:
No such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:371: sys/stat.h:
No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:400: sys/time.h:
No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:407: sys/times.h:
No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:420: errno.h: No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:465: sys/ioctl.h:
No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:488: dirent.h: No
such file or directory
In file included from
/usr/local/lib/gcc-lib/sparc-sun-solaris2.6/2.8.1/include/syslimits.h:7,
from
/usr/local/lib/gcc-lib/sparc-sun-solaris2.6/2.8.1/include/limits.h:11,
from
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:679,
from MD5.xs:20:
/usr/local/lib/gcc-lib/sparc-sun-solaris2.6/2.8.1/include/limits.h:116:
limits.h: No such file or directory
In file included from
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:917,
from MD5.xs:20:
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/unixish.h:85: signal.h:
No
such file or directory
In file included from
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:1214,
from MD5.xs:20:
/usr/local/lib/gcc-lib/sparc-sun-solaris2.6/2.8.1/include/math.h:5:
math.h: No such file or directory
gmake: *** [MD5.o] Error 1
************* ---- end of debug output---------************
------------------------------------------------
Very funny Scotty . No beam my clothes up.
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
Date: Thu, 22 Oct 1998 01:09:06 GMT
From: hermeet11@my-dejanews.com
Subject: MD5 module installation problems
Message-Id: <70m0ji$e88$1@nnrp1.dejanews.com>
Hi Guys !
i am trying to install the MD501.7 module for perl , the version of perl i
have is 5.004_04
I am not able to install it , is gives me errors while doing a make
************* ---- start of debug output---------************
cc -c -I/usr/local/include -O -DVERSION=\"1.7\" -DXS_VERSION=\"1.7\"
-fpic -I/usr/local/lib/perl5/sun4-solaris/5.00404/CORE
-DPERL_BYTEORDER=4321 MD5.c
In file included from MD5.xs:20:
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:163: sys/types.h:
No
such file or directory
In file included from
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perlio.h:23,
from
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:181,
from MD5.xs:20:
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perlsdio.h:5: stdio.h: No
such file or directory
In file included from MD5.xs:20:
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:192: ctype.h: No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:200: locale.h: No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:217: setjmp.h: No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:223: sys/param.h:
No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:229: stdlib.h: No
such file or directory
In file included from MD5.xs:20:
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:271: string.h: No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:361: netinet/in.h:
No such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:371: sys/stat.h:
No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:400: sys/time.h:
No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:407: sys/times.h:
No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:420: errno.h: No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:465: sys/ioctl.h:
No
such file or directory
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:488: dirent.h: No
such file or directory
In file included from
/usr/local/lib/gcc-lib/sparc-sun-solaris2.6/2.8.1/include/syslimits.h:7,
from
/usr/local/lib/gcc-lib/sparc-sun-solaris2.6/2.8.1/include/limits.h:11,
from
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:679,
from MD5.xs:20:
/usr/local/lib/gcc-lib/sparc-sun-solaris2.6/2.8.1/include/limits.h:116:
limits.h: No such file or directory
In file included from
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:917,
from MD5.xs:20:
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/unixish.h:85: signal.h:
No
such file or directory
In file included from
/usr/local/lib/perl5/sun4-solaris/5.00404/CORE/perl.h:1214,
from MD5.xs:20:
/usr/local/lib/gcc-lib/sparc-sun-solaris2.6/2.8.1/include/math.h:5:
math.h: No such file or directory
gmake: *** [MD5.o] Error 1
************* ---- end of debug output---------************
------------------------------------------------
Very funny Scotty . No beam my clothes up.
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
Date: Thu, 22 Oct 1998 00:59:04 GMT
From: tbates@spam-me-not.gi.com (Tom Bates)
Subject: Re: RegExp Hell!
Message-Id: <362f81e6.43441755@news>
Hopefully you got an answer to this already, but here's my $.02:
/.*\./
I believe this works because the beginning .* is greedy, so it sucks
up everything, including intervening dots. If you use:
/(.*)\./
then $1 will be what you want.
Tom
On Sat, 10 Oct 1998 00:05:30 GMT, nanobreath@my-dejanews.com wrote:
>Hello! Hopefully some kind soul will take mercy on me and defeat the reg exp
>demons ... I am trying to truncate a variable at one particular point, and
>then save everything before that point in another variable. For instance,
>
>domain.com
>
>What I want it to do is take everything *before* the . and put it into
>another variable. So, if before is domain.com, after is domain
>
>I swear, I have looked through several explanations of regular expressions
>& I think it is something I really have to sit down with. Unfortunately,
>I'm trying to get this done... this is NOT my homework, by the way.
>
>If possible, please send email to me at rbaguer@freenet.columbus.oh.us
>because I am physically unable to check newsgroups all the time. Thanks
>a lot.
>
>ryan
------------------------------
Date: Wed, 21 Oct 1998 18:20:50 -0700
From: Michael Peppler <mpeppler@mbay.net>
Subject: Re: Sybase DBD Solaris Probs
Message-Id: <362E8872.4CDAD189@mbay.net>
standj@earthlink.net wrote:
>
> I'm back, with _more_problems!
> Now I'm trying to get the sybase oc library installed correctly,
> and get DBI/DBD up and running.
> We have an existing Sybase server, Solaris 2.6, on a Enterprise4k.
> Since the client libs weren't installed, and we didn't
> want to stomp on the existing instalation, we
> installed them into a subdir of $SYBASE.
> /sybase
> /sybase/OpenClient
> Is this an acceptable way to do things, or can we install them under the
> base $SYBASE dir without stomping on anything?
You *should* install OpenClient in the normal Sybase directory - it
makes things
a lot easier.
>
> So, DBI compiled fine, and then in order to _compile_ DBD::Sybase, I had
> to add -I/sybase/OpenClient/include to all the gcc calls in the Makefile.
> Is this acceptable, or did I do something wrong?
You should edit CONFIG and set SYBASE=/sybase/OpenCLient, then rebuild
from scratch (start with perl Makefile.PL).
>
> It compiled fine, but when I tried to make test, the following errors
> were produced, it looks like it can't find the client header files again,
> but I can't find a place to add it.
>
> install_driver(Sybase) failed: Cant load
> './blib/arch/auto/DBD/Sybase/Sybase.so for module DBD::Sybase ld.so.1 failed:
> /bin/perl: fatal: relocation error: file
> /blib/arch/auto/DBD/Sybase/Sybase.sop: symbol ct_con_props: referenced symbol
> not found at /usr/local/lib/perl5/sun4-solaris/5.00404/DynaLoader.pm line 166
No - it found the header files OK. What it didn't find was the
libraries.
Now, if you keep things installed like this you *may* have incompatible
files
in /sybase/locales and/or /sybase/charsets that OpenClient needs. OC
will
use the $SYBASE env. variable to find the location of these directories,
and
it will also use that variable to find the interfaces file.
You could also place a softlink like this
ln -s /sybase/interfaces /sybase/OpenClient/interfaces
and then "setenv SYBASE /sybase/OpenClient" before running your perl
scripts...
Michael
--
Michael Peppler -||- Data Migrations Inc.
mpeppler@mbay.net -||- http://www.mbay.net/~mpeppler
Int. Sybase User Group -||- http://www.isug.com
Sybase on Linux mailing list: ase-linux-list@isug.com
------------------------------
Date: 22 Oct 1998 01:40:06 GMT
From: alfred@hw.stratus.com (Alfred von Campe)
Subject: What's the "best" way to call one Perl script from another?
Message-Id: <70m2dm$3ks@transfer.stratus.com>
I would have thought that this would be in the FAQ, but if it is
I couldn't find it. I've written a bunch of Perl scripts that
implement parts of our regression environment. Now I am writing
a script that will automate the entire process, and find myself
calling one Perl script from another often. So far, I've been
using the following:
$status = system("script_name param1 param2");
die "$0: script_name failed\n" if (($status >> 8) != 0);
But using system() to call another Perl script doesn't feel very
perlish - is there a better (more elegant) way to do it?
Alfred
--
+------------------------------------------------------------+
| Phone: H: 978.448.6214, W: 508.490.6306, fax: 508.229.4076 \
| Mail: Alfred von Campe, 402 Lowell Road, Groton, MA 01450 \
| Email: alfred@hw.stratus.com \
+----------------------------------+-----------------------------+
| Why is common sense so uncommon? | I'd rather be flying N4381Q |
+----------------------------------+-----------------------------+
------------------------------
Date: 12 Jul 98 21:33:47 GMT (Last modified)
From: Perl-Request@ruby.oce.orst.edu (Perl-Users-Digest Admin)
Subject: Special: Digest Administrivia (Last modified: 12 Mar 98)
Message-Id: <null>
Administrivia:
Special notice: in a few days, the new group comp.lang.perl.moderated
should be formed. I would rather not support two different groups, and I
know of no other plans to create a digested moderated group. This leaves
me with two options: 1) keep on with this group 2) change to the
moderated one.
If you have opinions on this, send them to
perl-users-request@ruby.oce.orst.edu.
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 4040
**************************************
