#!/usr/bin/perl

# calculates the percentage of papers hit within their first
# month of submission.

# Usage: firsthit < ../db/alldownloads
# Author: Ian Hickman

use strict;

use Time::Local;

my (@field, %firstsubmit);
my ($paperid, $paperfield, $paper);
my ($papers, $hits)=(0, 0);

my %months=( "Jan", 0, "Feb", 1, "Mar", 2, "Apr", 3, "May", 4, "Jun", 5, "Jul", 6, "Aug", 7, "Sep",
8, "Oct", 9, "Nov", 10, "Dec", 11 );

# open the list of papers and first submission dates
print"opening\n";
print" ".timelocal( 0, 0, 0, 24, 6, 1999)."\n";
print" ".timelocal( 0, 0, 0, 9, 4, 2000)."\n";
open( FILE, "../db/d_firstsubmit" ) || die"Cant open: $!";
while( <FILE> ){
  @field=split;
  if( $field[0] > timelocal( 0, 0, 0, 24, 6, 1999 ) &&
      $field[0] < timelocal( 0, 0, 0, 9, 4, 2000 ) ){
    $firstsubmit{$field[1]}=$field[0];
    $papers++;
  }
}
close( FILE ) || die"Cant close: $!";

print"$papers\n";

while( <STDIN> ){
  @field=split;

  # get the unique id of the paper
  $_=$field[6];
  ($paperid)=/(\d{7})/;
  if( $field[6]=~/papers/ ){
    ($paperfield)=/\/ftp\/([\w-]+)\.?.*\/papers/;
  } else {
    ($paperfield)=/\/.*\/([\w-]+)\.?.*\/\??\d{7}/;
  }
  $paper="$paperfield"."/"."$paperid";
  #print "$paper\n";
  
  if( exists( $firstsubmit{$paper} )){  
    # reformat time
    $_=$field[3];
    my ($day, $monthtext, $year, $hour, $min, $sec)=/.(\d\d).(\w+).(\d{4}).(\d\d).(\d\d).(\d\d)/;
    my $month=$months{$monthtext};
    my $hittime=timelocal( $sec, $min, $hour, $day, $month, $year );
    my $time=$firstsubmit{$paper}-$hittime;

    $time/=(3600*24);
    $time=roundoff($time);
    if( $time < 31 ){
      $hits++;
    }
    delete $firstsubmit{$paper};
  }
}

print"papers: $papers hits: $hits";

sub roundoff{
  my ($num)=@_;
  if( int($num) eq $num ){
    return $num;
  } elsif( ($num-int($num))<0.5 ){
    return int($num);
  } else {
    return (int($num)+1);
  }
}