[17433] in Perl-Users-Digest
Perl-Users Digest, Issue: 4853 Volume: 9
daemon@ATHENA.MIT.EDU (Perl-Users Digest)
Thu Nov 9 06:10:27 2000
Date: Thu, 9 Nov 2000 03:10:12 -0800 (PST)
From: Perl-Users Digest <Perl-Users-Request@ruby.OCE.ORST.EDU>
To: Perl-Users@ruby.OCE.ORST.EDU (Perl-Users Digest)
Message-Id: <973768212-v9-i4853@ruby.oce.orst.edu>
Content-Type: text
Perl-Users Digest Thu, 9 Nov 2000 Volume: 9 Number: 4853
Today's topics:
reconstituting hash from reference? (Chris)
Re: reconstituting hash from reference? (Martien Verbruggen)
Re: reconstituting hash from reference? (Chris)
Re: reconstituting hash from reference? <ianb@ot.com.au>
Re: reconstituting hash from reference? (Martien Verbruggen)
recursive directory list? <sidt@home.com>
Re: recursive directory list? (Martien Verbruggen)
Re: recursive directory list? (Martien Verbruggen)
Re: Redirect a user by country ? <emile@telegraafnet.nl>
Re: Rename file in Perl <bart.lateur@skynet.be>
simple question - help!! <athe@gte.net>
Re: Sockets? <eli@there-is-no-more-qzto.com>
Re: Sockets? <mahnke@fliegen-ist-schoener.de>
udp server - high volume - log switch fw58959@hotmail.com
Re: udp server - high volume - log switch (Martien Verbruggen)
Re: udp server - high volume - log switch (Chris Fedde)
Re: udp server - high volume - log switch (Garry Williams)
Digest Administrivia (Last modified: 16 Sep 99) (Perl-Users-Digest Admin)
----------------------------------------------------------------------
Date: Thu, 09 Nov 2000 05:51:45 GMT
From: cpr@intergate.ca (Chris)
Subject: reconstituting hash from reference?
Message-Id: <3a0a2eb4.42161895@news.intergate.ca>
I have a deep nest of subroutines. At various calling depths I need to
use and modify some data from the top level. Rather than create globals
or a bloated (and error-prone) list of arguments for each subroutine, I
decided to create a hash in the top-level element and pass it down. Now I
want to create a hash data element and pass a reference to it via the same
mechanism.
I have no problem passing the new hash inside the old using:
$data{"stuff"} = \%stuff;
and can access the data using
${$data{"stuff"}}{key} = value;
but this is an awkward notation to use all the time. I can extract the
reference to a scalar using:
my $stuff_ref = $data{"stuff"};
and then access the data using
${$stuff_ref}{key} = value;
or
$$stuff_ref{key} = value;
but that's only marginally better.
I thought I might get away with
%hash = $hashref;
but that doesn't work.
I read the perllol and perldsc pages but can't see any way to recreate the
original data as a "normal" hash....
What am I missing?
------------------------------
Date: Thu, 09 Nov 2000 05:55:37 GMT
From: mgjv@tradingpost.com.au (Martien Verbruggen)
Subject: Re: reconstituting hash from reference?
Message-Id: <slrn90kf2s.7ck.mgjv@verbruggen.comdyn.com.au>
On Thu, 09 Nov 2000 05:51:45 GMT,
Chris <cpr@intergate.ca> wrote:
>
> I have a deep nest of subroutines. At various calling depths I need to
> use and modify some data from the top level. Rather than create globals
> or a bloated (and error-prone) list of arguments for each subroutine, I
> decided to create a hash in the top-level element and pass it down. Now I
> want to create a hash data element and pass a reference to it via the same
> mechanism.
>
> I have no problem passing the new hash inside the old using:
> $data{"stuff"} = \%stuff;
> and can access the data using
> ${$data{"stuff"}}{key} = value;
Does
$data{stuff}->{key} = $value;
look better? Or even
$data{stuff}{key} = $value;
? Perl is pretty smart in working out what you want.
Martien
--
Martien Verbruggen |
Interactive Media Division | Can't say that it is, 'cause it
Commercial Dynamics Pty. Ltd. | ain't.
NSW, Australia |
------------------------------
Date: Thu, 09 Nov 2000 06:48:06 GMT
From: cpr@intergate.ca (Chris)
Subject: Re: reconstituting hash from reference?
Message-Id: <3a0b4051.46671490@news.intergate.ca>
On Thu, 09 Nov 2000 05:55:37 GMT, mgjv@tradingpost.com.au (Martien
Verbruggen) wrote in comp.lang.perl.misc:
>> I have no problem passing the new hash inside the old using:
>> $data{"stuff"} = \%stuff;
>> and can access the data using
>> ${$data{"stuff"}}{key} = value;
>
>Does
>
>$data{stuff}->{key} = $value;
>
>look better? Or even
>
>$data{stuff}{key} = $value;
>
>? Perl is pretty smart in working out what you want.
The problem I am having is that I can't extract keys from the hash using
keys($$data{stuff})
as perl complains:
Type of arg 1 to keys must be hash (not scalar deref)....
I can't find any way to recreate the original hash in a way that would
satisfy keys() other than to copy the hash into new memory. If I do that
then I can't change the data as seen by the calling code.
------------------------------
Date: Thu, 09 Nov 2000 17:34:01 +1100
From: Ian Boreham <ianb@ot.com.au>
Subject: Re: reconstituting hash from reference?
Message-Id: <3A0A4559.E02D9018@ot.com.au>
Chris wrote:
> The problem I am having is that I can't extract keys from the hash using
> keys($$data{stuff})
> as perl complains:
> Type of arg 1 to keys must be hash (not scalar deref)....
>
> I can't find any way to recreate the original hash in a way that would
> satisfy keys() other than to copy the hash into new memory. If I do that
> then I can't change the data as seen by the calling code.
You can either dereference as a hash using %{} and calling keys() on that,
or you can use typeglob aliasing to make a "normal" hash out of your
long-winded reference. See the following code for examples of each:
#!/usr/bin/perl -w
%original = (a => "A", b => "B", c => {d => "D"});
$hashref = \%original;
print "KEYS: ", keys(%{$hashref->{'c'}}), "\n";
*newouter = $hashref;
print "NEW: $newouter{'b'}\n";
*newinner = $hashref->{'c'};
print "NEW INNER: $newinner{'d'}\n";
Regards,
Ian
------------------------------
Date: Thu, 9 Nov 2000 21:47:15 +1100
From: mgjv@tradingpost.com.au (Martien Verbruggen)
Subject: Re: reconstituting hash from reference?
Message-Id: <slrn90l05j.22p.mgjv@martien.heliotrope.home>
On Thu, 09 Nov 2000 06:48:06 GMT,
Chris <cpr@intergate.ca> wrote:
[new, only vaguely related question, old quoted material removed]
> The problem I am having is that I can't extract keys from the hash
> using keys($$data{stuff}) as perl complains: Type of arg 1 to keys
> must be hash (not scalar deref)....
Ah, that's what you mean by 'reconstituting'. Ok. Whenever you have a
reference, you can 'cast' it back to its original type by putting one of
those thingies in front of it that correspond to that type. So, for an
array you use a @, for a hash a % and for a scalar a $.
my $scalar_ref = \3;
print $$scalar_ref;
my $array_ref = [1, 2, 3];
@array = @$array_ref;
my $hash_ref = {foo => 1, bar => 2};
%hash = %$hash_ref;
As long as the reference is a simple identifier, that works nicely.
However, when it becomes anything more complex, like in your case an
array element, you need to disambiguate the construct. What I mean is
that
$hash{foo} = [1, 2, 3]; # store array ref in $hash{foo}
@array = @$hash{foo};
won't work, because Perl binds the @ tighter than the {foo}, which means
that you end up with @$hash, and $hash is not an array reference (in
fact, it doesn't exist).
The way to solve abiguities in Perl for references, is by grouping the
bit that is the reference together with curlies:
@array = @{$hash{foo}};
This is in fact more than juts disambiguation, because the block
following the @ is a real block, that can contain any expression, as
long as it returns an array reference. This leads to things like:
print "I can interpolate funciton calls: @{[ length $string ]}";
@array = @{ sub_returning_array_ref($arg1 $arg2) };
and other things.
The same goes for hash references, with a %{} block.
So, after this digression (and who the hell knows why I decided to blab
on about all this):
%hash = %{$data{stuff}};
@keys = keys %{$data{stuff}};
The perlref documentaiton contains more information, and so do perldsc
and perllol.
Martien
--
Martien Verbruggen |
Interactive Media Division | If it isn't broken, it doesn't have
Commercial Dynamics Pty. Ltd. | enough features yet.
NSW, Australia |
------------------------------
Date: Thu, 09 Nov 2000 05:29:23 GMT
From: "Sidney Tupper" <sidt@home.com>
Subject: recursive directory list?
Message-Id: <TCqO5.594862$8u4.8468732@news1.rdc1.bc.home.com>
I'd like to get a list of pathnames from WIN98 into an array.
@files = system "dir /s/b/a-d $ARGV[0]"; doesn't work (why not?).
thanks
------------------------------
Date: Thu, 09 Nov 2000 05:37:43 GMT
From: mgjv@tradingpost.com.au (Martien Verbruggen)
Subject: Re: recursive directory list?
Message-Id: <slrn90ke1a.7ck.mgjv@verbruggen.comdyn.com.au>
On Thu, 09 Nov 2000 05:29:23 GMT,
Sidney Tupper <sidt@home.com> wrote:
> I'd like to get a list of pathnames from WIN98 into an array.
> @files = system "dir /s/b/a-d $ARGV[0]"; doesn't work (why not?).
From the documentation on system():
This is not what you
want to use to capture the output from a command,
for that you should use merely backticks or
`qx//', as described in the section on "`STRING`"
in the perlop manpage.
But I think you really want to use glob(), or opendir() and readdir().
All of them are described in the perlfunc documentation.
Not knowing that those options to dir do, I suspect that you want:
my @files = glob("$ARGV[0]/*");
(The file names will include the path)
or something like that. I'd actually prefer:
opendir(DIR, $ARGV[0]) or die "Cannot opendir '$ARGV[0]': $!";
my @files = readdir DIR;
closedir DIR;
(The file name do not include the path)
Martien
--
Martien Verbruggen |
Interactive Media Division |
Commercial Dynamics Pty. Ltd. | What's another word for Thesaurus?
NSW, Australia |
------------------------------
Date: Thu, 09 Nov 2000 05:51:44 GMT
From: mgjv@tradingpost.com.au (Martien Verbruggen)
Subject: Re: recursive directory list?
Message-Id: <slrn90kerj.7ck.mgjv@verbruggen.comdyn.com.au>
On Thu, 09 Nov 2000 05:37:43 GMT,
Martien Verbruggen <mgjv@tradingpost.com.au> wrote:
> On Thu, 09 Nov 2000 05:29:23 GMT,
> Sidney Tupper <sidt@home.com> wrote:
>> I'd like to get a list of pathnames from WIN98 into an array.
>> @files = system "dir /s/b/a-d $ARGV[0]"; doesn't work (why not?).
>
> From the documentation on system():
>
> This is not what you
> want to use to capture the output from a command,
> for that you should use merely backticks or
> `qx//', as described in the section on "`STRING`"
> in the perlop manpage.
Whoops. I missed the word 'recursive' in the subject the first time
around. It can't really hurt to repeat that in the body of the text as
well, you know?
Anyway, the above still stands. If you want an internal solution, have
a look at the File::Find module.
Martien
--
Martien Verbruggen |
Interactive Media Division | Never hire a poor lawyer. Never buy
Commercial Dynamics Pty. Ltd. | from a rich salesperson.
NSW, Australia |
------------------------------
Date: Thu, 9 Nov 2000 07:01:53 +0100
From: "Emile van Gelderen" <emile@telegraafnet.nl>
Subject: Re: Redirect a user by country ?
Message-Id: <8uderl$ur3$1@enterprise.cistron.net>
"Erik" <koeleman@cistron.nl> schreef in bericht
news:8ucdb3$d20$1@voyager.cistron.net...
> Hello,
>
> I'm looking for a perl script that redirect a user by country when they
> arive on my site...
> Can someone help me with that?
>
> Erik
>
>
You can use $ENV{'HTTP_ACCEPT_LANGUAGE'} and $ENV{'REMOTE_HOST'} together.
------------------------------
Date: Thu, 09 Nov 2000 08:19:31 GMT
From: Bart Lateur <bart.lateur@skynet.be>
Subject: Re: Rename file in Perl
Message-Id: <ccnk0to17nsmhft161rgnk2e2f46au7jnl@4ax.com>
Bob Walton wrote:
>Hmmmm...you wouldn't happen to be renaming across filesystems, would
>you? The docs (perldoc -f rename) state that rename probably can't do
>that on most systems, even if mv can. You could copy and delete.
use File::Copy;
move $origFile, $newFile;
That will optimize to a rename() if both source and destination are on
the same filesystem.
File::Copy ispart ofthe standard distribution.
--
Bart.
------------------------------
Date: Thu, 09 Nov 2000 08:52:19 GMT
From: "Andrew" <athe@gte.net>
Subject: simple question - help!!
Message-Id: <7BtO5.3765$Hr4.477814@dfiatx1-snr1.gtei.net>
does any one know the keyboard function to type the cursor box
i think it is a cursor box.
begin 666 cursor.txt
K#0H-"@T*#0H-"@T*#0HB"B(@(" @(@HB(" B"B(@(@HB("(*(B @#0H-"@``
`
end
------------------------------
Date: 9 Nov 2000 05:44:14 GMT
From: Eli the Bearded <eli@there-is-no-more-qzto.com>
Subject: Re: Sockets?
Message-Id: <eli$0011090031@qz.little-neck.ny.us>
In comp.lang.perl.misc, <joekind@my-deja.com> wrote:
> Hi. I am learning how to use sockets and I'm doing some
> experimenting.
Sounds good.
> With the program below, I'm trying to create a new
> socket but it seems like it just hangs for ever because it wouldn't
> print anything. So what am I doing wrong because I copied this example
> from the perl cookbook.
I've not looked at the perl cookbook, but I've found the perlipc
page useful for basic socket scripts.
> use IO::Socket;
> $remote_port = "27688";
> $socket = IO::Socket::INET->new(LocalPort => $remote_port,
> Type => 'SOCK_STREAM',
> Reuse => 1,
> Listen => SOMAXCONN)
> or die "couldnt connect to: $@\n";
> while ($client = $socket->accept) {
> print "test";
> }
> close($socket);
Um, well, is there anything at the other end connecting to the
socket? It looks to me like you waiting for a connection.
> Also, once I do get this socket thing working:
> I want to be able to retrieve the content from 10 different sites all
> at once. So would I just run the above program in a loop and use
> LWP::UserAgent to retrieve the content? And will this be able to
> retrieve the content from each site at the same time rather than having
> to wait for one site to finish before moving on?
LWP::UserAgent is going to deal with sockets on it's own. You don't
need to dirty you hands.
If you really want to read from multiple sites "at once", you are
going to have to get your hands rather dirty with socket stuff,
I suspect. Basically you open a bunch of connections, then use
the four argument form of select to wait for databack from each
of them. Not really beginner's stuff, but wizardry either.
Elijah
------
has avoided the need for select for several years
------------------------------
Date: Thu, 9 Nov 2000 11:01:12 -0000
From: "Christian Mahnke" <mahnke@fliegen-ist-schoener.de>
Subject: Re: Sockets?
Message-Id: <8udsll$mtq$04$1@news.t-online.com>
You also could use Parallel::UserAgent, if you don't care about performance.
Otherwise you have to use the select call together with sockets. There is an
example of it in the Perl Cookbook as well.
mit freundlichen Gr廻en / With kind regards
Christian Mahnke
같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같
Fliegen-ist-Schoener.de
http://www.fliegen-ist-schoener.de
mahnke@fliegen-ist-schoener.de
Tel: 0179-5304585
같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같
in-House Partner:
http://www.billiger-telefonieren.de/ ~ http://www.stromseite.de/
http://www.billiger-surfen.de/ ~ http://www.tvinfo.de/
http://www.billiger-versichern.de/ ~ http://www.booklooker.de/
http://www.sms.de/ ~ http://www.movil24.com/
같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같같
"Eli the Bearded" <eli@there-is-no-more-qzto.com> schrieb im Newsbeitrag
news:eli$0011090031@qz.little-neck.ny.us...
> In comp.lang.perl.misc, <joekind@my-deja.com> wrote:
> > Hi. I am learning how to use sockets and I'm doing some
> > experimenting.
>
> Sounds good.
>
> > With the program below, I'm trying to create a new
> > socket but it seems like it just hangs for ever because it wouldn't
> > print anything. So what am I doing wrong because I copied this example
> > from the perl cookbook.
>
> I've not looked at the perl cookbook, but I've found the perlipc
> page useful for basic socket scripts.
>
> > use IO::Socket;
> > $remote_port = "27688";
> > $socket = IO::Socket::INET->new(LocalPort => $remote_port,
> > Type => 'SOCK_STREAM',
> > Reuse => 1,
> > Listen => SOMAXCONN)
> > or die "couldnt connect to: $@\n";
> > while ($client = $socket->accept) {
> > print "test";
> > }
> > close($socket);
>
> Um, well, is there anything at the other end connecting to the
> socket? It looks to me like you waiting for a connection.
>
> > Also, once I do get this socket thing working:
> > I want to be able to retrieve the content from 10 different sites all
> > at once. So would I just run the above program in a loop and use
> > LWP::UserAgent to retrieve the content? And will this be able to
> > retrieve the content from each site at the same time rather than having
> > to wait for one site to finish before moving on?
>
> LWP::UserAgent is going to deal with sockets on it's own. You don't
> need to dirty you hands.
>
> If you really want to read from multiple sites "at once", you are
> going to have to get your hands rather dirty with socket stuff,
> I suspect. Basically you open a bunch of connections, then use
> the four argument form of select to wait for databack from each
> of them. Not really beginner's stuff, but wizardry either.
>
> Elijah
> ------
> has avoided the need for select for several years
------------------------------
Date: Thu, 09 Nov 2000 05:08:27 GMT
From: fw58959@hotmail.com
Subject: udp server - high volume - log switch
Message-Id: <8udbg8$12d$1@nnrp1.deja.com>
I have a udp server which will does nothing more than log the data sent
to it in a file as per the following snippet of code.....
while ($server->recv($datagram, $MAX_TO_READ)) {
print FLOWLOG "$datagram\n";
}
I would like the most efficient method of switching to a new log file.
Because this is udp and because it will be handling a high volume of
messages I need this to be fast. I tried opening FLOWLOG as append and
then mv FLOWLOG FLOWLOG.old but it just kept writing to FLOWLOG.old - a
new FLOWLOG file was not automatically created. (Makes sense I suppose
- I am not a wizard when it comes to Unix). Are there any perl or Unix
(Solaris) tricks to do this? Your comments on this are appreciated.
Thanks in advance.
./CK
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
Date: Thu, 09 Nov 2000 05:39:52 GMT
From: mgjv@tradingpost.com.au (Martien Verbruggen)
Subject: Re: udp server - high volume - log switch
Message-Id: <slrn90ke5b.7ck.mgjv@verbruggen.comdyn.com.au>
On Thu, 09 Nov 2000 05:08:27 GMT,
fw58959@hotmail.com <fw58959@hotmail.com> wrote:
>
> I would like the most efficient method of switching to a new log file.
> Because this is udp and because it will be handling a high volume of
> messages I need this to be fast. I tried opening FLOWLOG as append and
> then mv FLOWLOG FLOWLOG.old but it just kept writing to FLOWLOG.old - a
> new FLOWLOG file was not automatically created. (Makes sense I suppose
Rename the file, then reopen it.
On Unix, open files have nothing to do with the name they had when
they were opened. Once opened, they stay opened, regardless of file
name changes, or even unlink. You can open a file, immediately remove
it, and keep using it until your program exits.
Martien
--
Martien Verbruggen |
Interactive Media Division | The world is complex; sendmail.cf
Commercial Dynamics Pty. Ltd. | reflects this.
NSW, Australia |
------------------------------
Date: Thu, 09 Nov 2000 05:47:04 GMT
From: cfedde@fedde.littleton.co.us (Chris Fedde)
Subject: Re: udp server - high volume - log switch
Message-Id: <sTqO5.259$Bf7.188636672@news.frii.net>
In article <8udbg8$12d$1@nnrp1.deja.com>, <fw58959@hotmail.com> wrote:
>I have a udp server which will does nothing more than log the data sent
>to it in a file as per the following snippet of code.....
>
>while ($server->recv($datagram, $MAX_TO_READ)) {
> print FLOWLOG "$datagram\n";
>}
>
>I would like the most efficient method of switching to a new log file.
>Because this is udp and because it will be handling a high volume of
>messages I need this to be fast. I tried opening FLOWLOG as append and
>then mv FLOWLOG FLOWLOG.old but it just kept writing to FLOWLOG.old - a
>new FLOWLOG file was not automatically created. (Makes sense I suppose
>- I am not a wizard when it comes to Unix). Are there any perl or Unix
>(Solaris) tricks to do this? Your comments on this are appreciated.
>
A common paradigm for Unix daemons is for the program to close the
current logging file and reopen it using the configured name when
it recieves the HUP signal. An external program (like newsyslog
under BSD) moves the logfile to a new name then sends a HUP to the
daemon.
In your case it might look something like this:
#!/usr/bin/perl
use strict;
use warnings;
my $GotHup = 0;
$SIG{HUP} = sub{ $GotHup++ };
sub checkhup {
if ($GotHup) {
close LOG;
open (LOG, ">>log") or die "can't open log: $!";
}
}
open (LOG, ">>log") or die "can't open log: $!";
my $fh = select LOG; $|=1; select $fh;
while (1) {
print LOG gmtime(time())."\n";
checkhup();
sleep rand() * 5;
}
--
This space intentionally left blank
------------------------------
Date: Thu, 09 Nov 2000 09:26:49 GMT
From: garry@ifr.zvolve.net (Garry Williams)
Subject: Re: udp server - high volume - log switch
Message-Id: <t5uO5.1154$FG.62261@eagle.america.net>
On Thu, 09 Nov 2000 05:47:04 GMT, Chris Fedde
<cfedde@fedde.littleton.co.us> wrote:
>In article <8udbg8$12d$1@nnrp1.deja.com>, <fw58959@hotmail.com>
>wrote:
...
>A common paradigm for Unix daemons is for the program to close the
>current logging file and reopen it using the configured name when
>it recieves the HUP signal. An external program (like newsyslog
>under BSD) moves the logfile to a new name then sends a HUP to the
>daemon.
>
>In your case it might look something like this:
>
> #!/usr/bin/perl
>
> use strict;
> use warnings;
>
> my $GotHup = 0;
>
> $SIG{HUP} = sub{ $GotHup++ };
>
> sub checkhup {
> if ($GotHup) {
> close LOG;
> open (LOG, ">>log") or die "can't open log: $!";
> }
> }
sub checkhup {
if ($GotHup) {
close LOG;
open (LOG, ">>log") or die "can't open log: $!";
$GotHup = 0;
}
}
--
Garry Williams
------------------------------
Date: 16 Sep 99 21:33:47 GMT (Last modified)
From: Perl-Users-Request@ruby.oce.orst.edu (Perl-Users-Digest Admin)
Subject: Digest Administrivia (Last modified: 16 Sep 99)
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.
| NOTE: The mail to news gateway, and thus the ability to submit articles
| through this service to the newsgroup, has been removed. I do not have
| time to individually vet each article to make sure that someone isn't
| abusing the service, and I no longer have any desire to waste my time
| dealing with the campus admins when some fool complains to them about an
| article that has come through the gateway instead of complaining
| to the source.
To submit articles to comp.lang.perl.announce, send your article to
clpa@perl.com.
To request back copies (available for a week or so), send your request
to almanac@ruby.oce.orst.edu with the command "send perl-users x.y",
where x is the volume number and y is the issue number.
For other requests pertaining to the digest, send mail to
perl-users-request@ruby.oce.orst.edu. Do not waste your time or mine
sending perl questions to the -request address, I don't have time to
answer them even if I did know the answer.
------------------------------
End of Perl-Users Digest V9 Issue 4853
**************************************
