nms-cgi-devel — Developers discussion for NMS CGI programs

[Nms-cgi-devel] Non-nms scripts

[Dave C. <da...@da...>]

I've had an email from Rich Bowen (DrBacchus from #perl) telling me about
a program he's written called IAMB (IAMB A Message Board). The project site
is at <http://sourceforge.net/projects/iamb/>. He was wondering if we'd like
to incorporate it into nms.

It was originally based on wwwboard, but has since grown far beyond that
(it now uses a mysql backend, for example). For that reason I don't think
it's right for nms as it currently stands, but I was wondering if we should
perhaps have a list of "other recommended CGI programs" somewhere on the 
web site. What do you think? And if you agree, do you kow of any other
programs we could include on the list?

Dave...

-- 

  Drugs are just bad m'kay

[Nms-cgi-devel] Hi

[Dan G. <da...@ca...>]

Hi Guys,

Just joined your team. So if anyone has anything they'd like me to do
just let me know. Sourceforge ID is 'dannyg'

Regards,

Dan.

-- 
Daniel Goscomb
Cashcade Ltd
Tel: 0845 050 2285
Fax: 0845 050 2280
Mobile: 0774 041 1951
e: da...@ca...
w: www.getminted.com
**********************************************************************
CONFIDENTIALITY NOTICE:
This email contains confidential information, which is intended only for
the
recipient(s) named above. If you are not the addressee, please delete
this
message and notify the sender; you should not copy or distribute this
message or disclose its contents to anyone.
Any views or opinions expressed in this message are those of the author
and
do not necessarily represent those of Cashcade Ltd or any of its
affiliates.
No reliance may be placed on this message without written confirmation
from
an authorized representative of its contents.
**********************************************************************

[Nms-cgi-devel] Re: nms scripts

[Dave C. <da...@da...>]

On Wed, Mar 06, 2002 at 05:27:51PM -0800, Lori Wong (wo...@ll...) wrote:
> Dave:
> 
> I think your revision of Matt's Script Archive is a good thing.  I was 
> using Matt's simple search and recently made changes to it to eliminate the 
> cross-site scripting vulnerability.  I came across your nms site and got a 
> copy of your revised simple search script and modified it for my 
> use.  However, I needed to update it to remove the cross-site scripting 
> vulnerability (see http://www.perl.com/pub/a/2002/02/20/css.html).  You 
> might want to update your script accordingly.

Lori,

Thanks for the mail. I was under the impression that we had removed all of
the cross-site scripting problems, but I may be wrong.

I've copied the nms developer's mailing list on this email. They'll be
very interestd in seeing any changes that you had to make.

Cheers,

Dave...

-- 

  "Don't you boys know any _nice_ songs?"

Re: [Nms-cgi-devel] counter.pl

[Jonathan S. <gel...@ge...>]

On Sat, 9 Mar 2002, Sam Smith wrote:

>
> The below diff adds support for multiple virtual hosts on the same
> machine to use the same counter file / director without falling over
> each other's counter files.
>

Oooh bugger, I just committed a very similar change before I read this ...

/J\
-- 
Jonathan Stowe                      |
<http://www.gellyfish.com>          |      This space for rent
                                    |

[Nms-cgi-devel] counter.pl -- updated diff

[Sam S. <sa...@us...>]

This is an updated patch so that $ENV{SERVER_NAME} has higher priority
than $ENV{HTTP_HOST} so that host aliases under Apache do what you
expect.



The below diff adds support for multiple virtual hosts on the same machine to
use the same counter file / director without falling over each other's counter
files.

It is wrapped being an $emulate_matts_scripts and also $virtual_host_support
for those who don't want VHS but want some of the NMS additions which we've
not added yet.

URL of a site where this is running in a test environment available privately
on request.

It's been tested on Apache 1.3.19 running on OpenBSD 3.0 with perl 5.6.1.
I can't see anything that would break under anything else, but it might --
this would most probably be the way of getting the virtual host out of
the server.



Sam



--- old/counter.pl	Tue Feb 26 20:57:01 2002
+++ new/counter.pl	Fri Mar  8 23:57:24 2002
@@ -94,6 +51,12 @@

 my @no_header_servers = qw(Xitami);

+my $emulate_matts_code= 0;
+
+my $virtual_host_support= 1; # means you can run the same counter script on
+                             # multiple virtual hosts without them
+ 			      # updating the same count files.
+
 # End configuration

 # We need finer control over what gets to the browser and the CGI::Carp
@@ -150,7 +113,9 @@

 # Some servers (notably Xitami) do not give $ENV{DOCUMENT_URI}

-my $count_page = $ENV{DOCUMENT_URI} || $ENV{SCRIPT_NAME};
+my $count_page  = $ENV{DOCUMENT_URI} || $ENV{SCRIPT_NAME};
+my $virtual_host= $ENV{SERVER_NAME} || $ENV{HTTP_HOST} || '';
+my $counter_filename;

 check_server_software();

@@ -169,8 +134,16 @@
   $data_dir .= '/';
 }

-if (-e "$data_dir$count_page") {
-   sysopen(COUNT, "$data_dir$count_page", O_RDWR)
+
+if ((not $emulate_matts_code) and ($virtual_host_support)) {
+   $counter_filename="${data_dir}_${virtual_host}_$count_page";
+}
+else  {
+    $counter_filename= "$data_dir$count_page";
+}
+
+if (-e "$counter_filename") {
+   sysopen(COUNT, "$counter_filename", O_RDWR)
      or die "Can't open count file: $!\n";
    flock(COUNT, LOCK_EX)
      or die "Can't lock count file: $!\n";
@@ -179,7 +152,7 @@

    ($date, $count) = split(/\|\|/,$line);
 } elsif ($auto_create) {
-   $date = create();
+   $date = &create($counter_filename);
 } else {
    die "Count file not found\n";
 }
@@ -235,12 +208,13 @@
 }

 sub create {
+  my $filename= shift;
   my $date = strftime('%B %d %Y', localtime);

-  sysopen(COUNT, "$data_dir$count_page", O_CREAT|O_RDWR)
-    or die "Can't create count file: $!\n";
+  sysopen(COUNT, "$filename", O_CREAT|O_RDWR)
+    or die "Can't create count file $filename: $!\n";
   flock(COUNT, LOCK_EX)
-    or die "Can't lock count file: $!\n";
+    or die "Can't lock count file $filename: $!\n";

   print COUNT "$date||0";
   return $date;



_______________________________________________
Nms-cgi-devel mailing list
Nms...@li...
https://lists.sourceforge.net/lists/listinfo/nms-cgi-devel

[Nms-cgi-devel] counter.pl

[Sam S. <sa...@us...>]

The below diff adds support for multiple virtual hosts on the same machine to
use the same counter file / director without falling over each other's counter
files.

It is wrapped being an $emulate_matts_scripts and also $virtual_host_support
for those who don't want VHS but want some of the NMS additions which we've
not added yet.

URL of a site where this is running in a test environment available privately
on request.

It's been tested on Apache 1.3.19 running on OpenBSD 3.0 with perl 5.6.1.
I can't see anything that would break under anything else, but it might --
this would most probably be the way of getting the virtual host out of
the server.



Sam



--- old/counter.pl	Tue Feb 26 20:57:01 2002
+++ new/counter.pl	Fri Mar  8 23:57:24 2002
@@ -94,6 +51,12 @@

 my @no_header_servers = qw(Xitami);

+my $emulate_matts_code= 0;
+
+my $virtual_host_support= 1; # means you can run the same counter script on
+                             # multiple virtual hosts without them
+ 			      # updating the same count files.
+
 # End configuration

 # We need finer control over what gets to the browser and the CGI::Carp
@@ -150,7 +113,9 @@

 # Some servers (notably Xitami) do not give $ENV{DOCUMENT_URI}

-my $count_page = $ENV{DOCUMENT_URI} || $ENV{SCRIPT_NAME};
+my $count_page  = $ENV{DOCUMENT_URI} || $ENV{SCRIPT_NAME};
+my $virtual_host= $ENV{HTTP_HOST} || $ENV{SERVER_NAME} || '';
+my $counter_filename;

 check_server_software();

@@ -169,8 +134,16 @@
   $data_dir .= '/';
 }

-if (-e "$data_dir$count_page") {
-   sysopen(COUNT, "$data_dir$count_page", O_RDWR)
+
+if ((not $emulate_matts_code) and ($virtual_host_support)) {
+   $counter_filename="${data_dir}_${virtual_host}_$count_page";
+}
+else  {
+    $counter_filename= "$data_dir$count_page";
+}
+
+if (-e "$counter_filename") {
+   sysopen(COUNT, "$counter_filename", O_RDWR)
      or die "Can't open count file: $!\n";
    flock(COUNT, LOCK_EX)
      or die "Can't lock count file: $!\n";
@@ -179,7 +152,7 @@

    ($date, $count) = split(/\|\|/,$line);
 } elsif ($auto_create) {
-   $date = create();
+   $date = &create($counter_filename);
 } else {
    die "Count file not found\n";
 }
@@ -235,12 +208,13 @@
 }

 sub create {
+  my $filename= shift;
   my $date = strftime('%B %d %Y', localtime);

-  sysopen(COUNT, "$data_dir$count_page", O_CREAT|O_RDWR)
-    or die "Can't create count file: $!\n";
+  sysopen(COUNT, "$filename", O_CREAT|O_RDWR)
+    or die "Can't create count file $filename: $!\n";
   flock(COUNT, LOCK_EX)
-    or die "Can't lock count file: $!\n";
+    or die "Can't lock count file $filename: $!\n";

   print COUNT "$date||0";
   return $date;

[Nms-cgi-devel] Diff for counter.pl

[Sam S. <sa...@us...>]

Move the changelog out of counter.pl and into a ChangeLog file.

And the person who owns the Copyright on counter.pl could do with
putting a licence on it.



Sam


--- old/ChangeLog	Thu Jan  1 00:00:00 1970
+++ new/ChangeLog	Fri Mar  8 23:38:41 2002
@@ -0,0 +1,46 @@
+#
+# $Log: counter.pl,v $
+# Revision 1.11  2002/02/26 20:57:01  gellyfish
+# * Added DOCUMENT_URI to NMSTest::TestRun
+# * Added a default test for textcounter.pl
+# * Made it easier to test textcounter
+#
+# Revision 1.10  2002/02/26 08:59:28  gellyfish
+# * Fixed imagecounter to something
+# * fixed typo in textcounter/README
+# * suppressed emission of headers in fatalsToBrowser if already done.
+#
+# Revision 1.9  2002/02/11 09:16:35  gellyfish
+# * provided method to turn off emission of headers
+# * Fixed the locking race
+# * Turned off uploads and POST
+#
+# Revision 1.8  2002/01/27 14:13:33  davorg
+# Removed Matt's docs.
+# Removed unnecessary reference to lock file.
+#
+# Revision 1.7  2001/12/01 19:45:22  gellyfish
+# * Tested everything with 5.004.04
+# * Replaced the CGI::Carp with local variant
+#
+# Revision 1.6  2001/11/26 13:40:05  nickjc
+# Added \Q \E around variables in regexps where metacharacters in the
+# variables shouldn't be interpreted by the regex engine.
+#
+# Revision 1.5  2001/11/25 11:39:40  gellyfish
+# * add missing use vars qw($DEBUGGING) from most of the files
+# * sundry other compilation failures
+#
+# Revision 1.4  2001/11/13 20:35:14  gellyfish
+# Added the CGI::Carp workaround
+#
+# Revision 1.3  2001/11/13 09:19:24  gellyfish
+# Added CGI::Carp
+#
+# Revision 1.2  2001/11/11 17:55:27  davorg
+# Small amount of post-import tidying :)
+#
+# Revision 1.1.1.1  2001/11/11 16:48:59  davorg
+# Initial import
+#
+
diff -uN old/counter.pl new/counter.pl
--- old/counter.pl	Tue Feb 26 20:57:01 2002
+++ new/counter.pl	Fri Mar  8 23:36:10 2002
@@ -2,50 +2,7 @@
 #
 # $Id: counter.pl,v 1.11 2002/02/26 20:57:01 gellyfish Exp $
 #
-# $Log: counter.pl,v $
-# Revision 1.11  2002/02/26 20:57:01  gellyfish
-# * Added DOCUMENT_URI to NMSTest::TestRun
-# * Added a default test for textcounter.pl
-# * Made it easier to test textcounter
-#
-# Revision 1.10  2002/02/26 08:59:28  gellyfish
-# * Fixed imagecounter to something
-# * fixed typo in textcounter/README
-# * suppressed emission of headers in fatalsToBrowser if already done.
-#
-# Revision 1.9  2002/02/11 09:16:35  gellyfish
-# * provided method to turn off emission of headers
-# * Fixed the locking race
-# * Turned off uploads and POST
-#
-# Revision 1.8  2002/01/27 14:13:33  davorg
-# Removed Matt's docs.
-# Removed unnecessary reference to lock file.
-#
-# Revision 1.7  2001/12/01 19:45:22  gellyfish
-# * Tested everything with 5.004.04
-# * Replaced the CGI::Carp with local variant
-#
-# Revision 1.6  2001/11/26 13:40:05  nickjc
-# Added \Q \E around variables in regexps where metacharacters in the
-# variables shouldn't be interpreted by the regex engine.
-#
-# Revision 1.5  2001/11/25 11:39:40  gellyfish
-# * add missing use vars qw($DEBUGGING) from most of the files
-# * sundry other compilation failures
-#
-# Revision 1.4  2001/11/13 20:35:14  gellyfish
-# Added the CGI::Carp workaround
-#
-# Revision 1.3  2001/11/13 09:19:24  gellyfish
-# Added CGI::Carp
-#
-# Revision 1.2  2001/11/11 17:55:27  davorg
-# Small amount of post-import tidying :)
-#
-# Revision 1.1.1.1  2001/11/11 16:48:59  davorg
-# Initial import
-#
+# For Changes, see the ChangeLog file.

 use strict;
 use CGI qw(header);

[Nms-cgi-devel] FormMail.pl diff

[Sam S. <sa...@us...>]

The below moves the user Config section above all the POD. Having it below
is worse than having all the ChangeLog stuff there since we
actually expect people to read the POD (well, we probably should) by
a means other than their favourite text editor.


Also, whoever checked in the removal of all the Changes lines
should probably also checkin the CHANGELOG file which isn't there
in CVS but is referenced from FormMail.pl .


Sam

Index: FormMail.pl
===================================================================
RCS file: /cvsroot/nms-cgi/formmail/FormMail.pl,v
retrieving revision 1.50
diff -u -r1.50 FormMail.pl
--- FormMail.pl	7 Mar 2002 23:00:05 -0000	1.50
+++ FormMail.pl	8 Mar 2002 23:27:04 -0000
@@ -24,6 +24,37 @@
 # For help on configuration or installation see README
 #

+
+# USER CONFIGURATION SECTION
+# --------------------------
+# Modify these to your own settings. You might have to
+# contact your system administrator if you do not run
+# your own web server. If the purpose of these
+# parameters seems unclear, please see the README file.
+#
+BEGIN { $DEBUGGING    = 1; }
+my $emulate_matts_code= 0;
+my $secure            = 1;
+my $mailprog          = '/usr/lib/sendmail -oi -t';
+my @referers          = qw(dave.org.uk 209.207.222.64 localhost);
+my @allow_mail_to     = qw(yo...@yo... som...@yo... localhost);
+my @recipients        = ();
+my @valid_ENV         = qw(REMOTE_HOST REMOTE_ADDR REMOTE_USER HTTP_USER_AGENT);
+my $date_fmt          = '%A, %B %d, %Y at %H:%M:%S';
+my $style             = '/css/nms.css';
+my $send_confirmation_mail = 0;
+my $confirmation_text = <<'END_OF_CONFIRMATION';
+From: yo...@yo...
+Subject: form submission
+
+Thank you for your form submission.
+
+END_OF_CONFIRMATION
+#
+# USER CONFIGURATION << END >>
+# ----------------------------
+# (no user serviceable parts beyond here)
+
 =head1 COPYRIGHT

 FormMail Version 1.00
@@ -447,37 +478,6 @@
 nms...@li...

 =cut
-
-
-# USER CONFIGURATION SECTION
-# --------------------------
-# Modify these to your own settings. You might have to
-# contact your system administrator if you do not run
-# your own web server. If the purpose of these
-# parameters seems unclear, please see the README file.
-#
-BEGIN { $DEBUGGING    = 1; }
-my $emulate_matts_code= 0;
-my $secure            = 1;
-my $mailprog          = '/usr/lib/sendmail -oi -t';
-my @referers          = qw(dave.org.uk 209.207.222.64 localhost);
-my @allow_mail_to     = qw(yo...@yo... som...@yo... localhost);
-my @recipients        = ();
-my @valid_ENV         = qw(REMOTE_HOST REMOTE_ADDR REMOTE_USER HTTP_USER_AGENT);
-my $date_fmt          = '%A, %B %d, %Y at %H:%M:%S';
-my $style             = '/css/nms.css';
-my $send_confirmation_mail = 0;
-my $confirmation_text = <<'END_OF_CONFIRMATION';
-From: yo...@yo...
-Subject: form submission
-
-Thank you for your form submission.
-
-END_OF_CONFIRMATION
-#
-# USER CONFIGURATION << END >>
-# ----------------------------
-# (no user serviceable parts beyond here)


 # We don't need file uploads or very large POST requests.

[Nms-cgi-devel] Re: [Nms-cgi-support] Re: Counter Set-up

[Jonathan S. <gel...@ge...>]

On Fri, 8 Mar 2002, Tim Lee wrote:

> Any idea if its possible to make the script collect the hit information
> from all pages from a site instead of single pages?
>

Yes, it is entirely possible but :

  a)  I don't want to make this change to the code without some more
      discussion among the developers.

  b)  <snip comment about having to do real work for a living>

  c)  OK.


/J\
-- 
Jonathan Stowe                      |
<http://www.gellyfish.com>          |      This space for rent
                                    |

[Nms-cgi-devel] Re: [Nms-cgi-support] FormMail.pl

[Jonathan S. <gel...@ge...>]

On Thu, 7 Mar 2002, Dru Morgan wrote:

> I have replaced Matt's script with the NMS script. I
> have added my domain and my email to the user
> definable parts. I think I have to change my mailprog
> to "Geocel.Mailer" due to the fact that I have an NT
> server. Either changing it or leaving it to the
> default, I get the following error...
>
> Application Error
> An error has occurred in the program
>
> Broken pipe at d:\sites\heavenly\cgi-bin\FormMail.pl
> line

Ah,  The problem here is that Geocel.Mailer is an OLE object and not an
executable , consequently it is not just a matter of changing $mailprog
but rather changing virtually all of the send_mail() subroutine.  You
could in principle replace the send_mail() subroutine in FormMail.pl with
one similar to the examples found on the Geocel web site at
http://www.geocel.com/ .  I wouldn't however recommend replacing the NMS
FormMail with the one they have there as this is based entirely on the MSA
version and still exhibits all of the problems with that script.

I Have Cc'd this to the developer list in case anyone who has a Windows
machine they can test this on wants to do a port of the NMS FormMail and
send it GeoCel :)


/J\
-- 
Jonathan Stowe                      |
<http://www.gellyfish.com>          |      This space for rent
                                    |

Re: [Nms-cgi-devel] Re: [Nms-cgi-support] Re: Counter Set-up

[Jonathan S. <gel...@ge...>]

On Wed, 6 Mar 2002, Sam Smith wrote:

> On Wed, 6 Mar 2002, Jonathan Stowe wrote:
> > It just occured to me that the HTTP_HOST wouldn't work because that will
> > always be the HTTP_HOST of the server that the counter runs on, then I
> > went to test what the environment of a CGI program run from an shtml page
> > on another host would look like and it wouldn't run ....  Upon reading the
> > Apache documentation I discover that for 'exec cgi':
> >        If the path does not begin with a (/), then it is taken to be
> >        relative to the current document.
>
> Unless the files are in a place accessible by all virtual hosts?
> Which, if we only wanted one copy of the script, they would have to be.
>
> Either way, encoding the Hostname in the name of the file is probably a good
> idea as it allows for multiple servers to write their counters to the same directory (and
> only have one dir writable by the webserver rather than more).
>
>

Ah, got ya, *virtual hosts* on the same server ... OK.

/J\
-- 
Jonathan Stowe                      |
<http://www.gellyfish.com>          |      This space for rent
                                    |

Re: [Nms-cgi-devel] Re: [Nms-cgi-support] Re: Counter Set-up

[Sam S. <sa...@us...>]

On Wed, 6 Mar 2002, Jonathan Stowe wrote:
> It just occured to me that the HTTP_HOST wouldn't work because that will
> always be the HTTP_HOST of the server that the counter runs on, then I
> went to test what the environment of a CGI program run from an shtml page
> on another host would look like and it wouldn't run ....  Upon reading the
> Apache documentation I discover that for 'exec cgi':
>        If the path does not begin with a (/), then it is taken to be
>        relative to the current document.

Unless the files are in a place accessible by all virtual hosts?
Which, if we only wanted one copy of the script, they would have to be.

Either way, encoding the Hostname in the name of the file is probably a good
idea as it allows for multiple servers to write their counters to the same directory (and
only have one dir writable by the webserver rather than more).




Sam

-- 
  Conformity: When People are Free to Do as They Please, They Usually
    Imitate Each Other

Re: [Nms-cgi-devel] Re: [Nms-cgi-support] Re: Counter Set-up

[Jonathan S. <gel...@ge...>]

On Wed, 6 Mar 2002, Sam Smith wrote:
>
> It would also be a nice feature to be able to use the same script  for
> multiple hosts by including the HTTP_HOST value somewhere in the
> encoding.
>

It just occured to me that the HTTP_HOST wouldn't work because that will
always be the HTTP_HOST of the server that the counter runs on, then I
went to test what the environment of a CGI program run from an shtml page
on another host would look like and it wouldn't run ....  Upon reading the
Apache documentation I discover that for 'exec cgi':

       If the path does not begin with a (/), then it is taken to be
       relative to the current document.

so that one isn't going to float.   And for the alternative 'include
virtual':

       The URL cannot contain a scheme or hostname, only a path and an
       optional query string. If it does not begin with a slash (/) then
       it is taken to be relative to the current document.

So neither will that.  Of course this could be entirely down to the Apache
SSI implementation and this might work on a whole load of servers that I
don't have access to test right now, or there might be an alternative SSI
implementation for Apache that does behave like this or .. or ..


Then I went down a blind alley and started to type :

    Anyhow in principle you could do what you suggest now if your server
    *does* support the inclusion of remote CGI programs but otherwise
    behaves similarly to Apache (of course this might work for all servers
    but again I have no way of testing ) by using extra path information
    i.e you do :


And then tested what I had been thinking and discovered it was totally
crack induced and untrue ...

Based on these observations I don't actually believe a remote counter
that uses SSI is going to work, of course I would be delighted if someone
could prove me wrong ...

/J\
-- 
Jonathan Stowe                      |
<http://www.gellyfish.com>          |      This space for rent
                                    |

Re: [Nms-cgi-devel] Re: [Nms-cgi-support] Re: Counter Set-up

[Jonathan S. <gel...@ge...>]

On Wed, 6 Mar 2002, Sam Smith wrote:

> On Wed, 6 Mar 2002, Jonathan Stowe wrote:
> > On Wed, 6 Mar 2002, Tim Lee wrote:
> > > Is it possible to set the parameters for the counter to reference a
> > > single counter for an entire site rather than individual pages?
> > >
> > > I would like to have a single text counter that records hits to all
> > > pages of a site rather than individual pages.
> > >
> > > Right now the script when executed on any given page creates a new
> > > data script for that page and records the hit. Can the script be
> > > directed to reference a single count script in the cgi-bin/data
> > > directory?
> > >
> >
> > This is certainly possible, I have Cc'd the developers list to
> > determine whether anyone thinks this is a silly idea before I do
> > anything about it :)
> >
> > It almost certainly would involve having a $use_single_counter
> > configuration and a $counter_file - none of this coming into play
> > unless $emulate_matts_code is false.
>
> It would also be a nice feature to be able to use the same script for
> multiple hosts by including the HTTP_HOST value somewhere in the
> encoding.
>
> That way you can have
>    www.foo.wibble.com/index.shtml
>    www.bar.wibble.com/index.shtml
>
> both call the same script and not both incrememnt the same counter.
>
>
> Although I can see the single file format becoming a bottleneck on
> busy servers (who probably should be doing something better anyway..)
>

Yep this looks like a good feature too though I'll have to have a play on
the train :)

/J\
-- 
Jonathan Stowe                      |
<http://www.gellyfish.com>          |      This space for rent
                                    |

Re: [Nms-cgi-devel] Re: [Nms-cgi-support] Re: Counter Set-up

[Sam S. <sa...@us...>]

On Wed, 6 Mar 2002, Jonathan Stowe wrote:
> On Wed, 6 Mar 2002, Tim Lee wrote:
> > Is it possible to set the parameters for the counter to reference a single
> > counter for an entire site rather than individual pages?
> >
> > I would like to have a single text counter that records hits to all pages of a
> > site rather than individual pages.
> >
> > Right now the script when executed on any given page creates a new data script
> > for that page and records the hit. Can the script be directed to reference a
> > single count script in the cgi-bin/data directory?
> >
>
> This is certainly possible, I have Cc'd the developers list to determine
> whether anyone thinks this is a silly idea before I do anything about it
> :)
>
> It almost certainly would involve having a $use_single_counter
> configuration and a $counter_file - none of this coming into play unless
> $emulate_matts_code is false.

It would also be a nice feature to be able to use the same script  for
multiple hosts by including the HTTP_HOST value somewhere in the
encoding.

That way you can have
   www.foo.wibble.com/index.shtml
   www.bar.wibble.com/index.shtml

both call the same script and not both incrememnt the same counter.


Although I can see the single file format becoming a bottleneck on
busy servers (who probably should be doing something better anyway..)


Sam

-- 
  Mistakes: It Could be that the Purpose of Your Life is Only to Serve as
    a Warning to Others

[Nms-cgi-devel] Re: [Nms-cgi-support] Re: Counter Set-up

[Jonathan S. <gel...@ge...>]

On Wed, 6 Mar 2002, Tim Lee wrote:

> Is it possible to set the parameters for the counter to reference a single
> counter for an entire site rather than individual pages?
>
> I would like to have a single text counter that records hits to all pages of a
> site rather than individual pages.
>
> Right now the script when executed on any given page creates a new data script
> for that page and records the hit. Can the script be directed to reference a
> single count script in the cgi-bin/data directory?
>

This is certainly possible, I have Cc'd the developers list to determine
whether anyone thinks this is a silly idea before I do anything about it
:)

It almost certainly would involve having a $use_single_counter
configuration and a $counter_file - none of this coming into play unless
$emulate_matts_code is false.

/J\
-- 
Jonathan Stowe                      |
<http://www.gellyfish.com>          |      This space for rent
                                    |

RE: [Nms-cgi-devel] POD in script files (was: Search on the NMS web site.)

[Paul R. <pa...@ro...>]

OK, to get the ball rolling, I've checked in FormMail.pl with the README
text added as POD.  I haven't checked in the POD-generated README itself,
since (a) the actual content (though not the format) remains the same, and
(b) my guess is we'll want to do this on the fly when building a release.

Also moved the Log messages to the end as I'd suggested before -- sorry,
Iain, but this has been kicked around before, and until there's a consensus
I didn't want to remove the text.

-paul

Re: [Nms-cgi-devel] GIFs in Image Counter

[Jonathan S. <gel...@ge...>]

On Tue, 5 Mar 2002, Nicholas Clark wrote:
>
> And CGI scripts that call GD and stick digits together to make counters
> which only have 1 image, rather than an image per digit. Or does that wait
> until everything else is done? *and* servers have more CPU? And perl6 is
> bug free?
>

Implementing the functionality of GD within the design constraints of NMS
is almost certainly going to have to wait until perl6 is bug free *and*
XML::XSLT is fully compliant to the specification ;-}

/J\
-- 
Jonathan Stowe                      |
<http://www.gellyfish.com>          |      This space for rent
                                    |

Re: [Nms-cgi-devel] GIFs in Image Counter

[Dave C. <da...@da...>]

On Tue, Mar 05, 2002 at 07:50:35PM +0000, Jonathan Stowe (gel...@ge...) wrote:
> I have added an alternate set of PNG digits to Image Counter (well just
> run 'mogrify -format png *.gif' in the digits directory actually :) - I am
> of the opinion that we should remove the GIFs altogether at some point and
> have added a comment in the README that this is likely to happen sometime
> in the future.  Does anyone have any opinions on this matter ?

Hadn't thought of that but, yes, that's a great idea. All GIFs are bad.
Anyone who doesn't know why should read

<http://www.newarchitectmag.com/documents/s=5354/new1013637431/>
 
> While we are on the subject are these images still the same ones from the
> MSA image counter ?  If they are would it be possible for someone who has
> a hand with a graphics program to make a new bunch of 15x27 PNGs that we
> can distribute without upsetting anyone?  Maybe it would be nice to have a
> small library of them on the web site ;-}

Another good point. Using his GIFs is probably another area where Matt could
come after us and hit us with lawyers if he wanted to :)

Dave...

-- 

  Don't dream it... be it

Re: [Nms-cgi-devel] Search on the NMS web site.

[iain t. <ic...@eh...>]

* Paul Roub (pa...@ro...) [06 Mar 2002 09:47]:
> p.s. Might I also suggest that, since the POD would end up in the .pl file
> where we expect it to be obvious and useful, we might wish to move  the CVS
> log sections (# $Log: $...) to the *end* of the files?

Nono, remove the log sections.


cheers,
-- 
iain.                                          <http://eh.org/~koschei/>

Re: [Nms-cgi-devel] GIFs in Image Counter

[Nicholas C. <ni...@un...>]

On Tue, Mar 05, 2002 at 07:50:35PM +0000, Jonathan Stowe wrote:
> I have added an alternate set of PNG digits to Image Counter (well just
> run 'mogrify -format png *.gif' in the digits directory actually :) - I am
> of the opinion that we should remove the GIFs altogether at some point and
> have added a comment in the README that this is likely to happen sometime
> in the future.  Does anyone have any opinions on this matter ?

PNGs are good. GIFs are evil. [you said "opinion"]
Are there any compatibility issues with some browsers still not liking PNGs?
I had heard that some things that were fine with opaque PNGs, but didn't
like transparent PNGs. [not that I've looked at these]
But I heard this probably 2 years ago, so it might not be relevant.
My copy of links supports GIFs, PNGs and Coral Draw! files equally, so I
don't mind :-)

> can distribute without upsetting anyone?  Maybe it would be nice to have a
> small library of them on the web site ;-}

Frugal low colour ones that are damn fast to download,
pretty ones that might take longer.

And CGI scripts that call GD and stick digits together to make counters
which only have 1 image, rather than an image per digit. Or does that wait
until everything else is done? *and* servers have more CPU? And perl6 is
bug free?

Nicholas Clark
-- 
Even better than the real thing:	http://nms-cgi.sourceforge.net/

Re: [Nms-cgi-devel] Search on the NMS web site.

[Nicholas C. <ni...@un...>]

On Tue, Mar 05, 2002 at 08:02:22PM +0000, Jonathan Stowe wrote:
> Alternatively we could convert the READMEs to POD or XML and deploy some
> intermediate program to create both the distributed README and the HTML
> version.

To me this sounds like a better idea, as it should allow more flexibility
in the "richer" formats. [erk. how I hate using that word "rich" now that
Bill Gates and squillions of other marketdroids have soiled it in their
mouths]

Feel free to ignore me on the basis that I'm not volunteering to do this.

Nicholas Clark
-- 
Even better than the real thing:	http://nms-cgi.sourceforge.net/

RE: [Nms-cgi-devel] Search on the NMS web site.

[Paul R. <pa...@ro...>]

p.s. Might I also suggest that, since the POD would end up in the .pl file
where we expect it to be obvious and useful, we might wish to move  the CVS
log sections (# $Log: $...) to the *end* of the files?

The $Id: $ can (and probably should) remain up top for those who care.

-paul


-----Original Message-----
From: nms...@li...
[mailto:nms...@li...]On Behalf Of Paul Roub
Sent: Tuesday, March 05, 2002 4:39 PM
To: NMS Developers List
Subject: RE: [Nms-cgi-devel] Search on the NMS web site.


I'll third the motion...  and strongly advocate that POD is the way to go,
for all the obvious reasons.

-paul


> Alternatively we could convert the READMEs to POD or XML and deploy some
> intermediate program to create both the distributed README and the HTML
> version.
>



_______________________________________________
Nms-cgi-devel mailing list
Nms...@li...
https://lists.sourceforge.net/lists/listinfo/nms-cgi-devel

RE: [Nms-cgi-devel] Search on the NMS web site.

[Paul R. <pa...@ro...>]

I'll third the motion...  and strongly advocate that POD is the way to go,
for all the obvious reasons.

-paul


> Alternatively we could convert the READMEs to POD or XML and deploy some
> intermediate program to create both the distributed README and the HTML
> version.
> 

Re: [Nms-cgi-devel] Search on the NMS web site.

[Chris D. <cd...@ts...>]

On Tue, 5 Mar 2002, Jonathan Stowe wrote:

> Alternatively we could convert the READMEs to POD or XML and deploy some
> intermediate program to create both the distributed README and the HTML
> version.

Embedding POD sounds like an excellent idea. If NMS is supposed to help
teach proper Perl coding style, including at least some POD seems like it
ought to be part of the mix there. The scripts can also have a short "in
order to read the documentation for $foo, try 'perldoc $foo. On most
systems this will provide... yadda... if it doesn't work, you can just
read the top of the script itself."


-- 
Chris Devers                           chd...@ne...
Apache / mod_perl / http://homepage.mac.com/chdevers/resume/