#!/usr/local/bin/perl5
# ------------------------------------------------------------------------
#
#  readlogII  - a tool for reading part of the www-servers access log
#                 and displaying it in html 
#
#  Usage:  readlogII filename > log.html      (writes to std out)
#  in WWW: HREF = ".../readlogII?filename"
#
#  copyright (c) 1994  readlog.cgi by Brigitte Jellinek 
#  this program is available under the GNU General Public License
#  e-mail : bjelli@cosy.sbg.ac.at
#  HREF = "http://www.cosy.sbg.ac.at/people/bjelli.html"
#
#  copyright (c) 1998 readlogII by Christian rossi 
#  this program is available under the GNU General Public License
#  e-mail : rossi@loria.fr 
#  HREF = "http://www.loria.fr/~rossi"
#  country name, web link, traceroute...
#  based on readlog.cgi by Brigitte Jellinek 
#  Version 4.0  
#
#  version 3.0 -   cgi conform, no regex (those are handled by mreadlog)
#  version 2.6 -   sorts by date, backlinks to files
#  
#  use at your own risk !  - if you enhance it, please give me a copy
#
# ------------------------------------------------------------------------
# this script fgreps the access-log (found in $www_server_dir/logs )
# for the filename, and then gives a summary of the information found.


$SERVER_ROOT = "/usr/www/apache-httpd/logs";
$LOG_FILE = $SERVER_ROOT . "/access_log";

#$SERVER_ROOT = "/local/www/texte/statistiques/logs";
#$LOG_FILE = $SERVER_ROOT . "/access_log.0899";

#$SERVER_ROOT = "/local/www/cgi-bin/rossi";
#$LOG_FILE = $SERVER_ROOT . "/toto.txt";

$| = 1;
$date = `/bin/date`;

# a few arrays for storing the information we find:
@last = ();     # last access by host
@count = ();    # no of accesses by host
%index = ();    # index: host (as found in log), value: index of above arrays
%indexcountry = (); # index country : fr -> France
%indexwww = (); # index www : iecn.u-nancy.fr -> www.iecn.u-nancy.fr

# for sorting by date

%no_of_month = ( Jan, 1, Feb, 2, Mar, 3, Apr, 4, May, 5, Jun, 6,
                     Jul, 7, Aug, 8, Sep, 9, Oct,10, Nov,11, Dec,12 );

sub bydate     # for sorting

   {   ($amonth, $ano) = split(/ /, $last[$index{$a}]);   
       ($bmonth, $bno) = split(/ /, $last[$index{$b}]);

       if ( $amonth eq $bmonth ) 
       {  $ano <=> $bno }
       else  
       { $no_of_month{$amonth} <=> $no_of_month{$bmonth} };
   }

sub init_country 
{
 open(COUNTRYFILE, "listef.txt");
while ($code = <COUNTRYFILE>) {
chop($code);
$country=<COUNTRYFILE>;
chop($country);
$indexcountry{$code} = $country;
}
close(COUNTRYFILE);
#foreach $key (keys %indexcountry)
#{
#print "<li>at $key we have $indexcountry{$key}";
#}
}

sub init_www
{
open(WWWFILE, "listewww.txt");
while ($site = <WWWFILE>) {
chop($site);
$indexwww{$site} = 1;
}
close(WWWFILE);

#foreach $key (keys %indexwww)
#{
#print "<li>at $key we have $indexwww{$key}";
#}

}

# two little subroutins for formating output - really not important



sub witchcountry
{
$caller = $_[0];

$code=$indexcountry{$caller};
#print "code=$code et c=$caller \n";

while ($code eq "") { 
@adresse=split(/\./,$caller);
($nom,@adresse)=@adresse;
$caller=join(".",@adresse);
$code=$indexcountry{$caller};
#print "XXX n=$nom et ca=$caller et c=$code XXX\n";
if ($caller eq "") { last; } 

}
#print "code=$code et $caller \n";
if ($code eq "") {$code=Unknown}
$code;
}

#http://beacon.webtv.net/cgi-bin/nph-traceroute.cgi?
#http://hplyot.obspm.fr/cgi-bin/nph-traceroute?

sub witchwww
{
$caller = $_[0];

if ($caller =~ /.*[0-9]$/)
{
#$adrwww="hplyot.obspm.fr/cgi-bin/nph-traceroute?$caller";
#$adrwww="www.freenix.fr/cgi-bin/nph-traceroute?$caller";
$adrwww="www.nic.fr/cgi-bin/nph-prtraceroute?query=$caller&submit=Envoyer";

} else 
{
@adresse=split(/\./,$caller);

$nb_elem=@adresse;

while ($nb_elem > 2)
{
($nom,@adresse)=@adresse;
$nb_elem=@adresse;
$adrwww=join(".",@adresse);
$modif=$indexwww{$adrwww};
if ($modif)
{ @adresse = ($nom, @adresse); 
$adrwww=join(".",@adresse);
$adrwww= "www." . $adrwww;
last;
}
}
$adrwww=join(".",@adresse);
$adrwww= "www." . $adrwww;
}
}


sub pr
   {  $caller = $_[0];
      $i = $index{ $caller };
      print "$caller</a> called ";
      if    ($count[$i]==1) { print "<a href=\"http://www.loria.fr/cgi-bin/rossi/readlogII?$filename&$caller\">once, on $last[$i]</a>\n"; }
      elsif ($count[$i]==2) { print "<a href=\"http://www.loria.fr/cgi-bin/rossi/readlogII?$filename&$caller\">twice, last on $last[$i]</a>\n"; }
      else  {print "<a href=\"http://www.loria.fr/cgi-bin/rossi/readlogII?$filename&$caller\">$count[$i] times, last on $last[$i]</a>\n"};
    };


sub prli
   {  local($heading,$pattern) = @_;
      @l = grep(/$pattern/, sort bydate keys %index);
      if ($#l >= 0) 
       { $nombre = $#l + 1 ; 
print "<H1>$heading <font size=-1>($nombre sites)</font></H1>";
#print "$nombre sites\n"; 
         print "    <UL>\n";
         foreach (@l)
          { 
#  
    $country = &witchcountry( $_ );
$adrwww = &witchwww( $_ );
#$country=France;
print "    <LI><a href=\"http://$adrwww\" 
onMouseOver=\"window.status='$country'; return true\"
 >";
            &pr( $_ );
            delete $index{$_};
          };
        print "    </UL>\n";
      };
    }
# Main Program.

if ($ARGV[0] eq '')    # no argument (filename to search): introduction.
    { print <<'EndOfExplanation';
Content-type: text/html

<HTML> <HEAD>
    <TITLE>Read Log plus</title>
    <ISINDEX>
</HEAD> <BODY  "bgcolor=#ffffce">
     <!--#include virtual="/inc/backlink-cosy.htinc"-->
     <!--#include virtual="/inc/backlink-cosy-www-doku.htinc"-->
     <!--#include virtual="/inc/info.htinc"-->
     <!--#include virtual="/inc/help.htinc"-->
     <HR> 
      <H1>Read Log</H1>
      This is a tool for reading the httpd access log of documents.
      Enter the name of the document whos log you want to see.
      (the whole server specifiv part of the URL: e.g. '/people/bjelli.html'<P> 
      This only works if the document is on our local Server
      The <A HREF="http://www.loria.fr/~rossi/readlog">source code</A> 
      of this script is available. <P> 
      share and enjoy !<P>
EndOfExplanation
    }
else
    {
#A { text-decoration: none }  
print <<"EOM";
Content-type: text/html

<HTML> <HEAD>
    <TITLE>CCH Read Log</title>
 <STYLE type="text/css">
A {  text-decoration: none; link-color: #0000ff}  
.tdu { text-decoration: underline; link-color: #0000ff}  
 </STYLE>

</HEAD> <BODY LINK="#000000" bgcolor=#ffffce>
     <!--#include virtual="/inc/backlink-cosy.htinc"-->
     <!--#include virtual="/inc/backlink-cosy-www-doku.htinc"-->
     <!--#include virtual="/inc/info.htinc"-->
     <!--#include virtual="/inc/help.htinc"-->
     <HR> 
EOM


   $filenamemach = $ARGV[0];     # that's what we're looking for

      $filenamemach =~ s/\\//g;  # unescape characters
      $filenamemach =~ s/%(..)/pack('c',hex($1))/eg;  # unescape characters


($filename,$machine)=split(/&/,$filenamemach);
#print "    <H1> a=$ARGV[0] <br>f=$filename<br> m=$machine <br>f1=$filename1 </H1>\n";


if ( $machine eq "" )
{
print "    <H1>Log  d'activité du serveur WWW pour la partie '$filename'</H1>\n";
}
else 
{
print "    <H1>Log d'activité du serveur WWW pour la partie '$filename' <br>et la machine '$machine'</H1>\n"
}



#      print "    <H1>Log d'activité WWW du <a href=\"http://cch.loria.fr\">CCH</a> pour la partie '$filename'</H1>\n";
#      print " Dans $LOG_FILE <BR>\n" ;
#     $filename =~ s!\^! !g;
print "<b><i><ul> <li>le nom du site ou du pays se trouve dans la barre d'état lorsque l'on se place sur un nom
de machine\n";
print "<li>le serveur WEB des sites référencés est accessible en cliquant sur les noms de machine\n";
print "<li> pour les machines sans nom un traceroute est effectué depuis Paris\n";
print "<li> en cliquant sur le nombre de requête l'on obtient la liste de tous les fichier accédés
 par cette machine\n";
print "<p>\n";
print "<li> les fichiers accédés sont aussi disponibles, il suffit de cliquer dessus\n";
print "<li> en cliquant sur le nombre après le nom du fichier l'on obtient la liste de toutes les machines y ayant 
accédé\n"; 
print "</ul></i></b>\n";

&init_country;
&init_www;
$date = `/bin/date`;
#($day_of_month,$month,$year,$time_of_day) = split(/ /, $date);
($day_of_month,$month,$dayXX,$day,$time_of_day,$AREA1,$AERA2,$year) = split(/ /, $date);

#####################
#$month="Mar";
####################
#print "1X $month 2X $day_of_month 3X $day 4X $time_of_day 5X $AREA1 et $AERA2 6X $year 7X<br>";
#print "date : $date";
#$month=Aug;
#$machine="hugo.loria.fr"; 

if ( $machine eq "" )
{
      open(LOG, 
          "/bin/grep '/$month/.*' $LOG_FILE |grep $filename|");
#print " 1 -->  $filename<p>\n";
}
else 
{
 open(LOG, 
          "/bin/grep '$machine.*/$month/.*' $LOG_FILE |grep $filename|");
#print " 2 -->  $filename<p>\n";
}
#print " -->  $filename<p>\n";
#     $filename =~ s!^ !!;

      $global_count = 0;   # whole no of accesset to $filename

      while(<LOG>)
            {
            chop;
	    #print " --> $_ \n";
            ($caller,$id,$user,$date,$request,$status,$bytes) =
###	      $_ =~ /(\S*) (\S*) (\S*) \[([^\]]*)\] "* (*) *.*" (\S*) (\S*)/;
	      $_ =~ /(\S*) (\S*) (\S*) \[([^\]]*)\] "([^"]*)" (\S*) (\S*)/;
#print "date : $date <br>";
	   ($tmp, $file) = split( / /, $request) ;
           #print STDERR "+ request = $request | caller = $caller \n";
           #print STDERR "  date    = $date\n";

           # ($file) = $request =~ / (\S*)/;

            ($day_of_month,$month,$year,$time_of_day) =
              $date =~ m!^(..)/(...)/(....):(..:..):..!;

          #print "  $day_of_month,$month,$year,$time_of_day<br> \n";

            #if ($file =~ /(.*)\?/) { $file = $1 };

	    #print " --> $file ($filename) \n";
            if ( $file =~ /$filename/ )
             { $global_count++;
    
                $caller .= ".loria.fr"  unless $caller =~ /\./;
    
    
                if( defined $index{$caller} )  #caller already in array
                  { $i = $index{$caller};
                    $last[$i] = $month . " " . $day_of_month;
                    $count[$i]++;
                  }
                else
                  { 
                    push( @last, $month . " " . $day_of_month );
                    push( @count, 1 );
                    $i = $#count;
                    $index{ $caller } = $i;
                  };
		if ( defined $fichier{$file} ) {
		    $fichier{$file} ++ ;
		}
		else {
		    $fichier{$file} = 1 ;
		}
              };
            };

      chop $files_found; 

#      print "   Trouvé $global_count références pour : $filename\n";
$nombre = keys(%index) ;
      print "   Trouvé $global_count références et $nombre sites.\n";
#      print "<H1>LORIA</H1>\n";
#      $count_local = 0 ;
#	foreach ( keys %index) {
#	print "$_, " ;
#	}	
#	print "\n" ;
##      foreach ( grep(/.loria.fr/, keys %index) )
##             { if    ( /(.*)\.loria\.fr/     ) {$name = $1}
##               else                       {$name = $_};
## 	      $name =~ s/\W//g ;
## 	      $i = $index{$_};
##               printf "%s ($count[$i]), ", $name;
## 	      $count_local += $count[$i] ;
##               delete $index{$_};
##             };
##       print " $count_local connections locales.\n";

&prli("Le LORIA", '\.loria\.fr$'); 
#&prli("Nancy",'\..*nancy.*\.fr$|\.ciril\.fr$|\.nancy\.inra\.fr$|\.nancy\.inserm\.fr$|\.inist\.fr$');
      &prli("Nancy",'.*inalf\.cnrs\.fr$|.*nancy.*\.fr$|\.ciril\.fr$|\.inist\.fr$|\.crai.*\.fr$');
      &prli("La région", '.*metz.*\.fr$|\.*strasbg*\.fr$');
      &prli("L'INRIA", '\.inria.*\.fr$|\.irisa\.fr$');
      &prli("La France", '\.fr$');     
      &prli("L'Europe",'\.bg$|\.dk$|\.hr$|\.tr$|\.lt$|\.gb$|\.ro$|\.ie$|\.gr$|\.hu$|\.lu$|\.uk$|\.de$|\.at$|\.ch$|\.fi$|\.se$|\.es$|\.it$|\.no$|\.be$|\.pt$|\.pl$|\.ru$|\.nl$');
      &prli("Les USA", '\.us$|\.mil$|\.edu$|\.gov$');
      &prli("Canada", '\.ca$');
      &prli("Japon", '\.jp$');
#     &prli("Le Canada", '\.ca$');
      &prli("Les organisations", '\.org$');
      &prli("Sites commerciaux", '\.com$');
      &prli("Les networks", '\.net$');
      &prli("Sites inconnues", '\.*[0-9]$');
$nombre = keys(%index) ;
  if ($nombre > 0)
     {
      print "<H1>Les autres pays <font size=-1>($nombre sites)</font></H1>";
      print "    <UL>\n";
      foreach (sort bydate keys %index)
            { $i = $index{ $_ };
#              print "    <LI>";
$country = &witchcountry( $_ );
$adrwww = &witchwww( $_ );
#$country=France;
print "    <LI><a href=\"http://$adrwww\" 
onMouseOver=\"window.status='$country'; return true\"
 >";
              &pr($_);
            };
      print "    </UL>\n";
     }
      print "<H1>Liste des fichiers</H1>\n";
      print "    <UL>\n";
$PREF="http://www.loria.fr";
$PREFCGI="http://www.loria.fr/cgi-bin/rossi/readlogII?";
      foreach (sort keys %fichier) {
	  $nom = $_ ;
	  #$nom =~ s/\W//g ;
          $nom =~ s/<//g ;
	  print "    <LI> <a href=\"$PREF$nom\">$nom</a> : <a href=\"$PREFCGI$nom\">$fichier{ $_ }</a> \n" ;
      }
      print "    </UL>\n";
    }

$datefr = `/usr/local/bin/datefr`;


print <<"EndOfFooter";
    <HR>
    <ADDRESS>
        Script de  
    <A class=tdu HREF=${dq}http://www.cosy.sbg.ac.at/people/bjelli.html$dq>
    Brigitte Jellinek</A>, modifi&eacute; par <A class=tdu HREF="http://www.loria.fr/~rossi">Christian Rossi</a> </ADDRESS>
    <ADDRESS>Cette page est à jour au $day_of_month $month $year $time_of_day <br>
et a été crée le $datefr
</BODY> </HTML>
EndOfFooter
#
# That's all folks                                            share and enjoy !
###############################################################################