\documentclass{article}
\usepackage{times}%
\usepackage{multicol}%
\usepackage{noweb}%
\input nowebmargins%
\pagestyle{noweb}%

%\addtolength{\headwidth}{\marginparsep}
%\addtolength{\headwidth}{\marginparwidth}
\begin{document}
\title{Citation Statistics for Ingram Olkin}
\author{B. Narasimhan\\
        Department of Statistics\\
        Stanford University\\
        Stanford, CA 94305}
\date{\today}
\maketitle 

\section{Introduction}
\label{sec:intro}

Ingram Olkin wanted the following statistics for the period
1985--1994.
\begin{enumerate}
\item The number of single author papers, 
\item The number of multiple author papers with names in alphabetic
  order, and 
\item The number of multiple author papers with names not in
  alphabetic order.
\end{enumerate}

We present a quick-and-dirty perl hack to do this.

\section{The Perl Program}
\label{sec:prog}

The program is structured as follows.

<<*>>=
#! /usr/local/bin/perl
<<Copyright>>
<<Local variables>>
<<Process files>>
<<Print results>>
@

\subsection{Copyright}
\label{sec:copyright}

<<Copyright>>=
#
# $Revision: 1.1 $   
#
# Copyright (C) 1997, B. Narasimhan (naras@stat.Stanford.EDU)
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
#
#
@

\subsection{Local Variables}
\label{sec:local-vars}

We need to know where the CIS files are located. Let us use
[[\$cis_directory]] to denote the directory and [[\$cis_ext]] denote
the extension to the file names.

<<Local variables>>=
$cis_directory = '/usr/local/lib/cis/cis95/files';
$cis_ext = '.v95';
@ %def $cis_directory $cis_ext
@

The years in question.
<<Local variables>>=
@years = (85..94);
@ %def @years
@

Some variables for counts which we initialize to zero.
\begin{description}
\item[\texttt{\$sj}] Number of single author journal articles.
\item[\texttt{\$sp}] Number of single author proceedings articles.
\item[\texttt{\$mj}] Number of multiple author journal articles.
\item[\texttt{\$mp}] Number of multiple author proceedings articles.
\item[\texttt{\$mja}] Number of multiple author journal articles with
  author names in alphabetic order.
\item[\texttt{\$mpa}] Number of multiple author proceedings articles with
  author names in alphabetic order.
\end{description}

<<Local variables>>=
$sj = 0; # No of single author journal articles.
$sp = 0; # No of single author proceedings articles.
$mj = 0; # No of multiple author journal articles.
$mja = 0; # No of multiple author alphabetic journal articles.
$mp = 0;  # No of multiple author proceedings articles.
$mpa = 0;  # No of multiple author alphabetic proceedings articles.
@ %def $sj $sp $mj $mp $mja $mpa
@

\subsection{Processing the Files}
\label{sec:process-files}

We basically see if each file exists and is a text file and open it.

<<Process files>>=
YEAR:
  foreach $year (@years) {
    $filename = $cis_directory . '/' . 'cis' . $year . $cis_ext;
    next YEAR unless -T $filename;
    if (!open(FH, $filename)) {
      print STDERR "Can't open $filename---continuing...\n";
      next YEAR;
    }
    while (<FH>) {
      <<Split fields>>
      <<Skip irrelevant stuff>>
      <<Determine no of authors>>
      <<Update counts>>
    }
  }
@

Splitting the fields is trivial once we know the format of the
records. Notice that we ignore the trailing `garbage' in the records.
<<Split fields>>=
($null,$field1,$title,$authors) = split('#');
@ %def $null $field1 $title $authors
@

Skipping irrelevant stuff---we skip books, electronic publications and
other administrative records.

<<Skip irrelevant stuff>>=
$cite_ind = substr($field1,-1,1);
# Skip books, electronic literature or administrative records
next if ($cite_ind =~ /[Bb]/) || ($cite_ind =~ /C/) 
  || ($cite_ind =~ /Z/);
@ %def $cite_ind
@

Oh! there are a bunch of entries for reviews of books, which we should
skip too.

<<Skip irrelevant stuff>>=
# Skip reviews of books.            
next if ($authors =~ /\(Rev\)/);
@

Multiple authors are separated by semicolons in the author field.

<<Determine no of authors>>=
@authors = split(';', $authors);
$no_authors = $#authors + 1;
@ %def $no_authors
@

So we are now down to the last part: updating counts. To detect if the
authors are listed alphabetically, we sort the [[@author]] array and
pack it back into a string with semicolons, just as the field would
look if the authors were listed in alphabetic order. So, if the newly
constructed string matches the field obtained from the file, the
authors are indeed in alphabetic order! We only need to take care of
to update the relevant counters. 

<<Update counts>>=
if ($no_authors > 1) {
  $sorted_authors = join(';', sort @authors);
}
if ($cite_ind =~ /[Jj]/) { # article in Journal
  if ($no_authors > 1) {
    $mj++;
    if ($sorted_authors eq $authors) {
      $mja++;
    }
  } else {
    $sj++;
  }
} else { # article in proceedings or edited book
  if ($no_authors > 1) {
    $mp++;
    if ($sorted_authors eq $authors) {
      $mpa++;
    }
  } else {
    $sp++;
  }
}
@


\section{Printing Results}
\label{sec:print-results}

This is straight-forward.

<<Print results>>=
print "Statistics from CIS for years @years \n";
printf "Single author papers in Journals: %d\n", $sj;
printf "Multiple author papers in Journals (alph): %d\n", $mja;
printf "Multiple author papers in Journals (Non-alph): %d\n", $mj - $mja;
printf "Single author papers in Proceedings: %d\n", $sp;
printf "Multiple author papers in Proceedings (alph): %d\n", $mpa;
printf "Multiple author papers in Proceedings (Non-alph): %d\n", $mp - $mpa;
@

\section{Index of Code Chunks}
\label{sec:code-chunks}

This list is generated automatically. The numeral is that of the first
definition of the chunk.
\nowebchunks

\section{Index of Identifiers}
\label{sec:identifiers}

Here is a list of the identifiers used, and where they appear.
Underlined entries indicate the place of definition.
This index is generated automatically.
\nowebindex

\end{document}

% Local Variables: %
% TeX-master: t %
% mode: LaTeX %
% cweb-prog-language: "PERL" %
% mode: cweb %
% End: %