[7737] in Perl-Users-Digest
Perl-Users Digest, Issue: 1362 Volume: 8
daemon@ATHENA.MIT.EDU (Perl-Users Digest)
Mon Nov 24 05:07:18 1997
Date: Mon, 24 Nov 97 02:00:20 -0800
From: Perl-Users Digest <Perl-Users-Request@ruby.OCE.ORST.EDU>
To: Perl-Users@ruby.OCE.ORST.EDU (Perl-Users Digest)
Perl-Users Digest Mon, 24 Nov 1997 Volume: 8 Number: 1362
Today's topics:
Re: [Q] Why does this behave this way? (Jonathan Feinberg)
Re: Atomic operations (was Re: exclusive file rights) <chemcd@jetson.uh.edu>
Re: Convert an array to hash? (Andrew M. Langmead)
Re: core dumped with split (E.None Archibald)
Re: core dumped with split (Ilya Zakharevich)
Re: core dumped with split (E.None Archibald)
Cron <nospam.tbsmith@mindspring.com>
dbmopen? can't make it work (Laurel Shimer)
getting output from system <mikihasa@worldnet.att.net>
help with macperl... <tjbiuso@redrose.net>
Re: How to: GD-1.14, Perl 5.004.01 & OS/2? samdie@ibm.net
Re: How to: GD-1.14, Perl 5.004.01 & OS/2? (Ilya Zakharevich)
Re: How to: GD-1.14, Perl 5.004.01 & OS/2? samdie@ibm.net
maintain unique elements in an array? no duplicates? <jt@cs.pdx.edu>
passing more than 1 variable in a href <vfox@nbnet.nb.ca>
Re: Safe use of flock() -- was Re: giving up on flock (Clay Irving)
Re: Socket Problems <sdavis@localIP.net>
Digest Administrivia (Last modified: 8 Mar 97) (Perl-Users-Digest Admin)
----------------------------------------------------------------------
Date: Sun, 23 Nov 1997 14:31:23 -0500
From: jdf@pobox.com (Jonathan Feinberg)
Subject: Re: [Q] Why does this behave this way?
Message-Id: <MPG.ee24e915eb490a989688@news.concentric.net>
lm@binary9.net said...
> @a=qw ( Hello );
> $a=sprintf @a;
Because the first argument to sprintf is taken in a scalar context,
you might as well have written
$a = sprintf scalar @a;
Which decomposes to
$a = sprintf 1;
So, your format string is '1', and there is no list
of strings to substitute into your format string.
See perlfun for the explanation of the sprintf function.
I'm guessing that it's not what you mean.
--
Jonathan Feinberg jdf@pobox.com Sunny Manhattan, NY
------------------------------
Date: Sun, 23 Nov 1997 15:04:57 -0800
From: Yong Huang <chemcd@jetson.uh.edu>
To: Bill Guindon <billg@networkapparel.com>
Subject: Re: Atomic operations (was Re: exclusive file rights)
Message-Id: <3478B699.15C2@jetson.uh.edu>
Bill Guindon wrote:
> is running on 95. On the same thought, I've been testing the SERVER_NAME to
> determine where I'm running, is there any way to get the OS so my test can
> be a bit more portable?
Do you want to know the operating system you're using? $^O does that.
Yong
------------------------------
Date: Sun, 23 Nov 1997 17:57:03 GMT
From: aml@world.std.com (Andrew M. Langmead)
Subject: Re: Convert an array to hash?
Message-Id: <EK41v3.50H@world.std.com>
designky <designky@sonnets.dot.com> writes:
>I am wondering is there an easy way to convert an array to hash with the array
>values be the keys, and assign 1's to the hash values.
>what i really want is to do this:
> %hash = split(/ /,$line)
You could say:
%hash = map { $_ => 1 } split / /,$line;
or
@array = split / /, $line;
@hash{@array} = (1) x scalar @array;
or
for $key (split / /, $line) {
$hash{$key} = 1;
}
The first takes the list returned from split, and then uses map to
transform it to a list with each element of its input followed by a
one. The hash then takes this list and sets each key to the original
array element and the value to one.
The second makes and array based on the output of split. It then makes
a list of ones the size of the array and assigns them to a hash slice
of with all the array elements as the keys.
The third makes a list of the output of split and iterates over it. In
each iteration, it assigns to the hash with the element from the
original list being the key and a one being the value.
Testing it with the benchmark module seems to show that the for loop
and the slice methods as close in speed (the slice being slightly
faster), with the map one falling significantly behind.
(How come the Benchmark module only measures efficiency in what the
Camel book calls "Time Efficiency", and does nothing about "Space
Efficiency", "Programmer Efficiency", "Maintainer Efficiency", "Porter
Efficiency", or "User Efficiency"?)
#!/usr/bin/perl -w
use Benchmark;
my $count = shift || die "usage: bench count";
my $line = join ' ', 0 .. $count;
timethese(1000, {
use_map => q(%hash = map { $_ => 1 } split / /,$line;undef %hash),
use_slice => q(@array = split / /, $line;
@hash{@array} = (1) x scalar @array;undef %hash),
use_loop => q(for $key (split / /, $line) {$hash{$key} = 1;};undef %hash;)
});
--
Andrew Langmead
------------------------------
Date: 23 Nov 1997 19:33:39 GMT
From: yevgene@xochi.tezcat.com (E.None Archibald)
Subject: Re: core dumped with split
Message-Id: <65a0ej$b5o$1@tepe.tezcat.com>
Edmund Grimley-Evans <etg10@cl.cam.ac.uk> wrote:
: Is there a good reason for the following?
: $ perl5 -e 'split /(a)|(b)/,"a";'
: Memory fault - core dumped
There seems to be a problem using alternation with separate capturing
parentheses. Funky. It doesn't seem to be an issue with the single "a",
since "sdfsdjfsdlkgjdlasdkldgsgjlkbsfjdfglkjaafgjfdlgb" dumps core as
well.
It's a trick question, though, because alternation is not what you want.
See below...
: How can I avoid the conditions that make this happen?
: I want to be able to tokenise the input in a robust and general
: manner and I'm beginning to think that I can't use split for
: this, unfortunately, but will have to reimplement something
: very similar with m/^.../ in a loop ...
Well - you can keep using split, or you can use a while loop (remember,
matches in //g mode return a -BOOLEAN- value; if you try to use foreach()
(or for(), they're the same thing) you will get unexpected results.)
$string = "ssgsgsdfsdfadfsgdfabdsbdsgfddfgsdsdfgagfsdgb";
@sspl = split(/([ab])/,$string); # split matches on the regex. capturing
# parens interpolate the matched 'a' or 'b'
# into the resultant array. [ab] is 'better'
# than (a)|(b) in any regex context anyway.
while ($string =~ /([^ab]*) # 0-? of anything that is -not- 'a' or 'b'
([ab]) # exactly one 'a' or 'b'
/gx) { # repeat, returning TRUE until it can't
# match anymore(//g). //x allows us to
# use free spacing and comments.
push @smat, $1, $2;
}
print "$string\n\n"; # print the original string for reference
for (my $i=0; $i <= $#sspl; $i++) { # prettyprint split() output
print "'$sspl[$i]' => '$sspl[++$i]'\n";
}
print "\n";
for (my $i=0; $i <= $#smat; $i++) { # prettyprint m//g output
print "'$smat[$i]' => '$smat[++$i]'\n";
}
hope this helps!
--eugene
------------------------------
Date: 23 Nov 1997 19:39:50 GMT
From: ilya@math.ohio-state.edu (Ilya Zakharevich)
Subject: Re: core dumped with split
Message-Id: <65a0q6$j13$1@agate.berkeley.edu>
In article <659dn4$9mp$1@lyra.csx.cam.ac.uk>,
Edmund Grimley-Evans <etg10@cl.cam.ac.uk> wrote:
> Is there a good reason for the following?
>
> $ perl5 -e 'split /(a)|(b)/,"a";'
> Memory fault - core dumped
Does not happen here with 5 randomly picked up versions of Perl. You
need to be more precise, can you say perl -V?
Ilya
------------------------------
Date: 23 Nov 1997 20:16:13 GMT
From: yevgene@xochi.tezcat.com (E.None Archibald)
Subject: Re: core dumped with split
Message-Id: <65a2ud$bc3$1@tepe.tezcat.com>
E.None Archibald <yevgene@xochi.tezcat.com> wrote:
: Edmund Grimley-Evans <etg10@cl.cam.ac.uk> wrote:
: : Is there a good reason for the following?
: : $ perl5 -e 'split /(a)|(b)/,"a";'
: : Memory fault - core dumped
: There seems to be a problem using alternation with separate capturing
: parentheses. Funky. It doesn't seem to be an issue with the single "a",
: since "sdfsdjfsdlkgjdlasdkldgsgjlkbsfjdfglkjaafgjfdlgb" dumps core as
: well.
Note: a 5.003 compile is exhibiting this behaviour (i.e., dumping core.)
5.004_01 does not.
-e.
------------------------------
Date: Sun, 23 Nov 1997 14:39:41 -0600
From: Todd Smith <nospam.tbsmith@mindspring.com>
Subject: Cron
Message-Id: <3478948D.FA4A64AA@mindspring.com>
Can someone tell me what this is, how to use it, and where to get it?
--
-----------------------------------------------------
Take the "nospam." off my email address to email me.
-Todd Smith
------------------------------
Date: Sun, 23 Nov 1997 11:09:22 -0700
From: autopen@autopen.com (Laurel Shimer)
Subject: dbmopen? can't make it work
Message-Id: <autopen-2311971109230001@dynamic20.pm05.mv.best.com>
I have been trying to figure out how to use dbm files in my perl script
using both the camel book and the man page. But there are several things
I cannot figure out the answer to. BTW I'm a C non-expert, though I have
tried looking at 'man dbm' to try to get some more clues about what dbm
is.
1) How to create and write to a dbm file. In the following logs you can
see what I've tried - both of which I knew didn't make sense (but I have
to start somewhere especially if I'm going to ask for help!).
In Thing 1 I tried issuing a dbmpopen and just writing data out to it. I
realize that the OUTPUT file handle should actually be a hash so I
understand why it didn't work.
2) In Thing 2 I tried setting up a hash to associate with the dbmopen,
filling the hash with data, and writing the hash out along with some
keys.However I couldn't figure out what filehandle I should print to -
since the dbmopen is connected to a hash and not a filehandle. What should
I use as a filehandle to print to. As you can see I just tried printing to
the hash - which I realized wouldn't work, and sure enough it didn't.
3) The other thing I don't get is why - since I 'rm' these files each
time, then open them as '0666' - why do the empty .dir and .pag files show
up as being -rw-r--r--? Is that because they are pointer files and not
really the actual data? If I actually managed to create data would those
files be -rw-rw-rw-?
4) How should I REALLY write to these files from my perl script?
------------------Thing 1 ----------------
dbmopen (OUTPUT, 'dbtype.data',0666)|| die 'cannot open dbtype.data\n';
print OUTPUT "1=dog\n";
print OUTPUT "2=frog\n";
print OUTPUT "3=log\n";
print OUTPUT "4=bog\n";
dbmclose(OUTPUT);
dbtype.data.dir dbtype.data.pag
shellx 3% ls -l dbtype.*
-rw-r--r-- 1 autopen vip 0 Nov 23 09:48 dbtype.data.dir
-rw-r--r-- 1 autopen vip 0 Nov 23 09:48 dbtype.data.pag
------------------Thing 2 ----------------
shellx 16% perl5
dbmopen (OUTPUT, 'dbtype.data',0666)|| die 'cannot open dbtype.data\n';
@OUTPUT =("dog","frog","log","bog");
while (($output) = each %OUTPUT) {
$i++;
print OUTPUT "$i=$output\n";
}
dbmclose(OUTPUT);
shellx 18% ls -l dbtype*
-rw-r--r-- 1 autopen vip 0 Nov 23 09:48 dbtype.data.dir
-rw-r--r-- 1 autopen vip 0 Nov 23 09:48 dbtype.data.pag
--
The Reader's Corner: Mystery, Romance, Fantasy
http://www.autopen.com/index.shtml
Subscribe to our free StoryBytes publication
Did you miss? The Pigeon, A St.John Bathshirker Mystery
http://www.autopen.com/pigeon.shtml
------------------------------
Date: Sun, 23 Nov 1997 16:03:00 -0500
From: "Michael R. Harper" <mikihasa@worldnet.att.net>
Subject: getting output from system
Message-Id: <65a5dh$is5@mtinsc02.worldnet.att.net>
I would like to capture (save as a variable) the output (STDOUT) from a
program that I execute (using the system command) during a Perl script.
Can anyone give this Perl newbie some help?
Michael R. Harper
mikihasa@worldnet.att.net
------------------------------
Date: Sun, 23 Nov 1997 11:15:38 -0500
From: MrPc <tjbiuso@redrose.net>
Subject: help with macperl...
Message-Id: <347856AA.A74FE475@redrose.net>
i recently made some scripts for my school's web server assuming it ran
unix when i learned that it was a mac (i hate those things) running
webstar. i looked for sites that gave tutorials on macperl but found
nothing helpful. can ne1 tell me the major differences between unix
perl and macperl? thanks in advance =+}...
------------------------------
Date: Sun, 23 Nov 97 12:23:35 -0500
From: samdie@ibm.net
Subject: Re: How to: GD-1.14, Perl 5.004.01 & OS/2?
Message-Id: <34786ef9$3$fnzqvr$mr2ice@news-s01.ny.us.ibm.net>
In <347413D5.10B5@bigfoot.com>, on 11/20/97
at 11:41 AM, Koos Pol <koos_pol@bigfoot.com> said:
> samdie@ibm.net wrote:
> >
> > Has anyone managed to install GD under os/2?
> I never have succeeded installing modules using the install makefile under
> OS/2.
> I always just copy the module to the /perl/lib directory. I has never failed
> me.
Very hartening (sigh).
Do you mean that "make install" fails or the initial "make"?
If the latter, I suppose that only means copying/moving some files around. But
if it's the basic make that fails (as it usually does for me), then one has to
hack the compiles, links, library maintenance, etc.
It's only mildly annoying that makemaker can't find pod2man since I can run
that easily enough myself (*I* have no trouble finding it; it's in \emx\bin
and on the "path").
In the specific case of GD, however, emxomfar refuses to build a library with
the same extern defined in more than one member. Seems to me it's right to
refuse and I'll have to redo the source code (it's not really an os/2 issue at
all). Then, when compiling gd.c, I pick up an include in perl.h for direct.h
which isn't anywhere in the emx+gcc bundle. Sorta looks like a typo in Ilya's
port but, if that's the case, wouldn't others have complained (or is there
only a handful of people running Perl under os/2?)?
--
-----------------------------------------------------------
samdie@ibm.net 199711231223 -0500
------------------------------
Date: 23 Nov 1997 19:34:33 GMT
From: ilya@math.ohio-state.edu (Ilya Zakharevich)
Subject: Re: How to: GD-1.14, Perl 5.004.01 & OS/2?
Message-Id: <65a0g9$ivs$1@agate.berkeley.edu>
In article <34786ef9$3$fnzqvr$mr2ice@news-s01.ny.us.ibm.net>,
<samdie@ibm.net> wrote:
> In the specific case of GD, however, emxomfar refuses to build a library with
> the same extern defined in more than one member. Seems to me it's right to
> refuse and I'll have to redo the source code (it's not really an os/2 issue at
> all).
Thanks, I hope you send a patch to the author.
> Then, when compiling gd.c, I pick up an include in perl.h for direct.h
> which isn't anywhere in the emx+gcc bundle. Sorta looks like a typo in Ilya's
> port but, if that's the case, wouldn't others have complained (or is there
> only a handful of people running Perl under os/2?)?
Grepping through 5.004_01 source I find only two occurence of
direct.h, both in win32 directory. What do you mean?
Ilya
------------------------------
Date: Sun, 23 Nov 97 15:04:46 -0500
From: samdie@ibm.net
Subject: Re: How to: GD-1.14, Perl 5.004.01 & OS/2?
Message-Id: <347898a6$5$fnzqvr$mr2ice@news-s01.ny.us.ibm.net>
In <65a0g9$ivs$1@agate.berkeley.edu>, on 11/23/97
at 07:34 PM, ilya@math.ohio-state.edu (Ilya Zakharevich) said:
> In article <34786ef9$3$fnzqvr$mr2ice@news-s01.ny.us.ibm.net>,
> <samdie@ibm.net> wrote:
> > In the specific case of GD, however, emxomfar refuses to build a library with
> > the same extern defined in more than one member. Seems to me it's right to
> > refuse and I'll have to redo the source code (it's not really an os/2 issue at
> > all).
> Thanks, I hope you send a patch to the author.
Well I've been operating on the assumption that my configuration is not
sufficiently *nix-like for Perl's tastes (or that my almost non-existent
understanding of *nix is the root of the problem). Since the pieces of GD date
back to Sept '96, if the problem is *really* with the package, wouldn't others
have pointed this out? Wouldn't it have been fixed by now? If the problem lies
with my general set-up, then, even if I hack the installation of GD, I'm going
to keep on having difficulties every time I try to add a package, since I'll
be attacking symptoms rather than causes.
> > Then, when compiling gd.c, I pick up an include in perl.h for direct.h
> > which isn't anywhere in the emx+gcc bundle. Sorta looks like a typo in Ilya's
> > port but, if that's the case, wouldn't others have complained (or is there
> > only a handful of people running Perl under os/2?)?
> Grepping through 5.004_01 source I find only two occurence of
> direct.h, both in win32 directory. What do you mean?
What I meant was line 482 of perl.h There is, however, a complication that
I'll have to investigate. I've installed Perl at home (Warp 4) and at work
(Warp 3). The two installations were unrelated (except that *I* did them) and,
though I intended them to be logically equivalent, I did enough fiddling with
them that there are bound to be some inconsistencies. When I (now, at home)
look at line 482, it's fine (includes dirent.h, as seems reasonable in the
given context). Last Thurs/Fri (at work) I was doing a compare of perl.h
(yours) with the one in the 5.004.04 release and noticed that the line was
different (direct.h vs dirent.h). So I'll have to go back and unzip perl_blb
again (at work).
--
-----------------------------------------------------------
samdie@ibm.net 199711230304 -0500
------------------------------
Date: Sun, 23 Nov 1997 13:14:16 -0800
From: Jon Turner <jt@cs.pdx.edu>
Subject: maintain unique elements in an array? no duplicates?
Message-Id: <34789CA8.B57C5980@cs.pdx.edu>
Is it possible to maintain an array that will always have non-redundant
elements in it? If a new element is to be added, it doesn't get added
if there is already an identical element in the array. Suppose I wanted
to keep an array of people's names that are being read from various
files, I don't want a persons name duplicated in the list. So, Is
there a method to prevent redundant entries? or a method to remove
redundancies? I guess a 'set' datatype is what I'm thinking of.
JT
------------------------------
Date: Thu, 20 Nov 1997 21:52:41 -0400
From: "Vaughn Fox" <vfox@nbnet.nb.ca>
Subject: passing more than 1 variable in a href
Message-Id: <652ptg$81l$1@usenet76.supernews.com>
Hi all
Thanks for all of your help on my last post (another s/// question)
Now I have another.....I'm trying to pass 2 variables in an href statement
from a web page. I've been working with variations of the following:
<a href="http://whatever/script.pl?variable1&variable2">link</a>
I pick up the first variable in my script no problem using the following:
if ($#ARGV >= 0)
{
$variablename1 = $ARGV[0];
}
else
{
whatever;
}
I've tried picking up the 2 variable using variations of this:
if ($#ARGV >= 1)
{
$variablename2 = $ARGV[1];
}
else
{
whatever;
}
I can't seem to pick up that 2 variable. What am I doing wrong?
Thanks again for any help you can offer
Vaughn
------------------------------
Date: 23 Nov 1997 13:34:57 -0500
From: clay@panix.com (Clay Irving)
Subject: Re: Safe use of flock() -- was Re: giving up on flock
Message-Id: <659t0h$7dn@panix.com>
In <EK076B.qL@world.std.com> aml@world.std.com (Andrew M. Langmead) writes:
>Mark Mielke <markm@nortel.ca> writes:
>>Oh yeah... Tom... I just found a section in the manpage that says that:
>> To avoid the possibility of mis-coordination, Perl
>> flushes FILEHANDLE before (un)locking it.
>> (man perlfunc - flock)
>but what if the flush fails?
Drano?
--
Clay Irving <clay@panix.com> http://www.panix.com/~clay/
------------------------------
Date: Thu, 20 Nov 1997 20:55:43 -0600
From: sdavis <sdavis@localIP.net>
Subject: Re: Socket Problems
Message-Id: <3474F82F.AC2D1BDF@localIP.net>
--------------0BC34122D3E3899950ECFCB4
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
mchorgh@ike.com wrote:
> < snip>
> socket(S,AF_INET,SOCK_STREAM,$proto);
> select(S); $|=1; select(STDOUT);
> unless(connect(S, $that_addr))
> {Error("No Connection");}
> $reply = <S>;
> print "$reply";
> while (S){print S "$UID\n";}
> $reply2 = <S>;
> print "$reply";
>
> The connection hangs forever after/at print S. $UID is specified.
> < snip >
>
> Richard <mchorgh@ike.com> Please CC: directly.
>
> -------------------==== Posted via Deja News ====-----------------------
> http://www.dejanews.com/ Search, Read, Post to Usenet
I may be missing something but wouldn't "while(S){...}" be an infinite loop since S being a socket (S != (0 || undef)) will always test out true?
-- Steven F. Davis
-- sdavis@lips.net
-- PGP Public Key: http://www.lips.net/~sdavis/pgpkey.asc
--------------0BC34122D3E3899950ECFCB4
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<HTML>
mchorgh@ike.com wrote:
<BR>
<BLOCKQUOTE TYPE=CITE>< snip></BLOCKQUOTE>
<BLOCKQUOTE TYPE=CITE>
<P> socket(S,AF_INET,SOCK_STREAM,$proto);
<BR> select(S); $|=1; select(STDOUT);
<BR> unless(connect(S, $that_addr))
<BR>
{Error("No Connection");}
<BR> $reply = <S>;
<BR> print "$reply";
<BR> while (S){print S "$UID\n";}
<BR> $reply2 = <S>;
<BR> print "$reply";
<P>The connection hangs forever after/at print S. $UID is specified.
<BR>< snip >
<P>Richard <mchorgh@ike.com> Please CC: directly.
<P>-------------------==== Posted via Deja News ====-----------------------
<BR> <A HREF="http://www.dejanews.com/">http://www.dejanews.com/</A>
Search, Read, Post to Usenet</BLOCKQUOTE>
<PRE></PRE>
<PRE> I may be missing something but wouldn't "while(S){...}" be an infinite loop since S being a socket (S != (0 || undef)) will always test out true?</PRE>
<PRE>-- Steven F. Davis
-- sdavis@lips.net
-- PGP Public Key: <A HREF="http://www.lips.net/~sdavis/pgpkey.asc">http://www.lips.net/~sdavis/pgpkey.asc</A></PRE>
</HTML>
--------------0BC34122D3E3899950ECFCB4--
------------------------------
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 1362
**************************************
