«

Andrew Palmer wrote:
> Gunnar Hjalmarsson wrote:
>> Joe Smith wrote:
>>> Gunnar Hjalmarsson wrote:
>>>>
>>>> my $count = 0;
>>>> $count ++ for split /\n/, $in{packageID};
>>>
>>> my $count = () = split /\n/,$in{packageID};
>>
>> my %in = ( packageID => "one\ntwo\nthree" );
>>
>> my $count = () = split /\n/, $in{packageID};
>> print "$count\n";
>>
>> $count = 0;
>> $count ++ for split /\n/, $in{packageID};
>> print "$count\n";
>
> my $count = split /\n/, $in{packageID};
> print "$count\n";
>
>> Outputs:
>> 1
>> 3
>
> 3

Hey, did you read the docs or something?

Thanks!

--
Gunnar Hjalmarsson
Email: http://www.gunnar.cc/cgi-bin/contact.pl

Gunnar Hjalmarsson <> wrote in comp.lang.perl.misc:
> Joe Smith wrote:
> > Gunnar Hjalmarsson wrote:
> >>
> >> my $count = 0;
> >> $count ++ for split /\n/, $in{packageID};
> >
> > my $count = () = split /\n/,$in{packageID};
>
> my %in = ( packageID => "one\ntwo\nthree" );
>
> my $count = () = split /\n/, $in{packageID};
> print "$count\n";
>
> $count = 0;
> $count ++ for split /\n/, $in{packageID};
> print "$count\n";
>
> Outputs:
> 1
> 3

That can be fixed:

my $count = () = split /\n/,$in{packageID}, -1;

Anno

Gunnar Hjalmarsson <> wrote in comp.lang.perl.misc:
> Andrew Palmer wrote:
> > Gunnar Hjalmarsson wrote:
> >> Joe Smith wrote:
> >>> Gunnar Hjalmarsson wrote:
> >>>>
> >>>> my $count = 0;
> >>>> $count ++ for split /\n/, $in{packageID};
> >>>
> >>> my $count = () = split /\n/,$in{packageID};
> >>
> >> my %in = ( packageID => "one\ntwo\nthree" );
> >>
> >> my $count = () = split /\n/, $in{packageID};
> >> print "$count\n";
> >>
> >> $count = 0;
> >> $count ++ for split /\n/, $in{packageID};
> >> print "$count\n";
> >
> > my $count = split /\n/, $in{packageID};
> > print "$count\n";
> >
> >> Outputs:
> >> 1
> >> 3
> >
> > 3
>
> Hey, did you read the docs or something?
>
> Thanks!

Have you run it under warnings? "Use of implicit split to @_ is
deprecated...".

This warning can not be suppressed by "no warnings 'deprecated'", nor
by any other warnings category I've tried. So this would end up as
something like

my $count = do { no warnings; split /\n/, $in{packageID} };

which is not very nice.

Facit: split() is far too clever for its own good. Ilya has recently
pointed this out too.

Anno

Anno Siegel wrote:
> Gunnar Hjalmarsson wrote:
>> Joe Smith wrote:
>>> Gunnar Hjalmarsson wrote:
>>>>
>>>> my $count = 0;
>>>> $count ++ for split /\n/, $in{packageID};
>>>
>>> my $count = () = split /\n/,$in{packageID};
>>
>> my %in = ( packageID => "one\ntwo\nthree" );
>>
>> my $count = () = split /\n/, $in{packageID};
>> print "$count\n";
>>
>> $count = 0;
>> $count ++ for split /\n/, $in{packageID};
>> print "$count\n";
>>
>> Outputs:
>> 1
>> 3
>
> That can be fixed:
>
> my $count = () = split /\n/,$in{packageID}, -1;

Is it possible to figure out from the docs that LIMIT makes a
difference in that respect?

Wonder why I suggested the

$count ++ for split /\n/, $in{packageID};

solution?

--
Gunnar Hjalmarsson
Email: http://www.gunnar.cc/cgi-bin/contact.pl

Gunnar Hjalmarsson <> wrote in comp.lang.perl.misc:
> Anno Siegel wrote:
> > Gunnar Hjalmarsson wrote:
> >> Joe Smith wrote:
> >>> Gunnar Hjalmarsson wrote:
> >>>>
> >>>> my $count = 0;
> >>>> $count ++ for split /\n/, $in{packageID};
> >>>
> >>> my $count = () = split /\n/,$in{packageID};
> >>
> >> my %in = ( packageID => "one\ntwo\nthree" );
> >>
> >> my $count = () = split /\n/, $in{packageID};
> >> print "$count\n";
> >>
> >> $count = 0;
> >> $count ++ for split /\n/, $in{packageID};
> >> print "$count\n";
> >>
> >> Outputs:
> >> 1
> >> 3
> >
> > That can be fixed:
> >
> > my $count = () = split /\n/,$in{packageID}, -1;
>
> Is it possible to figure out from the docs that LIMIT makes a
> difference in that respect?

Indirectly. The reason that we see $count = 1 is that split in list
context figures out how many elements are expected and, if there is
a bound, assumes a LIMIT of one more than that. (One more so that
the remainder of the string doesn't cling to the last element.)
Supplying an explicit, but ineffective, LIMIT of -1 stops that from
happening.

Actually, a LIMIT of 0 would be better because it behaves like no LIMIT
with respect to trailing empty fields. If the string is closed with a
line feed, as when it is read from a file, that makes a difference.

Did I mention split() is too clever for its own good?

> Wonder why I suggested the
>
> $count ++ for split /\n/, $in{packageID};
>
> solution?

Not at all.

my $count = @{ [ split /\n/,$in{packageID}]};

could also be used.

Anno

Anno Siegel wrote:
> Gunnar Hjalmarsson wrote:
>> Anno Siegel wrote:
>>>
>>> That can be fixed:
>>>
>>> my $count = () = split /\n/,$in{packageID}, -1;
>>
>> Is it possible to figure out from the docs that LIMIT makes a
>> difference in that respect?
>
> Indirectly. The reason that we see $count = 1 is that split in
> list context figures out how many elements are expected and, if
> there is a bound, assumes a LIMIT of one more than that. (One more
> so that the remainder of the string doesn't cling to the last
> element.) Supplying an explicit, but ineffective, LIMIT of -1 stops
> that from happening.
>
> Actually, a LIMIT of 0 would be better because it behaves like no
> LIMIT with respect to trailing empty fields. If the string is
> closed with a line feed, as when it is read from a file, that makes
> a difference.

Aha, thanks for the explanation.

> Did I mention split() is too clever for its own good?

Yes, you did, and unless you consider 'clever' eq 'complex', I
disagree. Such design is not cleverness to me.

--
Gunnar Hjalmarsson
Email: http://www.gunnar.cc/cgi-bin/contact.pl

"Gunnar Hjalmarsson" <> wrote in message
news:...
> Andrew Palmer wrote:
> > Gunnar Hjalmarsson wrote:
> >> Joe Smith wrote:
> >>> Gunnar Hjalmarsson wrote:
> >>>>
> >>>> my $count = 0;
> >>>> $count ++ for split /\n/, $in{packageID};
> >>>
> >>> my $count = () = split /\n/,$in{packageID};
> >>
> >> my %in = ( packageID => "one\ntwo\nthree" );
> >>
> >> my $count = () = split /\n/, $in{packageID};
> >> print "$count\n";
> >>
> >> $count = 0;
> >> $count ++ for split /\n/, $in{packageID};
> >> print "$count\n";
> >
> > my $count = split /\n/, $in{packageID};
> > print "$count\n";
> >
> >> Outputs:
> >> 1
> >> 3
> >
> > 3
>
> Hey, did you read the docs or something?

I wish I could claim such a high virtue, but obviously not!

From the second paragraph: "Use of split in scalar context is deprecated"


>
> Thanks!
>
> --
> Gunnar Hjalmarsson
> Email: http://www.gunnar.cc/cgi-bin/contact.pl

Tore Aursand <> wrote in newsan.2004.08.13.08.48.02.412737
@aursand.no:

> On Fri, 13 Aug 2004 10:26:56 +0200, Davidd Sargent wrote:
>> What I want is on the results page for it to count the number of
>> subdomains (each domain will be seperated by \n (a newline)).
>
> 1. Read the FORM data by using CGI.pm.
> 2. Get the lines by splitting on "\n"
> 3. Process each line, discard those which doesn't meet your "subdomain
> requirements".
> 4. Present the result for the user.
>
>

textarea comes back with \r as well.

John S wrote:
> Tore Aursand wrote:
>> On Fri, 13 Aug 2004 10:26:56 +0200, Davidd Sargent wrote:
>>
>>> What I want is on the results page for it to count the number of
>>> subdomains (each domain will be seperated by \n (a newline)).
>>
>> 1. Read the FORM data by using CGI.pm.
>> 2. Get the lines by splitting on "\n"
>> 3. Process each line, discard those which doesn't meet your "subdomain
>> requirements".
>> 4. Present the result for the user.
>
> textarea comes back with \r as well.

Only on Windows (of course), and not if you use a library that takes
care of it, such as CGI.pm (as Tore suggested) or cgi-lib.pl.
Otherwise you can binmode STDOUT to prevent the \r pecularity.

--
Gunnar Hjalmarsson
Email: http://www.gunnar.cc/cgi-bin/contact.pl

Gunnar Hjalmarsson <> wrote in news:2o9scnF8db61U1@uni-
berlin.de:

> John S wrote:
>> Tore Aursand wrote:
>>> On Fri, 13 Aug 2004 10:26:56 +0200, Davidd Sargent wrote:
>>>
>>>> What I want is on the results page for it to count the number of
>>>> subdomains (each domain will be seperated by \n (a newline)).
>>>
>>> 1. Read the FORM data by using CGI.pm.
>>> 2. Get the lines by splitting on "\n"
>>> 3. Process each line, discard those which doesn't meet your "subdomain
>>> requirements".
>>> 4. Present the result for the user.
>>
>> textarea comes back with \r as well.
>
> Only on Windows (of course), and not if you use a library that takes
> care of it, such as CGI.pm (as Tore suggested) or cgi-lib.pl.
> Otherwise you can binmode STDOUT to prevent the \r pecularity.
>

When this is run from an Apache server:
----------------------------------------------------
#!/usr/bin/perl -T

use strict;
use warnings;
use CGI;

my $q = new CGI;

my $fe = $q->param( 'friend_email' ); # text area

my $msg;
if ( $fe =~ /\r/ ){
$msg = 'found \r'
}
else{
$msg = '\r not found'
}

print $q -> header,
$q -> start_html( -title=>'test' ),
$q -> p( 'testing text area ),
$q -> p( $fe ),
$q -> p( $msg ),
$q -> end_html;
-----------------------------------------------------------

I get 'found \r'.
Normally it doesn't matter until, as in this case, it is a list of emails
for example. It took me ages to track it down.
Used IE 6 browser on XP.