package CONF;

#
# Configuration file for RePEc log summarizing script rstat.pl
#
# Optional sub LOG_PREPROC at end provides the possiblity of preprocessing
# log lines before rstat.pl takes a look at it.

# Service we are collecting stats for, any unique description will do
$SERVICE = 'EconomistsOnline';

# default to doing statistics for current month, -c option
#$CURRENT = 1;

# compress (gz) output file, -z option
#$COMPRESS = 1;

# FTP output file to specified location, -ftp option
# this will cause output to be gzipped.
#$FTP = 'username:password@server/dir';
#$FTP = 'logec:xxx@logec.repec.org/upload/logec';
# Must replace xxx with password

# IP-numbers to exclude from statistics
$STAT_IP_EXCLUDE = '';

# beginning of path to abstract html-files as shown in server log
# see sub Path_2_handle
#$SITE_PATH = '/ideas/data/Papers/'; # should suite IDEAS
#$SITE_PATH = '/(=?WoPEc|BibEc)/data/Papers/'; # should suite WoPEc and BibEc
$SITE_PATH = '';
#$SITE_PATH = '/[a-z]{6}/(?:abs|papers)/'; # for hhs

# path to redirection script
# see sub Redir_2_handle 
$SITE_REDIR_SCRIPT = '/adnetec-cgi-bin/get_doc.pl'; # should suite WoPEc
$SITE_REDIR_SCRIPT = 'DOWNLOAD_';
#$SITE_REDIR_SCRIPT = '/scripts/redir.pl'; # for hhs

# default log format
# W3C = W3C extended log file format as per http://www.w3.org/TR/WD-logfile.html
# NCSA = NCSA "common" log file format
# MS = Microsoft IIS log file format
$DEFLOG = 'NCSA';

# ignore stuff at start of line with NCSA log format
#$SHIFT = 1; # skips first item in log line useful when the server name appears first.

# IP is declared a robot IP if the number of hits from a single IP to URLs we 
# care about exceeds this.
# Expressed as a fraction of the URLs in the logs we care about. 
# IPs are also declared robot IPs if they hit /robots.txt
$ROBOT_FRACTION = 0.5;

$AUTHORITY = 'RePEc';

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

sub Redir_2_handle {}

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

sub Path_2_handle {
#
# Figure out the handle of a paper from url for abstract access
# (really URLs that match $SITE_PATH).
#
# Arguments are uri-stem and uri-query (normally empty, -) 
#
# Returns handle (lower cased) and 'a' if this is for the abstract
# or 'd' if this is for a download.
# Return two nul strings if we should not collect statistics for this
#

my( $stem, $query ) = @_;
  if ( $stem=~ m'ABS_VIEW_RePEc\:([a-z]{3})\:([a-z0-9]{6})\:(.+?)$'i ) {
     @return = split('ABS_VIEW_',$stem);
     warn("query: $return[1] ok.");
     return (lc($return[1]), 'a' );
  } elsif( $stem=~ m'DOWNLOAD_RePEc\:([a-z]{3})\:([a-z0-9]{6})\:(.+?)$'i ) {
     @return = split('DOWNLOAD_',$stem);
     warn("Download query: $return[1] ok.");
     return (lc($return[1]), 'd' );
  } else {
    warn("query: '$stem not ok.");
    return ( '', '' );
  }
}

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

#sub LOG_PREPROC {
#
# If defined this subroutine is called with the log line as 
# argument prior to any processing of the log line
#
#my( $line ) = @_;
#
#$line =~ s/foo/bar/;
#
#return $line;
#
#}

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

#sub IS_ROBOT {
#
# If defined this subroutine overrides the default
# check for robots which just checks for accesses to /robots.txt
# as below
#
#my( $stem, $query ) = @_;
#
#return $stem =~ "/robots.txt";
#
#}

1; # return true