#!/bin/bash

###
### AUTHORSTATS - Get author statistics from Internet Drafts and RFCs
###
### Version 2.5.2
###
### Written in 2005-2008 by Jari Arkko
### Donated to the public domain.
###
###   2.5.0   Cleaned up HTML
###   2.5.1   Added support for authors moving, fixed country and company capitalization
###   2.5.2   Fixed monthly publication graph style
###
### Usage:
###
###   authorstats draft-foo-00.txt ...
###

###
### Initialize
###

debug=0
doctype=draft
authordata=none
wgdata=none
oldwgdata=none
rfcdata=none
chairdata=none
iabdata=none
iaocdata=none
popdata=none
topic="active I-Ds"
tmpbase=/tmp/$$-tmp

###
### Process options
###

for i in $*
do
  case x$1 in

    (x--doctype)shift
                doctype=$1
                shift;;

    (x--debug)  debug=1;
                shift;;

    (x--data)   shift;
                authordata=$1;
                shift;;

    (x--wgs)    shift
                wgdata=$1;
                shift;;

    (x--oldwgs) shift
                oldwgdata=$1;
                shift;;

    (x--rfcs)   shift
                rfcdata=$1;
                shift;;

    (x--chairs) shift
                chairdata=$1;
                shift;;

    (x--iab)    shift
                iabdata=$1;
                shift;;

    (x--iaoc)   shift
                iaocdata=$1;
                shift;;

    (x--population)
                shift
                popdata=$1;
                shift;;

    (x--topic)  shift
                topic="$1";
                shift;;

    (x-*)       echo 'authorstats: Unrecognized option -- exit';
                exit 1;;

  esac
done

###
### Get author data from the drafts
###

if [ x$authordata = xnone ]
then
  authordata=${tmpbase}-a
  getauthors $* > $authordata
fi

###
### Test that we have author data
###

if [ -s $authordata ]
then
  ok=ok
else
  echo 'There is no author data in file '$authordata' -- exit'
  exit 1
fi

###
### Get WG data from the IETF site
###

if [ x$wgdata = xnone ]
then
  (cd /tmp;
   rm -f wg-dir.html;
   wget -q http://www.ietf.org/html.charters/wg-dir.html)
  wgdata=/tmp/wg-dir.html
fi

###
### Get old WG data from the IETF site
###

if [ x$oldwgdata = xnone ]
then
  (cd /tmp;
   rm -f oldindex.html;
   wget -q -O oldindex.html http://www.ietf.org/html.charters/OLD/index.html)
  oldwgdata=/tmp/oldindex.html
fi

###
### Get RFC data from the IETF tools site
###

if [ x$rfcdata = xnone ]
then
  (cd /tmp;
   rm -f rfc_id.txt;
   wget -q http://tools.ietf.org/id/rfc_id.txt)
  rfcdata=/tmp/rfc_id.txt
fi

###
### Get chair/AD data from the IETF site
###

if [ x$chairdata = xnone ]
then
  (cd /tmp;
   rm -f rfc_id.txt;
   wget -q http://www.ietf.org/ietf/1wg-summary.txt)
  chairdata=/tmp/1wg-summary.txt
fi

###
### Get IAB data from the IAB site
###

if [ x$iabdata = xnone ]
then
  (cd /tmp;
   rm -f rfc_id.txt;
   wget -q http://www.iab.org/about/members.html)
  iabdata=/tmp/members.html
fi

###
### Get IAOC data from the temporary IAOC site
###

if [ x$iaocdata = xnone ]
then
  (cd /tmp;
   rm -f index.html
   #wget -q http://koi.uoregon.edu/~iaoc/index.html
   )
  iaocdata=/tmp/index.html
fi

###
### Get country population data from the wikipedia site
###

if [ x$popdata = xnone ]
then
  (cd /tmp;
   rm -rf List_of_countries_by_population List_of_countries_by_population.html List_of_countries_by_population.files;
   wget -q http://en.wikipedia.org/wiki/List_of_countries_by_population)
  popdata=/tmp/List_of_countries_by_population
fi

###
### Process WG data
###

wgs=/tmp/wgdata.txt
rm -f $wgs

if [ $debug = 1 ]
then
  echo authorstats: parsing wg data
fi

awk < $wgdata '
BEGIN {
  area = "";
}

/.* Area<.h2><.a>$/ {
  i1 = index($0,"<h2>");
  area = substr($0,i1+4);
  i2 = index(area," Area<");
  area = substr(area,1,i2 - 1);
  next;
}

/-charter.html/ {
  if (area != "") {
    i1 = index($0,"html.charters/");
    wg = substr($0,i1 + 5 + 8 + 1);
    i2 = index(wg,"-charter.html");
    wg = substr(wg,1,i2 - 1);
    printf("0-wgareadefinition:%s:%s\n", area, wg);
  }
}

/.*/ {
  next;
}

END {
}
' > $wgs

###
### Process old WG data
###

oldwgs=/tmp/oldindex.txt
rm -f $oldwgs

if [ $debug = 1 ]
then
  echo authorstats: parsing old wg data
fi

awk < $oldwgdata '
BEGIN {
  area = "";
}

/.* Area<.[hH]2>/ {
  i1 = index($0,"<h2>");
  area = substr($0,i1+4);
  i2 = index(area," Area<");
  area = substr(area,1,i2 - 1);
  #printf("9-debug: saw area %s on line %s\n", area, $0);
}

/-charter.html/ {
  #printf("9-debug: saw potential wg on line %s\n", $0);
  if (area != "") {
    i1 = index($0,"HREF=");
    wg = substr($0,i1 + 5 + 1 );
    i2 = index(wg,"-charter.html");
    wg = substr(wg,1,i2 - 1);
    printf("0-oldwgareadefinition:%s:%s\n", area, wg);
  }
}

/.*/ {
  next;
}

END {
}
' > $oldwgs


###
### Process IAB data
###

if [ $debug = 1 ]
then
  echo authorstats: parsing iab data
fi

iab=/tmp/iab.txt
rm -f $iab
awk < $iabdata '
BEGIN {
  iniab = 0;
}
/IAB MEMBERS/ {
  iniab = 1;
  next;
}
/EX-OFFICIO/ {
  iniab = 0;
  next;
}
/vacancy/ {
  next;
}
/blue-dot/ {
  if (iniab) {
    z = $0;
    gsub(/^.*[<]b[>]/,"",z);
    gsub(/[<].b[>].*$/,"",z);
    printf("z3-iab:%s\n",z);
  }
  next;
}
/.*/ {
  next;
}
END {
}' |
sed 's/&auml;/a/g' |
sed 's/&ouml;/o/g' |
cat > $iab

###
### Process IAOC data
###

if [ $debug = 1 ]
then
  echo authorstats: parsing iaoc data
fi

iaoc=/tmp/iaoc.txt
rm -f $iaoc
awk < $iaocdata '
BEGIN {
  iniaoc = 0;
}

/IAOC Membership/ {
  iniaoc = 1;
  #printf("starting...\n");
}

/IAOC Responsibilities/ {
  iniaoc = 0;
  #printf("ending...\n");
}

/^[<]li[>] .*,/ {
  #printf("candidate: %s\n", $0);
  if (iniaoc) {
    z = $0;
    z = substr(z,6);
    z = substr(z,1,index(z,",") - 1);
    printf("z4-iaoc:%s\n", z);
  }
  next;
}

/.*/ {
  next;
}
END {
}' |
sed 's/Kurtis Lindquist/Kurtis Lindqvist/' |
cat > $iaoc

###
### Process chair data
###

if [ $debug = 1 ]
then
  echo authorstats: parsing chair data
fi

chairs=/tmp/chairdata.txt
rm -f $chairs
awk < $chairdata '
BEGIN {
  area = "";
  wg = "";
}

/IETF Working Group Summary .By Area./ {
  next;
}

/-----------------/ {
  next;
}

/^ +WG Mail:/ {
  next;
}

/^ +To Join:/ {
  next;
}

/^ +In Body:/ {
  next;
}

/^ +Archive:/ {
  next;
}

/^[A-Z].* Area .[a-z]+.$/ {
  pos = index($0," Area ");
  area = substr($0,1,pos - 1);
  wg = "";
  next;
}

/^[A-Z].* [(].*[)]$/ {
  z = $0;
  gsub(/^.*[(]/,"",z);
  gsub(/[)]$/,"",z);
  wg = z;
}

/^ +[A-Za-z .0-9():]+ [<].*[>]$/ {
  z = $0;
  gsub(/ +Chairs.s.: +/,"",z);
  gsub(/^ +/,"",z);
  gsub(/ [<].*$/,"",z);
  if (wg == "") {
    printf("z1-ad:%s:%s\n", area, z);
  } else {
    printf("z2-chair:%s:%s\n", wg, z);
  }
  next;
}

END {
}
' |
sed 's/Kurt Zeilenga/Kurt D. Zeilenga/' |
sed 's/Russ Housley/Russell Housley/' |
sed 's/Gregory M[.] Lebovitz/Gregory Lebovitz/' |
cat > $chairs

###
### Process population data
###

pops=/tmp/popdata.txt
rm -f $pops

cat $popdata |
tr -d "'" |
tr ABCDEFGHIJKLMNOPQRSTUVWXYZ abcdefghijklmnopqrstuvwxyz |
sed 's/[Ff]lag[ _]+of[ _]+//g' |
sed 's/[Ff]lag of //g' |
sed 's/svg:/:/g' |
fgrep -v 'cite_ref-overseas_france' | # avoid overriding france
tee /tmp/poptmp.txt |
awk '
function isalpha(c) {
  return((c >= "a" && c <= "z") ||
         (c >= "A" && c <= "Z"));
}

function findchar(c) {
  for (y = 1; y <= 255; y++) {
    buf = sprintf("%c", y);
    if (buf == c) return(y);
  }
  return(0);
}

function printstring(what,s) {
  printf("# debug 4: %s\n", what);
  for (h = 1; h <= length(s); h++) {
      printf("#          %s (%d)\n", substr(s,h,1), findchar(substr(s,h,1)));
  }
}

BEGIN {
  inskip=1;

  #
  # Not all names in Wikipedia are in the same form
  # as the ones used by getauthors. Map the special
  # ones so that they match.
  #

  mapit["belgium civil"] = "belgium";
  mapit["the czech republic"] = "czech republic";
  mapit["the solomon islands"] = "solomon islands";
  mapit["the peoples republic of china"] = "china";
  mapit["peoples republic of china"] = "china";
  mapit["the united states"] = "usa";
  mapit["united states"] = "usa";
  mapit["the united kingdom"] = "united kingdom";
  mapit["republic of ireland"] = "ireland";
  mapit["netherlands"] = "the netherlands";
  mapit["the republic of china"] = "taiwan";
  mapit["republic of china"] = "taiwan";
  mapit["the united arab emirates"] = "united arab emirates";
  mapit["the central african republic"] = "central african republic";
  mapit["the gambia"] = "gambia";
  mapit["the comoros"] = "comoros";
  mapit["the bahamas"] = "bahamas";
  mapit["the vatican city"] = "vatican city";
}

/^.td.1..td.$/ {
  inskip = 0;
  next;
}

/^.td align="left"./ {
  if (!inskip) {
    s = $0;
    while (length(s) > 0 && substr(s,1,6) != "title=") s = substr(s,2);
    s = substr(s,8);
    n = "";
    while (length(s) > 0 && substr(s,1,1) != sprintf("%c",34)) {
      if (isalpha(substr(s,1,1)) || substr(s,1,1) == " " ||
          substr(s,1,1) == "-" || substr(s,1,1) == "_") {
        n = n substr(s,1,1);
      }
      s = substr(s,2);
    }
    country = tolower(n);
    gsub(/[.]*svg/,"",country);
#   printf("# debug 1: %s - %s\n", n, country);
#   printf("# debug 2: %s - %s\n", country, mapit[country]);
#           country == "republic of ireland");
#    printstring("input", country);
#    printstring("const", "republic of ireland");
    if (mapit[country] != "") country = mapit[country];
  }
  next;
}

/^.td.[0-9][0-9.,e+]+..td.$/ {
  if (inskip == 0 && country != "") {
    p = substr($0,5);
    p = substr(p,1,length(p) - 5);
    gsub(/,/,"",p);
    pv = p + 0;
    printf("0-population:%s:%d:%d\n",country,pv,NR);
#   printf("# debug 4: %s\n", country);
    country = "";
  }
  next;
}

/^.td.[0-9][0-9,]+.sup id.*..sup...td.$/ {
  if (inskip == 0 && country != "") {
    p = substr($0,5);
    p = substr(p,1,index(p,"sup") - 2);
    gsub(/,/,"",p);
    printf("0-population:%s:%s:%d\n",country,p,NR);
#   printf("# debug 5: %s\n", country);
    country = "";
  }
  next;
}

/.*/ {
  next;
}

' > $pops

###
### Process RFC data
###

if [ $debug = 1 ]
then
  echo authorstats: parsing rfc data
fi

rfcs=/tmp/rfcdata.txt
rm -f $rfcs

###
### Safety check to avoid bad characters etc
###

if [ $debug = 1 ]
then
  echo authorstats: process the database
fi

(cat $wgs;
 cat $pops;
 cat $chairs;
 cat $iab;
 cat $iaoc;
 cat $authordata) |
sort |
tr -d '*?"{}/%";&<>\341\207\351\355' |
tr -d "'" |
tee /tmp/inputdb.txt |

###
### Process the database
###

awk -v topic="$topic" \
    -v doctype=$doctype \
    -v debug=$debug \
'
function inittabs()
{
  areaacronym["applications"] = "app";
  areaacronym["general"] = "gen";
  areaacronym["internet"] = "int";
  areaacronym["ip: next generation"] = "nextgen";
  areaacronym["network management area"] = "ops";
  areaacronym["operational requirements"] = "ops";
  areaacronym["operations and management"] = "ops";
  areaacronym["osi integration"] = "osi";
  areaacronym["real-time applications and infrastructure"] = "rai";
  areaacronym["routing"] = "rtg";
  areaacronym["security"] = "sec";
  areaacronym["sub-ip"] = "sub";
  areaacronym["transport"] = "tsv";
  areaacronym["user services"] = "user";

  monthname["1"] = "January";
  monthname["2"] = "February";
  monthname["3"] = "March";
  monthname["4"] = "April";
  monthname["5"] = "May";
  monthname["6"] = "June";
  monthname["7"] = "July";
  monthname["8"] = "August";
  monthname["9"] = "September";
  monthname["10"] = "October";
  monthname["11"] = "November";
  monthname["12"] = "December";

  isineu["finland"] = 1;
  isineu["sweden"] = 1;
  isineu["denmark"] = 1;
  isineu["united kingdom"] = 1;
  isineu["germany"] = 1;
  isineu["the netherlands"] = 1;
  isineu["france"] = 1;
  isineu["belgium"] = 1;
  isineu["italy"] = 1;
  isineu["spain"] = 1;
  isineu["portugal"] = 1;
  isineu["austria"] = 1;
  isineu["cyprys"] = 1;
  isineu["czech republic"] = 1;
  isineu["estonia"] = 1;
  isineu["greece"] = 1;
  isineu["hungary"] = 1;
  isineu["ireland"] = 1;
  isineu["latvia"] = 1;
  isineu["lithuania"] = 1;
  isineu["luxembourg"] = 1;
  isineu["malta"] = 1;
  isineu["poland"] = 1;
  isineu["slovakia"] = 1;
  isineu["slovenia"] = 1;
  isineu["monaco"] = 1;

  population["european union"] = 0;
  
  for (eucountry in isineu) {
    continentof[eucountry] = "europe";
  }
  continentof["switzerland"] = "europe";
  continentof["norway"] = "europe";
  continentof["bulgaria"] = "europe";
  continentof["romania"] = "europe";
  continentof["russia"] = "europe";
  continentof["st. lucia"] = "europe";
  continentof["ukraine"] = "europe";
  continentof["turkey"] = "europe";
  continentof["croatia"] = "europe";

  continentof["australia"] = "australia";
  continentof["new zealand"] = "australia";
  continentof["tonga"] = "australia";
  continentof["solomon islands"] = "australia";

  continentof["armenia"] = "asia";
  continentof["east timor"] = "asia";
  continentof["south korea"] = "asia";
  continentof["pakistan"] = "asia";
  continentof["vietnam"] = "asia";
  continentof["japan"] = "asia";
  continentof["china"] = "asia";
  continentof["israel"] = "asia";
  continentof["jordan"] = "asia";
  continentof["hong kong"] = "asia";
  continentof["thailand"] = "asia";
  continentof["india"] = "asia";
  continentof["syria"] = "asia";
  continentof["united arab emirates"] = "asia";
  continentof["singapore"] = "asia";
  continentof["saudi arabia"] = "asia";
  continentof["lebanon"] = "asia";
  continentof["mongolia"] = "asia";
  continentof["cocos (keeling) isl."] = "asia";
  continentof["taiwan"] = "asia";
  continentof["macau"] = "asia";
  continentof["malaysia"] = "asia";
  continentof["bhutan"] = "asia";

  continentof["sudan"] = "africa";
  continentof["tunisia"] = "africa";
  continentof["egypt"] = "africa";
  continentof["st. tome and principe"] = "africa";

  continentof["netherland antilles"] = "south america";
  continentof["venezuela"] = "south america";
  continentof["aruba"] = "south america";
  continentof["panama"] = "south america";
  continentof["chile"] = "south america";
  continentof["brazil"] = "south america";
  continentof["argentina"] = "south america";
  continentof["costa rica"] = "south america";
  continentof["uruguay"] = "south america";
  continentof["st. lucia"] = "south america";
  continentof["antigua and barbuda"] = "south america";

  continentof["mexico"] = "north america";
  continentof["usa"] = "north america";
  continentof["canada"] = "north america";
  continentof["greenland"] = "north america";

  continentof["antarctica"] = "antarctica"

  monthnameshort[1] = "jan";
  monthnameshort[2] = "feb";
  monthnameshort[3] = "mar";
  monthnameshort[4] = "apr";
  monthnameshort[5] = "may";
  monthnameshort[6] = "jun";
  monthnameshort[7] = "jul";
  monthnameshort[8] = "aug";
  monthnameshort[9] = "sep";
  monthnameshort[10] = "oct";
  monthnameshort[11] = "nov";
  monthnameshort[12] = "dec";
  
  malefirstname["abraham"] = 1;
  malefirstname["adam"] = 1;
  malefirstname["adrian"] = 1;
  malefirstname["ajit"] = 1;
  malefirstname["alan"] = 1;
  malefirstname["albert"] = 1;
  malefirstname["alec"] = 1;
  malefirstname["alex"] = 1;
  malefirstname["alexander"] = 1;
  malefirstname["alexandru"] = 1;
  malefirstname["alfred"] = 1;
  malefirstname["alper"] = 1;
  malefirstname["alpesh"] = 1;
  malefirstname["anders"] = 1;
  malefirstname["andre"] = 1;
  malefirstname["andrei"] = 1;
  malefirstname["andrew"] = 1;
  malefirstname["angelica"] = 1;
  malefirstname["angelos"] = 1;
  malefirstname["anthony"] = 1;
  malefirstname["anton"] = 1;
  malefirstname["antti"] = 1;
  malefirstname["ari"] = 1;
  malefirstname["art"] = 1;
  malefirstname["arthur"] = 1;
  malefirstname["artur"] = 1;
  malefirstname["atle"] = 1;
  malefirstname["avi"] = 1;
  malefirstname["barney"] = 1;
  malefirstname["barry"] = 1;
  malefirstname["basavaraj"] = 1;
  malefirstname["ben"] = 1;
  malefirstname["benjamin"] = 1;
  malefirstname["bernard"] = 1;
  malefirstname["bernhard"] = 1;
  malefirstname["bernie"] = 1;
  malefirstname["bert"] = 1;
  malefirstname["bertrand"] = 1;
  malefirstname["bill"] = 1;
  malefirstname["billy"] = 1;
  malefirstname["bjorn"] = 1;
  malefirstname["bob"] = 1;
  malefirstname["brad"] = 1;
  malefirstname["brent"] = 1;
  malefirstname["bret"] = 1;
  malefirstname["brett"] = 1;
  malefirstname["brian"] = 1;
  malefirstname["bruce"] = 1;
  malefirstname["bryan"] = 1;
  malefirstname["burt"] = 1;
  malefirstname["burton"] = 1;
  malefirstname["butler"] = 1;
  malefirstname["buz"] = 1;
  malefirstname["buzz"] = 1;
  malefirstname["carlo"] = 1;
  malefirstname["carl-uno"] = 1;
  malefirstname["carsten"] = 1;
  malefirstname["cesar"] = 1;
  malefirstname["charles"] = 1;
  malefirstname["charley"] = 1;
  malefirstname["charlie"] = 1;
  malefirstname["chip"] = 1;
  malefirstname["chris"] = 1;
  malefirstname["christian"] = 1;
  malefirstname["christina"] = 1;
  malefirstname["christophe"] = 1;
  malefirstname["christopher"] = 1;
  malefirstname["chuck"] = 1;
  malefirstname["cliff"] = 1;
  malefirstname["clifford"] = 1;
  malefirstname["colin"] = 1;
  malefirstname["craig"] = 1;
  malefirstname["cullen"] = 1;
  malefirstname["curt"] = 1;
  malefirstname["curtis"] = 1;
  malefirstname["dan"] = 1;
  malefirstname["daniel"] = 1;
  malefirstname["danny"] = 1;
  malefirstname["dave"] = 1;
  malefirstname["david"] = 1;
  malefirstname["dean"] = 1;
  malefirstname["dennis"] = 1;
  malefirstname["derek"] = 1;
  malefirstname["dick"] = 1;
  malefirstname["dimitri"] = 1;
  malefirstname["dimitrios"] = 1;
  malefirstname["dirk"] = 1;
  malefirstname["dirk-willem"] = 1;
  malefirstname["don"] = 1;
  malefirstname["donald"] = 1;
  malefirstname["dorian"] = 1;
  malefirstname["doug"] = 1;
  malefirstname["doughan"] = 1;
  malefirstname["douglas"] = 1;
  malefirstname["drew"] = 1;
  malefirstname["duncan"] = 1;
  malefirstname["dwight"] = 1;
  malefirstname["ed"] = 1;
  malefirstname["edgar"] = 1;
  malefirstname["edmund"] = 1;
  malefirstname["eduard"] = 1;
  malefirstname["eduardo"] = 1;
  malefirstname["edward"] = 1;
  malefirstname["edwin"] = 1;
  malefirstname["einar"] = 1;
  malefirstname["eliot"] = 1;
  malefirstname["elmer"] = 1;
  malefirstname["enrique"] = 1;
  malefirstname["eric"] = 1;
  malefirstname["erik"] = 1;
  malefirstname["ernesto"] = 1;
  malefirstname["eugene"] = 1;
  malefirstname["fabio"] = 1;
  malefirstname["farid"] = 1;
  malefirstname["farooq"] = 1;
  malefirstname["fernando"] = 1;
  malefirstname["florent"] = 1;
  malefirstname["francis"] = 1;
  malefirstname["francois"] = 1;
  malefirstname["frangois"] = 1;
  malefirstname["frank"] = 1;
  malefirstname["fred"] = 1;
  malefirstname["frederic"] = 1;
  malefirstname["fredrik"] = 1;
  malefirstname["gabriel"] = 1;
  malefirstname["gary"] = 1;
  malefirstname["geoff"] = 1;
  malefirstname["geoffrey"] = 1;
  malefirstname["georg"] = 1;
  malefirstname["george"] = 1;
  malefirstname["georgios"] = 1;
  malefirstname["gerald"] = 1;
  malefirstname["gerard"] = 1;
  malefirstname["gerben"] = 1;
  malefirstname["gerry"] = 1;
  malefirstname["gery"] = 1;
  malefirstname["gilbert"] = 1;
  malefirstname["gilles"] = 1;
  malefirstname["glen"] = 1;
  malefirstname["glenn"] = 1;
  malefirstname["gonzalo"] = 1;
  malefirstname["gopal"] = 1;
  malefirstname["goran"] = 1;
  malefirstname["gordon"] = 1;
  malefirstname["graham"] = 1;
  malefirstname["grant"] = 1;
  malefirstname["greg"] = 1;
  malefirstname["gregg"] = 1;
  malefirstname["gregor"] = 1;
  malefirstname["gregory"] = 1;
  malefirstname["guenter"] = 1;
  malefirstname["gunnar"] = 1;
  malefirstname["guy"] = 1;
  malefirstname["hannes"] = 1;
  malefirstname["hanns"] = 1;
  malefirstname["hans"] = 1;
  malefirstname["hans-bernhard"] = 1;
  malefirstname["hans-werner"] = 1;
  malefirstname["harald"] = 1;
  malefirstname["harri"] = 1;
  malefirstname["harry"] = 1;
  malefirstname["henning"] = 1;
  malefirstname["henrik"] = 1;
  malefirstname["henry"] = 1;
  malefirstname["hesham"] = 1;
  malefirstname["holger"] = 1;
  malefirstname["horst"] = 1;
  malefirstname["igor"] = 1;
  malefirstname["isaac"] = 1;
  malefirstname["ivano"] = 1;
  malefirstname["jack"] = 1;
  malefirstname["jacob"] = 1;
  malefirstname["jake"] = 1;
  malefirstname["jakob"] = 1;
  malefirstname["james"] = 1;
  malefirstname["jan"] = 1;
  malefirstname["jari"] = 1;
  malefirstname["jarkko"] = 1;
  malefirstname["jarno"] = 1;
  malefirstname["jason"] = 1;
  malefirstname["javier"] = 1;
  malefirstname["jean-louis"] = 1;
  malefirstname["jean-marc"] = 1;
  malefirstname["jean-michel"] = 1;
  malefirstname["jean-philippe"] = 1;
  malefirstname["jean-yves"] = 1;
  malefirstname["jed"] = 1;
  malefirstname["jeff"] = 1;
  malefirstname["jeffery"] = 1;
  malefirstname["jeffrey"] = 1;
  malefirstname["jeremy"] = 1;
  malefirstname["jerome"] = 1;
  malefirstname["jerry"] = 1;
  malefirstname["jesse"] = 1;
  malefirstname["jim"] = 1;
  malefirstname["jimmy"] = 1;
  malefirstname["jinhyeock"] = 1;
  malefirstname["jinmei"] = 1;
  malefirstname["joachim"] = 1;
  malefirstname["joe"] = 1;
  malefirstname["joel"] = 1;
  malefirstname["joerg"] = 1;
  malefirstname["johan"] = 1;
  malefirstname["johannes"] = 1;
  malefirstname["john"] = 1;
  malefirstname["johnny"] = 1;
  malefirstname["johnson"] = 1;
  malefirstname["jon"] = 1;
  malefirstname["jonathan"] = 1;
  malefirstname["jonne"] = 1;
  malefirstname["jordi"] = 1;
  malefirstname["jorge"] = 1;
  malefirstname["jose"] = 1;
  malefirstname["joseph"] = 1;
  malefirstname["josh"] = 1;
  malefirstname["joshua"] = 1;
  malefirstname["jouni"] = 1;
  malefirstname["juergen"] = 1;
  malefirstname["juha"] = 1;
  malefirstname["juha-pekka"] = 1;
  malefirstname["juin-hwey"] = 1;
  malefirstname["jukka"] = 1;
  malefirstname["jun-ichiro"] = 1;
  malefirstname["kaj"] = 1;
  malefirstname["karim"] = 1;
  malefirstname["karl"] = 1;
  malefirstname["keiichi"] = 1;
  malefirstname["keith"] = 1;
  malefirstname["ken"] = 1;
  malefirstname["kendall"] = 1;
  malefirstname["kenneth"] = 1;
  malefirstname["kenny"] = 1;
  malefirstname["kevin"] = 1;
  malefirstname["klaus"] = 1;
  malefirstname["krister"] = 1;
  malefirstname["kristofer"] = 1;
  malefirstname["kuntal"] = 1;
  malefirstname["kurt"] = 1;
  malefirstname["kurtis"] = 1;
  malefirstname["lakshminath"] = 1;
  malefirstname["lance"] = 1;
  malefirstname["larry"] = 1;
  malefirstname["lars"] = 1;
  malefirstname["lars-ake"] = 1;
  malefirstname["lars-erik"] = 1;
  malefirstname["leo"] = 1;
  malefirstname["leon"] = 1;
  malefirstname["leonard"] = 1;
  malefirstname["linus"] = 1;
  malefirstname["loa"] = 1;
  malefirstname["lou"] = 1;
  malefirstname["louis"] = 1;
  malefirstname["louis-nicolas"] = 1;
  malefirstname["luis"] = 1;
  malefirstname["luke"] = 1;
  malefirstname["lyndon"] = 1;
  malefirstname["madjid"] = 1;
  malefirstname["magnus"] = 1;
  malefirstname["marc"] = 1;
  malefirstname["marcello"] = 1;
  malefirstname["marco"] = 1;
  malefirstname["marcus"] = 1;
  malefirstname["mark"] = 1;
  malefirstname["markku"] = 1;
  malefirstname["markus"] = 1;
  malefirstname["marshall"] = 1;
  malefirstname["martin"] = 1;
  malefirstname["marvin"] = 1;
  malefirstname["masataka"] = 1;
  malefirstname["mathias"] = 1;
  malefirstname["mats"] = 1;
  malefirstname["matt"] = 1;
  malefirstname["matthew"] = 1;
  malefirstname["max"] = 1;
  malefirstname["maximilian"] = 1;
  malefirstname["michael"] = 1;
  malefirstname["michel"] = 1;
  malefirstname["mick"] = 1;
  malefirstname["miguel"] = 1;
  malefirstname["miguel-angel"] = 1;
  malefirstname["mika"] = 1;
  malefirstname["mikael"] = 1;
  malefirstname["mike"] = 1;
  malefirstname["mikhail"] = 1;
  malefirstname["mitchell"] = 1;
  malefirstname["mohamad"] = 1;
  malefirstname["mohamed"] = 1;
  malefirstname["mohan"] = 1;
  malefirstname["mohit"] = 1;
  malefirstname["morgan"] = 1;
  malefirstname["nathan"] = 1;
  malefirstname["nathaniel"] = 1;
  malefirstname["neal"] = 1;
  malefirstname["ned"] = 1;
  malefirstname["neil"] = 1;
  malefirstname["nevil"] = 1;
  malefirstname["nevin"] = 1;
  malefirstname["nic"] = 1;
  malefirstname["nick"] = 1;
  malefirstname["nico"] = 1;
  malefirstname["nicolas"] = 1;
  malefirstname["nigel"] = 1;
  malefirstname["nikolaj"] = 1;
  malefirstname["nils"] = 1;
  malefirstname["noel"] = 1;
  malefirstname["norm"] = 1;
  malefirstname["norman"] = 1;
  malefirstname["north"] = 1;
  malefirstname["olaf"] = 1;
  malefirstname["olafur"] = 1;
  malefirstname["oliver"] = 1;
  malefirstname["olivier"] = 1;
  malefirstname["osama"] = 1;
  malefirstname["oskar"] = 1;
  malefirstname["osmund"] = 1;
  malefirstname["owen"] = 1;
  malefirstname["paolo"] = 1;
  malefirstname["parviz"] = 1;
  malefirstname["pascal"] = 1;
  malefirstname["pasi"] = 1;
  malefirstname["patrick"] = 1;
  malefirstname["patrik"] = 1;
  malefirstname["paul"] = 1;
  malefirstname["pavel"] = 1;
  malefirstname["pedro"] = 1;
  malefirstname["pekka"] = 1;
  malefirstname["per"] = 1;
  malefirstname["perry"] = 1;
  malefirstname["pete"] = 1;
  malefirstname["peter"] = 1;
  malefirstname["petri"] = 1;
  malefirstname["phil"] = 1;
  malefirstname["philip"] = 1;
  malefirstname["philipp"] = 1;
  malefirstname["phill"] = 1;
  malefirstname["phillip"] = 1;
  malefirstname["pierre"] = 1;
  malefirstname["raj"] = 1;
  malefirstname["rajeev"] = 1;
  malefirstname["ralph"] = 1;
  malefirstname["rami"] = 1;
  malefirstname["ramon"] = 1;
  malefirstname["ran"] = 1;
  malefirstname["randall"] = 1;
  malefirstname["randy"] = 1;
  malefirstname["ray"] = 1;
  malefirstname["raymond"] = 1;
  malefirstname["reinaldo"] = 1;
  malefirstname["reiner"] = 1;
  malefirstname["rene"] = 1;
  malefirstname["rich"] = 1;
  malefirstname["richard"] = 1;
  malefirstname["rick"] = 1;
  malefirstname["rickard"] = 1;
  malefirstname["rob"] = 1;
  malefirstname["robert"] = 1;
  malefirstname["rodney"] = 1;
  malefirstname["roger"] = 1;
  malefirstname["rohan"] = 1;
  malefirstname["roland"] = 1;
  malefirstname["rolf"] = 1;
  malefirstname["rolland"] = 1;
  malefirstname["ron"] = 1;
  malefirstname["ronald"] = 1;
  malefirstname["ross"] = 1;
  malefirstname["roy"] = 1;
  malefirstname["rudolf"] = 1;
  malefirstname["russ"] = 1;
  malefirstname["russel"] = 1;
  malefirstname["russell"] = 1;
  malefirstname["ryan"] = 1;
  malefirstname["sam"] = 1;
  malefirstname["sami"] = 1;
  malefirstname["samuel"] = 1;
  malefirstname["scott"] = 1;
  malefirstname["sean"] = 1;
  malefirstname["sebastian"] = 1;
  malefirstname["shawn"] = 1;
  malefirstname["simon"] = 1;
  malefirstname["skip"] = 1;
  malefirstname["spencer"] = 1;
  malefirstname["stan"] = 1;
  malefirstname["stanislav"] = 1;
  malefirstname["stanley"] = 1;
  malefirstname["stefaan"] = 1;
  malefirstname["stefan"] = 1;
  malefirstname["stefano"] = 1;
  malefirstname["stephan"] = 1;
  malefirstname["stephen"] = 1;
  malefirstname["steqven"] = 1;
  malefirstname["steve"] = 1;
  malefirstname["steven"] = 1;
  malefirstname["stewart"] = 1;
  malefirstname["stig"] = 1;
  malefirstname["stuart"] = 1;
  malefirstname["suresh"] = 1;
  malefirstname["sven"] = 1;
  malefirstname["tatu"] = 1;
  malefirstname["ted"] = 1;
  malefirstname["tero"] = 1;
  malefirstname["terry"] = 1;
  malefirstname["theodore"] = 1;
  malefirstname["thierry"] = 1;
  malefirstname["thomas"] = 1;
  malefirstname["thommy"] = 1;
  malefirstname["tim"] = 1;
  malefirstname["timo"] = 1;
  malefirstname["timothy"] = 1;
  malefirstname["tom"] = 1;
  malefirstname["tomi"] = 1;
  malefirstname["ton"] = 1;
  malefirstname["toni"] = 1;
  malefirstname["tony"] = 1;
  malefirstname["trevor"] = 1;
  malefirstname["troy"] = 1;
  malefirstname["tuomas"] = 1;
  malefirstname["uri"] = 1;
  malefirstname["vern"] = 1;
  malefirstname["vernon"] = 1;
  malefirstname["vesa"] = 1;
  malefirstname["vic"] = 1;
  malefirstname["victor"] = 1;
  malefirstname["vijay"] = 1;
  malefirstname["vilho"] = 1;
  malefirstname["vince"] = 1;
  malefirstname["vincent"] = 1;
  malefirstname["vint"] = 1;
  malefirstname["vladimir"] = 1;
  malefirstname["walt"] = 1;
  malefirstname["walter"] = 1;
  malefirstname["warren"] = 1;
  malefirstname["wayne"] = 1;
  malefirstname["werner"] = 1;
  malefirstname["wesley"] = 1;
  malefirstname["weston"] = 1;
  malefirstname["wilfred"] = 1;
  malefirstname["will"] = 1;
  malefirstname["willem"] = 1;
  malefirstname["william"] = 1;
  malefirstname["wilson"] = 1;
  malefirstname["winston"] = 1;
  malefirstname["wolfgang"] = 1;
  malefirstname["yakov"] = 1;
  malefirstname["yoshihiro"] = 1;
  malefirstname["yves"] = 1;

  femalefirstname["alice"] = 1;
  femalefirstname["alison"] = 1;
  femalefirstname["allison"] = 1;
  femalefirstname["amy"] = 1;
  femalefirstname["angela"] = 1;
  femalefirstname["anita"] = 1;
  femalefirstname["anja"] = 1;
  femalefirstname["ann"] = 1;
  femalefirstname["anna"] = 1;
  femalefirstname["anne"] = 1;
  femalefirstname["annette"] = 1;
  femalefirstname["barbara"] = 1;
  femalefirstname["betty"] = 1;
  femalefirstname["carol"] = 1;
  femalefirstname["cathy"] = 1;
  femalefirstname["cecilia"] = 1;
  femalefirstname["celeste"] = 1;
  femalefirstname["cheryl"] = 1;
  femalefirstname["claire"] = 1;
  femalefirstname["claudia"] = 1;
  femalefirstname["cornelia"] = 1;
  femalefirstname["cristina"] = 1;
  femalefirstname["cyndi"] = 1;
  femalefirstname["cynthia"] = 1;
  femalefirstname["dana"] = 1;
  femalefirstname["debbie"] = 1;
  femalefirstname["debby"] = 1;
  femalefirstname["deborah"] = 1;
  femalefirstname["debra"] = 1;
  femalefirstname["deirdre"] = 1;
  femalefirstname["diana"] = 1;
  femalefirstname["donna"] = 1;
  femalefirstname["dorothy"] = 1;
  femalefirstname["eleanor"] = 1;
  femalefirstname["elena"] = 1;
  femalefirstname["elisabetta"] = 1;
  femalefirstname["elizabeth"] = 1;
  femalefirstname["elise"] = 1;
  femalefirstname["ellen"] = 1;
  femalefirstname["emily"] = 1;
  femalefirstname["eva"] = 1;
  femalefirstname["evan"] = 1;
  femalefirstname["eve"] = 1;
  femalefirstname["faye"] = 1;
  femalefirstname["geraldine"] = 1;
  femalefirstname["helen"] = 1;
  femalefirstname["hilarie"] = 1;
  femalefirstname["ileana"] = 1;
  femalefirstname["ingrid"] = 1;
  femalefirstname["irina"] = 1;
  femalefirstname["janet"] = 1;
  femalefirstname["jeanette"] = 1;
  femalefirstname["jeanne"] = 1;
  femalefirstname["jennifer"] = 1;
  femalefirstname["jessica"] = 1;
  femalefirstname["jill"] = 1;
  femalefirstname["joan"] = 1;
  femalefirstname["joann"] = 1;
  femalefirstname["joanne"] = 1;
  femalefirstname["judith"] = 1;
  femalefirstname["judy"] = 1;
  femalefirstname["julie"] = 1;
  femalefirstname["karen"] = 1;
  femalefirstname["kathleen"] = 1;
  femalefirstname["kathryn"] = 1;
  femalefirstname["laura"] = 1;
  femalefirstname["lauren"] = 1;
  femalefirstname["leena"] = 1;
  femalefirstname["leslie"] = 1;
  femalefirstname["lila"] = 1;
  femalefirstname["lili"] = 1;
  femalefirstname["linda"] = 1;
  femalefirstname["lisa"] = 1;
  femalefirstname["lixia"] = 1;
  femalefirstname["lorna"] = 1;
  femalefirstname["lorrie"] = 1;
  femalefirstname["lucy"] = 1;
  femalefirstname["madelyn"] = 1;
  femalefirstname["magdalena"] = 1;
  femalefirstname["marcia"] = 1;
  femalefirstname["margaret"] = 1;
  femalefirstname["maria"] = 1;
  femalefirstname["maria-carmen"] = 1;
  femalefirstname["marianne"] = 1;
  femalefirstname["marie"] = 1;
  femalefirstname["marilyn"] = 1;
  femalefirstname["marjorie"] = 1;
  femalefirstname["martha"] = 1;
  femalefirstname["mary"] = 1;
  femalefirstname["maryann"] = 1;
  femalefirstname["maureen"] = 1;
  femalefirstname["melinda"] = 1;
  femalefirstname["michelle"] = 1;
  femalefirstname["monique"] = 1;
  femalefirstname["nancy"] = 1;
  femalefirstname["nora"] = 1;
  femalefirstname["olga"] = 1;
  femalefirstname["patricia"] = 1;
  femalefirstname["peggy"] = 1;
  femalefirstname["petra"] = 1;
  femalefirstname["radia"] = 1;
  femalefirstname["rebecca"] = 1;
  femalefirstname["rosanna"] = 1;
  femalefirstname["ruth"] = 1;
  femalefirstname["sally"] = 1;
  femalefirstname["sandra"] = 1;
  femalefirstname["sandy"] = 1;
  femalefirstname["samita"] = 1;
  femalefirstname["sharon"] = 1;
  femalefirstname["sheila"] = 1;
  femalefirstname["silvia"] = 1;
  femalefirstname["sue"] = 1;
  femalefirstname["susan"] = 1;
  femalefirstname["suzanne"] = 1;
  femalefirstname["tanja"] = 1;
  femalefirstname["tracy"] = 1;
  femalefirstname["trudy"] = 1;
  femalefirstname["vicky"] = 1;
  femalefirstname["wendy"] = 1;
}

function initstats() {
  nbaddrafts = 0;
  nbadauthorlines = 0;
  nbadauthoremaillines = 0;

  ndrafts = 0;
  nauthors = 0;
  nietfdrafts = 0;
  nirtfdrafts = 0;
  niabdrafts = 0;
  niesgdrafts = 0;
  nindividualdrafts = 0;
  nauthoroccurrences = 0;
  ncountries = 0;
  ncompanies = 0;
  nunknowncountry = 0;
  nunknowncompany = 0;
  euauthors = 0;
  eudrafts = 0;
  nalldrafts = 0;
  nareas = 0;
  nwgs = 0;

  nietfdrafts = 0;
  nirtfdrafts = 0;
  niesgdrafts = 0;
  niabdrafts = 0;
  nrfceddrafts = 0;
  nindividualdrafts = 0;
  npostscriptdrafts = 0;
  npdfdrafts = 0;
  nnontextformats = 0;
  nfigures = 0;
  nformats = 0;
  nabnfs = 0;
  nasn1s = 0;
  ncodes = 0;
  nxmls = 0;
  nkwds = 0;
  nsecconss = 0;
  nianaconss = 0;
  nerratas = 0;

  startyear = 0;
  endyear = 0;


  nsorts = 0;
}

function html_file_css_head(title,csshtmlfile,slogan) {
  csshtmlfileseensection[csshtmlfile] = 0;
  if (substr(csshtmlfile,1,5) != "/dev/") close(csshtmlfile);
  printf("<!DOCTYPE html PUBLIC %c-//W3C//DTD XHTML 1.0 Transitional//EN%c %chttp://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd%c>\n", 34, 34, 34, 34) > csshtmlfile;
  printf("<html xmlns=%chttp://www.w3.org/1999/xhtml%c xml:lang=%cen%c lang=%cen%c>\n",
         34, 34, 34, 34, 34, 34) >> csshtmlfile;
  printf("<head>\n") >> csshtmlfile;
  printf("<meta http-equiv=%cContent-Type%c content=%ctext/html;charset=utf-8%c />\n",
         34, 34, 34, 34) >> csshtmlfile;
  printf("<link href=%c../admeasurements/admeasurements.css%c rel=%cstylesheet%c type=%ctext/css%c />\n",
         34, 34, 34, 34, 34, 34) >> csshtmlfile;
  printf("<title>%s</title>\n", title) >> csshtmlfile;
  printf("</head>\n") >> csshtmlfile;
  printf("<body>\n") >> csshtmlfile;
  printf("<div class=%cpage%c>\n", 34, 34) >> csshtmlfile;
  printf("<table border=%c0%c width=%c100%%%c>\n", 34, 34, 34, 34) >> csshtmlfile;
  printf("<tbody>\n") >> csshtmlfile;
  printf("<tr><td class=%cwglist%c>\n", 34, 34) >> csshtmlfile;
  printf("   <a href=%chttp://tools.ietf.org%c><img class=%clogo%c alt=%cIETF%c src=%chttp://tools.ietf.org/images/ietflogo3h.png%c/></a><br/><br/>\n", 34, 34, 34, 34, 34, 34, 34, 34) >> csshtmlfile;
  printf("   <div class=%cmenuitem%c><a href=%chttp://www.ietf.org/%c>IETF Home</a></div>\n", 34, 34, 34, 34) >> csshtmlfile;
  printf("   <div class=%cmenuitem%c><a href=%chttp://tools.ietf.org/about%c>About Tools</a></div>\n", 34, 34, 34, 34) >> csshtmlfile;
  printf("   <div class=%cmenuitem%c><a href=%chttp://tools.ietf.org/dailydose%c>News</a></div>\n", 34, 34, 34, 34) >> csshtmlfile;
  printf("   <div class=%cmenuitem%c><a href=%chttp://tools.ietf.org/html/%c>Documents</a></div>\n", 34, 34, 34, 34) >> csshtmlfile;
  printf("   <br/>\n") >> csshtmlfile;
  printf("   <div class=%ctopitem%c><a href=%chttp://www.arkko.com/tools/stats.html%c>Statistics</a>:</div>\n", 34, 34, 34, 34) >> csshtmlfile;
  printf("   <div class=%cmenuitem%c>&nbsp; &nbsp; <a href=%chttp://www.arkko.com/tools/docstats.html%c>Docs</a></div>\n", 34, 34, 34, 34) >> csshtmlfile;
  printf("   <div class=%cmenuitem%c>&nbsp; &nbsp; <a href=%chttp://www.iana.org/about/performance/%c>IANA</a></div>\n", 34, 34, 34, 34) >> csshtmlfile;
  printf("   <div class=%cmenuitem%c>&nbsp; &nbsp; <a href=%chttp://www.arkko.com/tools/admeasurements/%c>IESG</a></div>\n", 34, 34, 34, 34) >> csshtmlfile;
  printf("   <div class=%cmenuitem%c>&nbsp; &nbsp; <a href=%chttp://www.arkko.com/tools/lifecyclestats.html%c>Lifecycle</a></div>\n", 34, 34, 34, 34) >> csshtmlfile;
  printf("   <div class=%cmenuitem%c>&nbsp; &nbsp; <a href=%chttp://www.arkko.com/tools/rfceditorstats.html%c>RFC Ed</a></div>\n", 34, 34, 34, 34) >> csshtmlfile;
  printf("   <div class=%cmenuitem%c>&nbsp; &nbsp; <a href=%chttp://rtg.ietf.org/~fenner/iesg/%c>Misc</a></div>\n", 34, 34, 34, 34) >> csshtmlfile;
  printf("</td>\n") >> csshtmlfile;
  printf("\n") >> csshtmlfile;
  printf("<td>\n") >> csshtmlfile;
  printf("\n") >> csshtmlfile;
  printf("<div class=%cadmeasurements%c>\n", 34, 34) >> csshtmlfile;
  printf("<div class=%cadmeasurements_banner%c>\n", 34, 34) >> csshtmlfile;
  printf("<h1>%s</h1>\n", title) >> csshtmlfile;
  printf("\n") >> csshtmlfile;
  printf("<div class=%cadmeasurements_slogan%c>\n", 34, 34) >> csshtmlfile;
  printf("&#8220;%s&#8221;\n", slogan) >> csshtmlfile;
  printf("</div><!-- admeasurements_slogan -->\n") >> csshtmlfile;
  printf("\n") >> csshtmlfile;
  printf("</div><!-- admeasurements_banner -->\n") >> csshtmlfile;
  printf("</div><!-- admeasurements -->\n") >> csshtmlfile;
  printf("\n") >> csshtmlfile;
  printf("<div class=%cadmeasurements%c>\n", 34, 34) >> csshtmlfile;
  printf("\n") >> csshtmlfile;
  printf("<table class=%cadmeasurements_columns%c>\n", 34, 34) >> csshtmlfile;
  printf("<tbody>\n") >> csshtmlfile;
  printf("<tr>\n") >> csshtmlfile;
  printf("<td class=%cadmeasurements_topic%c>\n", 34, 34) >> csshtmlfile;
}

function html_file_css_section(title,csshtmlfile) {
  if (csshtmlfileseensection[csshtmlfile] != 0) {
    printf("\n<div style=%cclear:both;padding:1px;%c></div>\n", 34, 34) >> csshtmlfile;
  }

  csshtmlfileseensection[csshtmlfile] = 1;

  printf("\n<h2>%s</h2>\n\n", title) >> csshtmlfile;
}

function html_file_css_section_nextcol(title,csshtmlfile) {
  csshtmlfileseensection[csshtmlfile] = 1;
  printf("\n</td>\n\n<td class=%cadmeasurements_topic%c>\n", 34, 34) >> csshtmlfile;
  printf("\n<h2>%s</h2>\n\n", title) >> csshtmlfile;
}

function html_file_css_end(htmlfile) {
  printf("</td>\n") >> htmlfile;
  printf("</tr>\n") >> htmlfile;
  printf("</tbody>\n") >> htmlfile;
  printf("</table><!-- admeasurements_columns -->\n") >> htmlfile;
  printf("</div><!-- admeasurements -->\n") >> htmlfile;
  printf("</td>\n") >> htmlfile;
  printf("</tr>\n") >> htmlfile;
  printf("</tbody>\n") >> htmlfile;
  printf("</table><!-- page -->\n") >> htmlfile;
  printf("</div><!-- page -->\n") >> htmlfile;
  printf("</body>\n") >> htmlfile;
  printf("</html>\n") >> htmlfile;
  close(htmlfile);
}

function checkareaacronym(a) {
  a = tolower(a);
  if (areaacronym[a] == "") {
    printf("authorstats: Fatal error -- area %s acronym unknown\n", a);
    exit(1);
  }
}

function fontify(f,x) {
  if (f == "") return(x);
  else return(sprintf("<font size=%c%s%c>%s</font>", 34, f, 34, x));
}

function removever(x) {
  sub(/-[0-9][0-9][.]txt/,"",x);
  return(x);
}

function docprocessref(draftnam) {
  if (docgettype(draftnam) == "rfc") {
    if (!realdraftnameknown[draftnam]) {
      realdraftnameknown[draftnam] = 1;
      tmpreffile = "/tmp/docprocref.txt";
      system("getrfcdraftname --nover " draftnam " > " tmpreffile);
      getline realname < tmpreffile;
      close(tmpreffile);
      realdraftname[draftnam] = realname;
    }
    if (realdraftname[draftnam] == "") {
      return(fontify("","(no timeline available)"));
    } else {
      return(fontify("",sprintf("<a href=%chttp://www.arkko.com/tools/lifecycle/%s-timing.html%c>(timeline)</a>",
                                34, realdraftname[draftnam], 34)));
    }
  } else if (docgettype(draftnam) == "draft") {
    return(fontify("",sprintf("<a href=%chttp://www.arkko.com/tools/lifecycle/%s-timing.html%c>(timeline)</a>",
                              34, removever(draftnam), 34)));
  } else {
    return("");
  }
}

BEGIN {
  FS=":";
  
  hrefprefix = "../allstats/";
  quote = sprintf("%c",34);

  inittabs();
  initstats();

  activeauthor = 1;
  activecompany = 1;
  ntopcompanies = 15
  ntopcountries = 14;
  minimumcompletion = 0.1;

  if (doctype == "draft") {
    docname = "draft";
    docnameu = "Draft";
    docprefix = "http://tools.ietf.org/html/";
  } else if (doctype == "rfc") {
    docname = "RFC";
    docnameu = "RFC";
    docprefix = "http://tools.ietf.org/html/";
  } else if (doctype == "all") {
    docname = "document";
    docnameu = "Document";
    docprefix = "http://tools.ietf.org/html/";
  } else {
    print("authorstats: Unrecognized doctype\n");
    exit(1);
  }
}

/^0-wgareadefinition:/ {
  if (isarea[$2] == "") {
    isarea[$2] = $2;
    nareas++;
    checkareaacronym($2);
  }
  iswg[$3] = $3;
  nwgs++;
  wgarea[$3] = $2;
  next;
}

/^0-population:/ {
  if (population[$2] != "") {
    printf("authorstats: Redefinition of population for %s in Wikipedia from line %d -- exit\n", $2,$4);
    exit(1);
  }
  population[$2] = $3;
  if (isineu[$2]) {
    population["european union"] += $3;
  }
  next;
}

/^9-debug:/ {
  next;
}

/^z1-ad:/ {
  if (isarea[$2] == "") {
    printf("authorstats: error - area director defined for unknown area (%s)\n", $2);
  }
  if (authorad[$3] != "") {
    printf("authorstats: error - %s can not be area director for both %s and %s\n", $3, $2, authorad[$3]);
  }
  if (isauthor[$3] == "") {
    #printf("authorstats: area director %s is an unknown author (not published anything)\n", $3);
  }
  authorad[$3] = $2;
  next;
}

/^z2-chair:/ {
  if (iswg[$2] == "") {
    printf("authorstats: error - chair defined for unknown wg (%s)\n", $2);
  }
  if (isauthor[$3] == "") {
    #printf("authorstats: chair %s is an unknown author (not published anything)\n", $3);
  }
  if (authorchairs[$3] != "") authorchairs[$3] = authorchairs[$3] "%";
  authorchairs[$3] = authorchairs[$3] $2;
  next;
}

/^z3-iab:/ {
  if (isauthor[$2] == "") {
    #printf("authorstats: iab member %s is an unknown author (not published anything)\n", $2);
  }
  authoriab[$2] = $2;
  next;
}

/^z4-iaoc:/ {
  if (isauthor[$2] == "") {
    #printf("authorstats: iaoc member %s is an unknown author (not published anything)\n", $2);
  }
  authoriaoc[$2] = $2;
  next;
}

/UNRECOGNIZED/ {
  allstatinc($1,"");
  baddraft($1);
  badauthoremailline($1,"all");
  next;
}

/.*/ {
  #printf("authorstats: debug: Processing line %s...\n", $0);
  draft = $1;
  author = $2;
  company = $3;
  email = $4;
  country = $5;
  pages = $6;
  month = $7;
  year = $8;
  day = $9
  features = $10;
  goodstatinc(draft,author,country,company,pages,month,year,features);
  addauthor(draft,author);

  if (company == "UNKNOWN") badauthorline($1,$2,"unknown affiliation");
  if (country == "UNKNOWN") badauthorline($1,$2,"unknown location");
  if (pages == "UNKNOWN") badauthorline($1,$2,"unknown page count");
  if (month == "UNKNOWN") badauthorline($1,$2,"unknown publication month");
  if (year == "UNKNOWN") badauthorline($1,$2,"unknown publication year");
  
  if (email == "UNKNOWN") badauthoremailline($1,$2);
}

END {
  #printf("authorstats: debug: End of input...\n", $0);
  finalizeauthors();
  calculatelastyear();
  reportbaddrafts("baddrafts.html");
  reportbadauthorlines("badauthors.html");
  reportbadauthoremaillines("bademails.html");
  reportoverall("index.html");
  reportindividualauthors("authors.html");
  reportindividualcompanies("companies.html");
  reportindividualcountries("countries.html");
}

function calculatelastyear() {
  tmpfile = "/tmp/as.date";
  system("rm -f " tmpfile);
  system("date +%Y > " tmpfile);
  system("date +%m >> " tmpfile);
  system("date +%d >> " tmpfile);
  getline nowyear < tmpfile;
  getline nowmonth < tmpfile;
  getline nowday < tmpfile;
  if (nowyear < 2000 || nowyear > 3000) {
    printf("authorstats: error - can not find out current year -- exit\n");
    exit(1);
  }
  if (nowmonth < 1 || nowmonth > 12) {
    printf("authorstats: error - can not find out current month -- exit\n");
    exit(1);
  }
  if (nowday < 1 || nowday > 31) {
    printf("authorstats: error - can not find out current day -- exit\n");
    exit(1);
  }
  lastyearcompletion = (nowmonth - 1) / 12.0 + ((nowday - 1) / 31.0) / 12.0;
  #printf("authorstats: lastyear is %f complete (%d %d %d)\n", lastyearcompletion, nowyear, nowmonth, nowday);
  system("rm -f " tmpfile);
}

function addauthor(draft,author) {
  if (draft in draftauthors) {
    draftauthors[draft] = draftauthors[draft] "%" author;
    draftnauthors[draft]++;
  } else {
    draftauthors[draft] = author;
    draftnauthors[draft] = 1;
  }
  
  if (author in authordrafts) {
    #printf("authorstats: debug: Author %s known, adding doc %s...\n", author, draft);
    authordrafts[author] = authordrafts[author] "%" draft;
    authorndrafts[author]++;
    #printf("authorstats: debug: Doc list for him now %s...\n", authordrafts[author]);
  } else {
    #printf("authorstats: debug: Author %s not known, adding doc %s...\n", author, draft);
    authordrafts[author] = draft;
    authorndrafts[author] = 1;
  }
}

function allstatinc(draft,author) {
  if (isdraft[draft] == "") {
    isdraft[draft] = draft;
    nalldrafts = nalldrafts + 1;
  }
  #nbadauthorlines = nbadauthorlines + 1;
}

function goodstatinc(draft,author,country,company,pages,month,year,features) {
  if (isdraft[draft] == "") {
    isdraft[draft] = draft;
    ndrafts = ndrafts + 1;
    nalldrafts = nalldrafts + 1;
    draftpagecounts[draft] = pages;
    if (month != "UNKNOWN") {
      draftmonths[draft] = month;
    }
    if (year != "UNKNOWN") {
      if (year <= 0 || year > 2020) {
	printf("authorstats: Invalid year (%s) for %s -- exit\n", year, draft);
	exit(1);
      }
      draftyears[draft] = year;
      if (startyear == 0 || year < startyear) startyear = year;
      if (endyear == 0 || year > endyear) endyear = year;
    }
    if (pagecounts[pages] == "") {
      pagecounts[pages] = 1;
    } else {
      pagecounts[pages] = pagecounts[pages] + 1;
    }
    if (draft ~ /^draft-ietf-/) {
      nietfdrafts++;
    } else if (draft ~ /^draft-irtf-/) {
      nirtfdrafts++;
    } else if (draft ~ /^draft-iesg-/) {
      niesgdrafts++;
    } else if (draft ~ /^draft-iab-/) {
      niabdrafts++;
    } else if (draft ~ /^draft-rfc/) {
      nrfceddrafts++;
    } else {
      nindividualdrafts++;
    }
    if (index(features,"postscript")) {
      npostscriptdrafts++;
    }
    if (index(features,"pdf")) {
      npdfdrafts++;
    }
    if (index(features,"postscript") || index(features,"pdf")) {
      nnontextformats++;
    }
    if (index(features,"figure")) {
      nfigures++;
    }
    if (index(features,"format")) {
      nformats++;
    }
    if (index(features,"abnf")) {
      nabnfs++;
    }
    if (index(features,"kwd")) {
      nkwds++;
    }
    if (index(features,"asn1")) {
      nasn1s++;
    }
    if (index(features,"code")) {
      ncodes++;
    }
    if (index(features,"xml")) {
      nxmls++;
    }
    if (index(features,"seccons")) {
      nsecconss++;
    }
    if (index(features,"ianacons")) {
      nianaconss++;
    }
    if (index(features,"errata")) {
      nerratas++;
    }
  }

  lyear = (year == "UNKNOWN" ? 0 : year);

  if (isauthor[author] == "") {
    isauthor[author] = author;
    nauthors = nauthors + 1;
    nallauthors = nallauthors + 1;
    newauthor = 1;
  } else {
    newauthor = 0;
  }
  
  if (country != "UNKNOWN" && country != "") {
    if (iscountry[country] == "") {
      iscountry[country] = country;
      countryauthors[country] = 0;
      countrydrafts[country] = 0;
      ncountries++;
    }
    if (index(countrydraftslist[country],draft) == 0) {
      countrydrafts[country] = countrydrafts[country] + 1;
      if (countrydraftslist[country] == "") {
        countrydraftslist[country] = draft;
      } else {
        countrydraftslist[country] = countrydraftslist[country] "%" draft;
      }
      if (isineu[country] == 1) eudrafts++;
    }
    if (isauthorcountry[author] == "") {
      isauthorcountry[author] = country;
      #printf("setting isauthorcountry[%s] = %s\n", author, country);
      isauthorcountrytime[author] = lyear;
    } else if (isauthorcountry[author] == country) {
      if (isauthorcountrytime[author] < lyear) isauthorcountrytime[author] = lyear;
    } else if (isauthorcountrytime[author] < lyear) {
      if (index(isauthorcountryothers[author],isauthorcountry[author]) == 0) {
        if (isauthorcountryothers[author] != "") isauthorcountryothers[author] = isauthorcountryothers[author] "%";
        isauthorcountryothers[author] = isauthorcountryothers[author] isauthorcountry[author];
      }
      isauthorcountry[author] = country;
      #printf("re-setting isauthorcountry[%s] = %s\n", author, country);
      isauthorcountrytime[author] = lyear;
    } else if (index(isauthorcountryothers[author],country) == 0) {
      if (isauthorcountryothers[author] != "") isauthorcountryothers[author] = isauthorcountryothers[author] "%";
      isauthorcountryothers[author] = isauthorcountryothers[author] country;
    }
    #printf("isauthorcountry[%s] left as %s\n", author, isauthorcountry[author]);
#    if (newauthor) {
#      countryauthors[country] = countryauthors[country] + 1;
#      if (isineu[country] == 1) euauthors++;
#      if (countryauthorslist[country] == "") {
#        countryauthorslist[country] = author;
#      } else {
#        countryauthorslist[country] = countryauthorslist[country] ", " author;
#      }
#    }
#  } else {
#    if (newauthor) nunknowncountry++;
  } else {
     isauthorcountry[author] = "UNKNOWN";
     #printf("setting isauthorcountry[%s] = %s\n", author, "UNKNOWN");
  }
  
  if (company != "UNKNOWN" && company != "") {
    company = tolower(company);
    if (iscompany[company] == "") {
      iscompany[company] = company;
      companyauthors[company] = 0;
      companydrafts[company] = 0;
      ncompanies++;
    }

    if (isauthorcompany[author] == "") {

      isauthorcompany[author] = company;
      isauthorcompanytime[author] = lyear;
      #printf("  first company\n") >> "/tmp/oc.txt";

    } else if (isauthorcompany[author] == company) {

      if (isauthorcompanytime[author] < lyear) isauthorcompanytime[author] = lyear;
      #printf("  same company\n") >> "/tmp/oc.txt";

    } else if (isauthorcompanytime[author] < lyear) {

      if (index(isauthorcompanyothers[author],isauthorcompany[author]) == 0) {
        if (isauthorcompanyothers[author] != "") isauthorcompanyothers[author] = isauthorcompanyothers[author] "%";
        isauthorcompanyothers[author] = isauthorcompanyothers[author] isauthorcompany[author];
      }
      isauthorcompany[author] = company;
      isauthorcompanytime[author] = lyear;
      #printf("  later company\n") >> "/tmp/oc.txt";

    } else if (index(isauthorcompanyothers[author],company) == 0) {

      #printf("  unseen company\n") >> "/tmp/oc.txt";
      if (isauthorcompanyothers[author] != "") isauthorcompanyothers[author] = isauthorcompanyothers[author] "%";
      isauthorcompanyothers[author] = isauthorcompanyothers[author] company;

    }

    #printf("  post-company = %s, others = %s\n", isauthorcompany[author], isauthorcompanyothers[author]) >> "/tmp/oc.txt";
    if (index(companydraftslist[company],draft) == 0) {
      companydrafts[company] = companydrafts[company] + 1;
      if (companydraftslist[company] == "") {
        companydraftslist[company] = draft;
      } else {
        companydraftslist[company] = companydraftslist[company] "%" draft;
      }
    }
#    if (newauthor) companyauthors[company] = companyauthors[company] + 1;
#  } else {
#    if (newauthor) nunknowncompany++;
  } else {
     isauthorcompany[author] = "UNKNOWN";
  }

  nauthoroccurrences = nauthoroccurrences + 1;
}

function finalizeauthors() {
  #printf("finalizeauthors...\n");
  for (author in isauthor) {
    finalizeauthor(author);
  }
}

function finalizeauthor(author) {
  #printf("finalizeauthor(%s)...\n", author);
  country = isauthorcountry[author];
  if (country != "UNKNOWN"  && country != "") {
    countryauthors[country] = countryauthors[country] + 1;
    #printf("setting countryauthors[%s] = %d\n", country, countryauthors[country]);
    if (isineu[country] == 1) euauthors++;
    if (countryauthorslist[country] == "") {
      countryauthorslist[country] = author;
    } else {
      countryauthorslist[country] = countryauthorslist[country] ", " author;
    }
  } else {
    if (newauthor) nunknowncountry++;
  }

  company = isauthorcompany[author];
  if (company != "UNKNOWN" && company != "") {
    companyauthors[company] = companyauthors[company] + 1;
  } else {
    nunknowncompany++;
  }
}

function badauthorline(draft,author,problem) {
  nbadauthorlines++;
  badauthorlines[nbadauthorlines] = author " - " draft ": " problem;
}

function badauthoremailline(draft,author) {
  nbadauthoremaillines++;
  badauthoremaillines[nbadauthoremaillines] = author " - " draft;
}

function baddraft(s) {
  nbaddrafts = nbaddrafts + 1;
  baddrafts[nbaddrafts] = s;
}

function reportbaddrafts(file) {
  openhtml(file,"Unrecognized " docname "s");
  html_file_css_section("Problems",file);
  printf("<ul>\n") >> file;
  for (i in baddrafts) {
    printf("<li>%s</li>\n", baddrafts[i]) >> file;
  }
  printf("</ul>\n") >> file;
  closehtml(file);
}

function reportbadauthorlines(file) {
  openhtml(file,"Unrecognized authors in " docname "s");
  html_file_css_section("Problems",file);
  printf("<ul>\n") >> file;
  for (i in badauthorlines) {
    printf("<li>%s</li>\n", badauthorlines[i]) >> file;
  }
  printf("</ul>\n") >> file;
  closehtml(file);
}

function reportbadauthoremaillines(file) {
  openhtml(file,"Unrecognized or incomplete e-mail addresses in " docname "s");
  html_file_css_section("Problems",file);
  printf("<ul>\n") >> file;
  for (i in badauthoremaillines) {
    printf("<li>%s</li>\n", badauthoremaillines[i]) >> file;
  }
  printf("</ul>\n") >> file;
  closehtml(file);
}

function reportindividualcountries(topfile) {
  openhtml(topfile,"Countries");
  printf("<ul>\n") >> topfile;
  topfile = startsort(topfile);
  for (country in iscountry) {
    file = tofname(country);
    printf("<li><a href=%c%s%s%c>%s</a> (%d authors and %d %ss)</li>\n",
           34, hrefprefix, file, 34, countrycapitalize(country),
           countryauthors[country], countrydrafts[country],
           docname) >> topfile;
    reportindividualcountry(country,countryauthors[country],file);
  }
  topfile = stopsort(topfile);
  printf("</ul>\n") >> topfile;
  closehtml(topfile);
}

function reportindividualcountry(country,n,file) {
  openhtml(file,countrycapitalize(country) " Data");
  html_file_css_section("Basic Information",file);
  countrywgs = "";
  for (author in isauthorcompany) {
    if (isauthorcountry[author] == country) {
      s1 = authorwgs[author];
      split(s1,tempwgs,"%");
      for (tempwgi in tempwgs) {
        tempwg = tempwgs[tempwgi];
        if (index(countrywgs,tempwg) == 0) {
	  if (countrywgs == "")
	    countrywgs = tempwg;
	  else
	    countrywgs = countrywgs "%" tempwg;
        }
      }
    }
  }
  if (doctype == "draft" || doctype == "all") {
    printwglist(countrywgs,"<p>The authors in this country appear to be working in the following WG%s: ","</p>\n",file);
  }
  html_file_css_section("Authors",file);
  printf("<p>There are %d authors in %s:</p>\n", n, country) >> file;
  printf("<ul>\n") >> file;
  for (f = realmaxdrafts; f >= 1; f--) {
    for (author in isauthor) {
      if (isauthorcountry[author] == country && authorndrafts[author] == f) {
        hisdrafts = authordrafts[author];
        n = split(hisdrafts,hisdraftstab,"%");
        if (n >= activeauthor) {
          printf("<li><a href=%c%s%s%c>%s</a> (%d %ss)</li>\n",
                 34, hrefprefix, tofname(author), 34, author, n, docname) >> file;
        } else {
          printf("<li>%s (%d %ss)<li>\n", author, n, docname) >> file;
        }
      }
    }
  }
  printf("</ul>\n") >> file;
  closehtml(file);
}

function capitalize(s) {
  if (length(s) >= 1 && substr(s,1,1) >= "a" && substr(s,1,1) <= "z") {
    return(toupper(substr(s,1,1)) substr(s,2));
  } else {
    return(s);
  }
}

function countrycapitalize(cn) {
  if (cn == "usa") return("USA");
  cnn = split(cn,cntab," ");
  cnres = "";
  for (cni = 1; cni <= cnn; cni++) {
    if (cnres != "") cnres = cnres " ";
    cnres = cnres capitalize(cntab[cni]);
  }
  return(cnres);
}

function areacapitalize(ar) {
  return(countrycapitalize(ar));
}

function companycapitalize(cp) {
  if (cp == "vpnc" ||
      cp == "nsn" ||
      cp == "ibm" ||
      cp == "att" ||
      cp == "bbn" ||
      cp == "mit" ||
      cp == "sri" ||
      cp == "hp" ||
      cp == "isoc" ||
      cp == "isc" ||
      cp == "dod" ||
      cp == "nec" ||
      cp == "dec" ||
      cp == "mci" ||
      cp == "nasa" ||
      cp == "cnri" ||
      cp == "bt" ||
      cp ~ /^uc/ ||
      cp ~ /^us/ ||
      cp ~ /^ui/) return(toupper(cp));
  return(countrycapitalize(cp));
}

function coname(s) {
  if (s == "united arab emirates") s = "arab emirates";
  return(s);
}

function tofname(s) {
  s = tolower(s) ".html";
  gsub(/ /,"",s);
  gsub(/[()]/,"",s);
  return(s);
}

function tofname_comp(s) {
  s = "c_" tolower(s) ".html";
  gsub(/ /,"",s);
  gsub(/[()]/,"",s);
  return(s);
}

function reportindividualcompanies(topfile) {
  openhtml(topfile,"All companies");
  printf("<ul>\n") >> topfile;
  topfile = startsort(topfile);
  for (company in iscompany) {
    if (companydrafts[company] >= activecompany ||
        companyauthors[company] >= activecompany) {
      file = tofname_comp(company);
      reportindividualcompany(company,companyauthors[company],companydrafts[company],file);
      printf("<li><a href=%c%s%s%c>%s</a> (%d authors and %d %ss)</li>\n",
             34, hrefprefix, file, 34, capitalize(company),
             companyauthors[company], companydrafts[company],
             docname) >> topfile;
      if (debug && companyauthors[company] > 50) printf("processing company %s to %s\n", company, topfile);
    }
  }
  if (debug) system("wc -l " topfile);
  topfile = stopsort(topfile);
  printf("</ul>\n") >> topfile;
  closehtml(topfile);
}

function startsort(inputfile) {
  sortfiles[nsorts] = inputfile;
  sortfile = sprintf("sort%d.tmp",nsorts);
  nsorts++;
  if (debug) printf("debug: start sort %d in %s to %s\n", nsorts - 1, sortfile, sortfiles[nsorts-1]);
  system("rm -f " sortfile);
  system("touch " sortfile);
  return(sortfile);
}

function stopsort(sortfile) {
  inputfile = sortfiles[--nsorts];
  if (debug) printf("debug: stop sort %d in %s to %s\n", nsorts, sortfile, inputfile);
  close(sortfile);
  close(inputfile);
  system("sort < " sortfile " >> " inputfile);
  close(sortfile);
  return(inputfile);
}

function reportindividualcompany(company,repna,repnd,file) {
  openhtml(file,companycapitalize(company) " Data (" topic ")");
  companywgs = "";
  for (author in isauthorcompany) {
    if (isauthorcompany[author] == company) {
      s1 = authorwgs[author];
      split(s1,tempwgs,"%");
      for (tempwgi in tempwgs) {
        tempwg = tempwgs[tempwgi];
        if (index(companywgs,tempwg) == 0) {
	  if (companywgs == "")
	    companywgs = tempwg;
	  else
	    companywgs = companywgs "%" tempwg;
        }
      }
    }
  }
  html_file_css_section("WGs",file);
  printwglist(companywgs,"<p>The authors in this company appear to be working in the following WG%s: ","</p>\n",file);
  html_file_css_section("Authors",file);
  printf("<p>This company has %d authors:</p>\n",
         repna) >> file;
  printf("<ul>\n") >> file;
  for (f = realmaxdrafts; f >= 1; f--) {
    for (author in isauthorcompany) {
      if (isauthorcompany[author] == company && authorndrafts[author] == f) {
        if (f >= activeauthor) {
          printf("<li><a href=%c%s%s%c>%s</a> (%d %ss)</li>\n",
                  34, hrefprefix, tofname(author), 34, author, f, docname) >> file;
        } else {
          printf("<li>%s (%d %ss)</li>\n", author, n, docname) >> file;
        }
      }
    }
  }
  
  split(companydraftslist[company],compd,"%");
  printf("</ul>\n") >> file;
  reportlistofdocs("The authors from " company " have", n, compd, file);
  closehtml(file);
}

function reportindividualauthors(topfile) {
  openhtml(topfile,activeauthor == 1 ? "All authors" : "Most active authors");
  printf("<ul>\n") >> topfile;
  topfile = startsort(topfile);
  for (author in isauthor) {
    hisdrafts = authordrafts[author];
    #printf("authorstats: debug: Reporting active %s with drafts %s...\n", author, hisdrafts);
    n = split(hisdrafts,hisdraftstab,"%");
    if (n >= activeauthor) {
      file = tofname(author);
      reportindividualauthor(author,n,hisdraftstab,file);
      printf("<li><a href=%c%s%s%c>%s</a></li>\n",
              34, hrefprefix, file, 34, author) >> topfile;
    }
  }
  topfile = stopsort(topfile);
  printf("</ul>\n") >> topfile;
  closehtml(topfile);
}

function printwglist(string,format1,format2,file) {
  n = split(string,wgtab,"%");
  printf(format1,
         (n == 1 ? "" : "s")) >> file;
  if (n == 0) {
    printf("(none)") >> file;
  } else {
    for (q = 1; q <= n; q++) {
      if (n > 1 && q == n) printf(" and ") >> file;
      else if (q > 1) printf(", ") >> file;
      printf("<a href=%chttp://www.ietf.org/html.charters/%s-charter.html%c>%s</a>",
             34, wgtab[q], 34, wgtab[q]) >> file;
    }
  }
  printf(format2,
         (n == 1 ? "" : "s")) >> file;
}

function removestring(x,y) {
  rs = "";
  while (length(x) > 0) {
    if (length(x) >= length(y) &&
        substr(x,1,length(y)) == y) {
      x = substr(x,length(y) + 1);
    } else {
      rs = rs substr(x,1,1);
      x = substr(x,2);
    }
  }
  return(rs);
}

function reportindividualauthor(author,n,hisdraftstab,file) {
  openhtml(file,author " Data (" topic ")");
  html_file_css_section("Personal Information",file);
  split(author,authornameparts," ");
  if (malefirstname[tolower(authornameparts[1])]) pron = "he";
  else if (femalefirstname[tolower(authornameparts[1])]) pron = "she";
  else pron = "he or she";
  if (isauthorcountry[author] == "") {
    printf("<p>The location of this author is unknown ") >> file;
  } else {
    printf("<p>This author is in <a href=%c%s%s%c>%s</a> ",
           34, hrefprefix, tofname(isauthorcountry[author]), 34, countrycapitalize(isauthorcountry[author])) >> file;
  }
  if (isauthorcountrytime[author] != 0) printf(" (as of %d)", isauthorcountrytime[author]) >> file;
  if (isauthorcountryothers[author] != "") {
    printf(", previous locations include") >> file;
    tz = isauthorcountryothers[author];
    tz = removestring(tz,isauthorcountry[author] "%");
    tz = removestring(tz,"%" isauthorcountry[author]);
    cn = split(tz,compo,"%");
    for (oi in compo) {
      if (oi == 1) {
      } else if (oi == cn - 1) {
        printf(" and") >> file;
      } else {
        printf(",") >> file;
      }
      printf(" <a href=%c%s%s%c>%s</a>", 34, hrefprefix, tofname(compo[oi]), 34, countrycapitalize(compo[oi])) >> file;
    }
  }
  printf(".\n") >> file;
  if (isauthorcompany[author] == "") {
    printf("It is not known where this author works in. ") >> file;
  } else {
    printf("This author works for <a href=%c%s%s%c>%s</a>",
	   34, hrefprefix, tofname_comp(isauthorcompany[author]), 34, capitalize(isauthorcompany[author])) >> file;
  }
  if (isauthorcompanytime[author] != 0) printf(" (as of %d)", isauthorcompanytime[author]) >> file;
  printf(". \n") >> file;
  if (isauthorcompanyothers[author] != "") {
    printf("Previous employers include") >> file;
    tz = isauthorcompanyothers[author];
    tz = removestring(tz,isauthorcompany[author] "%");
    tz = removestring(tz,"%" isauthorcompany[author]);
    cn = split(tz,compo,"%");
    for (oi in compo) {
      if (oi == 1) {
      } else if (oi == cn - 1) {
        printf(" and") >> file;
      } else {
        printf(",") >> file;
      }
      printf(" <a href=%c%s%s%c>%s</a>", 34, hrefprefix, tofname_comp(compo[oi]), 34, capitalize(compo[oi])) >> file;
    }
    printf(".\n") >> file;
  }
  printf("</p>\n") >> file;
  if (doctype == "draft" || doctype == "all") {
    printwglist(authorwgs[author],"<p>The working group%s where " pron " is active appear to be ",".\n",file);
  }
  if (authorchairs[author] != "") {
    printwglist(authorchairs[author],capitalize(pron) " chairs the "," working group%s.\n",file);
  }
  if (authorad[author] != "") {
    adstats = sprintf("(click <a href=%chttp://www.arkko.com/tools/admeasurements/stat/%s_%s.html%c>here</a> for statistics related to this role)",
	              34, capitalize(authornameparts[2]), capitalize(authornameparts[1]), 34);
    printf(capitalize(pron) " is an area director for the %s area %s.\n", authorad[author], adstats) >> file;
  }
  if (authoriab[author] != "") {
    printf(capitalize(pron) " is in the IAB.\n") >> file;
  }
  if (authoriaoc[author] != "") {
    printf(capitalize(pron) " is in the IAOC.\n") >> file;
  }
  printf("</p>\n") >> file;
  reportlistofdocs(capitalize(pron) " has", n, hisdraftstab, file);
  closehtml(file);
}

function reportlistofdocs(who,n,hisdraftstab,file) {
  if (doctype == "all") {
    
    nd = 0;
    nr = 0;
    for (adraft in hisdraftstab) {
      if (docgettype(hisdraftstab[adraft]) == "rfc") {
        nr++;
      } else {
        nd++;
      }
    }

    #printf("authorstats: debug: Reporting author with nr=%d and nd=%d...\n", nr, nd);
    
    html_file_css_section("RFCs",file);
    if (nr == 0) {
      printf("<p>%s no RFCs.</p>\n", who) >> file;
    } else {
      printf("<p>%s the following %d RFCs:</p>\n", who, nr) >> file;
      printf("<ul>\n") >> file;
      system("rm -f docslisttmp.txt");
      system("touch docslisttmp.txt");
      for (adraft in hisdraftstab) {
        if (docgettype(hisdraftstab[adraft]) == "rfc") {
          printf("<li><a href=%c%s%s%c>%s</a> %s</li>\n",
                 34, docprefix, hisdraftstab[adraft], 34, draftname(hisdraftstab[adraft]),
                 docprocessref(hisdraftstab[adraft])) >> "docslisttmp.txt";
        }
      }
      close("docslisttmp.txt");
      close(file);
      system("sort < docslisttmp.txt >> " file);
      printf("</ul>\n") >> file;
      close(file);
      system("rm -f docslisttmp.txt");
    }
    html_file_css_section_nextcol("Drafts",file);
    if (nd == 0) {
      printf("<p>%s no drafts.</p>\n", who) >> file;
    } else {
      printf("<p>%s the following %d drafts:</p>\n", who, nd) >> file;
      printf("<ul>\n") >> file;
      system("rm -f docslisttmp.txt");
      system("touch docslisttmp.txt");
      for (adraft in hisdraftstab) {
        if (docgettype(hisdraftstab[adraft]) != "rfc") {
          printf("<li><a href=%c%s%s%c>%s</a> %s</li>\n",
                 34, docprefix, hisdraftstab[adraft], 34, draftname(hisdraftstab[adraft]),
                 docprocessref(hisdraftstab[adraft])) >> "docslisttmp.txt";
        }
      }
      close("docslisttmp.txt");
      close(file);
      system("sort < docslisttmp.txt >> " file);
      printf("</ul>\n") >> file;
      close(file);
      system("rm -f docslisttmp.txt");
    }

  } else {

    html_file_css_section(docname "s",file);
    printf("<p>%s the following %d %ss:</p>\n", who, n, docname) >> file;
    printf("<ul>\n") >> file;
    system("rm -f docslisttmp.txt");
    system("touch docslisttmp.txt");
    for (adraft in hisdraftstab) {
      printf("<li><a href=%c%s%s%c>%s</a> %s</li>\n",
             34, docprefix, hisdraftstab[adraft], 34, draftname(hisdraftstab[adraft]),
             docprocessref(hisdraftstab[adraft])) >> "docslisttmp.txt";
    }
    close("docslisttmp.txt");
    close(file);
    system("sort < docslisttmp.txt >> " file);
    printf("</ul>\n") >> file;
    close(file);
    system("rm -f docslisttmp.txt");
  }
}

function reportoverall(file) {
  openhtml(file,"IETF document statistics (" topic ")");

  html_file_css_section(docnameu "s", file);
  printf("<p>Total number of %ss is %d.\n",
         docname, nalldrafts) >> file;
  if (doctype == "draft") {
    printf("Of these %d (%.2f%%) are individual %ss, %d (%.2f%%) IETF, %d (%.2f%%) IRTF, %d (%.2f%%) IAB, %d (%.2f%%) RFC Editor and %d (%.2f%%) IESG drafts.\n",
	   nindividualdrafts, (100.0 * nindividualdrafts) / ndrafts,
	   docname,
	   nietfdrafts, (100.0 * nietfdrafts) / ndrafts,
	   nirtfdrafts, (100.0 * nirtfdrafts) / ndrafts,
	   niabdrafts, (100.0 * niabdrafts) / ndrafts,
	   nrfceddrafts, (100.0 * nrfceddrafts) / ndrafts,
	   niesgdrafts, (100.0 * niesgdrafts) / ndrafts) >> file;
  }
  printf("</p>\n") >> file;
  printf("<p>The distribution of %s ", docname) >> file;
  if (doctype == "draft") {
    printf("according to version numbers is shown <a href=%cversiondistr.html%c>here</a>,\n",
  	   34, 34) >> file;
    reportversions();
  }
  printf("and according to number of authors <a href=%cauthdistr.html%c>here</a>,\n",
	 34, 34) >> file;
  reportauthors();
  printf("%s page count distribution looks like <a href=%cpagedistr.html%c>this</a>.\n", docnameu, 34, 34) >> file;
  reportpages();
  printf("The used document format types and content features are shown <a href=%cformatdistr.html%c>here</a>.\n", 34, 34) >> file;
  reportformats();
  if (doctype == "rfc") {
    printf("The growth of publication rate per year is shown <a href=%cpubdistr.html%c>here</a>.\n", 34, 34) >> file;
    reportpubyears();
  } else if (doctype == "draft") {
    printf("The publication rate per month is shown <a href=%cpubdistr.html%c>here</a>.\n", 34, 34) >> file;
    reportpubyears();
  } else {
    # do nothing
  }
  printf("</p>\n") >> file;

  html_file_css_section("Authors", file);
  printf("<p>The distribution of authors according to the number of %ss\n", docname) >> file;
  printf("they have is shown <a href=%cauthactdistr.html%c>here</a>.\n", 34, 34) >> file;
  reportauthoractivities(0);
  if (doctype == "draft" || doctype == "all") {
    printf("The same distribution for IETF (or other official) %ss is shown <a href=%cauthactdistr-wg.html%c>here</a>.\n",
           docname, 34, 34) >> file;
    reportauthoractivities(1);
    printf("The authors work in one or multiple WGs according <a href=%cauthcrosswg.html%c>this distribution</a>.\n", 34, 34) >> file;
    reportauthorcrosswg();
    printf("The same distribution for cross-area work is <a href=%cauthcrossarea.html%c>here</a>.\n", 34, 34) >> file;
    reportauthorcrossarea();
    printf("Most active authors per area can be seen <a href=%cauthactareadistr.html%c>here</a>.\n",
           34, 34) >> file;
    reportauthoractivitiesperarea();
  }
  printf("</p>\n") >> file;

  if (doctype == "draft" || doctype == "all") {
    html_file_css_section("Areas and WGs", file);
    printf("<p>The distribution of WGs per areas is shown in <a href=%careawgdistr.html%c>here</a>.\n",
           34, 34) >> file;
    reportareawgs();
    printf("The distribution of %ss according to WGs is shown <a href=%cwgdistr.html%c>here</a>,\n",
  	   docname, 34, 34) >> file;
    reportwgs();
    printf("and according to areas is shown <a href=%careadistr.html%c>here</a>.\n",
	   34, 34) >> file;
    reportareas();
    printf("</p>\n") >> file;
  }

  html_file_css_section("Affiliations", file);
  printf("<p>Authors come from <a href=%ccompanydistr.html%c>these companies</a>, %d different companies in total. \n",
	 34, 34, ncompanies) >> file;
  reportcompanies();
  if (endyear - startyear > 2 && topic != "active I-Ds") {
    printf("Over the years, the situation has changed like <a href=%ccompanydistrhist.html%c>this</a>\n",
	   34, 34) >> file;
    reporttopcompanies(0);
    # Normalization: OK
    printf(" (<a href=%ccompanydistrhist_norm.html%c>normalized</a>).\n",
	   34, 34) >> file;
    reporttopcompanies(1);
  }
  printf("</p>\n") >> file;

  html_file_css_section_nextcol("Countries and Continents", file);
  reportcountries_draft(0);
  if (endyear - startyear > 2 && topic != "active I-Ds") {
    printf("Over the years, the situation has developed like <a href=%ccountrydistrhist.html%c>this</a>",
	   34, 34) >> file;
    reporttopcountries(0);
    # Normalization: OK
    printf(" (<a href=%ccountrydistrhist_norm.html%c>normalized</a>)",
	   34, 34) >> file;
    reporttopcountries(1);
    printf(".\n") >> file;
  }
  printf("Considering EU as a country, the distribution would look like <a href=%cd-countryeudistr.html%c>this</a> instead.\n",
	 34, 34) >> file;
  reportcountries_draft(1);
  printf("The same distribution for continents can be found <a href=%cd-contdistr.html%c>here</a>.\n", 34, 34) >> file;
  reportcontinents_draft();
  if (endyear - startyear > 2 && topic != "active I-Ds") {
    printf("Over the years, the situation has developed like <a href=%cd-contdistrhist.html%c>this</a>\n",
	   34, 34) >> file;
    reporttopcontinents(0);
    # Normalization: OK
    printf(" (<a href=%cd-contdistrhist_norm.html%c>normalized</a>)",
	   34, 34) >> file;
    reporttopcontinents(1);
    printf(".\n") >> file;
  }
  printf("<p>Authors come from <a href=%ccountrydistr.html%c>these countries</a>.\n",
	 34, 34, ncountries) >> file;
  reportcountries(0);
  printf("Considering EU as a country, the distribution would look like <a href=%ccountryeudistr.html%c>this</a> instead.\n",
	 34, 34) >> file;
  reportcountries(1);
  printf("The same distribution for continents can be found <a href=%ccontdistr.html%c>here</a>.\n", 34, 34) >> file;
  reportcontinents();
  printf("</p>\n") >> file;
  
  html_file_css_section("Detailed data", file);
  printf("<p>All active authors can be seen <a href=%cauthors.html%c>here</a>.\n", 34, 34) >> file;
  printf("Countries can be seen <a href=%ccountries.html%c>here</a>.\n", 34, 34) >> file;
  printf("And active companies can be seen <a href=%ccompanies.html%c>here</a>.</p>\n", 34, 34) >> file;

  html_file_css_section("Data Freshness and Reliability", file);
  printf("<p>Out of the total %d %ss, %d or %.2f%% are unparseable by this tool.\n",
         nalldrafts, docname, nalldrafts - ndrafts, (100.0 * (nalldrafts - ndrafts)) / nalldrafts) >> file;
  printf("See <a href=%cbaddrafts.html%c>here</a> for a list of %ss with problems.\n", 34, 34, docname) >> file;
  printf("Out of a total of %d authors entries, %d or %.2f%% had limited information. ",
	 nauthoroccurrences, nbadauthorlines,
	 (100.0 * nbadauthorlines) / nauthoroccurrences) >> file;
  printf("See <a href=%cbadauthors.html%c>here</a> for a list of authors with problems. \n", 34, 34) >> file;
  printf("See <a href=%cbademails.html%c>here</a> for a list of %ss with incomplete or non-working e-mail addresses. \n", 34, 34, docname) >> file;
  printf("Warning: See the <a href=%chttp://www.arkko.com/tools/authorstats.html#quality%c>tool description</a> for detailed information about what can or can not be assumed about the quality of the results.</p>\n", 34, 34) >> file;
  closehtml(file);
}

function reportcontinents() {
  base = "contdistr";
  htmlfile = base ".html";
  epsfile = base ".eps";
  jpgfile = base ".jpg";
  datafile = base ".dat";
  gplfile = base".txt";

  openhtml(htmlfile,"Distribution of authors on continents");
  html_file_css_section("Statistics",htmlfile);
  printf("<img src=%c%s%c style=%cborder:0%c alt=%cdistribution%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  html_file_css_section("Details",htmlfile);
  printf("<ul>\n") >> htmlfile;

  continentauthors["antarctica"] = 0;
  for (country in countryauthors) {
    continent = continentof[country];
    if (continent == "") {
      printf("authorstats: Warning - Country %s unrecognized (authors %s)\n", country, countryauthors[country]);
      continent = "OTHER";
    }
    if (continentauthors[continent] == "") continentauthors[continent] = 0;
    continentauthors[continent] = continentauthors[continent] + countryauthors[country];
  }

  nth = 1;
  maxn = 0;
  printf("\n") > datafile;
  extra = "set xtics rotate (";
  for (continent in continentauthors) {
    if (continentauthors[continent] > 0 || continent != "OTHER") {
      if (maxn < continentauthors[continent]) maxn = continentauthors[continent];
      printf("<li>%d authors (%.2f%%) come from %s.</li>\n",
             continentauthors[continent],
	     (100.0 * continentauthors[continent]) / nauthors,
	     countrycapitalize(continent)) >> htmlfile;
      printf("%d	%d\n", nth, continentauthors[continent]) >> datafile;
      continentt = countrycapitalize(continent);
      sub(/ /,"\\n",continentt);
      extra = extra quote continentt quote " " nth ", ";
      nth++;
    }
  }
  printf("<li>unknown author location for %d authors (%.2f%%).</li>\n",
         nunknowncountry, (100.0 * nunknowncountry) / nauthors) >> htmlfile;
  printf("%d	%d\n", nth, nunknowncountry) >> datafile;
  
  extra = extra quote "(unknown)" quote " " nth ")\n";
  #extra = extra " font " quote "Times-Roman,40" quote "\n";
  extra = extra "set xrange [0 : " nth + 1 "]\n";
  extra = extra "set yrange [0 : " roundup(maxn) "]\n";
  basicgnuplotsettings(gplfile,epsfile,datafile,"Continent","# of Authors","Number of authors per continent",
                       0,extra);

  plottojpg(gplfile,epsfile,jpgfile);

  printf("</ul>\n") >> htmlfile;
  printf("<p>Location data is calculated from the first occurrence of an author.</p>\n") >> htmlfile;
  closehtml(htmlfile);
}

function reporttopcompanies(normalized) {
  base = "companydistrhist";
  if (normalized) base = base "_norm";
  htmlfile = base ".html";
  openhtml(htmlfile,"Comparison of companies over the years");

  delete reporttopcompanies_seencompany;
  delete multifunctiontrack;
  for (company in topcompanies) {
    multifunctiontrack[company] = company;
  }

  delete multifunctiontab;
  for (year = startyear; year <= endyear; year++) {
    yeardata = "";
    for (company in topcompanies) {
      n = 0;
      split(companydraftslist[company],compd,"%");
      for (drafti in compd) {
	draft = compd[drafti];
	if (year == draftyears[draft]) {
	  n++;
        }
      }
      if (reporttopcompanies_seencompany[company] != "" ||
	  n != 0) {
	if (reporttopcompanies_seencompany[company] == "") {
          if (multifunctiontab[year-1] != "") {
            multifunctiontab[year-1] = multifunctiontab[year-1] "%";
          }
          multifunctiontab[year-1] = multifunctiontab[year-1] company ":0";
        }
        if (yeardata != "") yeardata = yeardata "%";
        if (year == nowyear) {
          if (lastyearcompletion >= minimumcompletion) {
            yeardata = yeardata company ":" (n * (1/lastyearcompletion));
          }
        } else {
          yeardata = yeardata company ":" n;
        }
	reporttopcompanies_seencompany[company] = "seen";
      }
    }
    multifunctiontab[year] = yeardata;
  }
  
  mavgalpha = 0.4;
  genericmultifunctionreport(base,htmlfile,startyear,endyear,0,normalized,mavgalpha,
                             "Comparison of companies over the years","Year",
			     normalized ? "% of " docnameu "s" : "# of " docnameu "s");
  printf("<p>This tracks publication of %ss with authors from most active companies per year. Company data is calculated from the first occurrence of an author. ISI has been excluded from this graph, to make it clearer.\n", docname) >> htmlfile;
  if (normalized) printf("The graph is normalized to 100%% representing sum of the top companies.\n") >> htmlfile;
  printf("The data has been smoothed using an exponential moving average with alpha = %.2f.\n", mavgalpha) >> htmlfile;
  if (lastyearcompletion >= minimumcompletion) {
    printf("The last year (%d) has been adjusted assuming the same publication rates as has happened during first part of the year (now %.2f %% complete).\n", nowyear, 100 * lastyearcompletion) >> htmlfile;
  } else {
    printf("The last year (%d) has been excluded because of lack of sufficiently reliable data.\n", nowyear) >> htmlfile;
  }
  printf("</p>\n") >> htmlfile;
  closehtml(htmlfile);
}

function reporttopcountries(normalized) {
  base = "countrydistrhist";
  if (normalized) base = base "_norm";
  htmlfile = base ".html";
  if (normalized)
    openhtml(htmlfile,"Comparison of countries over the years");
  else
    openhtml(htmlfile,"Comparison of countries over the years (normalized)");
  
  #
  # Calculate the main results
  #

  delete reporttopcountries_seencountry;
  delete multifunctiontrack;
  for (country in topcountries) {
    multifunctiontrack[country] = country;
  }

  delete multifunctiontab;
  for (year = startyear; year <= endyear; year++) {
    yeardata = "";
    for (country in topcountries) {
      n = 0;
      split(countrydraftslist[country],compd,"%");
      for (drafti in compd) {
	draft = compd[drafti];
	if (year == draftyears[draft]) {
	  n++;
        }
      }
      if (reporttopcountries_seencountry[country] != "" ||
	  n != 0) {
	if (reporttopcountries_seencountry[country] == "") {
          if (multifunctiontab[year-1] != "") {
            multifunctiontab[year-1] = multifunctiontab[year-1] "%";
          }
          multifunctiontab[year-1] = multifunctiontab[year-1] country ":0";
        }
        if (yeardata != "") yeardata = yeardata "%";
        if (year == nowyear) {
          if (lastyearcompletion >= minimumcompletion) {
            yeardata = yeardata country ":" (n * (1/lastyearcompletion));
          }
        } else {
          yeardata = yeardata country ":" n;
        }
	reporttopcountries_seencountry[country] = "seen";
      }
    }
    multifunctiontab[year] = yeardata;
  }

  #
  # Output the graphs
  #
  
  mavgalpha = 0.4;
  genericmultifunctionreport(base,htmlfile,startyear,endyear,1,normalized,mavgalpha,
                             "Comparison of countries over the years","Year",
			     normalized ? ("% of " docnameu "s") : ("# of " docnameu "s"));
  printf("<p>This tracks publication of %ss with authors from a given country. Country data is calculated from the first occurrence of an author.\n",
         docname) >> htmlfile;
  printf("The scale is logarithmic, ") >> htmlfile;
  if (normalized) printf("normalized to 100%% representing sum of the top countries, ") >> htmlfile;
  printf("and data has been smoothed using an exponential moving average with alpha = %.2f.\n", mavgalpha) >> htmlfile;
  if (lastyearcompletion >= minimumcompletion) {
    printf("The last year (%d) has been adjusted assuming the same publication rates as has happened during first part of the year (now %.2f %% complete).\n", nowyear, 100 * lastyearcompletion) >> htmlfile;
  } else {
    printf("The last year (%d) has been excluded because of lack of sufficiently reliable data.\n", nowyear) >> htmlfile;
  }
  printf("</p>\n") >> htmlfile;
  closehtml(htmlfile);
}

function reporttopcontinents(normalized) {
  base = "d-contdistrhist";
  if (normalized) base = base "_norm";
  htmlfile = base ".html";
  openhtml(htmlfile,"Comparison of continents over the years");

  delete reporttopcontinents_seencontinent;
  delete multifunctiontrack;
  for (continent in continentdrafts) {
    multifunctiontrack[continent] = continent;
  }

  delete multifunctiontab;
  for (year = startyear; year <= endyear; year++) {
    yeardata = "";
    for (continent in continentdrafts) {
      n = 0;
      split(continentdraftslist[continent],compd,"%");
      for (drafti in compd) {
	draft = compd[drafti];
	if (year == draftyears[draft]) {
	  n++;
        }
      }
      if (reporttopcontinents_seencontinent[continent] != "" ||
	  n != 0) {
	if (reporttopcontinents_seencontinent[continent] == "") {
          if (multifunctiontab[year-1] != "") {
            multifunctiontab[year-1] = multifunctiontab[year-1] "%";
          }
          multifunctiontab[year-1] = multifunctiontab[year-1] continent ":0";
        }
        if (yeardata != "") yeardata = yeardata "%";
        if (year == nowyear) {
          if (lastyearcompletion >= minimumcompletion) {
            yeardata = yeardata continent ":" (n * (1/lastyearcompletion));
          }
        } else {
          yeardata = yeardata continent ":" n;
        }
	reporttopcontinents_seencontinent[continent] = "seen";
      }
    }
    multifunctiontab[year] = yeardata;
  }
  
  mavgalpha = 0.4;
  genericmultifunctionreport(base,htmlfile,startyear,endyear,1,normalized,mavgalpha,
                             "Comparison of continents over the years","Year",
			     normalized ? "% of " docnameu "s" : "# of " docnameu "s");
  printf("<p>This tracks publication of %ss with authors from a given continent.\n",
         docname) >> htmlfile;
  printf("Location data is calculated from the first occurrence of an author.\n") >> htmlfile;
  printf("The scale is logarithmic, ") >> htmlfile;
  if (normalized) printf("normalized to 100%% representing sum of the top companies, ") >> htmlfile;
  printf("and data has been smoothed using an exponential moving average with alpha = %.2f.\n", mavgalpha) >> htmlfile;
  if (lastyearcompletion >= minimumcompletion) {
    printf("The last year (%d) has been adjusted assuming the same\n",
           nowyear) >> htmlfile;
    printf("publication rates as has happened during first part of the year (now %.2f %% complete).\n",
	   100 * lastyearcompletion) >> htmlfile;
  } else {
    printf("The last year (%d) has been excluded because of lack of sufficiently reliable data.\n", nowyear) >> htmlfile;
  }
  printf("</p>\n") >> htmlfile;
  closehtml(htmlfile);
}

function genericmultifunctionreport(base,htmlfile,startx,endx,islog,isnormalized,mavgalpha,title,xlabel,ylabel) {
  pngfile = base ".png";
  epsfile = base ".eps";
  datafilebase = "-" base ".dat";
  gplfile = base".txt";

  html_file_css_section("Statistics",htmlfile);
  printf("<img src=%c%s%c style=%cborder:0%c alt=%cstatistics%c/>\n", 34, pngfile, 34, 34, 34, 34, 34) >> htmlfile;
  html_file_css_section("Details",htmlfile);

  for (track in multifunctiontrack) {
    trackfile = track datafilebase;
    printf("\n") > trackfile;
  }

  #
  # Calculate the per-x sums, if normalization needed
  #

  delete multifunctionsum;
  for (i = startx; i <= endx; i++) {
    sum = 0;
    e = multifunctiontab[i];
    split(e,multifunctionint,"%");
    for (vi in multifunctionint) {
      v = multifunctionint[vi];
      split(v,multifunctionint2,":");
      y = multifunctionint2[2];
      sum += y;
    }
    multifunctionsum[i] = sum;
  }

  #
  # Output the main per-track plotting data
  #

  maxy = 0;  
  for (i = startx; i <= endx; i++) {
    e = multifunctiontab[i];
    split(e,multifunctionint,"%");
    for (vi in multifunctionint) {
      v = multifunctionint[vi];
      split(v,multifunctionint2,":");
      track = multifunctionint2[1];
      y = multifunctionint2[2];
      if (!(track in multifunctiontrack)) {
	printf("authorstats: Error: track %s is unknown at %d for %s -- exit\n", track, i, htmlfile);
	exit(1);
      }
      trackfile = track datafilebase;
      if (isnormalized) {
        printf("%d	%6.4f\n",
	       i,
	       multifunctionsum[i] == 0 ? 0.0 : (y * 100.0) / multifunctionsum[i]) >> trackfile;
      } else {
        printf("%d	%d\n", i, y) >> trackfile;
      }
      if (y > maxy) maxy = y;
    }
  }
  
  for (track in multifunctiontrack) {
    trackfile = track datafilebase;
    close(trackfile);
  }

  if (mavgalpha > 0.0) {
    for (track in multifunctiontrack) {
      trackfile = track datafilebase;
      origtrackfile = "orig-" trackfile;
      cmd = sprintf("mv %c%s%c %c%s%c", 34, trackfile, 34, 34, origtrackfile, 34);
      #printf("doing %s...\n", cmd);
      system(cmd);
      mavg(origtrackfile,trackfile,mavgalpha);
    }
  }

  printf("set output %c%s%c\n", 34, epsfile, 34) > gplfile;
  #printf("set terminal png\n") >> gplfile;
  printf("set terminal postscript eps enhanced %cTimes-Roman%c 28 color solid\n", 34, 34) >> gplfile;
  printf("set data style lines\n") >> gplfile;
  #printf("set data style linespoints\n") >> gplfile;
  #printf("set style linespoints linewidth 4\n") >> gplfile;
  printf("set title %c%s%c\n", 34, title, 34) >> gplfile;
  printf("set xlabel %c%s%c\n", 34, xlabel, 34) >> gplfile;
  printf("set ylabel %c%s%c\n", 34, ylabel, 34) >> gplfile;
  printf("set size 2.7,2.5\n") >> gplfile;
  #printf("set size 1.7,1.5\n") >> gplfile;
  if (islog) {
    printf("set logscale y 2\n") >> gplfile;
    if (isnormalized) {
      printf("set yrange [ 0.5 : 100]\n") >> gplfile;
    } else {
      printf("set yrange [ 1 : %d]\n", roundup(maxy)) >> gplfile;
    }
    printf("set ytics ( %c1%c 1, %c5%c 5, %c20%c 20, %c10%c 10, %c50%c 50, %c100%c 100, %c200%c 200 )\n",
           34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34, 34) >> gplfile;
  }
  printf("set key top left\n") >> gplfile;
  printf("set pointsize 2\n") >> gplfile;
  #printf("set style fill solid\n") >> gplfile;
  printf("set view ,,2\n") >> gplfile;

  f = 0;
  printf("plot ") >> gplfile;
  for (track in multifunctiontrack) {
    trackfile = track datafilebase;
    if (f > 0) {
      printf(", ") >> gplfile;
    }
    f++;
    printf("%c%s%c", 34, trackfile, 34) >> gplfile;
    #printf(" smooth cspline") >> gplfile;
    printf(" with lines lt %d lw 4", f+1) >> gplfile;
    printf(" title %c%s%c", 34, countrycapitalize(track), 34) >> gplfile;
  }
  printf("\n") >> gplfile;

  plottopng(gplfile,epsfile,pngfile);
}

function mavg(fromfile,tofile,alpha) {

  # Read data

  delete mavgtab1a;
  delete mavgtab1b;
  i = 1;
  while ((getline frome < fromfile) == 1) {
    if (length(frome) > 1 &&
        split(frome,mavgtab2,"\t") == 2) {
      mavgtab1a[i] = mavgtab2[1]+0.0;
      mavgtab1b[i] = mavgtab2[2]+0.0;
      i++;
    }
  }
  n = i;

  # Calculate moving average
  delete mavgtab3;
  mavgtab3[1] = mavgtab1b[1];
  for (i = 2; i < n; i++) {
    mavgtab3[i] = mavgtab3[i-1] + alpha * (mavgtab1b[i] - mavgtab3[i-1]);
  }

  # Output data
  printf("\n") > tofile;
  for (i = 1; i < n; i++) {
    printf("%6.2f	%6.2f	# orig was %6.2f\n", mavgtab1a[i], mavgtab3[i], mavgtab1b[i]) >> tofile;
  }
}

function reportcompanies() {
  base = "companydistr";
  htmlfile = base ".html";
  epsfile = base ".eps";
  jpgfile = base ".jpg";
  datafile = base ".dat";
  gplfile = base".txt";

  openhtml(htmlfile,"Distribution of authors from companies");
  html_file_css_section("Statistics",htmlfile);
  printf("<img src=%c%s%c style=%cborder:0%c alt=%cdistribution%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  html_file_css_section("Details",htmlfile);
  printf("<ul>\n") >> htmlfile;

  maxauthors = 0;
  for (company in companyauthors) if (companyauthors[company] > maxauthors) maxauthors = companyauthors[company];

  nth = 1;
  extra = "set xtics rotate (";
  lasti = 0;
  printf("\n") > datafile;
  for (i = maxauthors; i >= 1; i--) {
    for (company in companyauthors) {
      if (companyauthors[company] == i) {
        printf("<li><a href=%c%s%s%c>%s</a> has %d (%.2f%%) authors.</li>\n",
               34, hrefprefix, tofname_comp(company), 34, companycapitalize(company),
               i, (100.0 * i) / nauthors) >> htmlfile;
	if (nth <= ntopcompanies + 1 && company != "isi") {
	  topcompanies[company] = company;
        }
	if (nth < 40) {
          printf("%d	%d\n", nth, i) >> datafile;
          cot = companycapitalize(company);
          sub(/ /,"\\n",cot);
          extra = extra quote cot quote " " nth ", ";
	  nth++;
	} else {
	  if (i > lasti) lasti = i;
        }
      }
    }
  }
  if (lasti > 0) {
    printf("%d	%d\n", nth, 2) >> datafile;
    extra = extra quote "(1-" lasti " authors)" quote " " nth++ ", ";
  }
  printf("<li>unknown author company for %d authors (%.2f%%).</li>\n",
         nunknowncompany, (100.0 * nunknowncompany) / nauthors) >> htmlfile;
  printf("%d	%d\n", nth, nunknowncompany) >> datafile;
  extra = extra quote "(unknown)" quote " " nth ")";
  extra = extra " font " quote "Times-Roman,30" quote "\n";
  #extra = extra "set xrange [0 : " nth + 1 "]\n";
  extra = extra "set yrange [0 : " roundup(maxauthors) "]\n";
  extra = extra "set xrange [0.5 : " nth+0.5 "]\n";
  basicgnuplotsettings(gplfile,epsfile,datafile,
                       "Company","# of Authors","Number of authors in a company",
                       0,extra);

  plottojpg(gplfile,epsfile,jpgfile);
  printf("</ul>\n") >> htmlfile;
  printf("<p>Company data is calculated from the first occurrence of an author.</p>\n") >> htmlfile;
  closehtml(htmlfile);
}

function reportcountries(useeu) {
  if (useeu) {
    base = "countryeudistr";
  } else {
    base = "countrydistr";
  }
  htmlfile = base ".html";
  epsfile = base ".eps";
  jpgfile = base ".jpg";
  datafile = base ".dat";
  gplfile = base".txt";

  openhtml(htmlfile,"Distribution of authors from countries");
  html_file_css_section("Statistics",htmlfile);
  printf("<img src=%c%s%c style=%cborder:0%c alt=%cdistribution%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  html_file_css_section("Details",htmlfile);
  printf("<ul>\n") >> htmlfile;

  if (useeu) maxauthors = euauthors; else maxauthors = 0;
  for (country in countryauthors) if (countryauthors[country] > maxauthors) maxauthors = countryauthors[country];

  nth = 1; printf("\n") > datafile;
  seeneu = 0;
  extra = "set xtics rotate (";
  for (i = maxauthors; i >= 1; i--) {
    for (country in countryauthors) {
      na = countryauthors[country];
      co = country;
      if (useeu && isineu[country]) {
        if (seeneu) {
          continue;
        } else {
          co = "european union";
          na = euauthors;
          seeneu = (na == i);
        }
      }
      if (na == i) {
        printf("<li>%d authors (%.2f%%) come from <a href=%c%s%s%c>%s</a>",
               i, (100.0 * i) / nauthors,
               34, hrefprefix, tofname(co), 34, countrycapitalize(co)) >> htmlfile;
        if (i <= 5 && co != "european union") {
          printf(" (%s)", lastnames(countryauthorslist[country])) >> htmlfile;
        }
        printf(".</li>\n") >> htmlfile;
        printf("%d	%d\n", nth, i) >> datafile;
        cot = countrycapitalize(coname(co));
        #sub(/ /,"\\n",cot);
        extra = extra quote cot quote " " nth ", ";
	nth++;
      }
    }
  }
  printf("<li>unknown author location for %d authors (%.2f%%).</li>\n",
         nunknowncountry, (100.0 * nunknowncountry) / nauthors) >> htmlfile;
  printf("%d	%d\n", nth, nunknowncountry) >> datafile;

  extra = extra quote "(unknown)" quote " " nth ")";
  if (useeu) {
    extra = extra " font " quote "Times-Roman,30" quote "\n";
  } else {
    extra = extra " font " quote "Times-Roman,24" quote "\n";
  }
  extra = extra "set xrange [0 : " nth + 1 "]\n";
  extra = extra "set yrange [0 : " maxauthors "]\n";
  basicgnuplotsettings(gplfile,epsfile,datafile,"Country","# of Authors","Number of authors in a country",0,extra);

  plottojpg(gplfile,epsfile,jpgfile);
  printf("</ul>\n") >> htmlfile;
  printf("<p>Location data is calculated from the first occurrence of an author.</p>\n") >> htmlfile;
  closehtml(htmlfile);
}

function reportcontinents_draft() {
  base = "d-contdistr";
  htmlfile = base ".html";
  epsfile = base ".eps";
  jpgfile = base ".jpg";
  datafile = base ".dat";
  gplfile = base".txt";

  openhtml(htmlfile,"Distribution of " docname "s according to the continents of their authors");
  html_file_css_section("Statistics",htmlfile);
  printf("<img src=%c%s%c style=%cborder:0%c alt=%cdistribution%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  html_file_css_section("Details",htmlfile);
  printf("<ul>\n") >> htmlfile;

  continentauthors["antarctica"] = 0;
  for (country in countryauthors) {
    continent = continentof[country];
    if (continent == "") continent = "OTHER";
    if (continentdrafts[continent] == "") continentdrafts[continent] = 0;
    continentdrafts[continent] = continentdrafts[continent] + countrydrafts[country];
    if (countrydraftslist[country] != "") {
      if (continentdraftslist[continent] != "") continentdraftslist[continent] = continentdraftslist[continent] "%";
      continentdraftslist[continent] = continentdraftslist[continent] countrydraftslist[country];
    }
  }
  
  i = 0;
  maxn = 0;
  printf("\n") > datafile;
  extra = "set xtics rotate (";
  for (continent in continentdrafts) {
    if (continentdrafts[continent] > 0 || continent != "OTHER") {
      if (maxn < continentdrafts[continent]) maxn = continentdrafts[continent];
      printf("<li>%d %ss (%.2f%%) have authors from %s.</li>\n",
             continentdrafts[continent],
             docname,
	     (100.0 * continentdrafts[continent]) / ndrafts,
             countrycapitalize(continent)) >> htmlfile;
      printf("%d	%d\n", i, continentdrafts[continent]) >> datafile;
      continentt = countrycapitalize(continent);
      sub(/ /,"\\n",continentt);
      extra = extra quote continentt quote " " i ", ";
      i++;
    }
  }
  extra = extra quote "(unknown)" quote " " i ")\n";
  #extra = extra " font " quote "Times-Roman,40" quote "\n";
  extra = extra "set xrange [-1 : " i + 1 "]\n";
  extra = extra "set yrange [0 : " roundup(maxn) "]\n";
  basicgnuplotsettings(gplfile,epsfile,datafile,"Continent","# of " docnameu "s","Continent",0,extra);

  plottojpg(gplfile,epsfile,jpgfile);

  printf("</ul>\n") >> htmlfile;
  printf("<p>Wondering why the total is greater than 100%%? %ss with multiple authors may be counted multiple times, if the authors are from different countries.</p>\n", docnameu) >> htmlfile;
  closehtml(htmlfile);
}

function reportcompanies_draft() {
  base = "d-companydistr";
  htmlfile = base ".html";
  epsfile = base ".eps";
  jpgfile = base ".jpg";
  datafile = base ".dat";
  gplfile = base".txt";

  openhtml(htmlfile,"Distribution of " docname "s according to the affiliation of their authors");
  html_file_css_section("Statistics",htmlfile);
  printf("<img src=%c%s%c style=%cborder:0%c alt=%cdistribution%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  html_file_css_section("Details",htmlfile);
  printf("<ul>\n") >> htmlfile;

  maxdrafts = 0;
  for (company in companydrafts) {
    n = split(companydrafts[company],compd,"%");
    if (n > maxdrafts) maxdrafts = n;
  }

  nth = 1; printf("\n") > datafile;
  for (i = maxdrafts; i >= 1; i--) {
    for (company in companydrafts) {
      if (companydrafts[company] == i) {
        n = split(companydrafts[company],compd,"%");
        printf("<li><a href=%c%s%s%c>%s</a> has %d (%.2f%%) drafts.</li>\n",
               34, hrefprefix, tofname_comp(company), 34, companycapitalize(company),
               i, (100.0 * i) / ndrafts) >> htmlfile;
        printf("%d	%d\n", nth++, i) >> datafile;
      }
    }
  }
  basicgnuplotsettings(gplfile,epsfile,datafile,"Company","# of " docnameu,"Number of " docname "s by a company",0,"");

  plottojpg(gplfile,epsfile,jpgfile);
  printf("</ul>\n") >> htmlfile;
  printf("<p>Wondering why the total is greater than 100%%? %ss with multiple authors may be counted multiple times, if the authors are from different companies.</p>\n", docnameu) >> htmlfile;
  closehtml(htmlfile);
}

function reportcountries_draft(useeu) {
  if (useeu) {
    base = "d-countryeudistr";
  } else {
    base = "d-countrydistr";
  }
  htmlfile = base ".html";
  epsfile = base ".eps";
  jpgfile = base ".jpg";
  datafile = base ".dat";
  gplfile = base".txt";

  openhtml(htmlfile,"Distribution of " docname "s according to the countries of their authors");
  html_file_css_section("Statistics",htmlfile);
  printf("<img src=%c%s%c style=%cborder:0%c alt=%cdistribution%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  html_file_css_section("Details",htmlfile);

  printf("<p>%ss come from <a href=%cd-countrydistr.html%c>these countries</a>, %d different countries in total. \n",
	 docnameu, 34, 34, ncountries) >> htmlfile;
  printf("</p>\n") >> htmlfile;
  printf("<ul>\n") >> htmlfile;
  if (useeu) maxdrafts = eudrafts; else maxdrafts = 0;
  for (country in countryauthors) if (countrydrafts[country] > maxdrafts) maxdrafts = countrydrafts[country];

  nth = 1; printf("\n") > datafile;
  seeneu = 0;
  extra = "set xtics rotate (";
  for (i = maxdrafts; i >= 1; i--) {
    for (country in countrydrafts) {
      nd = countrydrafts[country];
      co = country;
      if (useeu && isineu[country]) {
        if (seeneu) {
          continue;
        } else {
          co = "european union";
          nd = eudrafts;
          seeneu = (nd == i);
        }
      }
      if (nd == i) {
        printf("<li>%d %ss (%.2f%%) have authors from <a href=%c%s%s%c>%s</a>",
               i, docname, (100.0 * i) / ndrafts,
               34, hrefprefix, tofname(co), 34, countrycapitalize(co)) >> htmlfile;
	if (nth < ntopcountries && !useeu) {
	  topcountries[co] = co;
        }
        if (i <= 2 && co != "european union") {
          v = countrydraftslist[country];
	  split(v,draftc,"%");
          printf(" (") >> htmlfile;
          commayet = 0;
          for (g in draftc) {
            if (commayet) {
              printf(", ") >> htmlfile;
            } else {
	      commayet = 1;
            }
            printf("<a href=%c%s%s%c>%s</a>", 34, docprefix, draftc[g], 34, draftc[g]) >> htmlfile;
          }
          printf(")", v) >> htmlfile;
        }
        printf(".</li>\n") >> htmlfile;
        printf("%d	%d\n", nth, i) >> datafile;
        cot = countrycapitalize(coname(co));
        #sub(/ /,"\\n",cot);
        extra = extra quote cot quote " " nth ", ";
	nth++;
      }
    }
  }
  extra = extra quote "(unknown)" quote " " nth ")";
  if (useeu) {
    extra = extra " font " quote "Times-Roman,30" quote "\n";
  } else {
    extra = extra " font " quote "Times-Roman,20" quote "\n";
  }
  extra = extra "set xrange [0 : " nth + 1 "]\n";
  extra = extra "set yrange [0 : " maxdrafts "]\n";
  basicgnuplotsettings(gplfile,epsfile,datafile,"","# of " docnameu "s",
                       "Number of " docname "s with authors from a country",
		       0,extra);

  plottojpg(gplfile,epsfile,jpgfile);
  printf("</ul>\n") >> htmlfile;
  printf("<p>Wondering why the total is greater than 100%%? %ss with multiple authors may be counted multiple times, if the authors are from different countries.</p>\n", docnameu) >> htmlfile;
  printf("<p>See also the ranking of the countries <a href=%c%scap.html%c>per capita</a>.</p>\n",
         34, base, 34) >> htmlfile;
  closehtml(htmlfile);
  reportcountries_draft_percap(useeu);
}

function reportcountries_draft_percap(useeu) {
  if (useeu) {
    base = "d-countryeudistrcap";
  } else {
    base = "d-countrydistrcap";
  }
  htmlfile = base ".html";
  epsfile = base ".eps";
  jpgfile = base ".jpg";
  datafile = base ".dat";
  gplfile = base".txt";

  openhtml(htmlfile,"Distribution of " docname "s according to the countries of their authors, per capita");
  html_file_css_section("Statistics",htmlfile);
  printf("<img src=%c%s%c style=%cborder:0%c alt=%cdistribution%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  html_file_css_section("Details",htmlfile);
  printf("<ul>\n") >> htmlfile;
  printf("\n") > datafile;

  if (useeu) {
    #printf("rcdpc eu\n");
    checkpopulation("european union");
    maxdrafts = eudrafts / (1.0 * population["european union"]);
    #printf("max document density is %f for %s\n", maxdrafts, "european union");
  } else {
    maxdrafts = 0.0;
  }
  for (country in countryauthors) {
    #printf("rcdpc2 %s\n", country, countryauthors[country]);
    checkpopulation(country);
    forthis = 1000 * 1000 * countrydrafts[country] / (1.0 * population[country]);
    #printf("%s: %d at %d (%f)\n", country, countrydrafts[country], population[country], forthis);
    if (forthis > maxdrafts) {
      maxdrafts = forthis;
      #printf("max document density is %f for %s\n", maxdrafts, country);
    }
  }

  seeneu = 0;
  nth = 1;
  extra = "set xtics rotate (";
  delta = maxdrafts/500.0;
  for (i = maxdrafts; delta > 0.0 && i >= -delta; i -= delta) {
    #printf("stepping at %.20f\n", i);
    for (country in countrydrafts) {
      #printf("testing %s\n", country);
      nd = countrydrafts[country];
      co = country;
      if (useeu && isineu[country]) {
        if (seeneu) {
	  #printf("already seen eu\n");
          continue;
        } else {
          co = "european union";
          nd = eudrafts;
	  #printf("testing now eu\n");
        }
      }
      nd = nd;
      forthis = 1000 * 1000 * nd / (1.0 * population[co]);
      if (forthis < i + delta && forthis >= i) {
        printf("<li>%.2f %ss per million people from <a href=%c%s%s%c>%s</a> (%.2f million people)",
               forthis, docname,
               34, hrefprefix, tofname(co), 34, countrycapitalize(co),
               population[co] / (1000 * 1000)) >> htmlfile;
        printf(".</li>\n") >> htmlfile;
        printf("%d	%f\n", nth, forthis) >> datafile;
        cot = countrycapitalize(coname(co));
        #sub(/ /,"\\n",cot);
        extra = extra quote cot quote " " nth ", ";
	nth++;
        if (useeu && co == "european union") seeneu = 1;
      }
    }
  }
  extra = extra quote "(unknown)" quote " " nth ")";
  if (useeu) {
    extra = extra " font " quote "Times-Roman,30" quote "\n";
  } else {
    extra = extra " font " quote "Times-Roman,20" quote "\n";
  }
  extra = extra "set xrange [0 : " nth + 1 "]\n";
  extra = extra "set yrange [0 : " maxdrafts "]\n";
  basicgnuplotsettings(gplfile,epsfile,datafile,"",
                       "# of " docnameu "s per million capita",
                       "Number of " docname "s with authors from a country, per capita",
		       0,extra);

  plottojpg(gplfile,epsfile,jpgfile);
  printf("</ul>\n") >> htmlfile;
  printf("<p>Source of population data is Wikipedia.</p>\n") >> htmlfile;
  closehtml(htmlfile);
}

function checkpopulation(co) {
  if (population[co] == "" ||
      population[co] == 0) {
    fatalerror("Population for " co " is unknown");
  }
}

function fatalerror(s) {
  printf("authorstats: Fatal error: %s -- exit\n", s);
  exit(1);
}

function lastnames(s) {
  nk = split(s,nameparts,", ");
  result = lastname(nameparts[1]);
  for (k = 2; k <= nk; k++) {
    result = result ", " lastname(nameparts[k]);
  }
  return(result);
}

function lastname(y) {
  nr = split(y,lnamecomps," ");
  res = lnamecomps[nr];
  if (res == "3rd" || res == "2nd" || res == "III" || res == "II") res = lnamecomps[nr-1];
  if (substr(res,1,1) == "(") res = lnamecomps[nr-1];
  return(res);
}

function official(s) {
  return(s ~ /draft-ietf/ ||
         s ~ /draft-iab/ ||
         s ~ /draft-rfc/ ||
         s ~ /draft-iesg/ ||
         s ~ /draft-irtf/);
}

function isinarea(draft,area) {
  if (isarea[area] == "") return(0);
  wg = getofficialwg(draft);
  if (wg == "") wg = getrelatedwg(draft);
  if (wg == "") return(0);
  if (iswg[wg] == "") return(0);
  if (wgarea[wg] != area) return(0);
  return(1);
}

function reportauthoractivitiesperarea() {
  base = "authactareadistr";
  htmlfile = base ".html";
  openhtml(htmlfile,"Most active authors per area");
    
  for (area in isarea) {

    html_file_css_section(area " Area", htmlfile);
    printf("<ul>\n") >> htmlfile;
    maxdrafts = 0;
    delete numbersofdrafts;
    delete numbersofdraftsauthor;

    for (i = 0; i < 500; i++) {
      numbersofdrafts[n] = 0;
      numbersofdraftsauthor[n] = "";
    }

    for (author in authordrafts) {
      split(authordrafts[author],acomps,"%");
      n = 0;
      delete acomps2;
      for (drafti in acomps) {
        if (isinarea(acomps[drafti],area)) {
	   acomps2[++n] = acomps[drafti];
        }
      }
      if (n > 0) {
        numbersofdrafts[n] = numbersofdrafts[n] + 1;
        if (numbersofdraftsauthor[n] == "") {
          numbersofdraftsauthor[n] = author;
        } else {
  	  numbersofdraftsauthor[n] = numbersofdraftsauthor[n] "%" author;
        }
        if (n > maxdrafts) maxdrafts = n;
      }
    }
    
    printedauthors = 0;
    for (i = maxdrafts; i >= 0 && printedauthors < 20; i--) {
      if (numbersofdrafts[i] > 0) {
        split(numbersofdraftsauthor[i],guiltyauthors,"%");
        for (h in guiltyauthors) {
          printf("<li><a href=%c%s%s%c>%s</a> (%d %ss).</li>",
                 34, hrefprefix, tofname(guiltyauthors[h]), 34,
                 guiltyauthors[h], i,
                 docname) >> htmlfile;
	  printedauthors++;
        }
      }
    }
    
    printf("</ul>\n") >> htmlfile;
  }
  
  closehtml(htmlfile);
}

function reportauthoractivities(iswg) {
  if (iswg) {
    base = "authactdistr-wg";
  } else {
    base = "authactdistr";
  }
  htmlfile = base ".html";
  epsfile = base ".eps";
  jpgfile = base ".jpg";
  datafile = base ".dat";
  gplfile = base".txt";

  openhtml(htmlfile,"Distribution of authors according to how many documents they have");
  html_file_css_section("Statistics",htmlfile);
  printf("<img src=%c%s%c style=%cborder:0%c alt=%cdistribution%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  html_file_css_section("Details",htmlfile);
  printf("<p>Total number of distinct authors is %d, the different %ss together had a total of %d author entries.</p>",
	 nauthors, docname, nauthoroccurrences) >> htmlfile;
  printf("<ul>\n") >> htmlfile;
  maxdrafts = 0;
  delete numbersofdrafts;
  delete numbersofdraftsauthor;

  for (i = 0; i < 500; i++) {
    numbersofdrafts[n] = 0;
    numbersofdraftsauthor[n] = "";
  }

  for (author in authordrafts) {
    n = split(authordrafts[author],acomps,"%");
    if (n > 0) {
      if (iswg) {
        minus = 0;
        for (j = 1; j <= n; j++) {
          if (!official(acomps[j])) minus++;
        }
        #printf("for %s, its %d - %d = %d\n", author, n, minus, n - minus) >> "/tmp/debug";
        n = n - minus;
      }
      numbersofdrafts[n] = numbersofdrafts[n] + 1;
      if (numbersofdraftsauthor[n] == "") {
        numbersofdraftsauthor[n] = author;
      } else {
	numbersofdraftsauthor[n] = numbersofdraftsauthor[n] "%" author;
      }
      if (n > maxdrafts) maxdrafts = n;
    }
  }

  printf("\n") > datafile;
  for (i = maxdrafts; i >= 0; i--) {
    if (numbersofdrafts[i] > 0 || (iswg && i == 0)) {
      printf("<li>%d authors (%.2f%%) have %d %ss", numbersofdrafts[i],
  	     (100.0 * numbersofdrafts[i]) / nauthors, i, docname) >> htmlfile;
      if (numbersofdrafts[i] > 0 && numbersofdrafts[i] <= 14) {
        printf(" (") >> htmlfile;
	commayet = 0;
        split(numbersofdraftsauthor[i],guiltyauthors,"%");
        for (h in guiltyauthors) {
          if (commayet) {
            printf(", ") >> htmlfile;
          } else {
            commayet = 1;
          }
          printf("<a href=%c%s%s%c>%s</a>",
                 34, hrefprefix, tofname(guiltyauthors[h]), 34,
                 lastname(guiltyauthors[h])) >> htmlfile;
        }
        printf(")") >> htmlfile;
      }
      printf("</li>\n") >> htmlfile;
    }
    printf("%d	%.2f\n", i, (100.0 * numbersofdrafts[i]) / nauthors) >> datafile;
  }

  basicgnuplotsettings(gplfile,epsfile,datafile,
                       "# of " docnameu "s","% of Authors","Number of " docname "s by author",
		       1,"");

  plottojpg(gplfile,epsfile,jpgfile);
  printf("</ul>\n") >> htmlfile;
  if (iswg) printf("<p>(The percentages have been calculated against all authors, not just WG document authors.)</p>\n") >> htmlfile;
  closehtml(htmlfile);

  if (!iswg) realmaxdrafts = maxdrafts;
}

function reportauthorcrosswg() {
  base = "authcrosswg";
  htmlfile = base ".html";
  #epsfile = base ".eps";
  #jpgfile = base ".jpg";
  #datafile = base ".dat";
  #gplfile = base".txt";

  openhtml(htmlfile,"Author cross-WG activities");
  #html_file_css_section("Statistics",htmlfile);
  #printf("<img src=%c%s%c style=%cborder:0%c alt=%cstatistics%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  html_file_css_section("Details",htmlfile);
  printf("<ul>\n") >> htmlfile;

  maxwgs = 0;
  for (author in authordrafts) {
    split(authordrafts[author],acomps,"%");
    authorwgs[author] = "";
    count = 0;
    for (drafti in acomps) {
      draft = acomps[drafti];
      wg = getofficialwg(draft);
      if (wg == "") wg = getrelatedwg(draft);
      if (debug) printf("draftwg(%s)=%s\n", draft, wg);
      if (wg != "") {
        if (authorwgs[author] == "") {
          authorwgs[author] = wg;
	  count++;
        } else if (index(authorwgs[author], wg) == 0) {
          authorwgs[author] = authorwgs[author] "%" wg;
	  count++;
        }
      }
    }
    if (count > maxwgs) maxwgs = count;
    if (authorswithnwgs[count] == "") {
      authorswithnwgs[count] = author;
    } else {
      authorswithnwgs[count] = authorswithnwgs[count] "%" author;
    }
  }
  
  #printf("#\n") > datafile;
  for (i = maxwgs; i >= 0; i--) {
    n = split(authorswithnwgs[i],acomps,"%");
    if (n > 0) {
      #printf("%d	%d\n", n, i) >> datafile;
      printf("<li>%d authors work in %d WGs", n, i) >> htmlfile;
      if (n < 15) {
        printf(" (") >> htmlfile;
        commayet = 0;
        for (authori in acomps) {
          author = acomps[authori];
          if (commayet) {
            printf(", ") >> htmlfile;
          } else {
  	    commayet = 1;
         }
          printf("<a href=%c%s%s%c>%s</a>", 34, hrefprefix, tofname(author), 34, lastname(author)) >> htmlfile;
        }
        printf(")") >> htmlfile;
      }
      printf("</li>\n") >> htmlfile;
    }
  }

  #extra = "set xrange [-1 : " nauthors "]\n";
  #extra = extra "set xlabel\n";
  #basicgnuplotsettings(gplfile,epsfile,datafile,
  #		       "Authors", "Number of WGs","Number of WGs with official or related " docname "s",
  #		       1,extra);
  #
  #plottojpg(gplfile,epsfile,jpgfile);

  printf("</ul>\n") >> htmlfile;
  closehtml(htmlfile);
}

function reportauthorcrossarea() {
  base = "authcrossarea";
  htmlfile = base ".html";
  #epsfile = base ".eps";
  #jpgfile = base ".jpg";
  #datafile = base ".dat";
  #gplfile = base".txt";

  openhtml(htmlfile,"Author cross-area activities");
  #html_file_css_section("Statistics",htmlfile);
  #printf("<img src=%c%s%c style=%cborder:0%c alt=%cstatistics%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  html_file_css_section("Details",htmlfile);
  printf("<ul>\n") >> htmlfile;

  maxareas = 0;
  for (author in authordrafts) {
    split(authordrafts[author],acomps,"%");
    authorareas[author] = "";
    count = 0;
    for (drafti in acomps) {
      draft = acomps[drafti];
      wg = getofficialwg(draft);
      if (wg == "") wg = getrelatedwg(draft);
      area = wgarea[wg];
      if (area != "") {
        if (authorareas[author] == "") {
          authorareas[author] = area;
	  count++;
        } else if (index(authorareas[author], area) == 0) {
          authorareas[author] = authorareas[author] "%" area;
	  count++;
        }
      }
    }
    if (count > maxareas) maxareas = count;
    if (authorswithnareas[count] == "") {
      authorswithnareas[count] = author;
    } else {
      authorswithnareas[count] = authorswithnareas[count] "%" author;
    }
  }
  
  #printf("#\n") > datafile;
  for (i = maxareas; i >= 0; i--) {
    n = split(authorswithnareas[i],acomps,"%");
    if (n > 0) {
      #printf("%d	%d\n", n, i) >> datafile;
      printf("<li>%d authors work in %d areas", n, i) >> htmlfile;
      if (n < 30) {
        printf(" (") >> htmlfile;
        commayet = 0;
        for (authori in acomps) {
          author = acomps[authori];
          if (commayet) {
            printf(", ") >> htmlfile;
          } else {
  	    commayet = 1;
         }
          printf("<a href=%c%s%s%c>%s</a>", 34, hrefprefix, tofname(author), 34, lastname(author)) >> htmlfile;
        }
        printf(")") >> htmlfile;
      }
      printf("</li>\n") >> htmlfile;
    }
  }

  #extra = "set xrange [-1 : " nauthors "]\n";
  #extra = extra "set xlabel\n";
  #basicgnuplotsettings(gplfile,epsfile,datafile,"Author", "Number of areas","Authors with cross-area activity",
  #		       1,extra);
  #
  #plottojpg(gplfile,epsfile,jpgfile);

  printf("</ul>\n") >> htmlfile;
  closehtml(htmlfile);
}

function getofficialwg(draft) {
  split(draft,dnamecomps,"-");
  if (dnamecomps[2] == "ietf" && iswg[dnamecomps[3]]) return dnamecomps[3];
  else return("");
}

function getrelatedwg(draft) {
  split(draft,dnamecomps,"-");
  if (dnamecomps[2] != "ietf" && iswg[dnamecomps[3]]) return(dnamecomps[3]);
  else return("");
}

function reportauthors() {
  base = "authdistr";
  htmlfile = base ".html";
  epsfile = base ".eps";
  jpgfile = base ".jpg";
  datafile = base ".dat";
  gplfile = base".txt";

  openhtml(htmlfile,"Distribution of the number of authors per " docname);
  html_file_css_section("Statistics",htmlfile);
  printf("<img src=%c%s%c style=%cborder:0%c alt=%cdistribution%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  html_file_css_section("Details",htmlfile);
  printf("<ul>\n") >> htmlfile;
  maxauthors = 0;

  for (draft in draftauthors) {
    n = split(draftauthors[draft],acomps,"%");
    if (n > 0) {
      numbersofauthors[n] = numbersofauthors[n] + 1;
      numbersofauthorsdraft[n] = draft;
      if (n > maxauthors) maxauthors = n;
    }
  }

  printf("\n") > datafile;
  for (i = maxauthors; i >= 0; i--) {
    if (numbersofauthors[i] > 0) {
      printf("<li>%d %ss (%.2f%%) have %d authors", numbersofauthors[i], docname,
  	     (100.0 * numbersofauthors[i]) / ndrafts, i) >> htmlfile;
      if (numbersofauthors[i] == 1) {
        printf(" (<a href=%c%s%s%c>%s</a>)",
               34, docprefix, numbersofauthorsdraft[i], 34, draftname(numbersofauthorsdraft[i])) >> htmlfile;
      }
      printf("</li>\n") >> htmlfile;
    }
    printf("%d	%.2f\n", i, (100.0 * numbersofauthors[i]) / ndrafts) >> datafile;
  }

  basicgnuplotsettings(gplfile,epsfile,datafile,"# of Authors", "% of " docname "s","Number of authors in " docname "s",
		       0,"set xtics 1, 1\n");

  plottojpg(gplfile,epsfile,jpgfile);
  printf("</ul>\n") >> htmlfile;
  closehtml(htmlfile);
}

function reportpubyears() {
  base = "pubdistr";
  htmlfile = base ".html";
  epsfile = base ".eps";
  jpgfile = base ".jpg";
  datafile = base ".dat";
  gplfile = base".txt";

  if (doctype == "draft") {
    openhtml(htmlfile,"Publication rate per month");
  } else if (doctype == "rfc") {
    openhtml(htmlfile,"Publication rate per year");

    html_file_css_section("Yearly Statistics",htmlfile);
    printf("<img src=%c%s%c style=%cborder:0%c alt=%cstatistics%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
    html_file_css_section("Details",htmlfile);
    printf("<ul>\n") >> htmlfile;
  
    printf("\n") > datafile;
    maxn = 0;
    for (i = startyear; i <= endyear; i++) {
      n = 0;
      for (draft in draftyears) if (draftyears[draft] == i) n++;
      if (n > maxn) maxn = n;
      if (n > 0) {
        printf("<li>%d %ss in %d.", n, docname, i) >> htmlfile;
        printf("</li>\n") >> htmlfile;
      }
      printf("%d	%d\n", i, n) >> datafile;
    }
  
    extra = "set xrange [" startyear-1 " : " endyear+1 "]\n";
    extra = extra "set yrange [ 0 : " roundup(maxn) "]\n";
    extra = extra "set xtics ";
    if (endyear - startyear > 5) {
      extra = extra "rotate ";
    }
    extra = extra startyear ", 1, " endyear;
    if (endyear - startyear > 20) {
      extra = extra " font " quote "Times-Roman,20" quote "\n";
    } else if (endyear - startyear > 5) {
      extra = extra " font " quote "Times-Roman,30" quote "\n";
    } else {
      extra = extra "\n";
    }
    basicgnuplotsettings(gplfile,epsfile,datafile,"Year", "# of " docnameu "s","Publication rate per year",
  		         0,extra);
    plottojpg(gplfile,epsfile,jpgfile);
    printf("</ul>\n") >> htmlfile;
    printf("<br/>\n") >> htmlfile;
    close(epsfile);close(jpgfile);close(datafile);close(gplfile);
  }

  epsfile = base "2.eps";
  jpgfile = base "2.jpg";
  datafile = base "2.dat";
  gplfile = base "2.txt";
  printf("\n") > datafile;
  maxn = 0;
  for (i = startyear; i <= endyear; i++) {
    for (j = 1; j <= 12; j++) {
      n = 0;
      for (draft in draftyears) if (draftyears[draft] == i && draftmonths[draft] == monthnameshort[j]) n++;
      if (n > maxn) maxn = n;
      printf("%4.2f	%d\n", i - 0.5 + (j-1)/12.0, n) >> datafile;
    }
  }

  html_file_css_section("Monthly Statistics",htmlfile);
  printf("<img src=%c%s%c style=%cborder:0%c alt=%cstatistics%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  extra = "set xrange [" startyear-0.5 " : " endyear+0.5 "]\n";
  extra = extra "set yrange [ 0 : " roundup(maxn) "]\n";
  extra = extra "set xtics ";
  if (endyear - startyear > 5) {
    extra = extra "rotate ";
  }
  extra = extra startyear ", 1, " endyear;
  if (endyear - startyear > 20) {
    extra = extra " font " quote "Times-Roman,20" quote "\n";
  } else if (endyear - startyear > 5) {
    extra = extra " font " quote "Times-Roman,30" quote "\n";
  } else {
    extra = extra "\n";
  }
  extra = extra "set mxtics 0.08\n";
  extra = extra "set boxwidth 0.07\n";
  basicgnuplotsettings(gplfile,epsfile,datafile,"Year and Month", "# of " docnameu "s","Publication rate per month",
		       0,extra);
  plottojpg(gplfile,epsfile,jpgfile);

  closehtml(htmlfile);
}

function roundup(n) {
  if (n < 50) {
    return( 10 * (n/10) + 10);
  } else {
    return( 100 * (n/100) + 100);
  }
}

function reportformats() {
  base = "formatdistr";
  htmlfile = base ".html";
  epsfile = base ".eps";
  jpgfile = base ".jpg";
  datafile = base ".dat";
  gplfile = base".txt";

  openhtml(htmlfile,"Document formats and features in " docname "s");

  html_file_css_section("Formats", htmlfile);
  printf("<p>These formats are used:</p>\n") >> htmlfile;
  printf("<ul>\n") >> htmlfile;
  printf("<li>Plain text only %d documents (%.2f%%).</li>\n",
          ndrafts - nnontextformats, (100.0 * (ndrafts - nnontextformats)) / ndrafts) >> htmlfile;
  printf("<li>PDF %d documents (%.2f%%).</li>\n",
          npdfdrafts, (100.0 * npdfdrafts) / ndrafts) >> htmlfile;
  printf("<li>Postcript %d documents (%.2f%%).</li>\n",
          npostscriptdrafts, (100.0 * npostscriptdrafts) / ndrafts) >> htmlfile;
  printf("</ul>\n") >> htmlfile;

  html_file_css_section_nextcol("Features", htmlfile);
  printf("<p>These features are used within the documents:</p>\n") >> htmlfile;
  printf("<ul>\n") >> htmlfile;
  printf("<li>Packet formats in %d documents (%.2f%%).</li>\n",
         nformats, (100.0 * nformats) / ndrafts) >> htmlfile;
  printf("<li>Figures in %d documents (%.2f%%).</li>\n",
         nfigures, (100.0 * nfigures) / ndrafts) >> htmlfile;
  printf("<li>ABNF in %d documents (%.2f%%).</li>\n",
         nabnfs, (100.0 * nabnfs) / ndrafts) >> htmlfile;
  printf("<li>XML in %d documents (%.2f%%).</li>\n",
         nxmls, (100.0 * nxmls) / ndrafts) >> htmlfile;
  printf("<li>ASN.1 in %d documents (%.2f%%).</li>\n",
         nasn1s, (100.0 * nasn1s) / ndrafts) >> htmlfile;
  printf("<li>C code in %d documents (%.2f%%).</li>\n",
         ncodes, (100.0 * ncodes) / ndrafts) >> htmlfile;
  printf("<li>RFC 2119 keywords in %d documents (%.2f%%).</li>\n",
         nkwds, (100.0 * nkwds) / ndrafts) >> htmlfile;
  printf("<li>Security considerations section in %d documents (%.2f%%).</li>\n",
         nsecconss, (100.0 * nsecconss) / ndrafts) >> htmlfile;
  printf("<li>IANA considerations section in %d documents (%.2f%%).</li>\n",
         nianaconss, (100.0 * nianaconss) / ndrafts) >> htmlfile;
  if (doctype == "rfc") {
    printf("<li>Errata exists for %d documents (%.2f%%).</li>\n",
           nerratas, (100.0 * nerratas) / ndrafts) >> htmlfile;
  }
  printf("</ul>\n") >> htmlfile;

  printf("<p>The document feature recognition is based on heuristics and is inherently unreliable.</p>\n") >> htmlfile;
  closehtml(htmlfile);
}

function reportpages() {
  base = "pagedistr";
  htmlfile = base ".html";
  epsfile = base ".eps";
  jpgfile = base ".jpg";
  datafile = base ".dat";
  gplfile = base".txt";

  openhtml(htmlfile,"Distribution of the number of pages in " docname "s");
  html_file_css_section("Statistics",htmlfile);
  printf("<img src=%c%s%c style=%cborder:0%c alt=%cdistribution%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  html_file_css_section("Details",htmlfile);
  printf("<ul>\n") >> htmlfile;

  maxpages = 0;
  for (draft in draftpagecounts) {
    if (draftpagecounts[draft] > maxpages) maxpages = draftpagecounts[draft];
  }

  printf("\n") > datafile;
  for (i = maxpages; i >= 0; i--) {
    if (pagecounts[i] > 0) {
      printf("<li>%d %ss (%.2f%%) have %d pages", pagecounts[i], docname,
  	     (100.0 * pagecounts[i]) / ndrafts, i) >> htmlfile;
      if (pagecounts[i] <= 3 || (i <= 3 && pagecounts[i] <= 6)) {
        printf(" (") >> htmlfile;
        commayet = 0;
        for (draft in draftpagecounts) {
	  if (draftpagecounts[draft] == i) {
	    if (commayet) {
	      printf(", ") >> htmlfile;
	    } else {
	      commayet = 1;
	    }
            printf("<a href=%c%s%s%c>%s</a>",
                   34, docprefix, draft, 34, draftname(draft)) >> htmlfile;
	  }
	}
	printf(")") >> htmlfile;
      }
      printf("</li>\n") >> htmlfile;
    }
    printf("%d	%.2f\n", i, pagecounts[i]) >> datafile;
  }

  basicgnuplotsettings(gplfile,epsfile,datafile,"# of Pages",
                       "# of " docnameu "s","Number of pages in " docname "s",
		       1,"");

  plottojpg(gplfile,epsfile,jpgfile);
  printf("</ul>\n") >> htmlfile;
  closehtml(htmlfile);
}

function docgettype(t) {
  if (t ~ /^draft/) {
    return("draft");
  } else {
    return("rfc");
  }
}

function draftname(t) {
  gsub(/^draft-/,"",t);
  gsub(/-[0-9][0-9][.]txt/,"",t);
  #gsub(/^rfc/,"",t);
  gsub(/[.]txt/,"",t);
  return(t);
}

function reportareawgs() {
  base = "areawgdistr";
  htmlfile = base ".html";
  epsfile = base ".eps";
  jpgfile = base ".jpg";
  datafile = base ".dat";
  gplfile = base".txt";
  
  openhtml(htmlfile,"Distribution of WGs to areas");
  html_file_css_section("Statistics",htmlfile);
  printf("<p>There are %d WGs and %d areas, an average of %.2f WGs/area.\n",
         nwgs, nareas, (1.0 * nwgs) / nareas) >> htmlfile;
  printf("<img src=%c%s%c style=%cborder:0%c alt=%careas and wgs%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  printf("</p>\n") >> htmlfile;
  html_file_css_section("Details",htmlfile);
  printf("<ul>\n") >> htmlfile;
  printf("\n") > datafile;
  
  nth = 1;
  maxn = 0;
  extra = "set xtics rotate (";
  for (area in isarea) {
    thiscount = 0;
    for (wg in iswg) {
      if (area == wgarea[wg]) {
        thiscount++;
      }
    }
    if (thiscount > maxn) maxn = thiscount;
    printf("<li>%s has %d (%.2f%%) working groups.</li>\n",
           area, thiscount, (100.0 * thiscount) / nwgs) >> htmlfile;
    printf("%d	%d\n", nth, thiscount) >> datafile;
    areat = splittolines(areacapitalize(area));
    if (nth > 1) extra = extra ", ";
    extra = extra quote areat quote " " nth;
    nth++;
  }
  
  extra = extra ")\n";
  extra = extra "set xrange [-0.5 : " nth+1 "]\n";
  basicgnuplotsettings(gplfile,epsfile,datafile,"Area","# of WGs","Number of WGs in an area",0,extra);
  
  plottojpg(gplfile,epsfile,jpgfile);
  printf("</ul>\n") >> htmlfile;
  closehtml(htmlfile);
}

function splittolines(inputstring) {
  sub(/ /,"\\n",inputstring);
  sub(/ /,"\\n",inputstring);
  return(inputstring);
}

function reportwgs() {
  base = "wgdistr";
  htmlfile = base ".html";
  epsfile = base ".eps";
  jpgfile = base ".jpg";
  datafile = base ".dat";
  gplfile = base".txt";
  
  openhtml(htmlfile,"Distribution of " docname "s according to WGs");
  html_file_css_section("Statistics",htmlfile);
  printf("<img src=%c%s%c style=%cborder:0%c alt=%cdistribution%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  
  maxcount = 0;
  for (draft in isdraft) {
    wg = getofficialwg(isdraft[draft]);
    if (wg != "") {
      if (wg in wgcounts) {
        wgcounts[wg] = wgcounts[wg] + 1;
      } else {
        wgcounts[wg] = 1;
      }
      if (wgcounts[wg] > maxcount) maxcount = wgcounts[wg];
    } else {
      wg = getrelatedwg(isdraft[draft]);
      if (wg != "") {
        if (wg in wgcountsother) {
          wgcountsother[wg] = wgcountsother[wg] + 1;
        } else {
          wgcountsother[wg] = 1;
        }
      }
    }
  }

  for (i = maxcount; i >= 0; i--) {  
    nwgswithdrafts[i] = 0;
    for (wg in wgcounts) {
      if (wgcounts[wg] == i) {
	nwgswithdrafts[i] = nwgswithdrafts[i] + 1;
      }
    }
  }
  
  printf("\n") > datafile;
  html_file_css_section("Distribution of " docname " amounts", htmlfile);
  printf("<ul>\n") >> htmlfile;

  for (i = maxcount; i >= 0; i--) {
    if (nwgswithdrafts[i] > 0) {
      printf("<li>%d WGs have %d %ss.</li>\n", nwgswithdrafts[i], i, docname) >> htmlfile;
    }
    printf("%d	%d\n", i, nwgswithdrafts[i]) >> datafile;
  }

  printf("</ul>\n") >> htmlfile;
  html_file_css_section("List of WGs", htmlfile);
  printf("<ul>\n") >> htmlfile;
  for (i = maxcount; i >= 0; i--) {  
    for (wg in wgcounts) {
      if (wgcounts[wg] == i) {
         printf("<li>%s has %d official %ss (%.2f%% of all) and %d related drafts (together %.2f%% of all).</li>",
                wg,
                wgcounts[wg], docname, (100.0 * wgcounts[wg]) / nietfdrafts,
                wgcountsother[wg], (100.0 * (wgcounts[wg] + wgcountsother[wg])) / ndrafts) >> htmlfile;
         #printf("%s	%d\n", wg, wgcounts[wg]) >> datafile;
      }
    }
  }
  printf("</ul>\n") >> htmlfile;

  basicgnuplotsettings(gplfile,epsfile,datafile,"# of " docnameu "s","# of WGs","Number of " docname "s in a WG",1,"");

  plottojpg(gplfile,epsfile,jpgfile);
  printf("<p>Definition of a related %s is that it matches the pattern draft-something-wg, where something is not ietf, and wg is the name of a working group.</p>\n", docname) >> htmlfile;
  closehtml(htmlfile);
}

function reportareas() {
  base = "areadistr";
  htmlfile = base ".html";
  epsfile = base ".eps";
  jpgfile = base ".jpg";
  datafile = base ".dat";
  gplfile = base".txt";
  
  openhtml(htmlfile,docnameu " area distribution");
  html_file_css_section("Statistics",htmlfile);
  printf("<img src=%c%s%c style=%cborder:0%c alt=%cdistribution%c/>\n", 34, jpgfile, 34, 34, 34, 34, 34) >> htmlfile;
  
  maxcount = 0;
  for (draft in isdraft) {
    wg = getofficialwg(isdraft[draft]);
    if (wg != "") {
      area = wgarea[wg];
      if (area in areacounts) {
        areacounts[area] = areacounts[area] + 1;
      } else {
        areacounts[area] = 1;
      }
      if (areacounts[area] > maxcount) maxcount = areacounts[area];
    } else {
      wg = getrelatedwg(isdraft[draft]);
      if (wg != "") {
        area = wgarea[wg];
        if (area in areacountsother) {
          areacountsother[area] = areacountsother[area] + 1;
        } else {
          areacountsother[area] = 1;
        }
      }
    }
  }

  printf("\n") > datafile;

  html_file_css_section("List of areas", htmlfile);
  printf("<ul>\n") >> htmlfile;
  nth = 1;
  extra = "set xtics rotate (";
  areacountsother[""] = 0;
  for (i = maxcount; i >= 0; i--) {  
    for (area in areacounts) {
      if (areacounts[area] == i) {
         if (area == "") areaname = "Unknown"; else areaname = area;
         printf("<li>%s has %d %ss (%.2f%% of all) and %d related drafts (together %.2f%% of all).</li>",
                areaname,