#!/usr/bin/perl

# ****
#
# getpaper.pl v0.1
#
# (c) Anthony D. Joseph, UC Berkeley ... 2003
# (c) MONET Research Group ... 2001
#
# Using function "GetField" from bibsearch.pl (c) 1994 Andy Wood.
#
# This script will respond to a call with the following fields:
#
# 'action' - either listing or searching paper
# 1) Listing papers:
#      'area' - area of research papers, e.g. resource management, routing, etc.
#      'type' - type of publications, e.g. conference, journal, etc.
#      'interleave' - interleave types (default is no)
#      'order' - print order: normal or reverse (default is normal)
#      'highlight' - case-insensitive name to highlight (default is none)
#      'shortAuthor' - shorten and reverse author names (default is no)
#      'header' - add file header and footer (default is yes)
#      'bold' - use bold instead of italic for format (default is yes)
#      'startDate' - start date (YYYYMM)
#      'endDate' - end date (YYYYMM)
#      'recent' - show n most recent papers (sets interleave yes)
#
# 2) Searching papers:
#      'term' - search term, e.g. qos, multimedia, etc.
#      'searchfield' - search field, e.g. author, title, keywords, etc.
#      'interleave' - interleave types (always on, can't override)
#      'order' - print order: normal or reverse (default is normal)
#      'highlight' - case-insensitive name to highlight (default is none)
#      'shortAuthor' - shorten and reverse author names (default is no)
#      'header' - add file header and footer (default is yes)
#      'number' - add paper number (default is no)
#      'startDate' - start date (YYYYMM)
#      'endDate' - end date (YYYYMM)
#      'recent' - show n most recent papers (sets interleave yes)
#
# When listing papers, it reads the BibTex file corresponding to a certain 
# research area, and writes out to html nicely. When searching papers, it 
# reads the BibTex files one by one and perform search on the requested field. 
# When a matched paper is found, it writes out to html. The key of each bibtex 
# entry is the file name of that paper. Use a key of "null.ps" for papers with
# no paper file.
#
# MONET Research Group
# Computer Science Department
# Univ of Illinois at Urbana-Champaign
# Urbana, Illinois, USA.
# Homepage: http://www-monet.cs.uiuc.edu
# Comments: Kai Chen <kaichen@cs.uiuc.edu>
#
# ****

require "getopts.pl";

# ****
#
# Configurations
#
# ****

#@CONFIGFILES = ( "/export/sahara/d/prof/adj/public_html/publications/config" );
@CONFIGFILES = ( "./config" );

$BIBFILEDIR = "";
$PAPERFILEDIR = "";
$HEADERFILE = "";
$FOOTERFILE = "";
@bibfiles = ();
$bibcount = 0;
@fonts = ();
$fontcount = 0;
$pubcount = 0;

# month to number mapping
%months = ("January", "01",   "February", "02",
	   "March", "03",     "April", "04",
	   "May", "05",       "June", "06",
	   "July", "07",      "August", "08",
	   "September", "09", "October", "10",
	   "November", "11",  "December", "12");

# names of publication types
%pubnames = ("book", "Book / Book Chapter",
	     "journal", "Journal / Magazine",
	     "conference", "Conference Paper",
	     "report", "Technical Report",
	     "phd", "Ph.D. Thesis",
	     "master", "Master's Thesis");

# define every possible bib field and html output sequence.
@elements = ( "author", "editor", "title", "booktitle", "journal", "edition", 
	      "chapter", "volume", "number", "pages", "series", "type", 
	      "organization", "school", "institution", "publisher", "address", 
	      "month", "year", "note");


# ****
#
# Main part of the script
#
# It first parses the input parameters, then calls the subroutines
# according to the desired action (list or search paper).
#
# ****

# Get user arguments
&Getopts("e:");	# e <exp> for command line use with expression

# Read in the parameters
if ($ENV{'REQUEST_METHOD'} eq "GET") {
    $in = $ENV{'QUERY_STRING'};
} elsif ($ENV{'REQUEST_METHOD'} eq "POST") {
    read(STDIN,$in,$ENV{'CONTENT_LENGTH'});
} elsif ($opt_e) {
    $in = $opt_e;
}
 

# parse input pairs to retrieve 'area' and 'type' (default to "all")
$user = 0;
$action = "list";
$areaid = "all";
$type = "all";
$term = "";
$searchfield = "";
$interleave = "no";
$order = "normal";
$highlight = "";
$shortAuthor = "no";
$header = "yes";
$number = "no";
$startDate = "";
$endDate = "";
$recent = "";
$font = 0;
$bold = "yes";

$titleBold = "<B>";
$titleEndBold = "</B>";

@in = split(/&/,$in);
foreach $i (0 .. $#in) {
    $in[$i] =~ s/\+/ /g;   # convert plus's to spaces
    ($key, $val) = split(/=/, $in[$i], 2);
    if ( $key eq "action" )  { $action = $val; }
    elsif ( $key eq "area" ) { $areaid = $val; }
    elsif ( $key eq "type" ) { $type = $val; }
    elsif ( $key eq "term" ) { $term = $val; }
    elsif ( $key eq "searchfield" ) { $searchfield = $val; }
    elsif ( $key eq "user" ) { $user = $val; }
    elsif ( $key eq "interleave" ) { $interleave = $val; }
    elsif ( $key eq "order" ) { $order = $val; }
    elsif ( $key eq "header" ) { $header = $val; }
    elsif ( $key eq "bold" ) { $bold = $val; }
    elsif ( $key eq "number" ) { $number = $val; }
    elsif ( $key eq "startDate" ) { $startDate = $val; }
    elsif ( $key eq "endDate" ) { $endDate = $val; }
    elsif ( $key eq "font" ) { $font = $val; }
    elsif ( $key eq "recent" ) { 
	$recent = $val; 
	$interleave = "yes";
    } 
    elsif ( $key eq "highlight" ) {
	$highlight = $val; 
	$titleBold = "";
	$titleEndBold = "";
    }
    elsif ( $key eq "shortAuthor" ) { $shortAuthor = $val; }
}

if ( $bold ne "yes" ) {
    $titleBold = "<I>";
    $titleEndBold = "</I>";
}

if ( ($user > $#CONFIGFILES) || ($user < 0) ) {
    print "getpaper.pl: illegal user number: $user\n";
}

&LoadConfig( $CONFIGFILES[$user] );
$numfiles = $#bibfiles / 4;
#print "numfiles is $numfiles\n";
#print "BIBFILEDIR = $BIBFILEDIR\n";
#print "PAPERFILEDIR = $PAPERFILEDIR\n";
#print "HEADERFILE = $HEADERFILE\n";
#print "FOOTERFILE = $FOOTERFILE\n";
#foreach $i (0 .. $#bibfiles) {
#    print "bibfile $i = $bibfiles[$i]\n";
#}
#foreach $i (0 .. $#fonts) {
#    print "font $i = $fonts[$i]\n";
#}



# output header file
if ( $header eq "yes" ) { 
    &OutputFile( $HEADERFILE );
}

if ( ($font > $#fonts) || ($font < 0) ) {
    print "getpaper.pl: illegal font number: $font\n";
}

# take one of the actions
if ( $action eq "list" ) {
    &ListPaper( $areaid, $type );
} elsif ( $action eq "search" ) {
    &SearchPaper( $term, $searchfield );
} else {
    print("getpaper.pl: action $action not supported.\n");
}

# output footer file
if ( $header eq "yes" ) { 
    print "<font face=$fonts[$font]>Found $pubcount publications.</font><br>\n";
    &OutputFile( $FOOTERFILE );
}

exit;


sub LoadConfig
{
    local( $filename ) = @_;

    open ( FILE, $filename) ||
	print "getpaper.pl: Couldn't open config file - $filename\n";
    local( $key, $val );

    $/ = "\n";  # use new line as record seperator
    while ( $record = <FILE> ) {
	# print "Read |$record| \n";
	next if ( $record =~ /^\#.*/ );          # skip comment line
	next if ( $record =~ /^\s*$/ );          # skip blank lines

	if ( $record =~ /^\$(.*)\s*=\s*\"(.*)\".*/ ) {
	    # ($key, $val) = split(/=/, $record, 2);
	    $key = $1;
	    $val = $2;
	    # print "Key is $key, Value is $val \n";
	    # $val =~ s/^.*\"(.*)\".*$/$1/;      # retrieve the value
	     
	    if ( $key =~ /$BIBFILEDIR\s*/i )  { $BIBFILEDIR = $val; }
	    elsif ( $key =~ /$PAPERFILEDIR\s*/i ) { $PAPERFILEDIR = $val; }
	    elsif ( $key =~ /$HEADERFILE\s*/i ) { $HEADERFILE = $val; }
	    elsif ( $key =~ /$FOOTERFILE\s*/i ) { $FOOTERFILE = $val; }
	    else { print "getpaper.pl: Unknown config file line - $record\n"; }
	} elsif ( $record =~ /^\@bibfile\s*=\s*\"(.*)\",\s*\"(.*)\",\s*\"(.*)\",\s*\"(.*)\".*/ ) {
	    $bibfiles[$bibcount++] = $1;
	    $bibfiles[$bibcount++] = $2;
	    $bibfiles[$bibcount++] = $3;
	    $bibfiles[$bibcount++] = $4;
	} elsif ( $record =~ /^\@font\s*=\s*\"(.*)\".*/ ) {
	    $fonts[$fontcount++] = "\"$1\"";
	} else { print "getpaper.pl: Got config file line - $record\n"; }
    }

    close FILE;
}

# ****
#
# SearchPaper
#
# Search papers with a specific term (first argument) in a
# specific field (second argument). All the bibtex files are
# read and scanned. For each entry found, it prints out html text.
#
# ****

sub SearchPaper
{
    local( $term, $searchfield) = @_;

    if ( $header eq "yes" ) { 
	print "<h3><font face=$fonts[$font]>Search results with term \"$term\" in $searchfield field:</font></h3>\n";
    }

  # bib files to be processed
    foreach $i (0 .. $numfiles) {
	$files[$i] = $bibfiles[$i * 4 + 1];
	$url[$i] = $bibfiles[$i * 4 + 3];
	if ($url[$i] ne "") {
	    $areanames[$i] = "<a href = \"" . $url[$i] . "\">" . $bibfiles[$i * 4 + 2] . "</a>";
	} else {
	    $areanames[$i] = $bibfiles[$i * 4 + 2];
	}
    }

  # Set output count to 0 for 0 entry
    $typeCount[0] = 0;
    $typeList[0][0] = "";

  # search the papers in the bib files
    local( $foundSearch ) = 0;
    foreach $i ( 0 .. $#files ) {
    # print "Searching file $files[$i] ... \n";
    # print "<P><H2>$areanames[$i]</H2>\n";
	if ( $recent eq "" ) {
	    # Set output count to 0 for 0 entry
	    $typeCount[0] = 0;
	    $typeList[0][0] = "";
	}
	$foundSearch |= 
	    &SearchBibFile( $files[$i], $areanames[$i], $term, $searchfield );
    }
    if ( $foundSearch == 0 ) { 
	if ( $header eq "yes" ) { 
	    print "<font face=$fonts[$font] size=3><B>No publications found.</B></font><BR>"; 
	}
    } elsif ( $recent ne "" ) {
	if ($typeCount[0] > 0) {
	    # Output papers for each type
	    if ($number eq "yes" ) {
		print "<ol>\n";
	    } else {
		print "<ul>\n";
	    }
	    &OutputPapers(0);
	    if ($number eq "yes" ) {
		print "</ol>\n";
	    } else {
		print "</ul>\n";
	    }
	}
    }
}

# ****
#
# SearchBibFile
#
# Search the bib file (first argument) with specific term
# (second argument) in the requested field (third argument). 
# For each bib entry, it prints out the correspondent html 
# text as output.
#
# ****

sub SearchBibFile
{
    local( $filename, $aname, $term, $searchfield ) = @_;
    $filename = $BIBFILEDIR.$filename;

    open ( FILE, $filename) || print "<P><H2>Couldn't open file - $filename</H2><HR>\n";
    local( $found ) = 0;

    $/ = "";  # use blank lines as record seperator
    while ( $record = <FILE> ) {
	if ( not ( $record =~ /@.*{/ )) { next; }  # skip non-bibtex record

	     local( $junk, $bibtex ) = split( '@', $record, 2 );
	     $bibtex = '@'.$bibtex;
	 
	     $fieldcontent = &GetField( $searchfield, $bibtex );
	     if ( $fieldcontent =~ /\b($term)\b/i ) {  # exact match and case insensitive
		 if ( $found == 0 ) {
		     if ( $recent eq "" ) {
			 print "<P><BR><font face=$fonts[$font] size=3><B>$aname</B></font><BR><BR>\n";
		     }
		     $found = 1; # now have found
		 }
		 local( $firstline, $rest ) = split( ',', $bibtex, 2 );
		 local( $bibtype, $key ) = split( '{', $firstline, 2 );
		 $bibtype =~  s/^@(.*)\s*$/$1/;   # retrieve the publication type
		 $key =~ s/^\s*(.*)\s*$/$1/;      # retrieve the key

		 &MarkupOnePaper($key, $bibtex, $bibtype, 0 );
	     }
	 } # end of while
    
    close FILE;

  # if ( $found == 0 ) { print "No publication found.\n"; }

    if ( $recent eq "" ) {	
	if ($typeCount[0] > 0) {
	    # Output papers for each type
	    if ($number eq "yes" ) {
		print "<ol>\n";
	    } else {
		print "<ul>\n";
	    }
	    &OutputPapers(0);
	    if ($number eq "yes" ) {
		print "</ol>\n";
	    } else {
		print "</ul>\n";
	    }
	}
    }
	
    return $found;
}

# ****
#
# ListPaper
#
# List papers in certain research area (first argument), and
# with certain publication type (second argument). For each
# entry found, it prints out html text.
#
# ****

sub ListPaper
{
    local( $areaid, $type ) = @_;

  # bib files to be processed
    if ( $areaid eq "all") {
	foreach $i (0 .. $numfiles) {
	    $files[$i] = $bibfiles[$i * 4 + 1];
	    $url[$i] = $bibfiles[$i * 4 + 3];
	    if ($url[$i] ne "") {
		$areanames[$i] = "<a href = \"" . $url[$i] . "\">" . $bibfiles[$i * 4 + 2] . "</a>";
	    } else {
		$areanames[$i] = $bibfiles[$i * 4 + 2];
	    }
	    # print "File $i $files[$i] $areanames[$i]\n";
	}
    }
    else {
	foreach $i (0 .. $numfiles) {
	    if ($areaid eq $bibfiles[$i * 4]) { 
		$files[0] = $bibfiles[$i * 4 + 1]; 
		$url[0] = $bibfiles[$i * 4 + 3];
		if ($url[0] ne "") {
		    $areanames[0] = "<a href = \"" . $url[0] . "\">" . $bibfiles[$i * 4 + 2] . "</a>";
		} else {
		    $areanames[0] = $bibfiles[$i * 4 + 2];
		}
	    }
	}
    }

  # type of publications to be processed
    if ( $type eq "all") {
	@types = ("book", "journal", "conference", "report", "phd", "master");
	if ( $header eq "yes" ) {
	    if ( $areaid eq "all" ) {
		print "<h3><font face=$fonts[$font]>List all publications in all research areas.</font></h3>\n";
	    } else {
		print "<h3><font face=$fonts[$font]>List all publications in the following research area.</font></h3>\n";
	    }
	}
    } 
    else {
	@types = ($type);
	if ( $areaid eq "all" ) {
	    if  ( $areaid eq "all" ) {
		print "<h3><font face=$fonts[$font]>List $pubnames{$type} in all research areas.</font></h3>\n";
	    } else {
		print "<h3><font face=$fonts[$font]>List $pubnames{$type} in the following research area.</font></h3>\n";
	    }
	}
    }

  # Set output count to 0 for each type
    local ( $ty );
    foreach $ty (0 .. $#types) {
	$typeCount[$ty] = 0;
	$typeList[$ty][0] = "";
    }

  # process the type of publication in each file
    foreach $i ( 0 .. $numfiles ) {

	local ( $foundarea ) = 0;
	foreach $j ( 0 .. $#types ) {
	    # print "File $files[$i], Type $types[$j], J $j, FoundArea $foundarea <br>\n";
	  $foundarea = &ProcessBibFile($files[$i], $types[$j], $j, $foundarea);
        }
	
	if ( $foundarea && ($interleave eq "no" )) {
	    print "<P><BR><font face=$fonts[$font] size=4><B>$areanames[$i]</B></font>\n";
	    local ($k);
	    foreach $k (0 .. $#types) {
		if ($typeCount[$k] > 0) {
		    print "<P><BR><font face=$fonts[$font] size=3><B><I>$pubnames{$types[$k]}</B></I></font><BR><BR>\n";
		    if ($number eq "yes" ) {
			print "<ol>\n";
		    } else {
			print "<ul>\n";
		    }
		    &OutputPapers($k);
		    if ($number eq "yes" ) {
			print "</ol>\n";
		    } else {
			print "</ul>\n";
		    }
		    $typeCount[$k] = 0;
		    $typeList[$k][0] = "";
		}
	    }
	}
    }

  # Output papers for each type
    if ( $interleave eq "yes" ) {
	if ( $recent ne "" ) {
	    local ( $i, $k );
	    foreach $i (1 .. $#types) {
		if ($typeCount[$i] > 0) {
		    foreach $k (0 .. ($typeCount[$i] - 1)) {
			$typeList[0][$typeCount[0]++] = $typeList[$i][$k];
		    }
		}
	    }
	    if ($number eq "yes" ) {
		print "<ol>\n";
	    } else {
		print "<ul>\n";
	    }
	    &OutputPapers(0);
	    if ($number eq "yes" ) {
		print "</ol>\n";
	    } else {
		print "</ul>\n";
	    }
	} else {
	    foreach $i (0 .. $#types) {
		if ($typeCount[$i] > 0) {
		    print "<P><BR><font face=$fonts[$font] size=3><B><I>$pubnames{$types[$i]}</B></I></font><BR><BR>\n";
		    if ($number eq "yes" ) {
			print "<ol>\n";
		    } else {
			print "<ul>\n";
		    }
		    &OutputPapers($i);
		    if ($number eq "yes" ) {
			print "</ol>\n";
		    } else {
			print "</ul>\n";
		    }
		}
	    }
	}
    }
}


# ****
#
# ProcessBibFile
#
# Process the bib file (first argument) with specific type 
# of publications (second argument) for type number (third argument)
# and the flag of whether this area has found at least one
# paper (third argument). For each bib entry, it prints out 
# the correspondent html text as output.
#
# ****

sub ProcessBibFile 
{
    local( $filename, $pubtype, $typeNum, $foundarea ) = @_;
    $filename = $BIBFILEDIR.$filename;

    open ( FILE, $filename) || print "<P><H2>Couldn't open file - $filename</H2><HR>\n";
    local( $found ) = 0;

    $/ = "";  # use blank lines as record seperator
    while ( $record = <FILE> ) {
	if ( not ( $record =~ /@.*{/ )) { next; } # skip non-bibtex record
#	print "Record: $record\n";

	local( $junk, $bibtex ) = split( '@', $record, 2 );
	$bibtex = '@'.$bibtex;
	local( $firstline, $rest ) = split( ',', $bibtex, 2 );
	local( $bibtype, $key ) = split( '{', $firstline, 2 );
	$bibtype =~  s/^@(.*)\s*$/$1/;   # retrieve the publication type
	$key =~ s/^\s*(.*)\s*$/$1/;      # retrieve the key

	$bibtype = lc($bibtype);
	# print "$filename bibtype: $bibtype (looking for $pubtype)<br>\n";
        # only process the correct publiction type
	next if ( $pubtype eq "book" ) && ( $bibtype ne "book" ) && ( $bibtype ne "inbook");
	next if ( $pubtype eq "journal" ) && ( $bibtype ne "article" );
	next if ( $pubtype eq "conference" ) && ( $bibtype ne "inproceedings" );
	next if ( $pubtype eq "report" ) && ( $bibtype ne "techreport" );
	next if ( $pubtype eq "master" ) && ( $bibtype ne "mastersthesis" );
	next if ( $pubtype eq "phd" ) && ( $bibtype ne "phdthesis" );

	if ( $found == 0 ) {
	    #if ( $interleave eq "no") {
	       # if ( $foundarea == 0 ) {  # has this area found at least one paper ?
	           #print "<P><BR><font face=$fonts[$font] size=4><B>$areanames[$i]</B></font>\n";
	       # }
	       #print "<P><BR><font face=$fonts[$font] size=3><B><I>$pubnames{$pubtype}</B></I></font><BR><BR>\n";
	    #}
	    $found = 1;
	}
	&MarkupOnePaper($key, $bibtex, $bibtype, $typeNum );

    } # end of while 

    close FILE;
    
    if ( $foundarea == 0 ) { 
	return $found; 
    } else { 
	return $foundarea; 
    }
}


# ****
#
# MarkupOnePaper
#
# Print out html text for a paper. The first arguemnt is the key
# of the bib entry (same as file name). The second arguemnt is the
# whole bib entry. The third augument is the bib entry type. 
# The fourth argument is the paper type number.
#
# ****

sub MarkupOnePaper
{
    local( $key, $bibtex, $bibtype, $typeNumber ) = @_;

    ( $junk, $file_format ) = split( '\.', $key, 2 );

    local ( $fieldcontent, $fieldname, $output, $month, $year );
    $month = "";
    $year = "";
    $output = "<LI><font face=$fonts[$font] size=2>";  # each paper is a list item
    foreach $i (0 .. $#elements) {
	$fieldname = $elements[$i];
	$fieldcontent = &GetField( $fieldname, $bibtex );
	if ( $fieldcontent eq "" ) { next; } # non-exist field

	$filedname = lc($fieldname);
    # print out the html stuff for this paper
	if ( $fieldname eq "author" ) {
	    $output .= &ParseAuthor($fieldcontent) . ", ";
	} elsif ( $fieldname eq "editor" ) {
	    if ( $fieldcontent =~ /,/ ) { $output .= "$fieldcontent (Eds), "; } # more than one editor
	    else { $output .= "$fieldcontent (Ed), "; }  # only one editor
	} elsif ( $fieldname eq "title" ) {
	    # use bold font for title
	    $output .= $titleBold . $fieldcontent . $titleEndBold . ", "; 
	} elsif ( $fieldname eq "booktitle" ) {
	    if ( $bibtype eq "conference") { $output .= "<I>In the Proceedings of $fieldcontent</I>, "; }
	    else {  $output .= "<I>$fieldcontent</I>, "; } # others like books, etc.
	} elsif ( $fieldname eq "journal" ) {
	    $output .= "<I>$fieldcontent</I>, ";
	} elsif ( $fieldname eq "edition" ) {
	    $output .= "$fieldcontent edition, ";
	} elsif ( $fieldname eq "chapter" ) {
	    $output .= "chapter $fieldcontent, ";
	} elsif ( $fieldname eq "volume" ) {
	    $output .= "vol. $fieldcontent, ";
	} elsif ( $fieldname eq "number" ) {
	    if ( $bibtype eq "techreport" ) { $output .= "$fieldcontent, "; }
	    else { $output .= "num. $fieldcontent, "; }
	} elsif ( $fieldname eq "pages" ) {
	    $output .= "pp. $fieldcontent, ";
	} elsif ( $fieldname eq "series" ) {
	    $output .= "$fieldcontent, ";
	} elsif ( $fieldname eq "type" ) {
	    $output .= "$fieldcontent, ";
	} elsif ( $fieldname eq "organization" ) {
	    $output .= "$fieldcontent, ";
	} elsif ( $fieldname eq "school" ) {
	    $output .= "$fieldcontent, ";
	} elsif ( $fieldname eq "institution" ) {
	    $output .= "$fieldcontent, ";
	} elsif ( $fieldname eq "publisher" ) {
	    $output .= "$fieldcontent, ";
	} elsif ( $fieldname eq "address" ) {
	    $output .= "$fieldcontent, ";
	} elsif ( $fieldname eq "month" ) {
	    $output .= "$fieldcontent, ";
	    $month = &ParseMonth($fieldcontent);
	} elsif ( $fieldname eq "year" ) {
	    $output .= "$fieldcontent. ";
	    $year = $fieldcontent;
	} elsif ( $fieldname eq "note" ) {
	    $output .= "(note: $fieldcontent) ";
	} 
    } # end of foreach
  
    # link up the paper file, if it exists
    # if ( $key eq "null.ps" ) {
    if ( $key =~ /^null.*/i ) {
	$output .= "</font><BR><BR>\n";
    } else {
	$output .= "<a href=\"$PAPERFILEDIR$key\">$file_format</a></font><BR><BR>\n";
    }
    
    if ( $year eq "" ) {
	$year = "1900";
    }
    if ( $month eq "" ) {
	$month = "13";
    }
    #print "Saving $output<br>\n";
    $typeList[$typeNumber][$typeCount[$typeNumber]++] = 
	$year . $month . "_" . $output;
    #print "Setting typeList and incrementing type $typeNumber to $typeCount[$typeNumber]<BR><BR>\n";
    $pubcount++;
}

sub ParseAuthor
{
    local( $field ) = @_;                   # Argument: author field

    $field =~ s/, and /, /g;                # Convert ', and' to ', '
    $field =~ s/ and /, /g;                 # Convert ' and' to ', '

    local ( @authors ) = split( ',' , $field );
    local ( $result ) = "";
    local ( $first ) = 1;
    local ( $sep ) = ", ";
    local ( $final ) = ", and ";
    local ( $i, $contents );
    local ( @name );
    foreach $i (0 .. $#authors) {
	if ($authors[$i] =~ m/^\s*(.*)\s*$/) {     # filter spaces around name
	    $authors[$i] = $1;
	}

	# print "Author is `$authors[$i]'   ";
	if ($shortAuthor eq "yes") {
	    $sep = "; ";
	    $final = "; ";
	    @name = split( ' ', $authors[$i]);
	    #print "Field count is $#name   ";
	    if ( $#name == 1 ) {
		$authors[$i] = $name[1] . ", " . &Abbrev($name[0]);
	    } else {
		local ( $j );
		$authors[$i] = $name[$#name] . ",";
		foreach $j (0 .. ($#name - 1)) {
		    #print "Field is $name[$j]  ";
		    $authors[$i] .= " " . &Abbrev($name[$j]);
		}
	    }
	}

	#print "Author is `$authors[$i]'<br>";
	if ($highlight ne "" ) {
	    $contents = $authors[$i];
	    if ( $contents =~ /.*($highlight).*/i ) {
		$authors[$i] = "<B>" . $authors[$i] . "</B>";
	    }
	}
	if ( $first == 1 ) {
	    $result = $authors[$i];
	    $first = 0;
	} else {
	    if ( $i == $#authors) {
		$result .= $final . $authors[$i];
	    } else {
		$result .= $sep . $authors[$i];
	    }
	}
    }
    return $result;
}


sub OutputPapers
{
    local( $idx ) = @_;                   # Argument: index
    local ( $j, $start, $end, $inc, $list, @slist, $first, $last );
    $#list = 0;
    foreach $j (0 .. ($typeCount[$idx] - 1)) {
	$list[$j] = $typeList[$idx][$j];
    }
    @slist = sort @list;
    $first = 0;
    $last = $typeCount[$idx];
    #print "Type $idx, Start $first, Last $last <br>";
    if (($startDate ne "") || ($endDate ne "")) {
	#print "Checking against date range `$startDate' and `$endDate'<br>";
	#print "Original range is $first to $last<br>";
	for ($j = $first; $j != $last; $j++) {
	    local ( $f ) = substr($slist[$j], 0, 6);
	    #print "Comparing $j<br>";
	    if ( $startDate ne "") {
		#print "Got start $startDate<br>";
		#print "substr $f<br>";
		if ( $startDate gt $f ) {
		    #print "Found new start at $j (date is $f)<br>";
		    $first = $j+1;
		}
	    }
	    if ( $endDate ne "") {
		if ( $endDate lt $f ) {
		    #print "Found new end at $j (date is $f)<br>";
		    $last = $j;
		    last;
		}
	    }
	}	
    }

    if ( $recent ne "" ) {
	local ( $s ) = $last - $recent;
	#print "Recent: $recent, S: $s, first: $first, last: $last <br>";
	if ( $s > $first ) { $first = $s; }
	#print "first now is $first <br>";
    }

    if ( $order eq "reverse" ) {
	$start = $first;
	$end = $last;
	$inc = 1;
    } else {
	$start = $last - 1;
	$end = $first -1;
	$inc = -1;
    }
    for ($j = $start; $j != $end; $j += $inc) {
	# print "list value is $slist[$j] <br>\n";
	local ( $rest, $value ) = split ( '_', $slist[$j], 2);
	print $value;
    }
}

sub Abbrev
{
    local( $name ) = @_;                   # Argument: name
    if ( index($name, '.') == -1 ) {
	return (substr($name, 0, 1) . ".");
    } else {
	return ($name);
    }
}


sub ParseMonth
{
    local( $month ) = @_;                   # Argument: month field
    local( $monthField, $rest ) = split( '([ ,/])', $month, 2 );
    local ( $mon ) = $months{$monthField};
    # if ($mon eq "" ) {
	# print "Month $monthField not found<br>\n";
    # } else {
	# print "Month $monthField is $mon<br>\n";
    # }
    return $mon;
}

# ****
#
# GetField
# 
# Gets the field specified in the first argument and strips it of quotes and/or
# squiggly brackets, removes excess spaces and returns it.
#
# From bibsearch.pl - Copyright 1994 Andy Wood
# http://www.cs.bham.ac.uk/~amw
#
# ****

sub GetField
{
    local( $field, $contents ) = @_;                   # Arguments: field name, bibtex entry

    $contents =~ s/\n/ /g;                             # Remove all \n's

  if ( $contents =~ /.*\b($field)\s*=\s*"([^"]*)"\s*/i ) # KaiChen: don't match "," at the end
    {
    $contents = $2;
    }
  elsif ( $contents =~ /.*\b($field)\s*=\s*{(.*)}\s*/i ) # KaiChen: don't match "," at the end
    {
    $contents = $2;                                  # Contains remaining fields too
    $contents =~ s/}\s*,.*//g;                       # So remove everything after },
    }
  elsif ( $contents =~ /.*\b($field)\s*=\s*(\d*)\s*/i ) # KaiChen: don't match "," at the end
    {
    $contents = $2;
    }
  else
    {
    $contents = "";
    }

  $contents =~ s/"|{|}//g;                           # Remove ""`s and {}'s
$contents =~ s/\s+/ /g;                            # Make lots of spaces into 1.
return $contents;
}

# ****
#
# OutputFile
# 
# Read and output the whole file (header and footer) which is 
# already formatted as html text.
#
# ****

sub OutputFile
{
    undef $/;
    open ( FILE, $_[0] ) || print "<P><H2>getpaper.pl: incorrect file $_[0]</H2><HR>\n";
    local ( $header ) = <FILE>;
    print $header;
    close FILE;
}