[8994] in Perl-Users-Digest
Perl-Users Digest, Issue: 2612 Volume: 8
daemon@ATHENA.MIT.EDU (Perl-Users Digest)
Fri May 15 17:07:40 1998
Date: Fri, 15 May 98 14:00:52 -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 Fri, 15 May 1998 Volume: 8 Number: 2612
Today's topics:
Re: Does Perl have a IDE?I don't like command line. (Greg Bacon)
Re: Does Perl have a IDE?I don't like command line. (Ilya Zakharevich)
Re: Does Perl have a IDE?I don't like command line. <ebohlman@netcom.com>
Re: Does Perl have a IDE?I don't like command line. (Ken Fox)
Re: Does Perl have a IDE?I don't like command line. (Ken Fox)
Re: Does Perl have a IDE?I don't like command line. (Ilya Zakharevich)
help with quoting patterns, please <dmk@bell-labs.com>
Re: localtime function and Y2K compliance (brian d foy)
Re: localtime function and Y2K compliance (John Erjavec V)
Re: localtime function and Y2K compliance (Joel Coltoff)
Re: localtime function and Y2K compliance <ebohlman@netcom.com>
LWP vs Perl/Tk/Sockets for database interface. (Bryan Miller)
Re: Objects <ebohlman@netcom.com>
Re: perl install debate . . . (brian d foy)
Re: Problem with spaces in directory path names in Perl <lr@hpl.hp.com>
Re: Problem with spaces in directory path names in Perl <buzz@bear.com>
Re: reading file into an array of hash (brian d foy)
Re: Tip: Filehandles <tchrist@mox.perl.com>
Re: yet another test <brianm@kodak.com>
Digest Administrivia (Last modified: 8 Mar 97) (Perl-Users-Digest Admin)
----------------------------------------------------------------------
Date: 15 May 1998 19:51:04 GMT
From: gbacon@cs.uah.edu (Greg Bacon)
Subject: Re: Does Perl have a IDE?I don't like command line.
Message-Id: <6ji6b8$kru$1@info.uah.edu>
In article <6jht3q$nn$1@ultranet.com>,
fsg@ultranet.com writes:
: gbacon@cs.uah.edu (Greg Bacon) writes:
: >Believe it or not, I once used an IDE. It was Borland C++ version 3.5
: >on a DOS machine. I used it because I could have multiple files open at
: >once and flip around between them. I could compile my code without
: >leaving the editor. The IDE was necessary because of the chains in
: >which the operating system held me. Claiming the need for an IDE is
: >admitting to using a very challenged system.
:
: IDEs do more than flip windows; for instance, in Borland's C++Builder 3
: product, which has a great IDE, I can put my cursor over a variable
: in the source of a running program and see what the value of the variable
: is. Sounds odd, but it's an immense timesaver.
Perhaps I'm a child of the free software generation or maybe I'm just a
pessimist. When I see a glitzy shrinkwrapped box with the latest
greatest Buck-Bang-Pow compiler/debugger/IDE/whatever, I usually wonder
whether the cutesy bells and whistles are really worth the money when I
can have software with equivalent if not better (and probably more
reliable) software for an infinitely cheaper amount? If such glamour
were really so vital to programming, wouldn't it appear in free
development environments too?
I find that having to go to the mouse too frequently really limits my
productivity. IDEs seem to want you to click, click, click your way to
happiness.
: The IDE also makes
: the annoying and tedious part of writing win32 programs (designing the
: interface) trivial, which is good, because I've got better things to worry
: about than implementing a button.
What of Tk? Tk code will run on many, many, many more platforms than
the hoarde fodder will.
: And of course, all of the tools (editor,
: debugger, compiler, libraries) work together in harmony so I don't have to
: worry about tool-to-tool communication conflicts.
I've never heard of vi and perl -d exchanging even one cross word.
: It's much the same reason why people buy Macs. Macs are truly excellent
: machines. I love Macs. I would never be caught dead using a Mac, but
: like win32, like Unix, like Perl, like C++, like COBOL, everything really
: does have a time and a place. You should accept that.
I don't know why you'd believe that I don't. There are serious systems
for serious users and then there's everything else for everyone else.
Concept based visual interfaces are best suited for people who have
little need or desire for what actually happens so long as they can
express to the machine the task they want it to perform. Such novice
friendly systems are necessarily expert unfriendly.
It would be nice if there were a system that could change with the
user's needs. Desktop environments like the KDE try to provide a novice
friendly environment that can be left behind or at least pushed aside as
the user needs.
The only time and place for C++ is as an example of how *not* to
retrof^Wdesign a programming language. See the Stroustrup
``interview''.
: >But I suspect many of these people aren't programmers. They're
: >assembling black boxes to make one slow, monolothic black box. This is
: >no more programming that swapping parts in a car is automobile design.
:
: It seems that you and Tom are both bothered more by nomenclature and
: what the people call themselves than anything else. If they called
: themselves 'access forms designers', and yet kept doing the exact same
: thing, would you be as vocal in your abuse? If not, why not?
No, I wouldn't. How thrilled would you be if I called myself an
Anglophile but went around bashing England and everything related to it.
I see no value in lying about one's qualifications.
: >They're fine for the people who need the crutch. Programmers should
: >learn to program without the crutch or choose less challenged systems.
:
: Perhaps the problem is that you're inexperienced with IDEs. Tell you
: what -- why don't you go down to the university store and get Delphi 3,
: or C++Builder 3, at a student discount. Play with it for a few days;
: if you don't like it, I'll send you a check for the full amount of the
: purchase.
Could you please drop the fixation on the fact that I'm still in school?
It adds nothing to your argument and weakens your credibility. Second,
why should I pay to replace software that already works the way God and
Dennis intended? Thirdly, I don't possess the tool-challenged systems
on which to run such software.
Thanks for the offer, though.
: IDEs are a 'crutch' in the same way that the regex engine is a 'crutch'.
: Much of the work they hide from you is work that you simply never want
: to do by hand, no matter what; and the fact that I can no longer code a
: regex engine by hand is of no interest or import to me. I've just got
: better things to do.
Like Russ, I don't see the harm in using software or libraries that
perform the trivial or menial tasks that I could do anyway. I think
it's safe to say that any programmer could implement, say, a linked list
library, but the process has been done a million times and can be error
prone. It's nice to have wheels you can pull from a shelf without
having to endure the invention process all over again. Really, Perl is
a lot like a library that contains code for all the things programmers
hate to code up again and again. However, Perl has the added benefit of
doing lots of these things portably too.
Software libraries and IDEs are apples and oranges. Software libraries
still require a programmer's knowledge and skill to assemble them
properly. If the boxes are too big and too opaque, the end result will
be slovenly software, despite the lies IDEs would have you believe.
: As for the idea of choosing less challenged systems -- that's one reason
: why I suggested that you'll learn a lot when you leave school. Not because
: I think you're stupid, but because you really haven't seen a lot of the
: world yet.
Would you work in a place that subjected small children to sixteen hour
workdays? Would you work for someone for whom you couldn't find even an
ounce of respect in your entire person? I realize that hard times do
visit, but this is not the state of our industry. There are too many
Dev^H^H^HBill worshippers in this world for me to waste even one
keystroke in an environment that will always fight against those used to
programmer friendly systems.
: I've been a professional hacker for a number of years now, and only through
: the grace of God have I been able to choose my platform. Most of the people
: I work with are told what platform to work with. Sometimes the end customer
: even specifies the platform explicitly as part of the requirements document.
: The idea that 1/5th, if that many, of the programmers out there can 'choose
: less challenged systems' is quite seriously a riot.
One's freedom to provide one's own direction is directly proportional to
the skills and knowledge one has mastered. Every employer will always
need good hackers. The fact that you've been able to choose your
platform suggests to me that you've been able to select from the most
appealing jobs.
: >xterm &
:
: And how do you type that in? Do you use a mouse pointer or a window
: manager on that X server?
One of the Windows platform's hugest mistakes is that, as a design goal,
it seeks to minimize use of one of the highest bandwidth computer input
devices, the keyboard. Having to go to the mouse more frequently than
to tell my window manager, ``hey, I'm interested in this window now'' is
irritating and inefficient.
: The point is, you're talking personal preference (a little hypocritically,
: but that's of no matter), not objective reality. When I was a CS college
: student, I often spoke about computers in terms of objective reality. I've
: moved on to speaking about everything else in terms of objective reality,
: but computers, no sir. So I know where you're coming from, but from
: personal experience, I know that it doesn't work.
The difference in subjectivity usually amounts to the granularity of
control sought. Do you believe differently?
: >I never said that a Win32 user is necessarily clueless. In fact, I've
: >openly stated that there are Win32 users of clue.
:
: Again, I suggest you go back and read your posts. You strain credulity.
Please provide specific examples.
: >By their actions, they detract from the usefulness of a great resource.
:
: So set up another resource.
We all wait like hungry predators for clp.moderated.
: Or figure out a way to solve the problem.
: Use Perl. Use killfiles.
Again, ignoring the problem doesn't make it go away. What happens to
the new Perl user who comes to clpm to see 250+ messages, most of which
is junk? He'll likely think, ``what a load of shit'' and move on. I
feel an obligation to limit the requests for FAQtime stories and other
noisy junk to make this newsgroup as valuable a resource as it was when
I first came.
: >Why do you suppose Larry has left comp.lang.perl.misc? If you kill
: >clpm, you kill one of Perl's biggest selling points and much of its
: >usefulness.
:
: Nonsense. clpm is cool and all, but Perl is Perl, and clpm does not
: contribute significantly to the 'usefulness' of the language.
False. What other language can claim such a useful resource with highly
knowledgable developers who are very eagar to help people free of
charge?
: I suppose Larry left clpm because he didn't have the time.
I hear he saw this coming.
: You will
: notice that he didn't start responding to every post with a scathing
: diatribe about how their platform sucks.
I am told by many that Larry despises conflict. In fact, I've never
seen him get involved in any of the p5p skirmishes except to put an
end to them. Different people handle problems in very different ways.
: In fact, he's even actively
: attempted to bring the win32 people into the fold, and has helped out
: trying to get more win32 cooperation and involvement going in the core.
If someone stinks, view it as a reason to help them, not a reason to
avoid them.
-- Larry Wall in <199702111730.JAA28598@wall.org>
Even Larry isn't able to tolerate the stench around here.
: How is your way of blasting everyone more successful than his way of
: working within the system?
I challenge you to show how I blast everyone. I'd like to see an
example how I've blasted someone who presented a non-FAQ, non-homework
problem and actually attempted to solve their problem on their own
first.
: 1. Don't take it personally. You're likely to notice the outliers on the
: curve -- the people who want spoonfeeding and who rant and scream that you
: suck if they don't get it. You probably don't notice the hundreds of people
: that you help, because by definition they're off busily using the great info
: to do their job.
I'm not taking it personally. When someone defecates near the food at
the picnic, it tends to spoil everyone's appetite and kind of ruin the
outing. I hate to see people shit all over our newsgroup.
: 2. If you find yourself taking it personally, do something else. clpm
: doesn't need you, or Tom, or Tom, or Chip, or Randal, or me. Nothing
: in the Perl distribution you received mandates that you spend your time
: getting aggravated.
If we all took this defeatist attitude, I could not imagine the slum
clpm would be now. You criticize me for not seeing things through the
eyes of others, yet you don't understand that I'd rather take the
problem on instead of ignoring it.
: Keep trying, or get out of the way and stop embarrassing the rest of us.
How am I embarrassing anyone?
: >Ignoring the problem won't make it go away.
:
: Yes, it will. It will make it go away _for_ _you_.
That would be a very selfish attitude to take. It gives some people
great joy to see and participate in a successful potluck dinner.
: Maybe that'll be a good thing, if you must attack the newbies.
You paint with quite a broad brush. Please outline what lead you to
believe that I attack all newbies as a matter of personal policy.
Greg
--
open(G,"|gzip -dc");$_=<<EOF;s/[0-9a-f]+/print G pack("h*",$&)/eg
f1b88000b620f22320303fa2d2e21584ccbcf29c84d2258084
d2ac158c84c4ece4d22d1000118a8d5491000000
EOF
------------------------------
Date: 15 May 1998 20:35:15 GMT
From: ilya@math.ohio-state.edu (Ilya Zakharevich)
Subject: Re: Does Perl have a IDE?I don't like command line.
Message-Id: <6ji8u4$aff$1@mathserv.mps.ohio-state.edu>
[A complimentary Cc of this posting was sent to Tom Christiansen
<tchrist@mox.perl.com>],
who wrote in article <6jhhmq$bd2$1@csnews.cs.colorado.edu>:
> --tom, still wondering how to pass arguments to a mouse click
Observe Tom rediscovering dragndrop by himself!
Ilya
------------------------------
Date: Fri, 15 May 1998 20:25:21 GMT
From: Eric Bohlman <ebohlman@netcom.com>
Subject: Re: Does Perl have a IDE?I don't like command line.
Message-Id: <ebohlmanEt0M29.JA8@netcom.com>
Tom Christiansen <tchrist@mox.perl.com> wrote:
: [courtesy cc of this posting sent to cited author via email]
: In comp.lang.perl.misc,
: smcdow@arlut.utexas.edu (Stuart McDow) writes:
: :How many free IDEs does MS give away? Why is it so hard to
: :program (as in C/C++, Java, etc, etc) under NT without a (costly) IDE?
: Great questions. But I'm not ready to go all the way to the conspiracy
: theory here. I think they are just too stupid to come up with simple,
: small, clean, and intuitive interfaces. Any team of the mediocre can
: create something complex. It takes minds of singular genius to produce
: something simultaneously of startling simplicity and power.
Another possibility. The majority of Microsoft's software (and, to be
fair, that of its commercial competitors as well) isn't bought by
developers. It isn't bought by people who do word processing or desktop
publishing. It isn't bought by people who do spreadsheet calculations.
It's bought by the people that the developers and users report to. You
know, those people with the two peaks in their hair. Managers.
Microsoft's products are designed to impress managers. Managers don't
use those products, so they can't directly evaluate how well those
products work. They can indirectly evaluate them by getting feedback
from their employees, but unfortunately that feedback takes the form of
"how good were the decisions I made" and social norms about hierarchical
relations step in. So that kind of feedback is a minor part of the
evaluation process. They can look at feature counts. Managers certainly
know how to count. They can look at screen shots and decide which are
the most impressive.
The upshot of all this is that some of the normal principles of the free
market don't hold when you're talking about mass-produced commercial
software. Free-market theory assumes that producers and consumers come to
agreements based on their interests. It starts to break down when the
producers and consumers don't communicate directly, but rather through an
intermediary that bases its decisions on a third set of mostly unrelated
interests. For an even more extreme example, look at the US health-care
system, where the people who receive services, the people who choose
services and the people who pay for services are generally three different
groups.
------------------------------
Date: 15 May 1998 20:34:49 GMT
From: kfox@pt0204.pto.ford.com (Ken Fox)
Subject: Re: Does Perl have a IDE?I don't like command line.
Message-Id: <6ji8t9$f4s2@eccws1.dearborn.ford.com>
Keywords: If you're happy and you know it, clunk your chains.
Tom Christiansen <tchrist@mox.perl.com> writes:
> In comp.lang.perl.misc, chris+usenet@netmonger.net writes:
> : Some of them [emacs users] WROTE those tools you claim they don't
> : know how to use.
>
> Really? Are you sure? No one I know from the Unix group at Bell
> Labs
Does anybody use a Bell Labs variant of Unix anymore? ;)
> nor CSRG at Berkeley ever used emacs. These are the people who
> wrote virtually everything in /bin and /usr/bin, the core tools I'm
> referring to.
Unless you have a GNU system, right? Those CSRG guys made a pretty
nice prototype, but it took the emacs users to get it right. ;)
- Ken
--
Ken Fox (kfox@ford.com) | My opinions or statements do
| not represent those of, nor are
Ford Motor Company, Powertrain | endorsed by, Ford Motor Company.
Analytical Powertrain Methods Department |
Software Development Section | "Is this some sort of trick
| question or what?" -- Calvin
------------------------------
Date: 15 May 1998 20:22:52 GMT
From: kfox@pt0204.pto.ford.com (Ken Fox)
Subject: Re: Does Perl have a IDE?I don't like command line.
Message-Id: <6ji86s$f4s1@eccws1.dearborn.ford.com>
"Keith G. Murphy" <keithmur@mindspring.com> writes:
> One question to ponder: as a machine becomes an appliance (think "car",
> for instance), are *more* or *fewer* people qualified to
> make/modify/repair it?
Lots of people think they're qualified to drive them too. ;)
If we had to lift the hood to program, it'd cut down a *lot* on the
idea that anybody can program. "Gosh, lifting the hood is hard, maybe
I better not do that."
- Ken
--
Ken Fox (kfox@ford.com) | My opinions or statements do
| not represent those of, nor are
Ford Motor Company, Powertrain | endorsed by, Ford Motor Company.
Analytical Powertrain Methods Department |
Software Development Section | "Is this some sort of trick
| question or what?" -- Calvin
------------------------------
Date: 15 May 1998 20:53:52 GMT
From: ilya@math.ohio-state.edu (Ilya Zakharevich)
Subject: Re: Does Perl have a IDE?I don't like command line.
Message-Id: <6jia10$c61$1@mathserv.mps.ohio-state.edu>
[A complimentary Cc of this posting was sent to Tina Marie Holmboe
<tina@tech.scandinaviaonline.se>],
who wrote in article <6jhf1d$hu8$5@news1.sol.no>:
> In article <6jbf1c$au2$1@mathserv.mps.ohio-state.edu>,
> ilya@math.ohio-state.edu (Ilya Zakharevich) writes:
>
> > Yet another proof of my point. One cannot use Perl docs on *nix
> > without fluency with (4 in your examples) other tools.
>
> I'm abit lost in this thread, but... have you ever had a look at the
> tkman program ? If so, is that something you find more helpful ?
On my home machine, I have IBM online books. On department, tkman is
not on the path.
The questions are simple:
a) Does tkman has hierarchical expandable-collapsible
table-of-context?
b) Quick powerful global search?
c) Hyperlinks? (Perl docs in INF form have 3500 (? do not
remember exactly) hyperlinks, I would think that 40% of them
are fake - going to "this" page, but the rest is very useful)
d) Hierachical searchable index?
Ilya
------------------------------
Date: Fri, 15 May 1998 15:47:22 -0400
From: Dave Kristol <dmk@bell-labs.com>
Subject: help with quoting patterns, please
Message-Id: <355C9BCA.1631@bell-labs.com>
[Please respond by email -- I don't peruse the news group.]
This question pertains to Perl 5.003 on Solaris.
Consider the following Perl script. I want to be able to specify a
pattern as a string and to be able to use it as in the first two "print"
lines in the example below.
================================
#!/opt/default/bin/perl
$test = '\continuous_directive_with_nospaces';
$alldir_p = '^\\\S+$'; # want to match what literal pattern matches
print "matches var " . $alldir_p . "\n" if $test =~ $alldir_p;
print "matches pat " . $alldir_p . "\n" if $test =~ /$alldir_p/;
print "matches literal\n" if $test =~ /^\\\S+$/;
================================
The Perl 5 book implies (p.67) that the quoting in $alldir_p is
sufficient to make all three tests equivalent, but only the last one
actually matches my test string! Why? I'm trying to match backslash,
followed by one or more non-whitespace characters (as in the third print
line).
I've tried the following alternatives for $alldir_p (while changing
nothing else):
1) $alldir_p = '^\\\\S+$';
Does same thing as above
2) $alldir_p = '^\\\(\S+)$';
All three tests match(!).
3) $alldir_p = '^\\(\S+)$';
Fails: /^\(\S+)$/: unmatched () in regexp at y.pl line 8.
Evidently there's something I don't understand, either about quoting or
about pattern matching, or about the combination.
Can you help me? Thanks.
Dave Kristol
------------------------------
Date: Fri, 15 May 1998 16:11:07 -0400
From: comdog@computerdog.com (brian d foy)
Subject: Re: localtime function and Y2K compliance
Message-Id: <comdog-ya02408000R1505981611070001@news.panix.com>
Keywords: from just another new york perl hacker
In article <355C9AB2.183A869@moen.com>, Jennifer Amon <jamon@moen.com> posted:
> localtime
>If that is the case, then 2000 would come back as 100, and
>2001 would come back as 101, etc.
>
>Is this really the case, and is there any plan for that
>to change?
that's really the case and why do you think it should change?
--
brian d foy <comdog@computerdog.com>
CGI Meta FAQ <URL:http://computerdog.com/CGI_MetaFAQ.html>
Comprehensive Perl Archive Network (CPAN) <URL:http://www.perl.com>
Perl Mongers <URL:http://www.pm.org>
------------------------------
Date: 15 May 1998 20:18:16 GMT
From: jev@pconline.com (John Erjavec V)
Subject: Re: localtime function and Y2K compliance
Message-Id: <6ji7u8$8be$1@bell.pconline.com>
Jennifer Amon (jamon@moen.com) wrote:
: If that is the case, then 2000 would come back as 100, and
: 2001 would come back as 101, etc.
:
: Is this really the case, and is there any plan for that
: to change?
Jennifer-
This is the case, and as far as I know, there is no plan for this to
change. Why would it? You said yourself that the year value returned is
the current year minus 1900, so to get the current year, you are supposed
to add 1900, right. Why would this need to be changed for the year 2000?
Isn't 1900 plus 100 2000, and 1900 plus 101 2001? If you could give
reasons why this is a Y2K problem, then maybe people would think about it,
but I don't think that there is a problem in the implementation.
-JEV
--
John Erjavec V PGP fingerprint:
jev@pconline.com 7593 1B5A AE11 C0FE BA09 EB5E 8DE9 D2E5 BF5B 87AD
http://www.pconline.com/~jev/index.html
------------------------------
Date: Fri, 15 May 1998 20:43:51 GMT
From: joel@wmi0.wmi.com (Joel Coltoff)
Subject: Re: localtime function and Y2K compliance
Message-Id: <6ji9d9$1cl@netaxs.com>
In article <355C9AB2.183A869@moen.com>, Jennifer Amon <jamon@moen.com> wrote:
>
>If that is the case, then 2000 would come back as 100, and
>2001 would come back as 101, etc.
Yes that is really the case. I hope They don't change it. It works
fine as is and if changed there is a small chance that They will
introduce a Y2K bug. You wouldn't want that to happen so close
to the year 2000 now would you? If it ain't broke don't fix it.
--
Joel Coltoff
I'd explain it, but there's a lot of math. -- Calvin
------------------------------
Date: Fri, 15 May 1998 20:51:35 GMT
From: Eric Bohlman <ebohlman@netcom.com>
Subject: Re: localtime function and Y2K compliance
Message-Id: <ebohlmanEt0nA0.KFq@netcom.com>
Jennifer Amon <jamon@moen.com> wrote:
: My trusty Camel book (2nd Edition) says, and I quote:
: localtime
: localtime EXPR
[snip]
: If that is the case, then 2000 would come back as 100, and
: 2001 would come back as 101, etc.
: Is this really the case, and is there any plan for that
: to change?
Yes.
No.
It's up to you to handle years intelligently.
------------------------------
Date: 15 May 1998 20:52:37 GMT
From: millerb@millerb.fc.hp.com (Bryan Miller)
Subject: LWP vs Perl/Tk/Sockets for database interface.
Message-Id: <6ji9ul$fsv@fcnews.fc.hp.com>
I am in the design phase for a front end to a miniSQL-ish
database. The database will be fairly small and used over
100BT by 15 to 20 engineers. Small = less than 10,000 records
of 30 string and integer fields. No blobs.
I am interested in input from Perl hackers as two which is
simpler/cleaner: using LWP and POSTing to a CGI script or
simply writing a small Perl/Tk form that submits/retrieves
data directly to/from the SQL database?
I currently have a CGI script that works reasonably well but
I would like to have better control such as preventing users
from hitting return and posting data to the database, etc.
I welcome any advice, anecdotal information, horror stories...
cheers,
Bryan Miller
------------------------------
Date: Fri, 15 May 1998 20:49:39 GMT
From: Eric Bohlman <ebohlman@netcom.com>
Subject: Re: Objects
Message-Id: <ebohlmanEt0n6r.K85@netcom.com>
Shane <shane@techplus.com> wrote:
: Is there a way to make an array of pointers in perl which all point to
: another array of data. If so how? I can do this in java but not perl.
In Perl, we don't got no steenkin' pointers, we got references. Take a
look at perlref to get oriented and then perllol to see an example of
exactly what you want to do. But don't stop there; you'll be on a roll
and you'll want to look at perldsc; by the time you're done with that,
you should be able to figure out how to represent any arbitrary data
structure in Perl.
------------------------------
Date: Fri, 15 May 1998 16:54:18 -0400
From: comdog@computerdog.com (brian d foy)
Subject: Re: perl install debate . . .
Message-Id: <comdog-ya02408000R1505981654180001@news.panix.com>
Keywords: from just another new york perl hacker
In article <355C965C.8BD118C6@mail.umb.com>, wes hedrick <wes@mail.umb.com> posted:
>My sysadmin guy is convinced that our complete and utter failure to install mod_perl correctly has
>nothing to do with the perl install being in the form of a solaris package. If configured properly
>is he right? Notwithstanding a seemingly flawless install(first try) after compiling perl from
>source, on the same box.
the first few times i compiled mod_perl i kinda flopped around like
a fish out of water. i've done it several times now, so it's a bit
easier to deal with.
where, specifically, are you having the problem? with which server
source tree are you working with? which perl are you using (how about
the output of perl -V)? which mod_perl? which compiler?
let's see if we can't hurry up and get you using it :)
--
brian d foy <comdog@computerdog.com>
CGI Meta FAQ <URL:http://computerdog.com/CGI_MetaFAQ.html>
Comprehensive Perl Archive Network (CPAN) <URL:http://www.perl.com>
Perl Mongers <URL:http://www.pm.org>
------------------------------
Date: Fri, 15 May 1998 13:16:11 -0700
From: "Larry Rosler" <lr@hpl.hp.com>
Subject: Re: Problem with spaces in directory path names in Perl on WIN32
Message-Id: <6ji7qm$r4r@hplntx.hpl.hp.com>
Jan Krynicky wrote in message <355D0BF4.2595@McCann.cz>...
...
>BTW, unless strictly necessary never install anything into a directory
>with
>a space in it's name. It's pretty annoying to have to enclose
everything
>into quotes.
These problems are caused by the command interpreter splitting the
string at the space. Unix shells behave exactly the same. Neither
32-bit Windows systems nor Unix systems have any problem with such
strings as arguments to 'open' or 'opendir'.
But -- of course -- Windows has a much restricted set of permissible
characters. I have identified (by experiment) the following set of
printable seven-bit ASCII characters that are not permitted:
" * + , / : ; < = > ? [ \ ] |
These are all OK on UNIX (modulo the special semantics for slash),
though if you use some of them without quoting, you are asking for
trouble from the shell.
All this is bitter experience derived from porting an application that
allows users to name files arbitrarily (the users don't know they are
filenames). On Unix, I have to map slash (I chose to map to hash #,
becuase it rhymes and looks a little alike :-). On Windows, I map all
of the above (except slash) to underscore (for no good reason). Upper-
and lower-case letters map themselves into each other in a somewhat
arbitrary way. Collisions from mapping draw warnings.
A very thin thread started here yesterday addresses portability problems
due to system-call deficiencies. This thread addresses file-system
naming limitations. Is there a source or repository for lore about
these or other such issues re Perl? Let's face reality instead of only
bitching about it!
--
Larry Rosler
Hewlett-Packard Laboratories
lr@hpl.hp.com
------------------------------
Date: Fri, 15 May 1998 16:26:50 -0400
From: Buzz Moschetti <buzz@bear.com>
Subject: Re: Problem with spaces in directory path names in Perl on WIN32
Message-Id: <355CA50A.67A@bear.com>
Jan Krynicky wrote:
> [...]
> You HAVE to enclose the whole .exe name into doublequotes:
>
> c:\> "C:\Program
> files\DevStudio\programs\preprocess\release\preprocess.exe" data.txt
Careful about doing that in perl code:
$path = "C:\Program files\DevStudio\programs\pre...";
Those backslashes are interpreted within double quotes. Use single
quotes instead:
$path = 'C:\Program files\DevStudio\programs\pre...';
------------------------------
Date: Fri, 15 May 1998 16:29:23 -0400
From: comdog@computerdog.com (brian d foy)
Subject: Re: reading file into an array of hash
Message-Id: <comdog-ya02408000R1505981629230001@news.panix.com>
Keywords: from just another new york perl hacker
In article <1d92p4a.1kp9dg4kpzdksN@slip166-72-108-62.ny.us.ibm.net>, kpreid@ibm.net (Kevin Reid) posted:
brian d foy <comdog@computerdog.com> stupidly wrote:
:: while( defined( $_ = <FILE> ) ) { ... }
>It's
> while ($var = <FILE>) {
>that's the problem.
oy - so it is. sorry about that :(
#!/usr/bin/perl -w
while( <DATA> ) { print "First loop: $_"; last }
while( my $line = <DATA> ) { print "Second loop: $line";last }
while( defined( my $line = <DATA> ) ) { print "Third loop: $line";last }
__DATA__
0
1
2
3
Value of <HANDLE> construct can be "0"; test with defined()
at test.pl line 5
First loop: 0
Second loop: 1
Third loop: 2
--
brian d foy <comdog@computerdog.com>
CGI Meta FAQ <URL:http://computerdog.com/CGI_MetaFAQ.html>
Comprehensive Perl Archive Network (CPAN) <URL:http://www.perl.com>
Perl Mongers <URL:http://www.pm.org>
------------------------------
Date: 15 May 1998 20:35:54 GMT
From: Tom Christiansen <tchrist@mox.perl.com>
Subject: Re: Tip: Filehandles
Message-Id: <6ji8va$ecq$1@csnews.cs.colorado.edu>
[courtesy cc of this posting sent to cited author via email]
In comp.lang.perl.misc,
Martin Gregory <mgregory@asc.sps.mot.com> writes:
:> Mea culpa. The drawback is that because filehandles aren't a basic
:> data type, you have to futz with typeglobs to store or pass them.
:Ditto. This is why I use them (FileHandle).
And what do you do about subroutines? It's the same
as with filehandles. Notice:
&callme(SOMETHING);
sub SOMETHING { print "here I am\n"; }
sub callme { $func = shift; &$func(); }
That works fine. Just as this works fine:
&something(STDOUT);
sub something { $fh = shift; print $fh "here I am\n"; }
But there are big problems there in both cases. Passing \*STDOUT is
much better, as is passing \&something is much better.
&callme(\&SOMETHING);
&something(\*STDOUT);
I don't understand what the big fear here is.
--tom
--
Though I'll admit readability suffers slightly...
--Larry Wall in <2969@jato.Jpl.Nasa.Gov>
------------------------------
Date: Fri, 15 May 1998 16:47:03 -0400
From: Brian Mathis <brianm@kodak.com>
To: Christian Matthias Zmasek <cmzmasek@artsci.wustl.edu>
Subject: Re: yet another test
Message-Id: <355CA9C7.2A6FD95D@kodak.com>
Christian Matthias Zmasek wrote:
>
> test
>
> test
Please do NOT post test messages to a "real" newsgroup. There are a
myriad of "test" newsgroups that are designed for this purpose, such as
misc.test
Test Failed.
Brian Mathis
------------------------------
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 2612
**************************************
