@rem = '-*- Perl -*-';
@rem = '
@echo off
perl -S %0.cmd %1 %2 %3 %4 %5 %6 %7 %8 %9
goto endofperl
';

# musage3.prl script derived from musage.txt (see below)
# I have radically altered Nick Phillips script (he asserts 
# copyright and I have no wish to deny him that and just hope
# he doesn't mind my having done this and making this hacked
# version publically available.  If I knew how to contact him
# to discuss this I would.  If he, or someone who knows his 
# whereabouts comes across this I hope they will let me know
# how to contact him 
#
#  Chris Evans <C.Evans@sghms.ac.uk> 21.ii.96
#
# (Last update of Nick Phillips script: 1st May 1995)
#
# musage.txt    --- script to calculate monthly usage statistics from
#                   log files generate by the EMWAC Windows NT World
#                   Wide Web and gopher servers (https and gophers).
#
# Copyright (c) 1995 Nick Phillips (N.R.Phillips@lse.ac.uk)
#
# Configuration variables are hard-coded in
# (see the section "CONFIGURE THESE VARIABLES").
#
# File and path names should be absolute (e.g. C:/LOGS/)
# or relative to current directory (e.g. ./LOGS/)
# Directory names must always end with trailing backslash
# (e.g. ./ for current directory - don't just leave blank)
# In Perl for Windows NT use forward slash (not backslash)
# in pathnames (e.g. C:/LOGS/)
#
# Uses the gethostbyaddr() function to look up the DNS name
# corresponding to the IP addresses of users recorded in the logs -
# this will only work if your version of Perl supports this function.
# It is best to run this at a time when the Internet is quiet as it
# could end up doing lots of DNS lookups.
#
# If you rename the script to MUSAGE.CMD then on an NT system you will
# be able to run it from the command line (i.e. you will be able to just
# type MUSAGE, rather than PERL MUSAGE.TXT) - note that MUSAGE.CMD is
# a batch file (with Perl script embedded in it), so if you invoke it
# from another batch file you will have to do CALL MUSAGE in order for
# control to return to the original batch file.
#
print "\n\n\n\n\n\nMUSAGE3.PRL adapted by Chris Evans from the following:\n\n";
print "     MUSAGE --- script to calculate monthly usage statistics from\n";
print "     v1.4      log files generated by the EMWAC Windows NT World\n";
print "               Wide Web and gopher servers (https and gophers).\n";
print "     Copyright (c) 1995 Nick Phillips\n";

&config;
&error_lab;     # provide labels to classify errors
if ($dated) { # provide date info.
   &uptonow;
}
&countries; # initialise list of names of countries (e.g. uk = United Kingdom etc.)
# read in all known domain names to save looking them up again
if ($lookupdnsnames) {
   &getdomains;
}

# set the parameters usually collected (for subdirectories)
# from musage.set files
$topfiles = 50;
$topips   = 30;
$inputdir    = "/usr1/www/log/";
$outputdir   = "/usr1/www/pages/mhs/psychotherapy/wwwstats/";
@focusonpath = ("/mhs/psychotherapy/");
$f_pathcomment = "files in any Psychotherapy Section directories considered";
@ignoreip = ("194.80.201.27","cevanpc.sghms.ac.uk","infoserv.sghms.ac.uk","192.153.12.1","192.153.12.30");
$i_ipcomment = "does not include my own accesses as administrator";

# O.K. now collect up data from the server log directory
$logfilepattern = "proxy-log.log";       # log files match this

$mylog = "p_mylog.log";
$errlog = "p_errlog.log";
$psylog = "p.cur.log.log";

open (PSYLOG, ">$outputdir$psylog") ||
        die "cannot open $outputdir$psylog";
open (MYLOG, ">$outputdir$mylog") ||
        die "cannot open $outputdir$mylog";
open (ERRLOG, ">$outputdir$errlog") ||
        die "cannot open $outputdir$errlog";


if (!$debug) {
   # get all log server log directory sorted by filename
   # (extract only those files that match $logfilepattern)
   opendir(LOGDIR, "$inputdir") || die "\nCannot open input directory $inputdir";
   print "Reading server log files from directory:\n   $inputdir";
   @logfiles = sort(grep(/$logfilepattern/i, readdir(LOGDIR)));
   closedir(LOGDIR);
}
$extractonly = 1;
&crunchlog;
$extractonly = 0;
close PSYLOG;
close MYLOG;
close ERRLOG;

$mylog = "ptotal.log";
print "\nConcatenating extracted records with historical records\n";
system("cat ./wwwstats/p.old.log.log ./wwwstats/p.cur.log.log  > ./wwwstats/$mylog");


# relates to getting file size data
$outputdir  = "/usr1/www/pages/mhs/psychotherapy/dir/";
$startdir   = "/mhs/psychotherapy";
$www_offset = "/usr1/www/pages";
@filemask   = ("*.*");
# sort out the location that the server will show for files
$http_rep_dir = $outputdir;
$http_rep_dir =~ s|$www_offset||o;  # o switch as the value of $www_offset is fixed
open (TIME, ">time.dat") || die "Cannot open file time.dat\n";
&getdata;



# now run through all subdirectories looking for config files
open(FIND, "find . -name \"musage.set\" -print |") || die "Couldn't run find: $!\n";
while ($filename = <FIND>)
{
   # reset the default config
   &config;
   # now get the local data from musage.set file
   open (CONFIG, "$filename") || die "Couldn't open $filename\n";
   print "\n\nProcessing config file:\n   $filename";
   while (<CONFIG>) {
      chop;
      eval; # evaluates the line in musage.set as Perl statement
   }
   &undefine;
   # open the output log files
   open (PSYLOG, ">$outputdir$psylog") ||
        die "cannot open $outputdir$psylog";
   open (ERRLOG, ">$outputdir$errlog") ||
        die "cannot open $outputdir$errlog";

   # now for the input log files
   # get all log files from current directory sorted by filename
   # (extract only those files that match $logfilepattern)
   opendir(LOGDIR, "$inputdir") || die "\nCannot open input directory $inputdir";
   print "\nReading log files from directory:\n   $inputdir";
   @logfiles = sort(grep(/$logfilepattern/i, readdir(LOGDIR)));
   closedir(LOGDIR);
   &crunchlog;  # now crunch the logs using the variable settings picked up
   close PSYLOG;
   close ERRLOG;
}


# write out all the domain names to save looking them up again
if ($lookupdnsnames) {
   &writedomains;
}



# --------------------------------------------------------------------------
# SUB-ROUTINES
# --------------------------------------------------------------------------
sub newmonth {
    # write report then undefine all the cumulative total variables
    &summaryofmonth if $currentmonth;
    # *** the transfers to totals were added by CE
    foreach (keys %domain) {
       $totdomain{$_} += $domain{$_};
    }
    undef %domain;

    $totaccess += $access;
    undef $access;
    $totbytes  += $totalbytes;
    undef $totalbytes;
    $toterror += $error;
    $tote304 += $e304;
    undef $error;
    undef %errcl;
    undef %error;


    foreach (keys %file) {
       $totfile{$_} += $file{$_};
    }
    undef %file;
    foreach (keys %ip) {
       $totip{$_} += $ip{$_};
    }
    undef %ip;

    $tothomeacc += $homeaccess;
    undef $homeaccess;
    $tothomebyt += $homebytes;
    undef $homebytes;
    foreach (keys %homefile) {
       $tothomefile{$_} += $homefile{$_};
    }
    undef %homefile;
    foreach (keys %homeip) {
       $tothomeip{$_} += $homeip{$_};
    }
    undef %homeip;

    $totoutacc += $outsideaccess;
    undef $outsideaccess;
    $totoutbyt += $outsidebytes;
    undef $outsidebytes;
    foreach (keys %outsidefile) {
       $totoutfile{$_} += $outsidefile{$_};
    }
    undef %outsidefile;
    foreach (keys %outsideip) {
       $totoutip{$_} += $outsideip{$_};
    }
    undef %outsideip;

    undef $currentmonth;
    undef $currentyear;
    undef $currentlogtype;
}   # &newmonth


sub undefine {
    # reset all the cumulating variables, monthly and total
    # runs before starting crunching for a new musage.set file
    undef %domain;
    undef %totdomain;
    undef $access;
    undef $totaccess;
    undef $totalbytes;
    undef $totbytes;
    undef $error;
    undef $toterror;
    undef %errcl;
    undef %error;
    undef %toterrcl;
    undef %toterror;
    undef $e304;
    undef $tote304;
    undef %file;
    undef %totfile;
    undef %ip;
    undef %totip;
    undef $homeaccess;
    undef $tothomeacc;
    undef $homebytes;
    undef $tothomebyt;
    undef %homefile;
    undef %tothomefile;
    undef %homeip;
    undef %tothomeip;
    undef $outsideaccess;
    undef $totoutacc;
    undef $outsidebytes;
    undef $totoutbyt;
    undef %outsidefile;
    undef %totoutfile;
    undef %outsideip;
    undef %totoutip;
    undef $currentmonth;
    undef $currentyear;
    undef $currentlogtype;
}   # &undefine




sub total {
    # *** added by CE, transfers all accumulated totals to usual variables
    # used to print monthly summaries
    %domain      = %totdomain;
    $access      = $totaccess;
    $totalbytes  = $totbytes;
    %file        = %totfile;
    %ip          = %totip;
    $homeaccess  = $tothomeacc;
    $homebytes   = $tothomebyt;
    %homefile    = %tothomefile;
    %homeip      = %tothomeip;
    $outsideaccess = $totoutacc;
    $outsidebytes  = $totoutbyt;
    %outsidefile   = %totoutfile;
    %outsideip     = %totoutip;
    $error         = $toterror;
    $e304          = $tote304;
}   # &total



# --------------------------------------------------------------------------
sub summaryofmonth {
    # make a new report file for that month
    # filename e.g. 1994-11.HTM

    ($_) = grep(/^$currentmonth/i, January01, February02, March03,
        April04, May05, June06, July07, August08, September09, October10,
        November11, December12);
    ($monthname, $mon) = /(\w+)(\d\d)$/;
    $report = "$currentlogtype$currentyear-$mon";

    $servertype = $servertype{$currentlogtype};

    open (SUMMARY, ">$outputdir$report.HTM") ||
        die "cannot open $outputdir$report.HTM";
    print SUMMARY "<HTML><HEAD>\n";
    print SUMMARY "<TITLE>$servername $servertype server usage log $monthname $currentyear</TITLE>\n";
    print SUMMARY "</HEAD><BODY>\n\n";
    print SUMMARY "<H1>$servername $servertype server usage log</H1>\n";

    if (!$dated) {
       print SUMMARY "<H1>$monthname $currentyear</H1>\n";
    } else {
       print SUMMARY "<H1>$monthname $currentyear</H1>\n";
       print SUMMARY "<P><I> $uptonow</I><P>\n";
    }
    
    print SUMMARY "[<A HREF=\"#files\">Files</A>, ";
    print SUMMARY "<A HREF=\"#sites\">sites</A>, ";
    print SUMMARY "<A HREF=\"#domains\">domains</A>, ";
    print SUMMARY "<A HREF=\"#errors\">errors</A>]\n";

    if (@focusonip != 0) {
        print SUMMARY "<H4>Focussing on accesses from the following IP addresses</H4>\n";
        print SUMMARY "<UL>\n";
        foreach $focusonip (@focusonip) {
            print SUMMARY "<LI>$focusonip</LI>\n";
        }
        print SUMMARY "</UL>\n";
    }
    if (@focusonpath != 0) {
        print SUMMARY "<H4>Focussing on accesses to the following paths or files</H4>\n";
        print SUMMARY "<UL>\n";
        foreach $focusonpath (@focusonpath) {
            print SUMMARY "<LI>$focusonpath</LI>\n";
        }
        print SUMMARY "</UL>\n";
    }
    if ($i_ipcomment) {
       print SUMMARY "<UL>\n";
       print SUMMARY "<LI>$i_ipcomment<BR>\n";
       if ($f_pathcomment) {
          print SUMMARY "<LI>$f_pathcomment<P>\n";
       }
       print SUMMARY "</UL>\n";
    }

    print SUMMARY "<H2>Total number of accesses</H2>\n";
    print SUMMARY "Every time a document is retrieved counts as one access.\n";
    print SUMMARY "<PRE>";
    $length = length($homename) + 12; $length = 20 if $length < 20;
    printf SUMMARY "     %-${length}s %18s\n", "Overall total:", $access;
    printf SUMMARY "     %-${length}s %18s\n", "$homename users:", $homeaccess;
    printf SUMMARY "     %-${length}s %18s\n", "Outside users:", $outsideaccess;
    if ($access) { $opct = (int(10000*$outsideaccess/$access))/100; }
    elsif ($outsideaccess) { $opct = 100; }
    else { $opct = "***"; }
    $width = 6; $dp = 2;
    printf SUMMARY "\nI.e. outside use is %${width}.${dp}f &#037;\n", $opct;
    print SUMMARY "</PRE>\n\n";



    print SUMMARY "<H2>Total number of users</H2>\n";
    print SUMMARY "Each different IP address that accesses the server\n";
    print SUMMARY "counts as one user.  N.B. This may represent a considerably\n";
    print SUMMARY "greater number of actual people\n";
    print SUMMARY "<PRE>";
    $length = length($homename) + 12; $length = 20 if $length < 20;
    printf SUMMARY "     %-${length}s %18s\n", "Overall total:",
        ($totcount = keys(%ip));
    printf SUMMARY "     %-${length}s %18s\n", "$homename users:",
        ($count = keys(%homeip));
    printf SUMMARY "     %-${length}s %18s\n", "Outside users:",
        ($outcount = keys(%outsideip));
    if ($totcount) {
       $opct = (int(10000*$outcount/$totcount))/100;
    } elsif ($outcount) {
       $opct = "100";
    } else {
       $opct = "***";
    }
    printf SUMMARY "\nI.e. outside addresses represent %${width}.${dp}f &#037;\n", $opct;
    print SUMMARY "</PRE>\n\n";

    if ($totalbytes) {
        print SUMMARY "<H2>Total number of bytes transferred</H2>\n";
        print SUMMARY "<PRE>";
        $length = length($homename) + 12; $length = 20 if $length < 20;
        printf SUMMARY "     %-${length}s %18s\n", "Overall total:",
            &putcommasinnumber($totalbytes);
        printf SUMMARY "     %-${length}s %18s\n", "$homename users:",
            &putcommasinnumber($homebytes);
        printf SUMMARY "     %-${length}s %18s\n", "Outside users:",
            &putcommasinnumber($outsidebytes);
        $opct = (int(10000*$outsidebytes/$totalbytes))/100;
        printf SUMMARY "\nI.e. outside use is %${width}.${dp}f &#037;\n", $opct;
        print SUMMARY "</PRE>\n\n";
    }



    print SUMMARY "<HR>\n";
    print SUMMARY "<H2><A NAME=\"files\">Files</A></H2>\n";
    
    print SUMMARY "This is based on the number of times each individual\n";
    print SUMMARY "file is accessed.\n\n";
    &topfile;

    print SUMMARY "<HR>\n";
    print SUMMARY "<H2><A NAME=\"sites\">Sites</A></H2>\n";
    
    print SUMMARY "This is based on the numbers of documents accessed\n";
    print SUMMARY "by users calling from each IP address.\n\n";
    &topip;

    print SUMMARY "<HR><A NAME=\"domains\"></A>\n";
    print SUMMARY "<H2>Domains</H2>\n";
    print SUMMARY "This is based on the numbers of documents accessed\n";
    print SUMMARY "by users calling from each domain\n\n";
    &topdomain;

    print SUMMARY "<A NAME=\"errors\"></A>\n";
    print SUMMARY "<H2>Errors</H2>";
    if (!$error) {
       print SUMMARY "Wonderful but frankly suspicious: no failed access attempts found!\n";
       if ($e304) {
          print ERRORS "There were $e304 \"not modified\" \"errors\" which I don't count\n";
          print ERRORS "as errors as I regard them as legitimate responses to enquiries\n";
       }
    } else {
       if ($error == 1) {
          print SUMMARY "One error only\n";
       } else {
          print SUMMARY "Total of $error errors\n";
       }
       print SUMMARY "<H3>Error class summary</H3>\n<UL>\n";
       foreach $class (sort keys(%errcl)) {
          if ($errcl{$class} == 1) {
             print SUMMARY "<LI>$errcl{$class} error of class $class, i.e. \"$errc_lab{$class}\"\n";
          } else {
             print SUMMARY "<LI>$errcl{$class} errors of class $class, i.e. \"$errc_lab{$class}\"\n";
          }
       }
       print SUMMARY "</UL>\n<H3>Error numbers</H3>\n<UL>\n";
       foreach $class (sort keys(%error)) {
          if ($error{$class} == 1) {
             print SUMMARY "<LI>$error{$class} error of number $class, i.e. \"$err_lab{$class}\"\n";
          } else {
             print SUMMARY "<LI>$error{$class} errors of number $class, i.e. \"$err_lab{$class}\"\n";
          }
       }
       print SUMMARY "</UL>\n";
       print SUMMARY "A total cumulative <A HREF=\"$errlog\">error log</A> is available for these files.\n";
    }


    print SUMMARY $btm_s;
    close SUMMARY;

    # print cumulative report and clear cumulative totals
    open (SUMMARY, ">$outputdir$report.TXT") ||
        die "cannot open $report.TXT";
    print SUMMARY "Complete $servername $servertype usage log\n\n";
    print SUMMARY "$monthname $currentyear\n\n";

    if (@focusonip != 0) {
        print SUMMARY "Focussing on accesses from the following IP addresses:-\n";
        foreach $focusonip (@focusonip) {
            print SUMMARY "     $focusonip\n";
        }
        print SUMMARY "\n";
    }

    if (@focusonpath != 0) {
        print SUMMARY "Focussing on accesses to the following paths or files:-\n";
        foreach $focusonpath (@focusonpath) {
            print SUMMARY "     $focusonpath\n";
        }
        if ($f_pathcomment) {
           print SUMMARY "$f_pathcomment\n";
        }
        print SUMMARY "\n";
    }
    if ($i_ipcomment) {
       print SUMMARY "$i_ipcomment\n";
       if ($f_pathcomment) {
          print SUMMARY "$f_pathcomment\n\n";
       }
    }



    print SUMMARY "Focussing on $focus\n\n" if $focus ne '';

    print SUMMARY "Total number of accesses: $access\n";
    print SUMMARY "Total number of users:    ", ($count = keys(%ip)), "\n";
    print SUMMARY "Total number of bytes:    ",
        &putcommasinnumber($totalbytes), "\n";

    print SUMMARY "\nBy file\n";
    print SUMMARY "=======\n";
    foreach $file (sort byfilename keys(%file)) {
        printf SUMMARY "     %9s  %-s\n", $file{$file}, $file;
    }

    print SUMMARY "\nBy ip\n";
    print SUMMARY "=====\n";
    foreach $ip (sort byip keys(%ip)) {
        printf SUMMARY "     %9s  %-18s %-s\n", $ip{$ip}, $ip, $dnsname{$ip};
    }

    print SUMMARY "\nBy domain\n";
    print SUMMARY "=========\n";
    $uktotal = $ustotal = $total = 0;
    foreach $domain (sort bydnsname keys(%domain)) {
        printf SUMMARY "     %9s  %-15s %-s\n",
            $domain{$domain}, $domain, $country{$domain};
            if ($domain eq "uk") {
                $uktotal = $domain{$domain};
            }
            if (($country{$domain} =~ /\*/) || ($domain eq "us")) {
                $ustotal += $domain{$domain};
            }
            $total += $domain{$domain};
    }
    $blank = " ";
    $totals = "Grand total";
    printf SUMMARY "\n     %9s  %-15s %-s\n",
            $total, $blank, $totals;

    print SUMMARY "\n     (* = mainly US based)\n";
    print SUMMARY "   $ustotal from USA or \"mostly USA\" sites,";
    if ($total) {
       $opct = (int(10000*$ustotal/$total))/100;
    } elsif ($ustotal) {
       $opct = "100";
    } else {
       $opct = "***";
    }
    $width = 6; $dp = 2;
    printf SUMMARY " i.e. %${width}.${dp}f &#037;\n", $opct;
    print SUMMARY "   $uktotal from non-local British sites";
    if ($total) {
       $opct = (int(10000*$uktotal/$total))/100;
    } elsif ($uktotal) {
       $opct = "100";
    } else {
       $opct = "***";
    }
    $width = 6; $dp = 2;
    printf SUMMARY " i.e. %${width}.${dp}f%%\n", $opct;
    close SUMMARY;

    # if menu document does not exist, starts a new one
    # looks to see if this report file has already been added to menu
    # looks for <UL> list (if not found, fatal error)
    # adds new report file as first item of list

    $menuname = $outputdir . $currentlogtype . "USAGE.HTM";
    # open the menu file (points to each month's info)
    open (MENU, "$menuname") || &makenewmenu;
    # scoop the whole thing into an array and close it
    @menu = <MENU>;
    close MENU;

    # now set up something that's in entry for the current month
    $srchitem =
        "<LI><A HREF=\"$report.HTM\">$monthname $currentyear<\/A>";
    # set up the replacement for that item
    if ($access) { $opct = (int(10000*$outsideaccess/$access))/100; }
    elsif ($outsideaccess) { $opct = 100; }
    else { $opct = "***"; }
    $failperc = int($failperc);
    if (!$access) { $access = "0"; }
    if (!$error) { $error = "0"; }
    if ($lastentry && ($month eq $mon_now)) {
       # flag up that this is likely to be an incomplete month (and give date of data collection)
       $newitem = 
         "<LI><A HREF=\"$report.HTM\">$monthname $currentyear<\/A> <I>(to $day_now $dat_now)</I> $access good accesses and $error ($failperc% of total) failures<\/LI>\n";
    } else {
       $newitem = 
          "<LI><A HREF=\"$report.HTM\">$monthname $currentyear<\/A> $access good accesses and $error ($failperc% of total) failures<\/LI>\n";
    }

    # now set up the other thing to search for, the statement about when the logs were searched
    $srchlast = "(.*)(log files last checked.*)(</I>.*)";
    $item =~ s/$srchlast/$1$uptonow$3/;  # use it!

    $flag_chge = 0;
    foreach $item (@menu) {
       if ($item =~ /$srchitem/i) {
          # that month is already there, update it
          $item = $newitem;
          # and flag the change
          $flag_chge = 1;
       }
    }
    if (!$flag_chge) {
       # entry wasn't there, so add entry for new usage report
       foreach (@menu) {
           # search only until you come to start of list
           # replace that with itself followed by newline with
           # the new item in it
           last if s/^<UL>/<UL>\n$newitem/i;
           $flag_chge = 1;
       }
    }
    # overwrite old file with new one
    open (MENU, ">$menuname") || die "cannot open $menuname";
    print MENU @menu;
    close MENU;

    undef (@menu);

}   # &summaryofmonth

# --------------------------------------------------------------------------



sub totsummary {
    $report = "TOTAL";
    open (SUMMARY, ">$outputdir$report.HTM") ||
        die "cannot open $outputdirTOTAL.HTM";
    print SUMMARY "<HTML><HEAD>\n";
    print SUMMARY "<TITLE>$servername $servertype server cumulative total usage log</TITLE>\n";
    print SUMMARY "</HEAD><BODY>\n\n";

    print SUMMARY "<H1>$servername $servertype server cumulative total usage log</H1>\n";
    print SUMMARY "<P><I>up to $uptonow</I>\n";

    print SUMMARY "<P>[<A HREF=\"#files\">Files</A>, ";
    print SUMMARY "<A HREF=\"#sites\">sites</A>, ";
    print SUMMARY "<A HREF=\"#domains\">domains</A>, \n";
    print SUMMARY "<A HREF=\"#errors\">errors</A>]\n";

    if (@focusonip != 0) {
        print SUMMARY "<H4>Focussing on accesses from the following IP addresses</H4>\n";
        print SUMMARY "<UL>\n";
        foreach $focusonip (@focusonip) {
            print SUMMARY "<LI>$focusonip</LI>\n";
        }
        print SUMMARY "</UL>\n";
    }
    
    if (@focusonpath != 0) {
        print SUMMARY "<H4>Focussing on accesses to the following paths or files</H4>\n";
        print SUMMARY "<UL>\n";
        foreach $focusonpath (@focusonpath) {
            print SUMMARY "<LI>$focusonpath</LI>\n";
        }
        print SUMMARY "</UL>\n";
    }
    if ($i_ipcomment) {
       print SUMMARY "<UL>\n";
       print SUMMARY "<LI>$i_ipcomment<BR>\n";
       if ($f_pathcomment) {
          print SUMMARY "<LI>$f_pathcomment<P>\n";
       }
       print SUMMARY "</UL>\n";
    }

    print SUMMARY "<H2>Total number of accesses</H2>\n";
    print SUMMARY "Every time a document is retrieved counts as one access.\n";
    print SUMMARY "<PRE>";
    $length = length($homename) + 12; $length = 20 if $length < 20;
    printf SUMMARY "     %-${length}s %18s\n", "Overall total:", $access;
    printf SUMMARY "     %-${length}s %18s\n", "$homename users:", $homeaccess;
    printf SUMMARY "     %-${length}s %18s\n", "Outside users:", $outsideaccess;
    if (!$access) { $opct = (int(10000*$outsideaccess/$access))/100; }
    elsif ($outsideaccess) { $opct = 100; }
    else { $opct = "***"; }
    $width = 6; $dp = 2;
    printf SUMMARY "\nI.e. outside use is %${width}.${dp}f &#037;\n", $opct;
    printf SUMMARY "\n";
    print SUMMARY "</PRE>\n\n";



    print SUMMARY "<H2>Total number of users</H2>\n";
    print SUMMARY "Each different IP address that accesses the server\n";
    print SUMMARY "counts as one user.\n";
    print SUMMARY "<PRE>";
    $length = length($homename) + 12; $length = 20 if $length < 20;
    printf SUMMARY "     %-${length}s %18s\n", "Overall total:",
        ($totcount = keys(%ip));
    printf SUMMARY "     %-${length}s %18s\n", "$homename users:",
        ($count = keys(%homeip));
    printf SUMMARY "     %-${length}s %18s\n", "Outside users:",
        ($outcount = keys(%outsideip));
    $opct = (int(10000*$outcount/$totcount))/100;
    $width = 6; $dp = 2;
    printf SUMMARY "\nI.e. outside use is %${width}.${dp}f &#037;\n", $opct;

    print SUMMARY "</PRE>\n\n";

    if ($totalbytes) {
        print SUMMARY "<H2>Total number of bytes transferred</H2>\n";
        print SUMMARY "<PRE>";
        $length = length($homename) + 12; $length = 20 if $length < 20;
        printf SUMMARY "     %-${length}s %18s\n", "Overall total:",
            &putcommasinnumber($totalbytes);
        printf SUMMARY "     %-${length}s %18s\n", "$homename users:",
            &putcommasinnumber($homebytes);
        printf SUMMARY "     %-${length}s %18s\n", "Outside users:",
            &putcommasinnumber($outsidebytes);
        $opct = (int(10000*$outsidebytes/$totalbytes))/100;
        $width = 6; $dp = 2;
        printf SUMMARY "\nI.e. outside use is %${width}.${dp}f &#037;\n", $opct;
        print SUMMARY "</PRE>\n\n";
    }



    print SUMMARY "<HR><A NAME=\"files\"></A>\n";
    print SUMMARY "<H2>Files in order of popularity</H2>\n";
    print SUMMARY "This is based on the number of times each individual\n";
    print SUMMARY "file is accessed.\n\n";
    &topfile;

    print SUMMARY "<HR><A NAME=\"sites\"></A>\n";
    print SUMMARY "<H2>Sites in order of frequency of visits</H2>\n";
    print SUMMARY "This is based on the numbers of documents accessed\n";
    print SUMMARY "by users calling from each IP address\n\n";
    &topip;

    print SUMMARY "<HR><A NAME=\"domains\"></A>\n";
    print SUMMARY "<H2>Domains</H2>";
    print SUMMARY "This is based on the numbers of documents accessed\n";
    print SUMMARY "by users calling from each domain\n\n";
    &topdomain;

    print SUMMARY "<HR><A NAME=\"errors\"></A>\n";
    print SUMMARY "<H2>Errors</H2>";
    if (!$toterror) {
       print SUMMARY "Wonderful but frankly suspicious: no failed access attempts found!\n";
       if ($e304) {
          print ERRORS "There were $e304 \"not modified\" \"errors\" which I don't count\n";
          print ERRORS "as errors as I regard them as legitimate responses to enquiries\n";
       }
    } else {
       if ($error == 1) {
          print SUMMARY "One error only\n";
       } else {
          print SUMMARY "Total of $error errors\n";
       }
       print SUMMARY "<H3>Error class summary</H3>\n<UL>\n";
       foreach $class (sort keys(%toterrcl)) {
          if ($toterrcl{$class} == 1) {
             print SUMMARY "<LI>$toterrcl{$class} error of class $class, i.e. $errc_lab{$class}\n";
          } else {
             print SUMMARY "<LI>$toterrcl{$class} errors of class $class, i.e. $errc_lab{$class}\n";
          }
       }
       print SUMMARY "</UL>\n<H3>Breakdown by actual error code</H3>\n<UL>\n";
       foreach $class (sort keys(%toterror)) {
          if ($toterror{$class} == 1) {
             print SUMMARY "<LI>$toterror{$class} error of class $class, i.e. $err_lab{$class}\n";
          } else {
             print SUMMARY "<LI>$toterror{$class} errors of class $class, i.e. $err_lab{$class}\n";
          }
       }
       print SUMMARY "</UL>\n";
    }

    $report = "TOTAL";
    print SUMMARY $btm_s;
    close SUMMARY;

    # print cumulative report and clear cumulative totals
    open (SUMMARY, ">$outputdir$report.TXT") ||
        die "cannot open $report.TXT";
    print SUMMARY "Complete $servername $servertype usage log\n\n";
    print SUMMARY "$monthname $currentyear\n\n";

    if (@focusonip != 0) {
        print SUMMARY "Focussing on accesses from the following IP addresses:-\n";
        foreach $focusonip (@focusonip) {
            print SUMMARY "     $focusonip\n";
        }
        print SUMMARY "\n";
    }

    if (@focusonpath != 0) {
        print SUMMARY "Focussing on accesses to the following paths or files:-\n";
        foreach $focusonpath (@focusonpath) {
            print SUMMARY "     $focusonpath\n";
        }
        print SUMMARY "\n";
    }
    if ($i_ipcomment) {
       print SUMMARY "<UL>\n";
       print SUMMARY "<LI>$i_ipcomment<BR>\n";
       if ($f_pathcomment) {
          print SUMMARY "<LI>$f_pathcomment<P>\n";
       }
       print SUMMARY "</UL>\n";
    }



    print SUMMARY "Focussing on $focus\n\n" if $focus ne '';

    print SUMMARY "Total number of accesses: $access\n";
    print SUMMARY "Total number of users:    ", ($count = keys(%ip)), "\n";
    print SUMMARY "Total number of bytes:    ",
        &putcommasinnumber($totalbytes), "\n";

    print SUMMARY "\nBy file\n";
    print SUMMARY "=======\n";
    foreach $file (sort byfilename keys(%file)) {
        printf SUMMARY "     %9s  %-s\n", $file{$file}, $file;
    }

    print SUMMARY "\nBy ip\n";
    print SUMMARY "=====\n";
    foreach $ip (sort byip keys(%ip)) {
        printf SUMMARY "     %9s  %-18s %-s\n", $ip{$ip}, $ip, $dnsname{$ip};
    }

    print SUMMARY "\nBy domain\n";
    print SUMMARY "=========\n";
    $uktotal = $ustotal = $total = 0;
    foreach $domain (sort bydnsname keys(%domain)) {
        printf SUMMARY "     %9s  %-15s %-s\n",
            $domain{$domain}, $domain, $country{$domain};
            if ($domain eq "uk") {
                $uktotal = $domain{$domain};
            }
            if (($country{$domain} =~ /\*/) || ($domain eq "us")) {
                $ustotal += $domain{$domain};
            }
            $total += $domain{$domain};
    }
    $blank = " ";
    $totals = "Grand total";
    printf SUMMARY "\n     %9s  %-15s %-s\n",
            $total, $blank, $totals;

    print SUMMARY "\n     (* = mainly US based)\n";
    print SUMMARY "   $ustotal from USA or \"mostly USA\" sites";
    if ($total) {
       $opct = (int(10000*$ustotal/$total))/100;
    } elsif ($ustotal) {
       $opct = "100";
    } else {
       $opct = "***";
    }
    $width = 6; $dp = 2;
    printf SUMMARY " i.e. %${width}.${dp}f &#037;\n", $opct;
    print SUMMARY "   $uktotal from non-local British sites";
    if ($total) {
       $opct = (int(10000*$uktotal/$total))/100;
    } elsif ($uktotal) {
       $opct = "100";
    } else {
       $opct = "***";
    }
    $width = 6; $dp = 2;
    printf SUMMARY " i.e. %${width}.${dp}f &#037;\n", $opct;
    close SUMMARY;

}   # &totsummary

sub errsummary {
    $report = "ERRORS.HTM";
    open (ERRORS, ">$outputdir$report") ||
        die "cannot open $outputdir$report";
    print ERRORS "<HTML><HEAD>\n";
    print ERRORS "<TITLE>$servername $servertype server cumulative error log</TITLE>\n";
    print ERRORS "</HEAD><BODY>\n\n";

    if (!$dated) {
       print ERRORS "<H1>$monthname $currentyear</H1>\n";
    } else {
       print ERRORS "<H1>$monthname $currentyear</H1>\n";
       print ERRORS "<P><I> $uptonow</I><P>\n";
    }

    if (!$toterror) {
       print ERRORS "Wonderful but frankly suspicious: no failed access attempts found!\n";
       if ($e304) {
          print ERRORS "There were $e304 \"not modified\" \"errors\" which I don't count\n";
          print ERRORS "as errors as I regard them as legitimate responses to enquiries\n";
       }
    } else {
       print ERRORS "<H1>Overall error summary</H1>\n";
       if ($i_ipcomment) {
          print ERRORS "<UL>\n";
          print ERRORS "<LI>$i_ipcomment<BR>\n";
          if ($f_pathcomment) {
             print ERRORS "<LI>$f_pathcomment<P>\n";
          }
          print ERRORS "</UL>\n";
       }
       if ($error == 1) {
          print ERRORS "One error only\n";
       } else {
          print ERRORS "Total of $error errors\n";
       }
       print ERRORS "<HR>\n<H3>Breakdown by error classes</H3>\n<UL>\n";
       foreach $class (sort keys(%toterrcl)) {
          if ($toterrcl{$class} == 1) {
             print ERRORS "<LI>$toterrcl{$class} error of class $class, i.e. \"$errc_lab{$class}\"\n";
          } else {
             print ERRORS "<LI>$toterrcl{$class} errors of class $class, i.e. \"$errc_lab{$class}\"\n";
          }
       }
       print ERRORS "</UL>\n<H3>Breakdown by actual error code</H3>\n<UL>\n";
       foreach $class (sort keys(%toterror)) {
          if ($toterror{$class} == 1) {
             print ERRORS "<LI>$toterror{$class} error of class $class, i.e. \"$err_lab{$class}\"\n";
          } else {
             print ERRORS "<LI>$toterror{$class} errors of class $class, i.e. \"$err_lab{$class}\"\n";
          }
       }
    }
    print ERRORS "</UL>\n";
    print ERRORS $btm_e;
    close ERRORS;

}   # &errsummary




# --------------------------------------------------------------------------
sub makenewmenu {
    # no menu (title) document found - start a new one
    # new menu will contain a blank list <UL></UL>
    # leave file open for reading

    open (MENU, ">$menuname") || die "cannot open $menuname";
    print MENU "<HTML><HEAD>\n";
    print MENU "<TITLE>Monthly $servertype server usage logs</TITLE>\n";
    print MENU "</HEAD><BODY>\n";
    print MENU "<H1>$servername $servertype server</H1>\n";
    print MENU "<P>Cumulative total \n<A HREF=\"ERRORS.HTM\">errors</A> and \n";
    print MENU "<A HREF=\"TOTAL.HTM\">usage</A><P>\n";

    if (!$dated) {
       print MENU "<P><I>From $monthname $currentyear<P>\n";
    } else {
       print MENU "<P><I>From $monthname $currentyear, $uptonow</I><P>\n";
    }
    
    if (@focusonip != 0) {
        print MENU "<H4>Focussing on accesses from the following IP addresses</H4>\n";
        print MENU "<!-- this is a vital comment spacer! --><UL>\n";
        foreach $focusonip (@focusonip) {
            print MENU "<LI>$focusonip</LI>\n";
        }
        print MENU "</UL>\n";
    }
    if (@focusonpath != 0) {
        print MENU "<H4>Focussing on accesses to the following paths or files</H4>\n";
        print MENU "<!-- this is a vital comment spacer! --><UL>\n";
        foreach $focusonpath (@focusonpath) {
            print MENU "<LI>$focusonpath</LI>\n";
        }
        print MENU "</UL>\n";
    }
    if ($i_ipcomment) {
       print MENU "<!-- this is a vital comment spacer! --><UL>\n";
       print MENU "<LI>$i_ipcomment<BR>\n";
       if ($f_pathcomment) {
          print MENU "<LI>$f_pathcomment<P>\n";
       }
       print MENU "</UL>\n";
    }



    print MENU "<H2>Monthly usage logs</H2>\n";
    print MENU "<UL>\n";
    print MENU "</UL>\n";
    print MENU $btm_m;
    close MENU;

    open (MENU, "$menuname") || die "cannot open $menuname";

}   # &makenewmenu




# --------------------------------------------------------------------------
sub addname {
    local($ip) = $_[0];
    undef $domain;
    # if this is already a DNS name (as opposed to IP address)
    # then nothing to look up - the punter is wasting my time!
    if ($ip =~ /[A-Z]/i) {
        # domain = last word of dns name (e.g. lse.ac.uk -> uk)
        ($domain) = $ip =~ /(\w+)$/;
    }
    # if domain name for this ip address not yet known, look it up
    else {
        if ($lookupdnsnames && !$dnsname{$ip}) {
            print "\nLooking up DNS name for $ip . . . \t" if $verbosemode;
            # if lookup fails, set to "unknown"
            ($a, $b, $c, $d) = split(/\./, $ip);
            $address = pack('C4', $a, $b, $c, $d);

            ($name, $aliases, $adrtype, $length, @serveraddr) =
                gethostbyaddr($address, 2);
            $name =~ tr/A-Z/a-z/ if $filenamestolowercase;
            $name = " " if $name eq '';
            print "$name\n" if $verbosemode;
            $dnsname{$ip} = $name;
        }
        # domain = last word of dns name (e.g. lse.ac.uk -> uk)
        ($domain) = $dnsname{$ip} =~ /(\w+)$/;
    }
    $domain;
}   # &addname




# --------------------------------------------------------------------------
sub topfile {
    # print out the top $topfiles files
    # file accesses are in the %file associative array
    # *** this test was added by CE to deal with small lists
    if ($topfiles && ($#files > $topfiles)) {
       print SUMMARY "<H4>Top $topfiles files accessed by all users</H4>\n";
    } else {
       print SUMMARY "<H4>Files accessed by all users in order of popularity</H4>\n";
    }
    print SUMMARY "<PRE>";
    local ($count) = 0;
    printf SUMMARY "     %10s %-s\n", "Accesses", "File";
    printf SUMMARY "     %10s %-s\n", "========", "====";
    foreach $file (sort byfileaccess keys(%file)) {
        if ($size{$file}) {
           printf SUMMARY "     %9s  <A HREF=\"%-s\">%-s</A>\n", $file{$file}, $file, $file;
       } else {
           printf SUMMARY "     %9s  %-s\n", $file{$file}, $file;
       }
       last if ++$count == $topfiles;
    }
    print SUMMARY "</PRE>\n\n";

    if ($topfiles && ($#homefile > $topfiles)) {
       print SUMMARY "<H4>Top $topfiles files accessed by $homename users</H4>\n";
    } else {
       print SUMMARY "<H4>Files accessed by $homename users in order of popularity</H4>\n";
    }
    print SUMMARY "<PRE>";
    $count = 0;
    printf SUMMARY "     %10s %-s\n", "Accesses", "File";
    printf SUMMARY "     %10s %-s\n", "========", "====";
    foreach $file (sort byhomefileaccess keys(%homefile)) {
        if ($size{$file}) {
           printf SUMMARY "     %9s  <A HREF=\"%-s\">%-s</A>\n", $homefile{$file}, $file, $file;
        } else {
           printf SUMMARY "     %9s  %-s\n", $homefile{$file}, $file;
        }
        last if ++$count == $topfiles;
    }
    print SUMMARY "</PRE>\n\n";

    if ($topfiles && ($#outsidefile > $topfiles)) {
       print SUMMARY "<H4>Top $topfiles files accessed by outside users</H4>\n";
    } else {
       print SUMMARY "<H4>Files accessed by outside users in order of popularity</H4>\n";
    }
    print SUMMARY "<PRE>";
    $count = 1;
    printf SUMMARY "     %10s %-s\n", "Accesses", "File";
    printf SUMMARY "     %10s %-s\n", "========", "====";
    foreach $file (sort byoutsidefileaccess keys(%outsidefile)) {

        if ($size{$file}) {
           printf SUMMARY "     %9s  <A HREF=\"%-s\">%-s</A>\n", $outsidefile{$file}, $file, $file;
        } else {
           printf SUMMARY "     %9s  %-s\n", $outsidefile{$file}, $file;
        }
        last if ++$count == $topfiles;
    }
    print SUMMARY "</PRE>\n\n";
}   # &topfile




# --------------------------------------------------------------------------
# SUB-ROUTINES - Sorting
# --------------------------------------------------------------------------
sub byfileaccess {
    # compare by number of accesses in order to sort
    # into reverse order by access
    $file{$b} <=> $file{$a};
}


# --------------------------------------------------------------------------
sub byfilename {
    # compare by filename - convert to lowercase to compare strings
    local($stringa) = $a;
    local($stringb) = $b;
    $stringa =~ tr/A-Z/a-z/;
    $stringb =~ tr/A-Z/a-z/;
    $stringa cmp $stringb;
}


# --------------------------------------------------------------------------
sub byhomefileaccess {
    # compare by number of accesses in order to sort
    # into reverse order by access
    $homefile{$b} <=> $homefile{$a};
}


# --------------------------------------------------------------------------
sub byoutsidefileaccess {
    # compare by number of accesses in order to sort
    # into reverse order by access
    $outsidefile{$b} <=> $outsidefile{$a};
}


# --------------------------------------------------------------------------
sub byipaccess {
    # compare by number of accesses in order to sort
    # into reverse order by access
    $ip{$b} <=> $ip{$a};
}


# --------------------------------------------------------------------------
sub byhomeipaccess {
    # compare by number of accesses in order to sort
    # into reverse order by access
    $homeip{$b} <=> $homeip{$a};
}


# --------------------------------------------------------------------------
sub byoutsideipaccess {
    # compare by number of accesses in order to sort
    # into reverse order by access
    $outsideip{$b} <=> $outsideip{$a};
}


# --------------------------------------------------------------------------
sub bydnsname {
    # compare by DNS name in order to sort into alphabetical order
    # by country
    $country{$a} cmp $country{$b};
}


# --------------------------------------------------------------------------
sub bydomainaccess {
    # compare by number of accesses in order to sort
    # into reverse order by access
    $domain{$b} <=> $domain{$a};
}


# --------------------------------------------------------------------------
sub byip {
    # compare by ip address in order to sort into
    # order by ip address or domain

    local($comp) = 1;
    local($dnsa) = $dnsname{$a};
    local($dnsb) = $dnsname{$b};

    $dnsa = $a if !$dnsa;
    $dnsb = $b if !$dnsb;

    if ($dnsa =~ /[A-Z]/i) {
        # $dnsa is a DNS name
        if ($dnsb =~ /[A-Z]/i) {
            # $dnsa and $dnsb are both DNS names
            $dnsa = join('.', reverse(split(/\./, $dnsa)));
            $dnsb = join('.', reverse(split(/\./, $dnsb)));
            $comp = $dnsa cmp $dnsb;
        }
        else {
            # $dnsa is a DNS name but $dnsb is an IP address
            $comp = -1;
        }
    }
    elsif ($dnsb =~ /[A-Z]/i) {
        # $dnsa is an IP address but $dnsb is a DNS name
        $comp = +1;
    }
    else {
        # #$dnsa and $dnsb are both IP addresses
        ($a1, $a2, $a3, $a4) = split(/\./, $a);
        ($b1, $b2, $b3, $b4) = split(/\./, $b);
        $comp = $a1 <=> $b1;
        $comp = $a2 <=> $b2 if $comp == 0;
        $comp = $a3 <=> $b3 if $comp == 0;
        $comp = $a4 <=> $b4 if $comp == 0;
    }
    return $comp;
}


# --------------------------------------------------------------------------
# SUB-ROUTINES - Printing out top sites
# --------------------------------------------------------------------------
sub topip {
    # print out the top $topips sites (IP addresses)
    # ip accesses are in the %ip associative array
    if ($topips && ($#ip > $topips)) {
       print SUMMARY "<H4>Overall top $topips sites</H4>\n";
    } else {
       print SUMMARY "<H4>Listing of sites in order of visiting frequency</H4>\n";
    }
    print SUMMARY "<PRE>";
    local ($count) = 0;
    printf SUMMARY "     %10s %-17s", "Accesses", "IP address";
    printf SUMMARY " %-s", "DNS name" if $lookupdnsnames;
    printf SUMMARY "\n";
    printf SUMMARY "     %10s %-17s", "========", "==========";
    printf SUMMARY " %-s", "========" if $lookupdnsnames;
    printf SUMMARY "\n";
    foreach $ip (sort byipaccess keys(%ip)) {
        printf SUMMARY "     %9s  %-17s %-s\n", $ip{$ip}, $ip, $dnsname{$ip};
        last if ++$count == $topips;
    }
    print SUMMARY "</PRE>\n\n";

    if ($topips && ($#homeip > $topips)) {
       print SUMMARY "<H4>Overall top $topips $homename sites</H4>\n";
    } else {
       print SUMMARY "<H4>Listing of $homename sites in order of visiting frequency</H4>\n";
    }
    print SUMMARY "<PRE>";
    local ($count) = 0;
    printf SUMMARY "     %10s %-17s", "Accesses", "IP address";
    printf SUMMARY " %-s", "DNS name" if $lookupdnsnames;
    printf SUMMARY "\n";
    printf SUMMARY "     %10s %-17s", "========", "==========";
    printf SUMMARY " %-s", "========" if $lookupdnsnames;
    printf SUMMARY "\n";
    foreach $ip (sort byhomeipaccess keys(%homeip)) {
        printf SUMMARY "     %9s  %-17s %-s\n", $homeip{$ip}, $ip, $dnsname{$ip};
        last if ++$count == $topips;
    }
    print SUMMARY "</PRE>\n\n";

    if ($topips && ($#outsideip > $topips)) {
       print SUMMARY "<H4>Overall top $topips outside sites</H4>\n";
    } else {
       print SUMMARY "<H4>Listing of outside sites in order of visiting frequency</H4>\n";
    }
    print SUMMARY "<PRE>";
    local ($count) = 0;
    printf SUMMARY "     %10s %-17s", "Accesses", "IP address";
    printf SUMMARY " %-s", "DNS name" if $lookupdnsnames;
    printf SUMMARY "\n";
    printf SUMMARY "     %10s %-17s", "========", "==========";
    printf SUMMARY " %-s", "========" if $lookupdnsnames;
    printf SUMMARY "\n";
    foreach $ip (sort byoutsideipaccess keys(%outsideip)) {
        printf SUMMARY "     %9s  %-17s %-s\n", $ip{$ip}, $ip, $dnsname{$ip};
        last if ++$count == $topips;
    }
    print SUMMARY "</PRE>\n\n";
}   # &topip




# --------------------------------------------------------------------------
sub topdomain {
    # print out the domains
    # domain accesses are in the %domain associative array
    print SUMMARY "<PRE>";
    local ($count) = 0;
    printf SUMMARY "     %10s %-15s %-s\n", "Accesses", "Domain", "Country";
    printf SUMMARY "     %10s %-15s %-s\n", "========", "======", "=======";
    $uktotal = $ustotal = $total = 0;
    foreach $domain (sort bydomainaccess keys(%domain)) {
        printf SUMMARY "     %9s  %-15s %-s\n",
            $domain{$domain}, $domain, $country{$domain};
            if ($domain eq "uk") {
                $uktotal = $domain{$domain};
            }
            if (($country{$domain} =~ /\*/) || ($domain eq "us")) {
                $ustotal += $domain{$domain};
            }
            $total += $domain{$domain};
    }
    $blank = " ";
    $totals = "Grand total";
    printf SUMMARY "\n     %9s  %-15s %-s\n",
            $total, $blank, $totals;

    print SUMMARY "\n     (* = mainly US based)\n";
    print SUMMARY "   $ustotal from USA or \"mostly USA\" sites";
    if ($total) {
       $opct = (int(10000*$ustotal/$total))/100;
    } elsif ($ustotal) {
       $opct = "100";
    } else {
       $opct = "***";
    }
    $width = 6; $dp = 2;
    printf SUMMARY " i.e. %${width}.${dp}f &#037;\n", $opct;
    print SUMMARY "   $uktotal from non-local British sites";
    if ($total) {
       $opct = (int(10000*$uktotal/$total))/100;
    } elsif ($uktotal) {
       $opct = "100";
    } else {
       $opct = "***";
    }
    $width = 6; $dp = 2;
    printf SUMMARY " i.e. %${width}.${dp}f &#037;\n", $opct;
    print SUMMARY "</PRE>";
}   # &topdomain




# --------------------------------------------------------------------------
# SUB-ROUTINES - Focus on or ignore specific accesses
# --------------------------------------------------------------------------
sub ishomeip {
    local($ip) = @_;
    local($found) = 0;

    # $ip may contain an IP address or a DNS name
    # look for a match in the @homeip list
    foreach $homeip (@homeip) {
        if ( (($ip !~ /[A-Z]/i) &&
            (($ip =~ /^$homeip/i) || ($dnsname{$ip} =~ /$homeip$/i)))
            || (($ip =~ /[A-Z]/i) && ($ip =~ /$homeip$/i))
            ) {
            $found = 1;
            last;
        }
    }
    $found;
}


# --------------------------------------------------------------------------
sub focusonthisip {
    local($ip) = @_;
    local($found) = 0;
    if (@focusonip == 0) {
        $found = 1;
    }
    else {
        # $ip may contain an IP address or a DNS name
        # look for a match in the @focusonip list
        foreach $focusonip (@focusonip) {
            if ( (($ip !~ /[A-Z]/i) &&
                (($ip =~ /^$focusonip/i)  || ($dnsname{$ip}=~ /$focusonip$/i)))
                || (($ip =~ /[A-Z]/i) && ($ip =~ /$focusonip$/i))
                ) {
                $found = 1;
                last;
            }
        }
    }
    $found;
}


# --------------------------------------------------------------------------
sub myownip {
    local($ip) = @_;
    local($found) = 0;
    # $ip may contain an IP address or a DNS name
    # lok for a match in the @ignoreip list
    foreach $myownip (@myownip) {
        if ( (($ip !~ /[A-Z]/i) &&
            (($ip =~ /^$myownip/i) || ($dnsname{$ip} =~ /$myownip$/i)))
            || (($ip =~ /[A-Z]/i) && ($ip =~ /$myownip$/i))
            ) {
            $found = 1;
            last;
        }
    }
    $found;
}




# --------------------------------------------------------------------------
sub ignorethisip {
    local($ip) = @_;
    local($found) = 0;
    # $ip may contain an IP address or a DNS name
    # lok for a match in the @ignoreip list
    foreach $ignoreip (@ignoreip) {
        if ( (($ip !~ /[A-Z]/i) &&
            (($ip =~ /^$ignoreip/i) || ($dnsname{$ip} =~ /$ignoreip$/i)))
            || (($ip =~ /[A-Z]/i) && ($ip =~ /$ignoreip$/i))
            ) {
            $found = 1;
            last;
        }
    }
    $found;
}


# --------------------------------------------------------------------------
sub focusonthispath {
    local($path) = @_;
    local($found) = 0;
    if (@focusonpath == 0) {
        $found = 1;
    }
    else {
        foreach $focusonpath (@focusonpath) {
            if ($path =~ /^$focusonpath/i) {
                $found = 1;
                last;
            }
        }
    }
    $found;
}


# --------------------------------------------------------------------------
sub ignorethispath {
    local($path) = @_;
    local($found) = 0;
    foreach $ignorepath (@ignorepath) {
        if ($path =~ /^$ignorepath/i) {
            $found = 1;
            last;
        }
    }
    $found;
}


# --------------------------------------------------------------------------
sub focusonthisext {
    local($path) = @_;
    local($found) = 0;
    if (@focusonext == 0) {
        $found = 1;
    }
    else {
        foreach $focusonext (@focusonext) {
            if ($path =~ /$focusonext$/i) {
                $found = 1;
                last;
            }
        }
    }
    $found;
}


# --------------------------------------------------------------------------
sub ignorethisext {
    local($path) = @_;
    local($found) = 0;
    foreach $ignoreext (@ignoreext) {
        if ($path =~ /$ignoreext$/i) {
            $found = 1;
            last;
        }
    }
    $found;
}


# --------------------------------------------------------------------------
# SUB-ROUTINES - General
# --------------------------------------------------------------------------
sub progressmeter {
    if ($progressmeter eq '-') {$progressmeter = '\\';}
    elsif ($progressmeter eq '\\') {$progressmeter = '|';}
    elsif ($progressmeter eq '|') {$progressmeter = '/';}
    else {$progressmeter = '-';}
    print "$progressmeter\b";
}


# --------------------------------------------------------------------------
sub putcommasinnumber {
    local($n) = @_;
    local($result);
    local($remainder);
    local($string) = "";
    do {
        $result = $n / 1000;
        $remainder = $n - (1000 * int($result));
        $n = int($result);
        if (!$n) {
            $string = $remainder . $string if $remainder;
        }
        else {
            while (length($remainder) < 3) {
                $remainder = "0" . $remainder;
            }
            $string = "," . $remainder . $string;
        }
    } until !$n;
    $string;
}


# --------------------------------------------------------------------------
#   for padding text & numbers to right-justify in column
#   arguments are string, width of column, and character to pad with
sub rightjustifystring {
    local($string) = $_[0];
    local($width) = $_[1];
    local($leadingchar) = $_[2];

    $leadingchar = ' ' if !$leadingchar;
    $width = length($string) if $width < length($number);

    for ($i = length($string); $i < $width; $i++) {
        $string = $leadingchar . $string;
    }
    return $string;
}

sub three_digits {
   local ($_) = @_;
   /(\d+)\.(\d*)/;
   $_ = $1;
   if (length($_) == 0) {
      $_ = "  0";
   } elsif (length($_) == 1) {
      $_ = "  ".$_;
   } elsif (length($_) == 2) {
      $_ = " ".$_;
   } elsif (length > 3) {
      die "Supplied a number of more than 2 digits to three_digits\n";
   }
   $_;
}

# --------------------------------------------------------------------------
sub countries {
    $country{$homedomain} = "($homename)";

    $country{"com"} = "US Commercial*";
    $country{"edu"} = "US Educational*";
    $country{"gov"} = "US Government*";
    $country{"mil"} = "US Military*";
    $country{"net"} = "US Network*";
    $country{"org"} = "US Non-Commercial*";
    $country{"int"} = "US International*";
    $country{"arpa"} = "US Arpanet*";
    $country{"nato"} = "US Nato*";

    $country{"ad"} = "Andorra";
    $country{"ae"} = "UAE";
    $country{"af"} = "Afghanistan";
    $country{"ag"} = "Antigua and Barbuda";
    $country{"ai"} = "Anguilla";
    $country{"al"} = "Albania";
    $country{"am"} = "Armenia";
    $country{"an"} = "Antilles";
    $country{"ao"} = "Angola";
    $country{"aq"} = "Antarctica";
    $country{"ar"} = "Argentina";
    $country{"as"} = "American Samoa";
    $country{"at"} = "Austria";
    $country{"au"} = "Australia";
    $country{"aw"} = "Aruba";
    $country{"az"} = "Azerbaijan";
    $country{"ba"} = "Bosnia and Herzegovina";
    $country{"bb"} = "Barbados";
    $country{"bd"} = "Bangladesh";
    $country{"be"} = "Belgium";
    $country{"bf"} = "Burkina Faso";
    $country{"bg"} = "Bulgaria";
    $country{"bh"} = "Bahrain";
    $country{"bi"} = "Burundi";
    $country{"bj"} = "Benin";
    $country{"bm"} = "Bermuda";
    $country{"bn"} = "Brunei Darussalam";
    $country{"bo"} = "Bolivia";
    $country{"br"} = "Brazil";
    $country{"bs"} = "Bahamas";
    $country{"bt"} = "Bhutan";
    $country{"bv"} = "Bouvet Island";
    $country{"bw"} = "Botswana";
    $country{"by"} = "Belarus";
    $country{"bz"} = "Belize";
    $country{"ca"} = "Canada";
    $country{"cc"} = "Cocos (Keeling) Islands";
    $country{"cf"} = "Central African Republic";
    $country{"cg"} = "Congo";
    $country{"ch"} = "Switzerland";
    $country{"ci"} = "Cote D\'Ivoire (Ivory Coast)";
    $country{"ck"} = "Cook Islands";
    $country{"cl"} = "Chile";
    $country{"cm"} = "Cameroon";
    $country{"cn"} = "China";
    $country{"co"} = "Colombia";
    $country{"cr"} = "Costa Rica";
    $country{"cs"} = "Czechoslovakia (former)";
    $country{"cu"} = "Cuba";
    $country{"cv"} = "Cape Verde";
    $country{"cx"} = "Christmas Island";
    $country{"cy"} = "Cyprus";
    $country{"cz"} = "Czech Republic";
    $country{"de"} = "Germany";
    $country{"dj"} = "Djibouti";
    $country{"dk"} = "Denmark";
    $country{"dm"} = "Dominica";
    $country{"do"} = "Dominican Republic";
    $country{"dz"} = "Algeria";
    $country{"ec"} = "Ecuador";
    $country{"ee"} = "Estonia";
    $country{"eg"} = "Egypt";
    $country{"eh"} = "Western Sahara";
    $country{"er"} = "Eritrea";
    $country{"es"} = "Spain";
    $country{"et"} = "Ethiopia";
    $country{"fi"} = "Finland";
    $country{"fj"} = "Fiji";
    $country{"fk"} = "Falkland Islands (Malvinas)";
    $country{"fm"} = "Micronesia";
    $country{"fo"} = "Faroe Islands";
    $country{"fr"} = "France";
    $country{"fx"} = "France, Metropolitan";
    $country{"ga"} = "Gabon";
    $country{"gb"} = "Great Britain (UK)";
    $country{"gd"} = "Grenada";
    $country{"ge"} = "Georgia";
    $country{"gf"} = "French Guiana";
    $country{"gh"} = "Ghana";
    $country{"gi"} = "Gibraltar";
    $country{"gl"} = "Greenland";
    $country{"gm"} = "Gambia";
    $country{"gn"} = "Guinea";
    $country{"gp"} = "Guadeloupe";
    $country{"gq"} = "Equatorial Guinea";
    $country{"gr"} = "Greece";
    $country{"gs"} = "S. Georgia and S. Sandwich Islands";
    $country{"gt"} = "Guatemala";
    $country{"gu"} = "Guam";
    $country{"gw"} = "Guinea-Bissau";
    $country{"gy"} = "Guyana";
    $country{"hk"} = "Hong Kong";
    $country{"hm"} = "Heard and McDonald Islands";
    $country{"hn"} = "Honduras";
    $country{"hr"} = "Croatia (Hrvatska)";
    $country{"ht"} = "Haiti";
    $country{"hu"} = "Hungary";
    $country{"id"} = "Indonesia";
    $country{"ie"} = "Ireland";
    $country{"il"} = "Israel";
    $country{"in"} = "India";
    $country{"io"} = "British Indian Ocean Territory";
    $country{"iq"} = "Iraq";
    $country{"ir"} = "Iran";
    $country{"is"} = "Iceland";
    $country{"it"} = "Italy";
    $country{"jm"} = "Jamaica";
    $country{"jo"} = "Jordan";
    $country{"jp"} = "Japan";
    $country{"ke"} = "Kenya";
    $country{"kg"} = "Kyrgyzstan";
    $country{"kh"} = "Cambodia";
    $country{"ki"} = "Kiribati";
    $country{"km"} = "Comoros";
    $country{"kn"} = "Saint Kitts and Nevis";
    $country{"kp"} = "Korea (North)";
    $country{"kr"} = "Korea (South)";
    $country{"kw"} = "Kuwait";
    $country{"ky"} = "Cayman Islands";
    $country{"kz"} = "Kazakhstan";
    $country{"la"} = "Laos";
    $country{"lb"} = "Lebanon";
    $country{"lc"} = "Saint Lucia";
    $country{"li"} = "Liechtenstein";
    $country{"lk"} = "Sri Lanka";
    $country{"lr"} = "Liberia";
    $country{"ls"} = "Lesotho";
    $country{"lt"} = "Lithuania";
    $country{"lu"} = "Luxembourg";
    $country{"lv"} = "Latvia";
    $country{"ly"} = "Libya";
    $country{"ma"} = "Morocco";
    $country{"mc"} = "Monaco";
    $country{"md"} = "Moldova";
    $country{"mg"} = "Madagascar";
    $country{"mh"} = "Marshall Islands";
    $country{"mk"} = "Macedonia";
    $country{"ml"} = "Mali";
    $country{"mm"} = "Myanmar";
    $country{"mn"} = "Mongolia";
    $country{"mo"} = "Macau";
    $country{"mp"} = "Northern Mariana Islands";
    $country{"mq"} = "Martinique";
    $country{"mr"} = "Mauritania";
    $country{"ms"} = "Montserrat";
    $country{"mt"} = "Malta";
    $country{"mu"} = "Mauritius";
    $country{"mv"} = "Maldives";
    $country{"mw"} = "Malawi";
    $country{"mx"} = "Mexico";
    $country{"my"} = "Malaysia";
    $country{"mz"} = "Mozambique";
    $country{"na"} = "Namibia";
    $country{"nc"} = "New Caledonia";
    $country{"ne"} = "Niger";
    $country{"nf"} = "Norfolk Island";
    $country{"ng"} = "Nigeria";
    $country{"ni"} = "Nicaragua";
    $country{"nl"} = "Netherlands";
    $country{"no"} = "Norway";
    $country{"np"} = "Nepal";
    $country{"nr"} = "Nauru";
    $country{"nt"} = "Neutral Zone";
    $country{"nu"} = "Niue";
    $country{"nz"} = "New Zealand";
    $country{"om"} = "Oman";
    $country{"pa"} = "Panama";
    $country{"pe"} = "Peru";
    $country{"pf"} = "French Polynesia";
    $country{"pg"} = "Papua New Guinea";
    $country{"ph"} = "Philippines";
    $country{"pk"} = "Pakistan";
    $country{"pl"} = "Poland";
    $country{"pm"} = "St. Pierre and Miquelon";
    $country{"pn"} = "Pitcairn";
    $country{"pr"} = "Puerto Rico";
    $country{"pt"} = "Portugal";
    $country{"pw"} = "Palau";
    $country{"py"} = "Paraguay";
    $country{"qa"} = "Qatar";
    $country{"re"} = "Reunion";
    $country{"ro"} = "Romania";
    $country{"ru"} = "Russian Federation";
    $country{"rw"} = "Rwanda";
    $country{"sa"} = "Saudi Arabia";
    $country{"sb"} = "Solomon Islands";
    $country{"sc"} = "Seychelles";
    $country{"sd"} = "Sudan";
    $country{"se"} = "Sweden";
    $country{"sg"} = "Singapore";
    $country{"sh"} = "St. Helena";
    $country{"si"} = "Slovenia";
    $country{"sj"} = "Svalbard and Jan Mayen Islands";
    $country{"sk"} = "Slovak Republic";
    $country{"sl"} = "Sierra Leone";
    $country{"sm"} = "San Marino";
    $country{"sn"} = "Senegal";
    $country{"so"} = "Somalia";
    $country{"sr"} = "Suriname";
    $country{"st"} = "Sao Tome and Principe";
    $country{"su"} = "USSR (former)";
    $country{"sv"} = "El Salvador";
    $country{"sy"} = "Syria";
    $country{"sz"} = "Swaziland";
    $country{"tc"} = "Turks and Caicos Islands";
    $country{"td"} = "Chad";
    $country{"tf"} = "French Southern Territories";
    $country{"tg"} = "Togo";
    $country{"th"} = "Thailand";
    $country{"tj"} = "Tajikistan";
    $country{"tk"} = "Tokelau";
    $country{"tm"} = "Turkmenistan";
    $country{"tn"} = "Tunisia";
    $country{"to"} = "Tonga";
    $country{"tp"} = "East Timor";
    $country{"tr"} = "Turkey";
    $country{"tt"} = "Trinidad and Tobago";
    $country{"tv"} = "Tuvalu";
    $country{"tw"} = "Taiwan";
    $country{"tz"} = "Tanzania";
    $country{"ua"} = "Ukraine";
    $country{"ug"} = "Uganda";
    $country{"uk"} = "United Kingdom";
    $country{"um"} = "US Minor Outlying Islands";
    $country{"us"} = "USA";
    $country{"uy"} = "Uruguay";
    $country{"uz"} = "Uzbekistan";
    $country{"va"} = "Vatican City State";
    $country{"vc"} = "Saint Vincent and the Grenadines";
    $country{"ve"} = "Venezuela";
    $country{"vg"} = "Virgin Islands (British)";
    $country{"vi"} = "Virgin Islands (U.S.)";
    $country{"vn"} = "Viet Nam";
    $country{"vu"} = "Vanuatu";
    $country{"wf"} = "Wallis and Futuna Islands";
    $country{"ws"} = "Samoa";
    $country{"ye"} = "Yemen";
    $country{"yt"} = "Mayotte";
    $country{"yu"} = "Yugoslavia";
    $country{"za"} = "South Africa";
    $country{"zm"} = "Zambia";
    $country{"zr"} = "Zaire";
    $country{"zw"} = "Zimbabwe";
}


# --------- *** added by CE to provide date information on stats -----------
sub uptonow {
   # *** added by CE, sort out today's date & time
   ($sec,$min,$hour,$mday,$mon,$year,$wday, $yday, $isdat) = localtime(time);
   $day_now = (Sunday,Monday,Tuesday,Wednesday,Thursday,Friday,Saturday,Sunday)[$wday];
   $mon_now = (Jan,Feb,Mar,Apr,May,Jun,Jul,Aug,Sep,Oct,Nov,Dec)[$mon];
   $thismon = $mon + 1;
   if (($mday % 10) == 1) {
      $ord = "st";
   } elsif (($mday % 10) == 2) {
      $ord = "nd";
   } elsif (($mday % 10) == 3) {
      $ord = "rd";
   } else {
      $ord = "th";
   }
   $dat_now   = $mday . $ord;
   $uptonow = "log files last checked: ".$hour.":".$min.":".$sec.",  ".$day_now.", the ".$dat_now." of ".$mon_now." ".$year;
} # sub uptonow

# --------- *** added by CE to deal with error classification --------------
sub error_lab {
   $err_lab{"300"} = "Multiple choices";
   $err_lab{"301"} = "Moved permanently";
   $err_lab{"302"} = "Moved temporarily";
   $err_lab{"304"} = "Not modified (I contend this <I>isn't</I> an error and don't add it to error total)";
   $err_lab{"400"} = "Bad request";
   $err_lab{"401"} = "Unauthorized";
   $err_lab{"403"} = "Forbidden";
   $err_lab{"404"} = "Not found";
   $err_lab{"500"} = "Internal server error";
   $err_lab{"501"} = "Not implemented";
   $err_lab{"502"} = "Bad gateway";
   $err_lab{"503"} = "Service unavailable";
   $errc_lab{"3"} = "Redirection";
   $errc_lab{"4"} = "Client error";
   $errc_lab{"5"} = "Server error";
}


# --------- *** moved here by CE to make easier to read -------------------
sub getdomains {
    # read in all known domain names to save looking them up again
    if (open (DOMAINS, "$dnsnames")) {
        print "\nReading domain names from:\n   $dnsnames\n\n";
        while (<DOMAINS>) {
            chop;
            ($ipaddress, $dnsname) = split(/\t/);
            $dnsname{$ipaddress} = $dnsname;
        }
        close DOMAINS;
    }
    else {
        print "\nCannot read from DNS names file $dnsnames!";
    }
} # sub getdomains



# --------- *** moved here by CE to make easier to read -------------------
sub writedomains {
   # write out all the domain names to save looking them up again
   print "\nWriting domain names back to:\n   $dnsnames\n";
   if (open (DOMAINS, ">$dnsnames")) {
      foreach $ipaddress (keys(%dnsname)) {
         print DOMAINS "$ipaddress\t$dnsname{$ipaddress}\n";
      }
      close DOMAINS;
   }
   else {
      print "\nCannot write to DNS names file $dnsnames\n";
   }
} # sub writedomains




sub getdata {
   # stores all size data in one associative array
   open(FIND, "find . -name \"*.*\" -print |") || die "Couldn't run find: $!\n";
   while ($filename = <FIND>)
   {
      undef $size;
      chop $filename;
      # get data on the file
      ($nsize,$time) = (stat($filename))[7,9]; #use the 7th & 9th items in the stat array
      # reformat the size
      if ($nsize <= 1000) {
         $size = "<1k";
      } elsif ($nsize < 1000000) {
         $size = int($nsize/1000);
         $size = $size."kb";
      } else {
         $size = int($nsize/1000000);
         $size = $size."Mb";
      }
      # reformat the filename to the http server form
      $href = $filename;
      $href =~ s|.|$startdir|o;
      # put this associative array for future use
      $size{$href} = $size;
   }
   close TIME;
} # &getdata


sub two_digits {
   local ($_) = @_;
   if (length($_) == 0) {
      $_ = "00";
   } elsif (length($_) == 1) {
      $_ = "0".$_;
   } elsif (length($_) > 2) {
      die "Supplied a number of more than 2 digits to two_digits\n";
   }
   $_;
}







# --------- *** moved here by CE to make easier to duplicate --------------
sub crunchlog {
   # debug or real data?
   if ($debug) {
      print "\n**** DEBUG MODE RUN ****\n";
      open (INPUT,$debuglog) || die "cannot open debug.log";
      print " \nInputting from:\n   $debuglog\n";
      $logtype = "p";
      $logtype =~ tr/a-z/A-Z/;
      # process log file line by line
      while (<INPUT>) {
         if ($verbosemode) { print ".";}
         # *** I'm commenting this out as I don't think I need it
         # *** else { &progressmeter; }
         $lastentry = 0;
         &doentry;
      }
   } else {
      # process each log file one at a time
      foreach $logfile (@logfiles) {
         open (INPUT, "$inputdir$logfile") || die "cannot open $logfile";
         print " \nInputting from:\n   $inputdir$logfile";
         $logtype = substr($logfile, 0, 1);
         $logtype =~ tr/a-z/A-Z/;
         # process log file line by line
         while (<INPUT>) {
            if ($verbosemode) { print ".";}
            # *** I'm commenting this out as I don't think I need it
            # *** else { &progressmeter; }
            $lastentry = 0;
            &doentry;
         }   # get next entry
      } # process next log file
   }

   # write details from current month to file
   print " \nFinished reading log files from:\n   $inputdir$logfile\n";
   if (!$extractonly) {
      $lastentry = 1;
      &newmonth;
      # *** added by CE, now print out the totals, before quitting
      $topfiles = 0;
      &total;
      &totsummary;
      &errsummary;
   }
   print "\n";
   close ERRLOG;
   close PSYLOG;
} # sub crunchlog




# --------- *** moved here by CE to make easier to duplicate --------------
sub doentry {
   chop;
   # split fields separated by spaces
   $bytes = 0;
   if ($commonlogformat) {
      ($bit1, $bit2, $bit3)      = split(/[\[\]]/);
      ($datetime, $zone)         = split(/[ ]+/,$bit2);
      ($date, $hr, $min, $sec)   = split(/:/,$datetime);
      ($mday, $month, $year)     = split (/\//,$date);
      ($ip, $logonid, $authname) = split(/[ ]+/,$bit1);
      ($tmp,$bit3a,$bit3b)       = split(/"/,$bit3);
      ($method, $file, $version) = split(/[ ]+/,$bit3a);
      ($tmp, $status, $bytes)    = split(/[ ]+/,$bit3b);
      $ip =~ tr/A-Z/a-z/;   # convert dns name to lowercase
   } else {
      ($wday, $month, $mday, $time, $year, $myip, $ip, $method, $file)
                = split;
   }
   $file =~ tr/A-Z/a-z/ if $filenamestolowercase;
   # *** added by CE to see if this will get rid of odd things that turn up in the log
   # may inflate the figures but I suspect that a lot of these things are handled
   # successfully by the browser/server
   $file =~ tr/"//d;    
   $file =~ s/\.html/\.htm/;
   $file =~ s/>.*//;
   # if this entry is not one that we are ignoring
   # and if it is one that we are focussing on
   # then process it
   if (
      !&ignorethisip($ip) &&
      !&ignorethispath($file) &&
      !&ignorethisext($file) &&
      &focusonthisip($ip) &&
      &focusonthispath($file) &&
      &focusonthisext($file)
    ) {
      # gopher log has one less field than http log
      $file = $method if $logtype =~ /^g/i;
      # look up domain for this ip address
      $domain = &addname($ip);

      if ($extractonly) {
         # log the record regardless of error status
         print PSYLOG "$_\n";
      } else {
         # if it was a failed access, put it in the error log
         $stat_class = int($status/100);            
         if ($stat_class != 2) {
            print ERRLOG "$_\n";
         } else {
            print PSYLOG "$_\n";
         }
         # also compile the statistics if not just extracting the data
	 if ($stat_class != 2) {
	    # sort out the error counting
	    if ($status != 304) {
               $error++;
               # @@@@@@@ this is where to add compilation of errlog.htm
            } else {
               $e304++;
            }
	    $errcl{$stat_class}++;
	    $error{$status}++;
	    $toterrcl{$stat_class}++;
	    $toterror{$status}++;
            if (!$errc_lab{$stat_class}) {
               $errc_lab{$stat_class} = "Unclassified error type";
            }
            if (!$err_lab{$status}) {
               $err_lab{$status} = "Unclassified error type";
            }

	 } 
    if (($stat_class == 2) || ($stat_class == 304)) {
         # not an error according to me 
         # so add this access to cumulative counts
         # 3 separate counts are kept - overall, home and outside
         $access++;  # note that e304 are added separately, see above
         $totalbytes += $bytes;
         $file{$file}++;
         $ip{$ip}++;
         if (&ishomeip($ip)) {
            $homeaccess++;
            $homebytes += $bytes;
            $homefile{$file}++;
            $homeip{$ip}++;
            $domain = $homedomain;
         }
         else {
            $outsideaccess++;
            $outsidebytes += $bytes;
            $outsidefile{$file}++;
            $outsideip{$ip}++;
         }
         $domain{$domain}++;
    } # end of 
	 # if month in current log entry not equal to month
         # being processed then it is the start of a new month
         if (($currentmonth ne $month) || ($currentlogtype ne $logtype)) {
               &newmonth;
               $currentmonth = $month;
               $currentyear = $year;
               $currentlogtype = $logtype;
         }



      } # end of if (!&extractonly)
   }   # end of processing entry
} # sub doentry
# --------------------------------------------------------------------------


# --------- *** moved here by CE to make easier to read -------------------
sub config {
   # --------------------------------------------------------------------------
   # CONFIGURE THESE VARIABLES
   #
   # File and path names should be absolute (e.g. C:/LOGS/)
   # or relative to current directory (e.g. ./LOGS/)
   # In Perl for Windows NT use forward slash (not backslash)
   # in pathnames (e.g. C:/LOGS/)
   # Directory names must always end with trailing slash
   # (e.g. ./ for current directory - don't just leave blank)
   #
   # Lists of items consist of a list surrounded by brackets, with
   # each item in the list separated by commas
   # e.g. @ignoreip = ('158.143.104.174', '158.143.104.209');
   # Lists of IP addresses can also include DNS names
   # e.g. @homeip = ('158.143', 'lse.ac.uk');
   #
   # Flags can be set to 1 (meaning 'true' or 'enabled') or
   # set to 0 or not defined (meaning 'false' or 'disabled')

   # @homeip           --  list of ip addresses of local users (e.g. ('158.143.') )
   # $homedomain       --  domain of local users (e.g. lse.ac.uk)
   #
   # $homename         --  name of home site (e.g. LSE, London School of Economics)
   # $servername       --  name of server (not always the same as the name of
   #                       the home site - may be name of PART of overall server
   #                       e.g. Law Department)
   #
   # $topfiles         --  present statistics for this many most popular files (e.g. 20)
   # $topips           --  present statistics for this many most attentive ip addresses
   #
   # $dnsnames         --  file containing DNS names (e.g. ./dnsnames.lst)
   #
   # $inputdir         --  directory containing https & gophers log files (e.g. /logs/)
   # $outputdir        --  directory to write usage statistics reports (e.g. ./)
   #
   # @ignoreip         --  list of IP addresses to ignore (e.g. ('158.143.104.174') )
   # @myownip          --  list of IP addresses which are administrative accesses
   # $i_ipcomment      --  comment about excluded IP addresses for reports
   # @ignorepath       --  list of paths/files to ignore (e.g. () )
   # @ignoreext        --  list of file extensions to ignore (e.g. ('.gif')
   # @focusonip        --  list of IP addresses to focus on (e.g. () )
   # @focusonpath      --  list of paths/files to focus on (e.g. () )
   # $f_pathcomment    --  comment abut path focus for reports
   # @focusonext       --  list of file extensions to focus on (e.g. ('.htm')
   #
   # $verbosemode      --  flag - disable if you don't want the detailed output
   #                       musage gives to indicate what it's doing
   # $lookupdnsnames   --  flag - disable if you don't want to look up dnsname of
   #                       each ip address accessing your server
   # $commonlogformat  --  flag - disable for use with EMWAC's free server, enable
   #                       to use MUSAGE with logs from an http server that uses
   #                       the common log format (e.g. the professional version
   #                       of EMWAC's server)
   # $filenamestolowercase
   #                   --  flag - enable to convert all filenames (URLs) to lower
   #                       case before processing them
   # $logfilepattern   --  the pattern used to recognise log files (e.g. EMWAC's
   #                       server's log files match *.log - translate this into
   #                       a Perl regular expression and you get .+\.log)
   #
   # %servertype       --  PERL associative array - MUSAGE looks at the
   #                       first letter of the log file to deduce what kind
   #                       of server it comes from (e.g. EMWAC gopher, WWW
   #                       & WAIS logs start with g, h & w respectively;
   #                       CERN3 httpd proxy, access & cache logs start
   #                       with p, a & c respectively)
   # $debug            --  invokes a debugging run on a small file
   # $debuglog         --  name of a debugging file
   $debug = 1;
   $debuglog = "/usr1/www/pages/mhs/psychotherapy/wwwstats/debug.log";
   $dnsnames    = "/usr1/www/pages/mhs/psychotherapy/wwwstats/dnsnames.lst";
   @homeip = ("192.153", "sghms.ac.uk");
   $homedomain = ("sghms.ac.uk");
   $homename = "SGHMS";
   $servername = "SGHMS Psychotherapy";
   @ignorepath = ();
   @ignoreext = ();
   @focusonip = ();
   @myownip  = ("194.80.201.27","cevanpc.sghms.ac.uk");
   @focusonext = ();
   $verbosemode = 0;
   $lookupdnsnames = 1;
   $commonlogformat = 1;
   $filenamestolowercase = 0;
   # this block are default settings usually overridden by musage.set data
   $topfiles = 50;
   $topips   = 30;
   $inputdir    = "/usr1/www/pages/psychotherapy/wwwstats/";
   $outputdir   = "/dev/null/";
   @focusonpath = ("/mhs/psychotherapy/");
   $f_pathcomment = "files in any Psychotherapy Section directories considered";
   @ignoreip = ("194.80.201.27","cevanpc.sghms.ac.uk","infoserv.sghms.ac.uk","192.153.12.1","192.153.12.30");
   $i_ipcomment = "does not include my own accesses as administrator";


   $servertype{"g"} = "Gopher";
   $servertype{"h"} = "WWW";
   $servertype{"w"} = "WAIS";
   $servertype{"p"} = "Proxy";
   $servertype{"a"} = "Access";
   $servertype{"c"} = "Cache";

   $logfilepattern = "ptotal.log";       # CE shifted this around to deal with local arrangements
   $psylog         = "psylog.log";
   $errlog         = "errlog.log";
   # *** added by CE to get date of statistics run into the output
   $dated = 1;


   # *** the next three variables contain text to go at the end of the menu, summary and error files
   # *** respectively

   $btm_m   = <<END;
   <HR>
   File created using Perl script:
   <A HREF="/mhs/psychotherapy/musage3.prl">musage3.prl</A> adapted by Chris Evans
   <A HREF="mailto:C.Evans@sghms.ac.uk">C.Evans@sghms.ac.uk</A> from MUSAGE.TXT by Nick Philips
   <UL>
   <LI><A HREF="/mhs/psychotherapy/stats.htm">overview of general visit statistics</A>
   <LI><A HREF="/mhs/psychotherapy/intro.htm">psychotherapy pages intro.</A>
   <LI><A HREF="http://www.sghms.ac.uk/">S.G.H.M.S. main WWW pages</A>
   </UL>
   </BODY>
   </HTML>
END

   $btm_s   = <<END;
   <P>
   A <A HREF="$report.TXT">complete usage</A> is also available.
   <HR>
   File created using Perl script:
   <A HREF="/mhs/psychotherapy/musage3.prl">musage3.prl</A> adapted by Chris Evans
   <A HREF="mailto:C.Evans@sghms.ac.uk">C.Evans@sghms.ac.uk</A> from MUSAGE.TXT by Nick Philips
   <UL>
   <LI><A HREF="/mhs/psychotherapy/stats.htm">overview of general visit statistics</A>
   <LI><A HREF="/mhs/psychotherapy/intro.htm">psychotherapy pages intro.</A>
   <LI><A HREF="http://www.sghms.ac.uk/">S.G.H.M.S. main WWW pages</A>
   </UL>
   </BODY>
   </HTML>
END

   $btm_e   = <<END;
   <P>
   An <A HREF="$errlog">error log</A> is available.
   <HR>
   File created using Perl script:
   <A HREF="/mhs/psychotherapy/musage3.prl">musage3.prl</A> adapted by Chris Evans
   <A HREF="mailto:C.Evans@sghms.ac.uk">C.Evans@sghms.ac.uk</A> from MUSAGE.TXT by Nick Philips
   <UL>
   <LI><A HREF="/mhs/psychotherapy/stats.htm">overview of general visit statistics</A>
   <LI><A HREF="/mhs/psychotherapy/intro.htm">psychotherapy pages intro.</A>
   <LI><A HREF="http://www.sghms.ac.uk/">S.G.H.M.S. main WWW pages</A>
   </UL>
   </BODY>
   </HTML>
END

} # sub config



__END__
:endofperl