Different results Command Line/CGI

[Owen]

I have this script stolen and modified from somewhere

#!/usr/bin/perl -w
use strict;
use File::Find;
print "Content-type: text/html\n\n";

my $u=shift;
my $sizes = 0;
# replace this with your absolute path
my $path = "/home/$u/";

find (sub {$sizes += -s ;}, $path);
print "$sizes\n";


I am logged onto my machine as user owen and am root. From the command line I execute

perl /var/www/cgi-bin/ff1.cgi owen with the result

--------------------------
Content-type: text/html

(warnings snipped)

504137000
---------------------------

with rcook as the user in /home the result is 988865

Now when I do 'links [url]http://localhost/cgi-bin/ff1.cgi?owen[/url]'

the result is 350200607 (vs 504137000 ???) and
for user rcook the result is 0 (vs 988865)

du gives me yet another set of answers as does other File::Find type programs I have found, but that is not the problem at the moment.

Why do I get 0 for rcook when I run it through the browser?
I suspect the difference is something to do with permissions of the cgi script when run as a cgi script but do not know.



Any advice/clues would be most welcome


TIA

--
Owen

[Ramprasad A Padmanabhan]

Owen wrote:

> I have this script stolen and modified from somewhere
>
> #!/usr/bin/perl -w
> use strict;
> use File::Find;
> print "Content-type: text/html\n\n";
>
> my $u=shift;
> my $sizes = 0;
> # replace this with your absolute path
> my $path = "/home/$u/";
>
> find (sub {$sizes += -s ;}, $path);
> print "$sizes\n";
>
>
> I am logged onto my machine as user owen and am root.

That is percisely why.




From the command line I execute

>
> perl /var/www/cgi-bin/ff1.cgi owen with the result
>
> --------------------------
> Content-type: text/html
>
> (warnings snipped)
>
> 504137000
> ---------------------------
>
> with rcook as the user in /home the result is 988865
>
> Now when I do 'links [url]http://localhost/cgi-bin/ff1.cgi?owen[/url]'
>
> the result is 350200607 (vs 504137000 ???) and
> for user rcook the result is 0 (vs 988865)
>
> du gives me yet another set of answers as does other File::Find type programs I have found, but that is not the problem at the moment.
>
> Why do I get 0 for rcook when I run it through the browser?
> I suspect the difference is something to do with permissions of the cgi script when run as a cgi script but do not know.
>
>
>
> Any advice/clues would be most welcome
>
>
> TIA
>

Your web server may not be running as user root and therefore it may not
have permissions to go to all directories.
Hence you would get a zero

Ram

[Dan Muey]

> I have this script stolen and modified from somewhere

Stealling is bad! :)

>
> #!/usr/bin/perl -w
> use strict;
> use File::Find;
> print "Content-type: text/html\n\n";
>
> my $u=shift;
> my $sizes = 0;
> # replace this with your absolute path
> my $path = "/home/$u/";
>
> find (sub {$sizes += -s ;}, $path);
> print "$sizes\n";
>
>
> I am logged onto my machine as user owen and am root. From
> the command line I execute
>
> perl /var/www/cgi-bin/ff1.cgi owen with the result
>
> --------------------------
> Content-type: text/html
>
> (warnings snipped)
>
> 504137000
> ---------------------------
>
> with rcook as the user in /home the result is 988865
>
> Now when I do 'links [url]http://localhost/cgi-bin/ff1.cgi?owen[/url]'
>
> the result is 350200607 (vs 504137000 ???) and
> for user rcook the result is 0 (vs 988865)
>

Permissions Permissions Permissions:

The user the webserver is runnign at obvioulsy doesn't have
permissions to "see" all the files that the user you where
logged in as did. And it is not allowed to see the rcook directory at
all aparently.

HTH

Dmuey

[Vishal Vasan]

Hi All,

I have a file with the following contents (temp.txt).
A1110
G1115
B1110
C1111
D1111
E1113
F1115
and so on. I have to read the contents from this file and create 2
seperate files. The 1st file contains the lines ending with 0 and 1
while the 2nd file contains the lines ending with 3 and 5. I can do this
easily by creating a seperate perl script. But I have a constraint. I
should not write a seperate file to do this processing.

I am calling the perl from a shell script. Is there any way of calling
the Perl program "in-line" with all the required parameters? "In-line"
means no seperate file containing the script.

Is it worth doing this? Is there a better way of doing it?

Regards,
Vishal Vasan.

[James Edward Gray II]

On Feb 13, 2004, at 6:24 AM, Vishal Vasan wrote:

> Hi All,
>
> I have a file with the following contents (temp.txt).
> A1110
> G1115
> B1110
> C1111
> D1111
> E1113
> F1115
> and so on. I have to read the contents from this file and create 2
> seperate files. The 1st file contains the lines ending with 0 and 1
> while the 2nd file contains the lines ending with 3 and 5. I can do
> this easily by creating a seperate perl script. But I have a
> constraint. I should not write a seperate file to do this processing.
>
> I am calling the perl from a shell script. Is there any way of calling
> the Perl program "in-line" with all the required parameters? "In-line"
> means no seperate file containing the script.
>
> Is it worth doing this? Is there a better way of doing it?

Is there a reason not to just use a Perl script, instead of a shell
script? We can definitely do all the work in one Perl script.

#!/usr/bin/perl

use strict;
use warnings;

die "Usage: perl script_name INPUT_FILE_NAME\n" unless @ARGV;
my $file = shift;
open INPUT, '<', $file or die "File error: $!";
open SMALLOUT, '>', "01$file" or die "File error: $!";
open BIGOUT, '>', "2+$file" or die "File error: $!";

while (<INPUT>) {
chomp;
if (/[01]\s*$/) { print SMALLOUT "$_\n"; }
else { print BIGOUT "$_\n"; }
}

close INPUT;
close SMALLOUT;
close BIGOUT;

__END__

Hope that helps.

James

[Benjamin Walkenhorst]

On Fri, 13 Feb 2004 17:54:54 +0530
Vishal Vasan <vishal.v@net-kraft.com> wrote:

> Hi All,
>
> I have a file with the following contents (temp.txt).
> A1110
> G1115
> B1110
> C1111
> D1111
> E1113
> F1115
> and so on. I have to read the contents from this file and create 2
> seperate files. The 1st file contains the lines ending with 0 and 1
> while the 2nd file contains the lines ending with 3 and 5. I can do
> this easily by creating a seperate perl script. But I have a
> constraint. I should not write a seperate file to do this processing.
>
> I am calling the perl from a shell script. Is there any way of calling
>
> the Perl program "in-line" with all the required parameters? "In-line"
>
> means no seperate file containing the script.
>
> Is it worth doing this? Is there a better way of doing it?

You can do all the stuff in one perl-script. You can just put it on top
of the data:

#!/path/to/perl -w

# Your perl-code goes here

DATA

A1110
G1115
B1110
C1111
D1111
E1113
F1115
....

--------------------------

Then you can treat the file's content after the line

DATA

as a file by using the filehandle <DATA> to read from it.


Kind regards,

Benjamin

[John McKown]

On Fri, 13 Feb 2004, Vishal Vasan wrote:

> Hi All,
>
> I have a file with the following contents (temp.txt).
> A1110
> G1115
> B1110
> C1111
> D1111
> E1113
> F1115
> and so on. I have to read the contents from this file and create 2
> seperate files. The 1st file contains the lines ending with 0 and 1
> while the 2nd file contains the lines ending with 3 and 5. I can do this
> easily by creating a seperate perl script. But I have a constraint. I
> should not write a seperate file to do this processing.
>
> I am calling the perl from a shell script. Is there any way of calling
> the Perl program "in-line" with all the required parameters? "In-line"
> means no seperate file containing the script.
>
> Is it worth doing this? Is there a better way of doing it?
>
> Regards,
> Vishal Vasan.

This may be hearsy here on the Perl list, but is Perl the best choice in
this case? If I had to imbed this in a shell script, I'd use grep twice.

grep '[01]$' input.file >output.file.01
grep '[35]$' input.file >output.file.35

--
Maranatha!
John McKown

[Tim]

The "deep" opinions are interspersed with your original message. An in-line
(pardon the "win32-ish-ness" quote escaping) could take the form:

perl -ne "BEGIN{open S, '>smf';open L, '>lgf'} chomp;if (/[01]\s*$/){print
S \"$_\n\";}else{print L \"$_\n\";}" f1.txt

where your data is in f1.txt, your "0" or "1" ending lines go into "smf",
and the others go into "lgf"

At 05:54 PM 2/13/04 +0530, you wrote:

>Hi All,
>
>I have a file with the following contents (temp.txt).
>A1110
>G1115
>B1110
>C1111
>D1111
>E1113
>F1115
>and so on. I have to read the contents from this file and create 2
>seperate files. The 1st file contains the lines ending with 0 and 1 while
>the 2nd file contains the lines ending with 3 and 5. I can do this easily
>by creating a seperate perl script. But I have a constraint. I should not
>write a seperate file to do this processing.
>
>I am calling the perl from a shell script. Is there any way of calling the
>Perl program "in-line" with all the required parameters? "In-line" means
>no seperate file containing the script.
>
>Is it worth doing this?

Only your boss and the deadline knows for sure

>Is there a better way of doing it?

As always, yes. That's the cool part about this language, but don't take
more than 3 minutes... ;-)

>Regards,
>Vishal Vasan.
>
>--
>To unsubscribe, e-mail: [email]beginners-unsubscribe@perl.org[/email]
>For additional commands, e-mail: [email]beginners-help@perl.org[/email]
><http://learn.perl.org/> <http://learn.perl.org/first-response>
>