The R Project SVN – R – /branches/R-exp-gnome/doc/html/faq.html

<html lang="en"><head>
<title>R FAQ</title>
<meta http-equiv="Content-Type" content="text/html">
<meta name=description content="R FAQ">
<meta name=generator content="makeinfo 4.0">
<link href="http://texinfo.org/" rel=generator-home>
</head><body>

<html>
<head>
<TITLE>R FAQ</TITLE>
</head>

<body>
<h1>R FAQ</h1>
<h2>Frequently Asked Questions on R</h2>
<h2>Version 0.90-2, 1999/11/25</h2>
<address>Kurt Hornik</address>
<p><p><hr><p>


<h1>Table of Contents</h1>
<ul>
<li><a href="#Introduction">1 Introduction</a>
<ul>
<li><a href="#Legalese">1.1 Legalese</a>
<li><a href="#Obtaining%20this%20document">1.2 Obtaining this document</a>
<li><a href="#Citing%20this%20document">1.3 Citing this document</a>
<li><a href="#Notation">1.4 Notation</a>
<li><a href="#Feedback">1.5 Feedback</a>
</ul>
<li><a href="#R%20Basics">2 R Basics</a>
<ul>
<li><a href="#What%20is%20R%3f">2.1 What is R?</a>
<li><a href="#What%20machines%20does%20R%20run%20on%3f">2.2 What machines does R run on?</a>
<li><a href="#What%20is%20the%20current%20version%20of%20R%3f">2.3 What is the current version of R?</a>
<li><a href="#How%20can%20R%20be%20obtained%3f">2.4 How can R be obtained?</a>
<li><a href="#How%20can%20R%20be%20installed%3f">2.5 How can R be installed?</a>
<ul>
<li><a href="#How%20can%20R%20be%20installed%20(Unix)">2.5.1 How can R be installed (Unix)</a>
<li><a href="#How%20can%20R%20be%20installed%20(Windows)">2.5.2 How can R be installed (Windows)</a>
<li><a href="#How%20can%20R%20be%20installed%20(Macintosh)">2.5.3 How can R be installed (Macintosh)</a>
</ul>
<li><a href="#Are%20there%20Unix%20binaries%20for%20R%3f">2.6 Are there Unix binaries for R?</a>
<li><a href="#What%20documentation%20exists%20for%20R%3f">2.7 What documentation exists for R?</a>
<li><a href="#Citing%20R">2.8 Citing R</a>
<li><a href="#What%20mailing%20lists%20exist%20for%20R%3f">2.9 What mailing lists exist for R?</a>
<li><a href="#What%20is%20CRAN%3f">2.10 What is CRAN?</a>
</ul>
<li><a href="#R%20and%20S">3 R and S</a>
<ul>
<li><a href="#What%20is%20S%3f">3.1 What is S?</a>
<li><a href="#What%20is%20S-PLUS%3f">3.2 What is S-PLUS?</a>
<li><a href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">3.3 What are the differences between R and S?</a>
<ul>
<li><a href="#Lexical%20scoping">3.3.1 Lexical scoping</a>
<li><a href="#Models">3.3.2 Models</a>
<li><a href="#Others">3.3.3 Others</a>
</ul>
<li><a href="#Is%20there%20anything%20R%20can%20do%20that%20S-PLUS%20cannot%3f">3.4 Is there anything R can do that S-PLUS cannot?</a>
</ul>
<li><a href="#R%20Web%20Interfaces">4 R Web Interfaces</a>
<li><a href="#R%20Add-On%20Packages">5 R Add-On Packages</a>
<ul>
<li><a href="#Which%20add-on%20packages%20exist%20for%20R%3f">5.1 Which add-on packages exist for R?</a>
<li><a href="#How%20can%20add-on%20packages%20be%20installed%3f">5.2 How can add-on packages be installed?</a>
<li><a href="#How%20can%20add-on%20packages%20be%20used%3f">5.3 How can add-on packages be used?</a>
<li><a href="#How%20can%20add-on%20packages%20be%20removed%3f">5.4 How can add-on packages be removed?</a>
<li><a href="#How%20can%20I%20create%20an%20R%20package%3f">5.5 How can I create an R package?</a>
<li><a href="#How%20can%20I%20contribute%20to%20R%3f">5.6 How can I contribute to R?</a>
</ul>
<li><a href="#R%20and%20Emacs">6 R and Emacs</a>
<ul>
<li><a href="#Is%20there%20Emacs%20support%20for%20R%3f">6.1 Is there Emacs support for R?</a>
<li><a href="#Should%20I%20run%20R%20from%20within%20Emacs%3f">6.2 Should I run R from within Emacs?</a>
</ul>
<li><a href="#R%20Miscellania">7 R Miscellania</a>
<ul>
<li><a href="#Why%20does%20R%20run%20out%20of%20memory%3f">7.1 Why does R run out of memory?</a>
<li><a href="#Why%20does%20sourcing%20a%20correct%20file%20fail%3f">7.2 Why does sourcing a correct file fail?</a>
<li><a href="#How%20can%20I%20set%20components%20of%20a%20list%20to%20NULL%3f">7.3 How can I set components of a list to NULL?</a>
<li><a href="#How%20can%20I%20save%20my%20workspace%3f">7.4 How can I save my workspace?</a>
<li><a href="#How%20can%20I%20clean%20up%20my%20workspace%3f">7.5 How can I clean up my workspace?</a>
<li><a href="#How%20can%20I%20get%20eval()%20and%20D()%20to%20work%3f">7.6 How can I get eval() and D() to work?</a>
<li><a href="#Why%20do%20my%20matrices%20lose%20dimensions%3f">7.7 Why do my matrices lose dimensions?</a>
<li><a href="#How%20does%20autoloading%20work%3f">7.8 How does autoloading work?</a>
<li><a href="#How%20should%20I%20set%20options%3f">7.9 How should I set options?</a>
<li><a href="#How%20do%20file%20names%20work%20in%20Windows%3f">7.10 How do file names work in Windows?</a>
<li><a href="#Why%20does%20plotting%20give%20a%20color%20allocation%20error%3f">7.11 Why does plotting give a color allocation error?</a>
<li><a href="#Is%20R%20Y2K-compliant%3f">7.12 Is R Y2K-compliant?</a>
</ul>
<li><a href="#R%20Programming">8 R Programming</a>
<ul>
<li><a href="#How%20should%20I%20write%20summary%20methods%3f">8.1 How should I write summary methods?</a>
<li><a href="#How%20can%20I%20debug%20dynamically%20loaded%20code%3f">8.2 How can I debug dynamically loaded code?</a>
<li><a href="#How%20can%20I%20inspect%20R%20objects%20when%20debugging%3f">8.3 How can I inspect R objects when debugging?</a>
</ul>
<li><a href="#R%20Bugs">9 R Bugs</a>
<ul>
<li><a href="#What%20is%20a%20bug%3f">9.1 What is a bug?</a>
<li><a href="#How%20to%20report%20a%20bug">9.2 How to report a bug</a>
</ul>
<li><a href="#Acknowledgments">10 Acknowledgments</a>
</ul>

<p><hr>
Node:<a name="Top">Top</a>,
Next:<a rel=next href="#Introduction">Introduction</a>,
Previous:<a rel=previous href="#(dir)">(dir)</a>,
Up:<a rel=up href="#(dir)">(dir)</a>
<br>

<ul>
<li><a href="#Introduction">Introduction</a>: 
<li><a href="#R%20Basics">R Basics</a>: 
<li><a href="#R%20and%20S">R and S</a>: 
<li><a href="#R%20Web%20Interfaces">R Web Interfaces</a>: 
<li><a href="#R%20Add-On%20Packages">R Add-On Packages</a>: 
<li><a href="#R%20and%20Emacs">R and Emacs</a>: 
<li><a href="#R%20Miscellania">R Miscellania</a>: 
<li><a href="#R%20Programming">R Programming</a>: 
<li><a href="#R%20Bugs">R Bugs</a>: 
<li><a href="#Acknowledgments">Acknowledgments</a>: 
</ul>

<p><hr>
Node:<a name="Introduction">Introduction</a>,
Next:<a rel=next href="#R%20Basics">R Basics</a>,
Previous:<a rel=previous href="#Top">Top</a>,
Up:<a rel=up href="#Top">Top</a>
<br>

<h1>1 Introduction</h1>

<p>This document contains answers to some of the most frequently asked
questions about R.

<ul>
<li><a href="#Legalese">Legalese</a>: 
<li><a href="#Obtaining%20this%20document">Obtaining this document</a>: 
<li><a href="#Citing%20this%20document">Citing this document</a>: 
<li><a href="#Notation">Notation</a>: 
<li><a href="#Feedback">Feedback</a>: 
</ul>

<p><hr>
Node:<a name="Legalese">Legalese</a>,
Next:<a rel=next href="#Obtaining%20this%20document">Obtaining this document</a>,
Previous:<a rel=previous href="#Introduction">Introduction</a>,
Up:<a rel=up href="#Introduction">Introduction</a>
<br>

<h2>1.1 Legalese</h2>

<p>This document 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, or (at your option) any
later version.

<p>This document 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.

<p>A copy of the GNU General Public License is available via WWW at

<pre><a href="http://www.gnu.org/copyleft/gpl.html">http://www.gnu.org/copyleft/gpl.html</a>.
</pre>

<p>You can also obtain it by writing to the Free Software Foundation, Inc.,
59 Temple Place -- Suite 330, Boston, MA 02111-1307, USA.

<p><hr>
Node:<a name="Obtaining%20this%20document">Obtaining this document</a>,
Next:<a rel=next href="#Citing%20this%20document">Citing this document</a>,
Previous:<a rel=previous href="#Legalese">Legalese</a>,
Up:<a rel=up href="#Introduction">Introduction</a>
<br>

<h2>1.2 Obtaining this document</h2>

<p>The latest version of this document is always available from

<pre><a href="http://www.ci.tuwien.ac.at/~hornik/R/">http://www.ci.tuwien.ac.at/~hornik/R/</a>
</pre>

<p>From there, you can obtain versions converted to
<a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.txt">plain ASCII text</a>, <a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.dvi.gz">DVI</a>,
<a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.info.gz">GNU info</a>,
<a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html">HTML</a>,
<a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.pdf">PDF</a>,
<a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.ps.gz">PostScript</a> as
well as the <a href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.texi">Texinfo source</a> used for creating all these formats using the
<a href="http://texinfo.org/">GNU Texinfo system</a>.

<p>You can also obtain the R FAQ from the <code>doc/FAQ</code> subdirectory of a
CRAN site (<a href="#What%20is%20CRAN%3f">What is CRAN?</a>).

<p><hr>
Node:<a name="Citing%20this%20document">Citing this document</a>,
Next:<a rel=next href="#Notation">Notation</a>,
Previous:<a rel=previous href="#Obtaining%20this%20document">Obtaining this document</a>,
Up:<a rel=up href="#Introduction">Introduction</a>
<br>

<h2>1.3 Citing this document</h2>

<p>In publications, please refer to this FAQ as Hornik (1999), "The R
FAQ" and give the above, <em>official</em> URL.

<p><hr>
Node:<a name="Notation">Notation</a>,
Next:<a rel=next href="#Feedback">Feedback</a>,
Previous:<a rel=previous href="#Citing%20this%20document">Citing this document</a>,
Up:<a rel=up href="#Introduction">Introduction</a>
<br>

<h2>1.4 Notation</h2>

<p>Everything should be pretty standard.  <code>R&gt;</code> is used for the R
prompt, and a <code>$</code> for the shell prompt (where applicable).

<p><hr>
Node:<a name="Feedback">Feedback</a>,
Previous:<a rel=previous href="#Notation">Notation</a>,
Up:<a rel=up href="#Introduction">Introduction</a>
<br>

<h2>1.5 Feedback</h2>

<p>Feedback is of course most welcome.

<p>In particular, note that I do not have access to Windows or Mac systems. 
Features specific to the Windows port of R are described in the
<a href="http://www.stats.ox.ac.uk/pub/R/rw-FAQ.html">"Frequently Asked Questions for R for Windows"</a>.  If you have information on Windows or
Mac systems that you think should be added to this document, please let
me know.

<p><hr>
Node:<a name="R%20Basics">R Basics</a>,
Next:<a rel=next href="#R%20and%20S">R and S</a>,
Previous:<a rel=previous href="#Introduction">Introduction</a>,
Up:<a rel=up href="#Top">Top</a>
<br>

<h1>2 R Basics</h1>

<ul>
<li><a href="#What%20is%20R%3f">What is R?</a>: 
<li><a href="#What%20machines%20does%20R%20run%20on%3f">What machines does R run on?</a>: 
<li><a href="#What%20is%20the%20current%20version%20of%20R%3f">What is the current version of R?</a>: 
<li><a href="#How%20can%20R%20be%20obtained%3f">How can R be obtained?</a>: 
<li><a href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>: 
<li><a href="#Are%20there%20Unix%20binaries%20for%20R%3f">Are there Unix binaries for R?</a>: 
<li><a href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>: 
<li><a href="#Citing%20R">Citing R</a>: 
<li><a href="#What%20mailing%20lists%20exist%20for%20R%3f">What mailing lists exist for R?</a>: 
<li><a href="#What%20is%20CRAN%3f">What is CRAN?</a>: 
</ul>

<p><hr>
Node:<a name="What%20is%20R%3f">What is R?</a>,
Next:<a rel=next href="#What%20machines%20does%20R%20run%20on%3f">What machines does R run on?</a>,
Previous:<a rel=previous href="#R%20Basics">R Basics</a>,
Up:<a rel=up href="#R%20Basics">R Basics</a>
<br>

<h2>2.1 What is R?</h2>

<p>R is a system for statistical computation and graphics.  It consists of
a language plus a run-time environment with graphics, a debugger, access
to certain system functions, and the ability to run programs stored in
script files.

<p>The design of R has been heavily influenced by two existing languages:
Becker, Chambers &amp; Wilks' S (see <a href="#What%20is%20S%3f">What is S?</a>) and Sussman's
<a href="http://www.cs.indiana.edu/scheme-repository/home.html">Scheme</a>. 
Whereas the resulting language is very similar in appearance to S, the
underlying implementation and semantics are derived from Scheme. 
See <a href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>, for a discussion of
the differences between R and S.

<p>R was initially written by <a href="mailto:Ross.Ihaka@r-project.org">Ross Ihaka</a>
and <a href="mailto:Robert.Gentleman@r-project.org">Robert Gentleman</a>, who are
Senior Lecturers at the Department of Statistics of the University of
Auckland in Auckland, New Zealand.  In addition, a large group of
individuals has contributed to R by sending code and bug reports.

<p>Since mid-1997 there has been a core group (the "R Core Team") who can
modify the R source code CVS archive.  The group currently consists of
Doug Bates, Peter Dalgaard, Robert Gentleman, Kurt Hornik, Ross Ihaka,
Friedrich Leisch, Thomas Lumley, Martin Maechler, Guido Masarotto, Paul
Murrell, Brian Ripley, Duncan Temple Lang, and Luke Tierney.

<p>R has a home page at <a href="http://stat.auckland.ac.nz/r/r.html">http://stat.auckland.ac.nz/r/r.html</a>.  It is
free software distributed under a GNU-style copyleft, and an official
part of the GNU project ("GNU S").

<p><hr>
Node:<a name="What%20machines%20does%20R%20run%20on%3f">What machines does R run on?</a>,
Next:<a rel=next href="#What%20is%20the%20current%20version%20of%20R%3f">What is the current version of R?</a>,
Previous:<a rel=previous href="#What%20is%20R%3f">What is R?</a>,
Up:<a rel=up href="#R%20Basics">R Basics</a>
<br>

<h2>2.2 What machines does R run on?</h2>

<p>R is being developed for the Unix, Windows and Mac families of operating
systems.

<p>The current version of R will configure and build under a number of
common Unix platforms including i386-freebsd, i386-linux, ppc-linux,
mips-sgi-irix, alpha-linux, alpha-dec-osf4, rs6000-ibm-aix,
hppa-hp-hpux, sparc-linux, and sparc-sun-solaris, see the file
<code>PLATFORMS</code> in the R distribution for more information.

<p>If you know about other platforms, please drop us a note.

<p><hr>
Node:<a name="What%20is%20the%20current%20version%20of%20R%3f">What is the current version of R?</a>,
Next:<a rel=next href="#How%20can%20R%20be%20obtained%3f">How can R be obtained?</a>,
Previous:<a rel=previous href="#What%20machines%20does%20R%20run%20on%3f">What machines does R run on?</a>,
Up:<a rel=up href="#R%20Basics">R Basics</a>
<br>

<h2>2.3 What is the current version of R?</h2>

<p>The current stable Unix/Windows version is 0.90.0, the
unstable one is 0.99.0.  Typically, new features are
introduced in the development versions; updates of stable versions are
for bug fixes mostly.  The version for the Mac is pre-alpha.

<p><hr>
Node:<a name="How%20can%20R%20be%20obtained%3f">How can R be obtained?</a>,
Next:<a rel=next href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>,
Previous:<a rel=previous href="#What%20is%20the%20current%20version%20of%20R%3f">What is the current version of R?</a>,
Up:<a rel=up href="#R%20Basics">R Basics</a>
<br>

<h2>2.4 How can R be obtained?</h2>

<p>Sources, binaries and documentation for R can be obtained via CRAN, the
"Comprehensive R Archive Network" (see <a href="#What%20is%20CRAN%3f">What is CRAN?</a>).

<p><hr>
Node:<a name="How%20can%20R%20be%20installed%3f">How can R be installed?</a>,
Next:<a rel=next href="#Are%20there%20Unix%20binaries%20for%20R%3f">Are there Unix binaries for R?</a>,
Previous:<a rel=previous href="#How%20can%20R%20be%20obtained%3f">How can R be obtained?</a>,
Up:<a rel=up href="#R%20Basics">R Basics</a>
<br>

<h2>2.5 How can R be installed?</h2>

<ul>
<li><a href="#How%20can%20R%20be%20installed%20(Unix)">How can R be installed (Unix)</a>: 
<li><a href="#How%20can%20R%20be%20installed%20(Windows)">How can R be installed (Windows)</a>: 
<li><a href="#How%20can%20R%20be%20installed%20(Macintosh)">How can R be installed (Macintosh)</a>: 
</ul>

<p><hr>
Node:<a name="How%20can%20R%20be%20installed%20(Unix)">How can R be installed (Unix)</a>,
Next:<a rel=next href="#How%20can%20R%20be%20installed%20(Windows)">How can R be installed (Windows)</a>,
Previous:<a rel=previous href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>,
Up:<a rel=up href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>
<br>

<h3>2.5.1 How can R be installed (Unix)</h3>

<p>If binaries are available for your platform (see <a href="#Are%20there%20Unix%20binaries%20for%20R%3f">Are there Unix binaries for R?</a>), you can use these, following the instructions that
come with them.

<p>Otherwise, you can compile and install R yourself, which can be done
very easily under a number of common Unix platforms (see <a href="#What%20machines%20does%20R%20run%20on%3f">What machines does R run on?</a>).  The file <code>INSTALL</code> that comes with the
R distribution contains instructions.

<p>Note that you need a FORTRAN compiler or <code>f2c</code> in addition to a C
compiler to build R.  Also, you need Perl version 5 to build the
documentation.  If this is not available on your system, you can obtain
precompiled documentation files via CRAN.

<p>In the simplest case, untar the R source code, cd to the directory thus
created, and issue the following commands (at the shell prompt):

<pre>$ ./configure
$ make
</pre>

<p>If these commands execute successfully, the R binary and a shell script
font-end called <code>R</code> are created and copied to the <code>bin</code>
directory.  You can copy the script to a place where users can invoke
it, for example to <code>/usr/local/bin</code>.  In addition, plain text help
pages as well as HTML and LaTeX versions of the documentation are
built.

<p>Use <kbd>make dvi</kbd> to obtain a dvi version of the R manual.  This
creates the files <code>Manual.dvi</code> (a start of a manual) and
<code>Reference.dvi</code> (an R object reference index) in the
<code>doc/manual</code> subdirectory.  These files can be previewed and
printed using standard programs such as <code>xdvi</code> and <code>dvips</code>. 
(Note that they have to be built in the source tree.)

<p>Finally, use <kbd>make check</kbd> to find out whether your R system works
correctly.

<p>You can also perform a "system-wide" installation using <kbd>make
install</kbd>.  By default, this will install to the following directories:

<dl>
<dt><code>${prefix}/bin</code>
<dd>(some) executables
<br><dt><code>${prefix}/man/man1</code>
<dd>man pages
<br><dt><code>${prefix}/share/R</code>
<dd>all the rest (libraries, on-line help system, <small>...</small>).  This is the "R
Home Directory" (<code>R_HOME</code>) of the installed system. 
</dl>

<p>In the above, <code>prefix</code> is determined during configuration
(typically <code>/usr/local</code>) and can be set by running <code>configure</code>
with the option

<pre>$ ./configure --prefix=/where/you/want/R/to/go
</pre>

<p>(E.g., the R executable will then be installed into
<code>/where/you/want/R/to/go/bin</code>.)

<p><hr>
Node:<a name="How%20can%20R%20be%20installed%20(Windows)">How can R be installed (Windows)</a>,
Next:<a rel=next href="#How%20can%20R%20be%20installed%20(Macintosh)">How can R be installed (Macintosh)</a>,
Previous:<a rel=previous href="#How%20can%20R%20be%20installed%20(Unix)">How can R be installed (Unix)</a>,
Up:<a rel=up href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>
<br>

<h3>2.5.2 How can R be installed (Windows)</h3>

<p>The <code>bin/windows/windows-NT</code> directory of a CRAN site contains the
latest binary distributions of R for 32 bit versions of MS Windows
(i.e., 95, 98 or NT), as well as binary distributions for a large number
of add-on packages from CRAN.  The Windows version of R was created by
Robert Gentleman, and is now being developed and maintained by
<a href="mailto:Guido.Masarotto@r-project.org">Guido Masarotto</a> and
<a href="mailto:Brian.Ripley@r-project.org">Brian D. Ripley</a>.

<p>For most installations the installer <code>rwinst.exe</code> will be the
easiest tool to use.

<p>See the <a href="http://www.stats.ox.ac.uk/pub/R/rw-FAQ.html">R Windows FAQ</a> for more details.

<p><hr>
Node:<a name="How%20can%20R%20be%20installed%20(Macintosh)">How can R be installed (Macintosh)</a>,
Previous:<a rel=previous href="#How%20can%20R%20be%20installed%20(Windows)">How can R be installed (Windows)</a>,
Up:<a rel=up href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>
<br>

<h3>2.5.3 How can R be installed (Macintosh)</h3>

<p>The Power Macintosh port is temporarily on hold, and currently no binary
distribution is available.  We hope that this will change soon.

<p><hr>
Node:<a name="Are%20there%20Unix%20binaries%20for%20R%3f">Are there Unix binaries for R?</a>,
Next:<a rel=next href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>,
Previous:<a rel=previous href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>,
Up:<a rel=up href="#R%20Basics">R Basics</a>
<br>

<h2>2.6 Are there Unix binaries for R?</h2>

<p>The <code>bin/linux</code> directory contains Debian 2.1 packages for the i386
platform (now part of the Debian distribution and maintained by Doug
Bates) as well as Red Hat 5.1 packages for the alpha and sparc platforms
(maintained by Nassib Nasser and Vin Everett, respectively), Red Hat 6.0
packages for the i386 and alpha platforms (maintained by Martyn Plummer
and Naoki Takebayashi, respectively), S.u.S.E. 5.3/6.0/6.2 i386 packages
by Albrecht Gebhardt, and RPMs for the ppc platform by Alex Buerkle.

<p>The <code>bin/osf</code> directory contains RPMs for alpha systems running
Digital Unix 4.0 by Albrecht Gebhardt.

<p>No other binary distributions have thus far been made publically
available.

<p><hr>
Node:<a name="What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>,
Next:<a rel=next href="#Citing%20R">Citing R</a>,
Previous:<a rel=previous href="#Are%20there%20Unix%20binaries%20for%20R%3f">Are there Unix binaries for R?</a>,
Up:<a rel=up href="#R%20Basics">R Basics</a>
<br>

<h2>2.7 What documentation exists for R?</h2>

<p>Online documentation for most of the functions and variables in R
exists, and can be printed on-screen by typing <kbd>help(<var>name</var>)</kbd>
(or <kbd>?<var>name</var></kbd>) at the R prompt, where <var>name</var> is the name of
the topic help is sought for.  (In the case of unary and binary
operators and control-flow special forms, the name may need to be be
quoted.)

<p>This documentation can also be made available as HTML, and as hardcopy
via LaTeX, see <a href="#How%20can%20R%20be%20installed%3f">How can R be installed?</a>.  An up-to-date HTML
version is always available for web browsing at
<a href="http://stat.ethz.ch/R/manual/">http://stat.ethz.ch/R/manual/</a>.

<p>An earlier attempt at an R manual ("Notes on R:  A Programming
Environment for Data Analysis and Graphics"), based on the "Notes on
S-PLUS" by <a href="mailto:venables@stats.adelaide.edu.au">Bill Venables</a> and
<a href="mailto:D.M.Smith@lancaster.ac.uk">David Smith</a>, can be obtained as
<code>Rnotes.tgz</code> (LaTeX source) in a CRAN <code>doc</code> directory. 
Further documentation on R and the R API are currently being written.

<p>In the absence of an R manual, documentation for S/S-PLUS (see <a href="#R%20and%20S">R and S</a>) can be used in combination with this FAQ (see <a href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>).  We recommend

<blockquote>
W. N. Venables and B. D. Ripley (1999), "Modern Applied Statistics with
S-PLUS.  Third Edition".  Springer, ISBN 0-387-98825-4. 
</blockquote>

<p>which has a home page at <a href="http://www.stats.ox.ac.uk/pub/MASS3/">http://www.stats.ox.ac.uk/pub/MASS3/</a>
providing additional material, in particular "R Complements" which
describe how to use the book with R.  These complements provide both
descriptions of some of the differences between R and S-PLUS, and the
modifications needed to run the examples in the book.  Its companion
volume on "S Programming", due in about April 2000, will provide an
in-depth guide to writing software in the S language which forms the
basis of both the commercial S-PLUS and the Open Source R data analysis
software systems.

<p>More introductory books are

<blockquote>
P. Spector (1994), "An introduction to S and S-PLUS", Duxbury Press.

<p>A. Krause and M. Olsen (1997), "The Basics of S and S-PLUS", Springer. 
</blockquote>

<p>Last, but not least, Ross' and Robert's experience in designing and
implementing R is described in:

<pre>@article{,
  author =    {Ross Ihaka and Robert Gentleman},
  title =     {R: A Language for Data Analysis and Graphics},
  journal =   {Journal of Computational and Graphical Statistics},
  year =      1996,
  volume =    5,
  number =    3,
  pages =     {299--314}
}
</pre>

<p><hr>
Node:<a name="Citing%20R">Citing R</a>,
Next:<a rel=next href="#What%20mailing%20lists%20exist%20for%20R%3f">What mailing lists exist for R?</a>,
Previous:<a rel=previous href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>,
Up:<a rel=up href="#R%20Basics">R Basics</a>
<br>

<h2>2.8 Citing R</h2>

<p>To cite R in publications, use the above Ihaka &amp; Gentleman (1996), "R:
A Language for Data Analysis and Graphics", <em>Journal of
Computational and Graphical Statistics</em>, <strong>5</strong>, 299-314.

<p><hr>
Node:<a name="What%20mailing%20lists%20exist%20for%20R%3f">What mailing lists exist for R?</a>,
Next:<a rel=next href="#What%20is%20CRAN%3f">What is CRAN?</a>,
Previous:<a rel=previous href="#Citing%20R">Citing R</a>,
Up:<a rel=up href="#R%20Basics">R Basics</a>
<br>

<h2>2.9 What mailing lists exist for R?</h2>

<p>Thanks to <a href="mailto:Martin.Maechler@r-project.org">Martin Maechler</a>, there
are three mailing lists devoted to R.

<dl>
<dt><code>r-announce</code>
<dd>This list is for announcements about the development of R and the
availability of new code. 
<br><dt><code>r-devel</code>
<dd>This list is for discussions about the future of R and pre-testing of
new versions.  It is meant for those who maintain an active position in
the development of R. 
<br><dt><code>r-help</code>
<dd>The `main' R mailing list, for announcements about the development of R
and the availability of new code, questions and answers about problems
and solutions using R, enhancements and patches to the source code and
documentation of R, comparison and compatibility with S and S-PLUS, and
for the posting of nice examples and benchmarks. 
</dl>

<p>Note that the r-announce list is gatewayed into r-help, so you don't
need to subscribe to both of them.

<p>Send email to <a href="mailto:r-help@lists.r-project.org">r-help@lists.r-project.org</a> to reach everyone on
the r-help mailing list.  To subscribe (or unsubscribe) to this list
send <code>subscribe</code> (or <code>unsubscribe</code>) in the BODY of the message
(not in the subject!)  to <a href="mailto:r-help-request@lists.r-project.org">r-help-request@lists.r-project.org</a>. 
Information about the list can be obtained by sending an email with
<code>info</code> as its contents to
<a href="mailto:r-help-request@lists.r-project.org">r-help-request@lists.r-project.org</a>.

<p>Subscription and posting to the other lists is done analogously, with
`r-help' replaced by `r-announce' and `r-devel', respectively.

<p>It is recommended that you send mail to r-help rather than only to the R
developers (who are also subscribed to the list, of course).  This may
save them precious time they can use for constantly improving R, and
will typically also result in much quicker feedback for yourself.

<p>Of course, in the case of bug reports it would be very helpful to have
code which reliably reproduces the problem.  Also, make sure that you
include information on the system and version of R being used.  See
<a href="#R%20Bugs">R Bugs</a> for more details.

<p>Archives of the above three mailing lists are made available on the net
in a monthly schedule via the <code>doc/mail/mail.html</code> file in CRAN. 
An HTML archive of the lists are available via
<a href="http://www.ens.gu.edu.au/robertk/R/">http://www.ens.gu.edu.au/robertk/R/</a>.

<p>The R Core Team can be reached at <a href="mailto:r-core@lists.r-project.org">r-core@lists.r-project.org</a>
for comments and reports.

<p><hr>
Node:<a name="What%20is%20CRAN%3f">What is CRAN?</a>,
Previous:<a rel=previous href="#What%20mailing%20lists%20exist%20for%20R%3f">What mailing lists exist for R?</a>,
Up:<a rel=up href="#R%20Basics">R Basics</a>
<br>

<h2>2.10 What is CRAN?</h2>

<p>The "Comprehensive R Archive Network" (CRAN) is a collection of sites
which carry identical material, consisting of the R distribution(s), the
contributed extensions, documentation for R, and binaries.

<p>The CRAN master site can be found at the URL

<blockquote>
<p><table><tr align="left"><td><a href="http://cran.r-project.org/">http://cran.r-project.org/</a>
<td>(Austria)
<br></tr></table>
</blockquote>

<p>(which is the same as <a href="http://cran.at.r-project.org/">http://cran.at.r-project.org/</a>) and is
currently being mirrored daily at

<blockquote>
<p><table><tr align="left"><td><a href="http://cran.dk.r-project.org/">http://cran.dk.r-project.org/</a>
<td>(Denmark)
<br></tr><tr align="left"><td><a href="http://cran.it.r-project.org/">http://cran.it.r-project.org/</a>
<td>(Italy)
<br></tr><tr align="left"><td><a href="http://cran.ch.r-project.org/">http://cran.ch.r-project.org/</a>
<td>(Switzerland)
<br></tr><tr align="left"><td><a href="http://cran.uk.r-project.org/">http://cran.uk.r-project.org/</a>
<td>(United Kingdom)
<br></tr><tr align="left"><td><a href="http://cran.us.r-project.org/">http://cran.us.r-project.org/</a>
<td>(USA/Wisconsin)
<br></tr></table>
</blockquote>

<p>Please use the CRAN site closest to you to reduce network load.

<p>From CRAN, you can obtain the latest official release of R, daily
snapshots of R (copies of the current CVS trees), as gzipped and bzipped
tar files or as two gzipped tar files (ready for 1.4M floppies), a
wealth of additional contributed code, as well as prebuilt binaries for
various operating systems (Linux, Nextstep, MacOS, MSWin) and
pre-formatted help pages.  CRAN also provides access to documentation on
R, existing mailing lists and the R Bug Tracking system.

<p>To "submit" to CRAN, simply upload to
<a href="ftp://cran.r-project.org/incoming">ftp://cran.r-project.org/incoming</a> and send an email to
<a href="mailto:wwwadmin@cran.r-project.org">wwwadmin@cran.r-project.org</a>.

<blockquote>
<strong>Note:</strong>  It is very important that you indicate the copyright
(license) information (GPL, BSD, Artistic, <small>...</small>) in your submission. 
</blockquote>

<p>Please always use the URL of the master site when referring to CRAN.

<p><hr>
Node:<a name="R%20and%20S">R and S</a>,
Next:<a rel=next href="#R%20Web%20Interfaces">R Web Interfaces</a>,
Previous:<a rel=previous href="#R%20Basics">R Basics</a>,
Up:<a rel=up href="#Top">Top</a>
<br>

<h1>3 R and S</h1>

<ul>
<li><a href="#What%20is%20S%3f">What is S?</a>: 
<li><a href="#What%20is%20S-PLUS%3f">What is S-PLUS?</a>: 
<li><a href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>: 
<li><a href="#Is%20there%20anything%20R%20can%20do%20that%20S-PLUS%20cannot%3f">Is there anything R can do that S-PLUS cannot?</a>: 
</ul>

<p><hr>
Node:<a name="What%20is%20S%3f">What is S?</a>,
Next:<a rel=next href="#What%20is%20S-PLUS%3f">What is S-PLUS?</a>,
Previous:<a rel=previous href="#R%20and%20S">R and S</a>,
Up:<a rel=up href="#R%20and%20S">R and S</a>
<br>

<h2>3.1 What is S?</h2>

<p>S is a very high level language and an environment for data analysis and
graphics.  S was written by Richard A. Becker, John M. Chambers, and
Allan R. Wilks of AT&amp;T Bell Laboratories Statistics Research Department.

<p>The primary references for S are two books by the creators of S.

<ul>
<li>Richard A. Becker, John M. Chambers and Allan R. Wilks (1988), "The New
S Language," Chapman &amp; Hall, London.

<p>This book is often called the "<em>Blue Book</em>". 
</p><li>John M. Chambers and Trevor J. Hastie (1992), "Statistical Models in
S," Chapman &amp; Hall, London.

<p>This is also called the "<em>White Book</em>". 
</ul>

<p>Version 4 of S, a major revision of S designed by John Chambers to
improve its usefulness at every stage of the programming process, is
described in <a href="http://cm.bell-labs.com/cm/ms/departments/sia/Sbook/">"Programming with Data"</a> by John M. Chambers (1998), Springer: New
York, ISBN 0-387-98503-4.

<p>In 1998, the Association for Computing Machinery presented its Software
System Award to John Chambers for the design of the S system. The ACM
citation stated that "S has forever altered the way people analyze,
visualize, and manipulate data <small>...</small>. S is an elegant, widely
accepted, and enduring software system, with conceptual integrity,
thanks to the insight, taste, and effort of John Chambers."  See
<a href="http://netlib.bell-labs.com/cm/ms/departments/sia/S/index.html">http://netlib.bell-labs.com/cm/ms/departments/sia/S/index.html</a>
for "Stages in the Evolution of S".

<p>There is a huge amount of user-contributed code for S, available at the
<a href="http://lib.stat.cmu.edu">S Repository</a> at CMU.

<p>The <a href="http://lib.stat.cmu.edu/S/faq">"Frequently Asked Questions about S"</a> contains further information about S, but is not up-to-date.

<p><hr>
Node:<a name="What%20is%20S-PLUS%3f">What is S-PLUS?</a>,
Next:<a rel=next href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>,
Previous:<a rel=previous href="#What%20is%20S%3f">What is S?</a>,
Up:<a rel=up href="#R%20and%20S">R and S</a>
<br>

<h2>3.2 What is S-PLUS?</h2>

<p>S-PLUS is a value-added version of S sold by Statistical Sciences,
Inc. (now a division of Mathsoft, Inc.).  Based on the S language,
S-PLUS provides functionality in a wide variety of areas, including
robust regression, modern non-parametric regression, time series,
survival analysis, multivariate analysis, classical statistical tests,
quality control, and graphics drivers.  Add-on modules add additional
capabilities for wavelet analysis, spatial statistics, GARCH models, and
design of experiments.

<p>See the <a href="http://www.mathsoft.com/splus/">MathSoft S-PLUS page</a> for
further information.

<p><hr>
Node:<a name="What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>,
Next:<a rel=next href="#Is%20there%20anything%20R%20can%20do%20that%20S-PLUS%20cannot%3f">Is there anything R can do that S-PLUS cannot?</a>,
Previous:<a rel=previous href="#What%20is%20S-PLUS%3f">What is S-PLUS?</a>,
Up:<a rel=up href="#R%20and%20S">R and S</a>
<br>

<h2>3.3 What are the differences between R and S?</h2>

<ul>
<li><a href="#Lexical%20scoping">Lexical scoping</a>: 
<li><a href="#Models">Models</a>: 
<li><a href="#Others">Others</a>: 
</ul>

<p><hr>
Node:<a name="Lexical%20scoping">Lexical scoping</a>,
Next:<a rel=next href="#Models">Models</a>,
Previous:<a rel=previous href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>,
Up:<a rel=up href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>
<br>

<h3>3.3.1 Lexical scoping</h3>

<p>Whereas the developers of R have tried to stick to the S language as
defined in "The New S Language" (Blue Book, see <a href="#What%20is%20S%3f">What is S?</a>),
they have adopted the evaluation model of Scheme.

<p>This difference becomes manifest when <em>free</em> variables occur in a
function.  Free variables are those which are neither formal parameters
(occurring in the argument list of the function) nor local variables
(created by assigning to them in the body of the function).  Whereas S
(like C) by default uses <em>static</em> scoping, R (like Scheme) has
adopted <em>lexical</em> scoping.  This means the values of free variables
are determined by a set of global variables in S, but in R by the
bindings that were in effect at the time the function was created.

<p>Consider the following function:

<pre>cube &lt;- function(n) {
  sq &lt;- function() n * n
  n * sq()
}
</pre>

<p>Under S, <code>sq()</code> does not "know" about the variable <code>n</code>
unless it is defined globally:

<pre>S&gt; cube(2)
Error in sq():  Object "n" not found
Dumped
S&gt; n &lt;- 3
S&gt; cube(2)
[1] 18
</pre>

<p>In R, the "environment" created when <code>cube()</code> was invoked is
also looked in:

<pre>R&gt; cube(2)
[1] 8
</pre>

<p>As a more "interesting" real-world problem, suppose you want to write
a function which returns the density function of the r-th order
statistic from a sample of size n from a (continuous) distribution. 
For simplicity, we shall use both the cdf and pdf of the distribution
as explicit arguments.  (Example compiled from various postings by Luke
Tierney.)

<p>The S-PLUS documentation for <code>call()</code> basically suggests the
following:

<pre>dorder &lt;- function(n, r, pfun, dfun) {
  f &lt;- function(x) NULL
  con &lt;- round(exp(lgamma(n + 1) - lgamma(r) - lgamma(n - r + 1)))
  PF &lt;- call(substitute(pfun), as.name("x"))
  DF &lt;- call(substitute(dfun), as.name("x"))
  f[[length(f)]] &lt;-
    call("*", con,
         call("*", call("^", PF, r - 1),
              call("*", call("^", call("-", 1, PF), n - r),
                   DF)))
  f
}
</pre>

<p>Rather tricky, isn't it?  The code uses the fact that in S,
functions are just lists of special mode with the function body as the
last argument, and hence does not work in R (one could make the idea
work, though).

<p>A version which makes heavy use of <code>substitute()</code> and seems to work
under both S and R is

<pre>dorder &lt;- function(n, r, pfun, dfun) {
  con &lt;- round(exp(lgamma(n + 1) - lgamma(r) - lgamma(n - r + 1)))
  eval(substitute(function(x) K * PF(x)^a * (1 - PF(x))^b * DF(x),
                  list(PF = substitute(pfun), DF = substitute(dfun),
                       a = r - 1, b = n - r, K = con)))
}
</pre>

<p>(the <code>eval()</code> is not needed in S).

<p>However, in R there is a much easier solution:

<pre>dorder &lt;- function(n, r, pfun, dfun) {
  con &lt;- round(exp(lgamma(n + 1) - lgamma(r) - lgamma(n - r + 1)))
  function(x) {
    con * pfun(x)^(r - 1) * (1 - pfun(x))^(n - r) * dfun(x)
  }
}
</pre>

<p>This seems to be the "natural" implementation, and it works because
the free variables in the returned function can be looked up in the
defining environment (this is lexical scope).

<p>Note that what you really need is the function <em>closure</em>, i.e., the
body along with all variable bindings needed for evaluating it.  Since
in the above version, the free variables in the value function are not
modified, you can actually use it in S as well if you abstract out the
closure operation into a function <code>MC()</code> (for "make closure"):

<pre>dorder &lt;- function(n, r, pfun, dfun) {
  con &lt;- round(exp(lgamma(n + 1) - lgamma(r) - lgamma(n - r + 1)))
  MC(function(x) {
       con * pfun(x)^(r - 1) * (1 - pfun(x))^(n - r) * dfun(x)
     },
     list(con = con, pfun = pfun, dfun = dfun, r = r, n = n))
}
</pre>

<p>Given the appropriate definitions of the closure operator, this works in
both R and S, and is much "cleaner" than a substitute/eval solution
(or one which overrules the default scoping rules by using explicit
access to evaluation frames, as is of course possible in both R and S).

<p>For R, <code>MC()</code> simply is

<pre>MC &lt;- function(f, env) f
</pre>

<p>(lexical scope!), a version for S is

<pre>MC &lt;- function(f, env = NULL) {
  env &lt;- as.list(env)
  if (mode(f) != "function")
    stop(paste("not a function:", f))
  if (length(env) &gt; 0 &amp;&amp; any(names(env) == ""))
    stop(paste("not all arguments are named:", env))
  fargs &lt;- if(length(f) &gt; 1) f[1:(length(f) - 1)] else NULL
  fargs &lt;- c(fargs, env)
  if (any(duplicated(names(fargs))))
    stop(paste("duplicated arguments:", paste(names(fargs)),
         collapse = ", "))
  fbody &lt;- f[length(f)]
  cf &lt;- c(fargs, fbody)
  mode(cf) &lt;- "function"
  return(cf)
}
</pre>

<p>Similarly, most optimization (or zero-finding) routines need some
arguments to be optimized over and have other parameters that depend on
the data but are fixed with respect to optimization.  With R scoping
rules, this is a trivial problem; simply make up the function with the
required definitions in the same environment and scoping takes care of
it.  With S, one solution is to add an extra parameter to the function
and to the optimizer to pass in these extras, which however can only
work if the optimizer supports this.

<p>Lexical scoping allows using function closures and maintaining local
state.  A simple example (taken from Abelson and Sussman) is obtained by
typing <kbd>demo(scoping)</kbd> at the R prompt.  Further information is
provided in the standard R reference "R: A Language for Data Analysis
and Graphics" (see <a href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>) and a paper on
"Lexical Scope and Statistical Computing" by Robert Gentleman and Ross
Ihaka which can be obtained from the <code>doc/misc</code> directory of a CRAN
site and will appear in the <em>Journal of Computational and Graphical
Statistics</em> around the beginning of 2000.

<p>Lexical scoping also implies a further major difference.  Whereas S
stores all objects as separate files in a directory somewhere (usually
<code>.Data</code> under the current directory), R does not.  All objects
in R are stored internally.  When R is started up it grabs a very large
piece of memory and uses it to store the objects.  R performs its own
memory management of this piece of memory.  Having everything in memory
is necessary because it is not really possible to externally maintain
all relevant "environments" of symbol/value pairs.  This difference
also seems to make R <em>faster</em> than S.

<p>The down side is that if R crashes you will lose all the work for the
current session.  Saving and restoring the memory "images" (the
functions and data stored in R's internal memory at any time) can be a
bit slow, especially if they are big.  In S this does not happen,
because everything is saved in disk files and if you crash nothing is
likely to happen to them.  (In fact, one might conjecture that the S
developers felt that the price of changing their approach to persistent
storage just to accommodate lexical scope was far too expensive.)  R is
still in a beta stage, and may crash from time to time.  Hence, for
important work you should consider saving often (see <a href="#How%20can%20I%20save%20my%20workspace%3f">How can I save my workspace?</a>).  Other possibilities are logging your sessions, or have
your R commands stored in text files which can be read in using
<code>source()</code>.

<blockquote>
<strong>Note:</strong>  If you run R from within Emacs (see <a href="#R%20and%20Emacs">R and Emacs</a>),
you can save the contents of the interaction buffer to a file and
conveniently manipulate it using <code>ess-transcript-mode</code>, as well as
save source copies of all functions and data used. 
</blockquote>

<p><hr>
Node:<a name="Models">Models</a>,
Next:<a rel=next href="#Others">Others</a>,
Previous:<a rel=previous href="#Lexical%20scoping">Lexical scoping</a>,
Up:<a rel=up href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>
<br>

<h3>3.3.2 Models</h3>

<p>There are some differences in the modeling code, such as

<ul>
<li>Whereas in S, you would use <code>lm(y ~ x^3)</code> to regress <code>y</code> on
<code>x^3</code>, in R, you have to insulate powers of numeric vectors (using
<code>I()</code>), i.e., you have to use <code>lm(y ~ I(x^3))</code>. 
<li>The glm family objects are implemented differently in R and S.  The same
functionality is available but the components have different names. 
<li>Terms objects are stored differently.  In S a terms object is an
expression with attributes, in R it is a formula with attributes.  The
attributes have the same names but are mostly stored differently.  The
major difference in functionality is that a terms object is
subscriptable in S but not in R.  If you can't imagine why this would
matter then you don't need to know. 
<li>Finally, in R <code>y~x+0</code> is an alternative to <code>y~x-1</code> for
specifying a model with no intercept.  Models with no parameters at all
can be specified by <code>y~0</code>. 
</ul>

<p><hr>
Node:<a name="Others">Others</a>,
Previous:<a rel=previous href="#Models">Models</a>,
Up:<a rel=up href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>
<br>

<h3>3.3.3 Others</h3>

<p>Apart from lexical scoping and its implications, R follows the S
language definition in the Blue Book as much as possible, and hence
really is an "implementation" of S.  There are some intentional
differences where the behavior of S is considered "not clean".  In
general, the rationale is that R should help you detect programming
errors, while at the same time being as compatible as possible with S.

<p>Some known differences are the following.

<ul>

<li>In R, if <code>x</code> is a list, then <code>x[i] &lt;- NULL</code> and <code>x[[i]]
&lt;- NULL</code> remove the specified elements from <code>x</code>.  The first of
these is incompatible with S, where it is a no-op.  (Note that you can
set elements to <code>NULL</code> using <code>x[i] &lt;- list(NULL)</code>.)

<li>In S, the functions named <code>.First</code> and <code>.Last</code> in the
<code>.Data</code> directory can be used for customizing, as they are executed
at the very beginning and end of a session, respectively.

<p>In R, the startup mechanism is as follows.  R first sources the system
startup file <code><code>$R_HOME</code>/library/base/R/Rprofile</code>.  Then, it
searches for a site-wide startup profile unless the command line option
<code>--no-site-file</code> was given.  The name of this file is taken from
the value of the <code>R_PROFILE</code> environment variable.  If that variable
is unset, the default is <code><code>$R_HOME</code>/etc/Rprofile</code>.  Then,
unless <code>--no-init-file</code> was given, R searches for a file called
<code>.Rprofile</code> in the current directory or in the user's home
directory (in that order) and sources it.  It also loads a saved image
from <code>.RData</code> in case there is one (unless <code>--no-restore</code>
was specified).  If needed, the functions <code>.First()</code> and
<code>.Last()</code> should be defined in the appropriate startup profiles.

</p><li>In R, <code>T</code> and <code>F</code> are just variables being set to <code>TRUE</code>
and <code>FALSE</code>, respectively, but are not reserved words as in S and
hence can be overwritten by the user.  (This helps e.g. when you have
factors with levels "T" or "F".)  Hence, when writing code you should
always use <code>TRUE</code> and <code>FALSE</code>.

<li>In R, <code>dyn.load()</code> can only load <em>shared libraries</em>, as
created for example by <kbd>R SHLIB</kbd>.

<li>In R, <code>attach()</code> currently only works for lists and data frames
(not for directories).  Also, you cannot attach at position 1.

<li>Categories do not exist in R, and never will as they are deprecated now
in S.  Use factors instead.

<li>In R, <code>For()</code> loops are not necessary and hence not supported.

<li>In R, <code>assign()</code> uses the argument <code>envir=</code> rather than
<code>where=</code> as in S.

<li>The random number generators are different, and the seeds have different
length.

<li>R passes integer objects to C as <code>int *</code> rather than <code>long *</code>
as in S.

<li>R has no single precision storage mode.  However, as of version 0.65.1,
there is a single precision interface to C/FORTRAN subroutines.

<li>By default, <code>ls()</code> returns the names of the objects in the current
(under R) and global (under S) environment, respectively.  For example,
given

<pre>x &lt;- 1; fun &lt;- function() {y &lt;- 1; ls()}
</pre>

<p>then <code>fun()</code> returns <code>"y"</code> in R and <code>"x"</code> (together with
the rest of the global environment) in S.

</p><li>R allows for zero-extent matrices (and arrays, i.e., some elements of
the <code>dim</code> attribute vector can be 0).  This has been determined a
useful feature as it helps reducing the need for special-case tests for
empty subsets.  For example, if <code>x</code> is a matrix, <code>x[, FALSE]</code>
is not <code>NULL</code> but a "matrix" with 0 columns.  Hence, such objects
need to be tested for by checking whether their <code>length()</code> is zero
(which works in both R and S), and not using <code>is.null()</code>.

<li>Named vectors are considered vectors in R but not in S (e.g.,
<code>is.vector(c(a = 1:3))</code> returns <code>FALSE</code> in S and <code>TRUE</code>
in R).

<li>Data frames are not considered as matrices in R (i.e., if <code>DF</code> is a
data frame, then <code>is.matrix(DF)</code> returns <code>FALSE</code> in R and
<code>TRUE</code> in S).

<li>R by default uses treatment contrasts in the unordered case, whereas S
uses the Helmert ones.  This is a deliberate difference reflecting the
opinion that treatment contrasts are more natural.

<li>In R, the last argument (which corresponds to the right hand side) of an
assignment function must be named <code>value</code>.  E.g., <code>fun(a) &lt;-
b</code> is evaluated as <code>(fun&lt;-)(a, value = b)</code>.

<li>In S, <code>substitute()</code> searches for names for substitution in the
given expression in three places: the actual and the default arguments
of the matching call, and the local frame (in that order).  R looks in
the local frame only, with the special rule to use a "promise" if a
variable is not evaluated. Since the local frame is initialized with the
actual arguments or the default expressions, this is usually equivalent
to S, until assignment takes place.

<li>In R, <code>eval(EXPR, sys.parent())</code> does not work.  Instead, one
should use <code>eval(EXPR, sys.frame(sys.parent())),</code> which also works
in S.

<li>In S, the index variable in a <code>for()</code> loop is local to the inside
of the loop.  In R it is local to the environment where the <code>for()</code>
statement is executed.

<li>In S, <code>tapply(simplify=TRUE)</code> returns a vector where R returns a
one-dimensional array (which can have named dimnames).

</ul>

<p>There are also differences which are not intentional, and result from
missing or incorrect code in R.  The developers would appreciate hearing
about any deficiencies you may find (in a written report fully
documenting the difference as you see it).  Of course, it would be
useful if you were to implement the change yourself and make sure it
works.

<p><hr>
Node:<a name="Is%20there%20anything%20R%20can%20do%20that%20S-PLUS%20cannot%3f">Is there anything R can do that S-PLUS cannot?</a>,
Previous:<a rel=previous href="#What%20are%20the%20differences%20between%20R%20and%20S%3f">What are the differences between R and S?</a>,
Up:<a rel=up href="#R%20and%20S">R and S</a>
<br>

<h2>3.4 Is there anything R can do that S-PLUS cannot?</h2>

<p>Since almost anything you can do in R has source code that you could
port to S-PLUS with little effort there will never be much you can do in
R that you couldn't do in S-PLUS if you wanted to.  (Note that using
lexical scoping may simplify matters considerably, though.)

<p>R offers several graphics features that S-PLUS does not, such as finer
handling of line types, more convenient color handling (via palettes),
gamma correction for color, and, most importantly, mathematical
annotation in plot texts, via input expressions reminiscent of TeX
constructs.  Unfortunately, this feature still is mostly undocumented. 
The paper "An Approach to Providing Mathematical Annotation in Plots"
by Paul Murrell and Ross Ihaka, which will soon appear in the
<em>Journal of Computational and Graphical Statistics</em>, has more
details on this.

<p><hr>
Node:<a name="R%20Web%20Interfaces">R Web Interfaces</a>,
Next:<a rel=next href="#R%20Add-On%20Packages">R Add-On Packages</a>,
Previous:<a rel=previous href="#R%20and%20S">R and S</a>,
Up:<a rel=up href="#Top">Top</a>
<br>

<h1>4 R Web Interfaces</h1>

<p><strong>Rcgi</strong> is a CGI WWW interface to R by <a href="mailto:h089@mth.uea.ac.uk">Mark J Ray</a>.  Recent version have the ability to use "embedded code":
you can mix user input and code, allowing the HTML author to do anything
from load in data sets to enter most of the commands for users without
writing CGI scripts.  Graphical output is possible in PostScript or GIF
formats and the executed code is presented to the user for revision.

<p>Demo and download are available from
<a href="http://www.mth.uea.ac.uk/~h089/Rcgi/">http://www.mth.uea.ac.uk/~h089/Rcgi/</a>.

<p><strong>Rweb</strong> is developed and maintained by
<a href="mailto:jeff@math.montana.edu">Jeff Banfield</a>.  The
<a href="http://www.math.montana.edu/Rweb">Rweb Home Page</a> provides access
to all three versions of Rweb--a simple text entry form that returns
output and graphs, a more sophisticated Javascript version that provides
a multiple window environment, and a set of point and click modules that
are useful for introductory statistics courses and require no knowledge
of the R language.  All of the Rweb versions can analyze Web accessible
datasets if a URL is provided.

<p>A paper on Rweb, providing a detailed explanation of the different
versions of Rweb and an overview of how Rweb works, was published in the
Journal of Statistical Software
(<a href="http://www.stat.ucla.edu/journals/jss/">http://www.stat.ucla.edu/journals/jss/</a>).

<p><hr>
Node:<a name="R%20Add-On%20Packages">R Add-On Packages</a>,
Next:<a rel=next href="#R%20and%20Emacs">R and Emacs</a>,
Previous:<a rel=previous href="#R%20Web%20Interfaces">R Web Interfaces</a>,
Up:<a rel=up href="#Top">Top</a>
<br>

<h1>5 R Add-On Packages</h1>

<ul>
<li><a href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>: 
<li><a href="#How%20can%20add-on%20packages%20be%20installed%3f">How can add-on packages be installed?</a>: 
<li><a href="#How%20can%20add-on%20packages%20be%20used%3f">How can add-on packages be used?</a>: 
<li><a href="#How%20can%20add-on%20packages%20be%20removed%3f">How can add-on packages be removed?</a>: 
<li><a href="#How%20can%20I%20create%20an%20R%20package%3f">How can I create an R package?</a>: 
<li><a href="#How%20can%20I%20contribute%20to%20R%3f">How can I contribute to R?</a>: 
</ul>

<p><hr>
Node:<a name="Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>,
Next:<a rel=next href="#How%20can%20add-on%20packages%20be%20installed%3f">How can add-on packages be installed?</a>,
Previous:<a rel=previous href="#R%20Add-On%20Packages">R Add-On Packages</a>,
Up:<a rel=up href="#R%20Add-On%20Packages">R Add-On Packages</a>
<br>

<h2>5.1 Which add-on packages exist for R?</h2>

<p>The R distribution comes with the following extra packages:

<dl>
<dt><strong>eda</strong>
<dd>Exploratory Data Analysis.  Currently only contains functions for robust
line fitting, and median polish and smoothing. 
<br><dt><strong>lqs</strong>
<dd>Resistant regression and covariance estimation. 
<br><dt><strong>modreg</strong>
<dd>MODern REGression: smoothing and local methods. 
<br><dt><strong>mva</strong>
<dd>MultiVariate Analysis.  Currently contains code for principal
components, canonical correlations, metric multidimensional scaling, and
hierarchical and k-means clustering. 
<br><dt><strong>nls</strong>
<dd>Nonlinear regression routines. 
<br><dt><strong>splines</strong>
<dd>Regression spline functions and classes. 
<br><dt><strong>stepfun</strong>
<dd>Code for dealing with step functions, including empirical cumulative
distribution functions. 
<br><dt><strong>ts</strong>
<dd>Time series. 
</dl>

<p>The following packages are available from the CRAN <code>src/contrib</code>
area.

<dl>
<dt><strong>KernSmooth</strong>
<dd>Functions for kernel smoothing (and density estimation) corresponding to
the book "Kernel Smoothing" by M. P. Wand and M. C. Jones, 1995. 
<br><dt><strong>MASS</strong>
<dd>Functions and datasets from the main library of Venables and Ripley,
"Modern Applied Statistics with S-PLUS".  Contained in <code>VR</code>. 
<br><dt><strong>NISTnls</strong>
<dd>A set of test nonlinear least squares examples from NIST, the
U.S. National Institute for Standards and Technology. 
<br><dt><strong>RmSQL</strong>
<dd>An interface between R and the mSQL database system. 
<br><dt><strong>Rnotes</strong>
<dd>The data sets for the exercises in Rnotes (see <a href="#What%20documentation%20exists%20for%20R%3f">What documentation exists for R?</a>). 
<br><dt><strong>acepack</strong>
<dd>ace (Alternating Conditional Expectations) and avas (Additivity and
VAriance Stabilization for regression) for selecting regression
transformations. 
<br><dt><strong>akima</strong>
<dd>Linear or cubic spline interpolation for irregularly gridded data. 
<br><dt><strong>ash</strong>
<dd>Programs for 1D and 2D density estimation. 
<br><dt><strong>bindata</strong>
<dd>Generation of correlated artificial binary data. 
<br><dt><strong>boot</strong>
<dd>Functions and datasets for bootstrapping from the book "Bootstrap
Methods and Their Applications" by A. C. Davison and D. V. Hinkley,
1997, Cambridge University Press. 
<br><dt><strong>bootstrap</strong>
<dd>Software (bootstrap, cross-validation, jackknife), data and errata for
the book "An Introduction to the Bootstrap" by B. Efron and
R. Tibshirani, 1993, Chapman and Hall. 
<br><dt><strong>cclust</strong>
<dd>Convex clustering methods, including k-means algorithm, on-line update
algorithm (Hard Competitive Learning) and Neural Gas algorithm (Soft
Competitive Learning) and calculation of several indexes for finding the
number of clusters in a data set. 
<br><dt><strong>chron</strong>
<dd>A package for working with chronological objects (times and dates). 
<br><dt><strong>class</strong>
<dd>Functions for classification (k-nearest neighbor and LVQ).  Contained in
<code>VR</code>. 
<br><dt><strong>cluster</strong>
<dd>Functions for cluster analysis. 
<br><dt><strong>coda</strong>
<dd>Output analysis and diagnostics for Markov Chain Monte Carlo (MCMC)
simulations. 
<br><dt><strong>ctest</strong>
<dd>A collection of classical tests, including the Bartlett, Fisher,
Kruskal-Wallis, Kolmogorov-Smirnov, and Wilcoxon tests. 
<br><dt><strong>date</strong>
<dd>Functions for dealing with dates.  The most useful of them accepts a
vector of input dates in any of the forms <code>8/30/53</code>,
<code>30Aug53</code>, <code>30 August 1953</code>, <small>...</small>, <code>August 30 53</code>, or
any mixture of these. 
<br><dt><strong>e1071</strong>
<dd>Miscellaneous functions used at the Department of Statistics at TU Wien
(E1071). 
<br><dt><strong>fracdiff</strong>
<dd>Maximum likelihood estimation of the parameters of a fractionally
differenced ARIMA(p,d,q) model (Haslett and Raftery, Applied Statistics,
1989). 
<br><dt><strong>funfits</strong>
<dd>An integrated set of functions for fitting curves and surfaces including
thin plate splines, kriging and neural networks. 
<br><dt><strong>gee</strong>
<dd>An implementation of the Liang/Zeger generalized estimating equation
approach to GLMs for dependent data. 
<br><dt><strong>gss</strong>
<dd>A comprehensive package for structural multivariate function estimation
using smoothing splines. 
<br><dt><strong>integrate</strong>
<dd>Adaptive quadrature in up to 20 dimensions. 
<br><dt><strong>leaps</strong>
<dd>A package which performs an exhaustive search for the best subsets of a
given set of potential regressors, using a branch-and-bound algorithm,
and also performs searches using a number of less time-consuming
techniques. 
<br><dt><strong>lmtest</strong>
<dd>A collection of tests on the assumptions of linear regression models
from the book "The linear regression model under test" by W. Kraemer
and H. Sonnberger, 1986, Physica. 
<br><dt><strong>locfit</strong>
<dd>Local Regression, likelihood and density estimation. 
<br><dt><strong>logspline</strong>
<dd>Logspline density estimation. 
<br><dt><strong>mclust</strong>
<dd>Model-based cluster analysis. 
<br><dt><strong>mda</strong>
<dd>Code for mixture discriminant analysis (MDA), flexible discriminant
analysis (FDA), penalized discriminant analysis (PDA), multivariate
additive regression splines (MARS), adaptive back-fitting splines
(BRUTO), and penalized regression. 
<br><dt><strong>mlbench</strong>
<dd>A collection of artificial and real-world machine learning benchmark
problems, including the Boston housing data. 
<br><dt><strong>multilm</strong>
<dd>A basic method for fitting and testing multivariate linear models,
including stabilized test procedures by Laeuter et. al. 
<br><dt><strong>multiv</strong>
<dd>Functions for hierarchical clustering, partitioning, bond energy
algorithm, Sammon mapping, PCA and correspondence analysis. 
<br><dt><strong>nlme</strong>
<dd>Fit and compare Gaussian linear and nonlinear mixed-effects models. 
<br><dt><strong>nnet</strong>
<dd>Software for single hidden layer perceptrons ("feed-forward neural
networks"), and for multinomial log-linear models.  Contained in
<code>VR</code>. 
<br><dt><strong>norm</strong>
<dd>Analysis of multivariate normal datasets with missing values. 
<br><dt><strong>oz</strong>
<dd>Functions for plotting Australia's coastline and state boundaries. 
<br><dt><strong>pls</strong>
<dd>Univariate Partial Least Squares Regression. 
<br><dt><strong>polymars</strong>
<dd>Polychotomous regression based on Multivariate Adaptive Regression
Splines. 
<br><dt><strong>polynom</strong>
<dd>A collection of functions to implement a class for univariate polynomial
manipulations. 
<br><dt><strong>princurve</strong>
<dd>Fits a principal curve to a matrix of points in arbitrary dimension. 
<br><dt><strong>pspline</strong>
<dd>Smoothing splines with penalties on order m derivatives. 
<br><dt><strong>quadprog</strong>
<dd>For solving quadratic programming problems. 
<br><dt><strong>quantreg</strong>
<dd>Compute regression quantiles and some related rank statistics. 
<br><dt><strong>rmeta</strong>
<dd>Functions for simple fixed and random effects meta-analysis for
two-sample comparison of binary outcomes. 
<br><dt><strong>rpart</strong>
<dd>Recursive Partitioning. 
<br><dt><strong>sgeostat</strong>
<dd>An object-oriented framework for geostatistical modeling. 
<br><dt><strong>sm</strong>
<dd>Software linked to the book "Applied Smoothing Techniques for Data
Analysis:  The Kernel Approach with S-PLUS Illustrations" by
A. W. Bowman and A. Azzalini (1997), Oxford University Press. 
<br><dt><strong>spatial</strong>
<dd>Functions for kriging and point pattern analysis from "Modern Applied
Statistics with S-PLUS" by W. Venables and B. Ripley.  Contained in
<code>VR</code>. 
<br><dt><strong>stable</strong>
<dd>Density, distribution, quantile and hazard functions of a stable
variate; generalized linear models for the parameters of a stable
distribution. 
<br><dt><strong>stataread</strong>
<dd>Read and write Stata v6 <code>.dta</code> files. 
<br><dt><strong>survival5</strong>
<dd>Functions for survival analysis, version 5 (requires <strong>splines</strong>
and suggests <strong>date</strong>), the main new feature being penalised
(partial) likelihood. 
<br><dt><strong>tree</strong>
<dd>Classification and regression trees. 
<br><dt><strong>tripack</strong>
<dd>A constrained two-dimensional Delaunay triangulation package. 
<br><dt><strong>tseries</strong>
<dd>Additional code for time series analysis. 
<br><dt><strong>wavethresh</strong>
<dd>Software to perform 1-d and 2-d wavelet statistics and transforms. 
<br><dt><strong>xgobi</strong>
<dd>Interface to the XGobi and XGvis programs for graphical data analysis. 
</dl>

<p>See CRAN <code>src/contrib/INDEX</code> for more information.

<p>There is also a CRAN <code>src/contrib/Devel</code> directory which contains
packages still "under development" or depending on features only
present in the current development versions of R.  Volunteers are
invited to give these a try, of course.  This area of CRAN currently
contains

<dl>
<dt><strong>bats</strong>
<dd>Basic time series modelling functions. 
<br><dt><strong>dopt</strong>
<dd>Finding D-optimal experimental designs. 
<br><dt><strong>tcltk</strong>
<dd>Basic interface with Tcl/Tk. 
<br><dt><strong>timeslab</strong>
<dd>Time series routines. 
<br><dt><strong>vtcl</strong>
<dd>Interface to Visual Tcl. 
</dl>

<a href="mailto:jlindsey@luc.ac.be">Jim Lindsey</a> has written a collection of R
packages for nonlinear regression and repeated measurements, consisting
of <strong>event</strong> (procedures for event history analysis), <strong>gnlm</strong>
(generalized nonlinear regression models), <strong>growth</strong> (normal
theory repeated measurements models), <strong>repeated</strong> (models for
non-normal repeated measurements.), and <strong>rmutil</strong> (tools for
repeated measurements).  Jim's packages were used for all analyses in
the new edition of his book "Models for Repeated Measurements" (1999,
Oxford University Press), and can be obtained from
<a href="http://www.luc.ac.be/~jlindsey/rcode.html">http://www.luc.ac.be/~jlindsey/rcode.html</a>.

<a href="mailto:pgilbert@bank-banque-canada.ca">Paul Gilbert</a> has written the
Dynamic Systems Estimation (DSE) Library, a multivariate time series
package which implements an object oriented approach to time series
models (using classes and methods in R/S).  The package provides
state-space models and the Kalman filter, VARMA and cointegration
models, and numerical differentiation.  It also contains Troll models as
another class (the Troll interface is not yet functional in R).  Paul's
package <strong>dse</strong> can be obtained via
<a href="http://www.bank-banque-canada.ca/pgilbert/dsedesc.htm">http://www.bank-banque-canada.ca/pgilbert/dsedesc.htm</a>.

<p>More code has been posted to the r-help mailing list, and can be
obtained from the mailing list archive.

<p><hr>
Node:<a name="How%20can%20add-on%20packages%20be%20installed%3f">How can add-on packages be installed?</a>,
Next:<a rel=next href="#How%20can%20add-on%20packages%20be%20used%3f">How can add-on packages be used?</a>,
Previous:<a rel=previous href="#Which%20add-on%20packages%20exist%20for%20R%3f">Which add-on packages exist for R?</a>,
Up:<a rel=up href="#R%20Add-On%20Packages">R Add-On Packages</a>
<br>

<h2>5.2 How can add-on packages be installed?</h2>

<p>(Unix only.)  The add-on packages on CRAN come as gzipped tar files
(which may contain more than one package).  First "unpack" the files
of interest.  If you have GNU tar, you can use <kbd>tar zxf <var>name</var></kbd>,
otherwise you can use <kbd>gunzip -c <var>name</var> | tar xf -</kbd>.

<p>Let <var>pkgdir_1</var>, <small>...</small>, <var>pkgdir_n</var> be the (relative or
absolute) path names of the packages to be installed.  (In the simplest
case, the unpacking creates a single package directory, and its name is
used.)  To install to the default R directory tree (the <code>library</code>
subdirectory of <code>R_HOME</code>), type

<pre>$ R INSTALL <var>pkgdir_1</var> <small>...</small> <var>pkgdir_n</var>
</pre>

<p>at the shell prompt.  To install to another tree (e.g., your private
one), use

<pre>$ R INSTALL -l <var>lib</var> <var>pkgdir_1</var> <small>...</small> <var>pkgdir_n</var>
</pre>

<p>where <var>lib</var> gives the path to the library tree to install to.

<p>As of R version 0.65.1, you no longer need to explicitly unpack for
installing; i.e., you can do <kbd>R INSTALL /path/to/pkg_ver.tar.gz</kbd>.

<p>You can use several library trees of add-on packages.  The easiest way
to tell R to use these is via the environment variable <code>R_LIBS</code>
which should be a colon-separated list of directories at which R library
trees are rooted.  You do not have to specify the default tree in
<code>R_LIBS</code>.  E.g., to use a private tree in <code>$HOME/lib/R</code> and a
public site-wide tree in <code>/usr/local/lib/R/site</code>, put

<pre>R_LIBS="$HOME/lib/R:/usr/local/lib/R/site"; export R_LIBS
</pre>

<p>into your (Bourne) shell profile or your <code>~/.Renviron</code> file.

<p><hr>
Node:<a name="How%20can%20add-on%20packages%20be%20used%3f">How can add-on packages be used?</a>,
Next:<a rel=next href="#How%20can%20add-on%20packages%20be%20removed%3f">How can add-on packages be removed?</a>,
Previous:<a rel=previous href="#How%20can%20add-on%20packages%20be%20installed%3f">How can add-on packages be installed?</a>,
Up:<a rel=up href="#R%20Add-On%20Packages">R Add-On Packages</a>
<br>

<h2>5.3 How can add-on packages be used?</h2>

<p>To find out which additional packages are available on your system, type

<pre>library()
</pre>

<p>at the R prompt.

<p>This produces something like

<pre>Packages in `/home/me/lib/R':

mystuff      My own R functions, nicely packaged but not documented

Packages in `/usr/local/lib/R/library':

MASS         Main package of Venables and Ripley's MASS
base         The R base package
class        Functions for classification
cluster      Functions for clustering
ctest        Classical tests
date         Functions for handling dates
eda          Exploratory Data Analysis
gee          Generalized Estimating Equation models
lme          Linear mixed effects library
locfit       Local regression, likelihood and density estimation
lqs          Resistant regression and covariance estimation
modreg       Modern regression: smoothing and local methods
mva          Classical Multivariate Analysis
nls          Nonlinear regression
nnet         Software for feed-forward neural networks with a single
             hidden layer and for multinomial log-linear models.
splines      Regression spline functions and classes
stepfun      Step functions, including empirical distributions
survival5    Survival analysis
ts           Time series functions
</pre>

<p>You can "load" the installed package <var>pkg</var> by

<pre>library(<var>pkg</var>)
</pre>

<p>You can then find out which functions it provides by typing one of

<pre>help(package = <var>pkg</var>)
library(help = <var>pkg</var>)
</pre>

<p>You can unload the loaded package <var>pkg</var> by

<pre>detach("package:<var>pkg</var>")
</pre>

<p><hr>
Node:<a name="How%20can%20add-on%20packages%20be%20removed%3f">How can add-on packages be removed?</a>,
Next:<a rel=next href="#How%20can%20I%20create%20an%20R%20package%3f">How can I create an R package?</a>,
Previous:<a rel=previous href="#How%20can%20add-on%20packages%20be%20used%3f">How can add-on packages be used?</a>,
Up:<a rel=up href="#R%20Add-On%20Packages">R Add-On Packages</a>
<br>

<h2>5.4 How can add-on packages be removed?</h2>

<p>To remove the packages <var>pkg_1</var>, <small>...</small>, <var>pkg_n</var> from the
default library or the library <var>lib</var>, do

<pre>$ R REMOVE <var>pkg_1</var> <small>...</small> <var>pkg_n</var>
</pre>

<p>or

<pre>$ R REMOVE -l <var>lib</var> <var>pkg_1</var> <small>...</small> <var>pkg_n</var>
</pre>

<p>respectively.

<p><hr>
Node:<a name="How%20can%20I%20create%20an%20R%20package%3f">How can I create an R package?</a>,
Next:<a rel=next href="#How%20can%20I%20contribute%20to%20R%3f">How can I contribute to R?</a>,
Previous:<a rel=previous href="#How%20can%20add-on%20packages%20be%20removed%3f">How can add-on packages be removed?</a>,
Up:<a rel=up href="#R%20Add-On%20Packages">R Add-On Packages</a>
<br>

<h2>5.5 How can I create an R package?</h2>

<p>A package consists of a subdirectory containing the files
<code>DESCRIPTION</code>, <code>INDEX</code>, and <code>TITLE</code>, and the
subdirectories <code>R</code>, <code>data</code>, <code>exec</code>, <code>inst</code>,
<code>man</code> and <code>src</code> (some of which can be missing).

<p>The <code>DESCRIPTION</code> file contains basic information about the package
in the following format:

<pre>Package: e1071
Version: 0.7-3
Author: Compiled by Fritz Leisch &lt;Friedrich.Leisch@ci.tuwien.ac.at&gt;.
Description: Miscellaneous functions used at the Department of
        Statistics at TU Wien (E1071).
Depends:
License: GPL version 2 or later
</pre>

<p>The license field should contain an explicit statement or a well-known
abbreviation (such as <code>GPL</code>, <code>LGPL</code>, <code>BSD</code> and
<code>Artistic</code>), maybe followed by a reference to the actual license
file.  It is very important that you include this information! 
Otherwise, it may not even be legally correct for others to distribute
copies of the package.

<p>The <code>TITLE</code> file contains a line giving the name of the package
and a brief description.  <code>INDEX</code> contains a line for each
sufficiently interesting object in the package, giving its name and a
description (functions such as print methods not usually called
explicitly might not be included).  Note that you can automatically
create this file using something like <code>R CMD Rdindex man/*.Rd &gt;
INDEX</code> provided that Perl is available on your system.

<p>The <code>R</code> subdirectory contains code files.  The code files to be
installed must start with a (lower- or uppercase) letter and have one of
the extensions <code>.R</code>, <code>.S</code>, <code>.q</code>, <code>.r</code>, or <code>.s</code>. 
We recommend using <code>.R</code>, as this extension seems to be not used by
any other software.  It should be possible to read in the files using
<code>source()</code>, so R objects must be created by assignments.  Note that
there has to be no connection between the name of the file and the R
objects created by it.  If necessary, one of these files (historically
<code>zzz.R</code>) should use <code>library.dynam()</code> <em>inside</em>
<code>.First.lib()</code> to load compiled code.

<p>The <code>man</code> subdirectory should contain documentation files for the
objects in the package.  The documentation files to be installed must
also start with a (lower- or uppercase) letter and have the extension
<code>.Rd</code> (the default) or <code>.rd</code>.

<p>C or FORTRAN source and optionally a <code>Makefile</code> for the compiled
code is in <code>src</code>.  Note that the <code>Makefile</code> most likely is not
needed.

<p>The <code>data</code> subdirectory is for additional data files the package
makes available for loading using <code>data()</code>.  Currently, data files
can have one of three types as indicated by their extension:  plain R
code (<code>.R</code> or <code>.r</code>), tables (<code>.tab</code>, <code>.txt</code>, or
<code>.csv</code>), or <code>save()</code> images (<code>.RData</code> or <code>.rda</code>). 
The subdirectory should contain a <code>00Index</code> file that describes
the datasets available.

<p>The contents of the <code>inst</code> subdirectory will be copied recursively
to the installation directory.

<p>Finally, <code>exec</code> could contain additional executables the package
needs, typically Shell or Perl scripts.  This mechanism is currently not
used by any package, and still experimental.

<p>You can also provide (Bourne) shell scripts <code>configure</code> and
<code>cleanup</code> for configuration before installation (such as checking
for additional software that might be needed by the package), and
removing the files created in this process.  See the add-on package
<strong>e1071</strong> on CRAN for an example of this mechanism.

<p>See the documentation for <code>library()</code> for more information.

<p>The web page <a href="http://www.biostat.washington.edu/~thomas/Rlib.html">http://www.biostat.washington.edu/~thomas/Rlib.html</a>
maintained by Thomas Lumley provides information on porting S packages
to R.

<p>See <a href="#What%20is%20CRAN%3f">What is CRAN?</a>, for information on uploading a package to CRAN.

<p><hr>
Node:<a name="How%20can%20I%20contribute%20to%20R%3f">How can I contribute to R?</a>,
Previous:<a rel=previous href="#How%20can%20I%20create%20an%20R%20package%3f">How can I create an R package?</a>,
Up:<a rel=up href="#R%20Add-On%20Packages">R Add-On Packages</a>
<br>

<h2>5.6 How can I contribute to R?</h2>

<p>R is in active development and there is always a risk of bugs creeping
in.  Also, the developers do not have access to all possible machines
capable of running R.  So, simply using it and communicating problems is
certainly of great value.

<p>One place where functionality is still missing is the modeling software
as described in "Statistical Models in S" (see <a href="#What%20is%20S%3f">What is S?</a>);
Generalized Additive Models (<strong>gam</strong>) and some of the nonlinear
modeling code are not there yet.

<p>See also the <code>PROJECTS</code> file in the top level R source directory.

<p>Many (more) of the packages available at the Statlib S Repository might
be worth porting to R.

<p>If you are interested in working on any of these projects, please notify
<a href="mailto:Kurt.Hornik@r-project.org">Kurt Hornik</a>.

<p><hr>
Node:<a name="R%20and%20Emacs">R and Emacs</a>,
Next:<a rel=next href="#R%20Miscellania">R Miscellania</a>,
Previous:<a rel=previous href="#R%20Add-On%20Packages">R Add-On Packages</a>,
Up:<a rel=up href="#Top">Top</a>
<br>

<h1>6 R and Emacs</h1>

<ul>
<li><a href="#Is%20there%20Emacs%20support%20for%20R%3f">Is there Emacs support for R?</a>: 
<li><a href="#Should%20I%20run%20R%20from%20within%20Emacs%3f">Should I run R from within Emacs?</a>: 
</ul>

<p><hr>
Node:<a name="Is%20there%20Emacs%20support%20for%20R%3f">Is there Emacs support for R?</a>,
Next:<a rel=next href="#Should%20I%20run%20R%20from%20within%20Emacs%3f">Should I run R from within Emacs?</a>,
Previous:<a rel=previous href="#R%20and%20Emacs">R and Emacs</a>,
Up:<a rel=up href="#R%20and%20Emacs">R and Emacs</a>
<br>

<h2>6.1 Is there Emacs support for R?</h2>

<p>There is an Emacs package which provides a standard interface between
statistical programs and statistical processes called ESS ("Emacs
Speaks Statistics").  It is intended to provide assistance for
interactive statistical programming and data analysis.  Languages
supported include: S dialects (S 3/4, S-PLUS 3.x/4.x/5.x, and R),
LispStat dialects (XLispStat, ViSta), SAS, Stata, SPSS dialects (SPSS,
Fiasco) and SCA.

<p>ESS grew out of the desire for bug fixes and extensions to S-mode 4.8
(which was a GNU Emacs interface to S/S-PLUS version 3 only).  The
current set of developers desired support for XEmacs, R, S4, and MS
Windows.  In addition, with new modes being developed for R, Stata, and
SAS, it was felt a unifying interface and framework for the user
interface, would benefit both the user and the developer, by helping
both groups conform to standard Emacs usage.  The end result is an
increase in efficiency for statistical programming and data analysis,
over the usual tools.

<p>R support contains code for editing R source code (syntactic indentation
and highlighting of source code, partial evaluations of code, loading
and error-checking of code, and source code revision maintenance) and
documentation (syntactic indentation and highlighting of source code,
sending Examples to running ESS process, and previewing), interacting
with an inferior R process from within Emacs (command-line editing,
searchable command history, command-line completion of R object and file
names, quick access to object and search lists, transcript recording,
and an interface to the help system), and transcript manipulation
(recording and saving transcript files, manipulating and editing saved
transcripts, and re-evaluating commands from transcript files).

<p>The latest versions of ESS are available from
<a href="http://ess.stat.wisc.edu/pub/ESS/">http://ess.stat.wisc.edu/pub/ESS/</a> or
<a href="ftp://ess.stat.wisc.edu/pub/ESS/">ftp://ess.stat.wisc.edu/pub/ESS/</a>, or via CRAN.  The HTML version
of the documentation can be found at <a href="http://stat.ethz.ch/ESS/">http://stat.ethz.ch/ESS/</a>.

<p>ESS comes with detailed installation instructions.

<p>For help with ESS, send email to <a href="mailto:ESS-help@stat.ethz.ch">ESS-help@stat.ethz.ch</a>.

<p>Please send bug reports and suggestions on ESS to
<a href="mailto:ESS-bugs@stat.math.ethz.ch">ESS-bugs@stat.math.ethz.ch</a>.  The easiest way to do this from is
within Emacs by typing <kbd>M-x ess-submit-bug-report</kbd> or using the
[ESS] or [iESS] pulldown menus.

<p><hr>
Node:<a name="Should%20I%20run%20R%20from%20within%20Emacs%3f">Should I run R from within Emacs?</a>,
Previous:<a rel=previous href="#Is%20there%20Emacs%20support%20for%20R%3f">Is there Emacs support for R?</a>,
Up:<a rel=up href="#R%20and%20Emacs">R and Emacs</a>
<br>

<h2>6.2 Should I run R from within Emacs?</h2>

<p>Yes, <em>definitely</em>.  Inferior R mode provides a readline/history
mechanism, object name completion, and syntax-based highlighting of the
interaction buffer using Font Lock mode, as well as a very convenient
interface to the R help system.

<p>Of course, it also integrates nicely with the mechanisms for editing R
source using Emacs.  One can write code in one Emacs buffer and send
whole or parts of it for execution to R; this is helpful for both data
analysis and programming.  One can also seamlessly integrate with a
revision control system, in order to maintain a log of changes in your
programs and data, as well as to allow for the retrieval of past
versions of the code.

<p>In addition, it allows you to keep a record of your session, which can
also be used for error recovery through the use of the transcript mode.

<p>To specify command line arguments for the inferior R process, use
<kbd>C-u M-x R</kbd> for starting R.  This prompts you for the arguments; in
particular, you can increase the memory size this way (see <a href="#Why%20does%20R%20run%20out%20of%20memory%3f">Why does R run out of memory?</a>).

<p><hr>
Node:<a name="R%20Miscellania">R Miscellania</a>,
Next:<a rel=next href="#R%20Programming">R Programming</a>,
Previous:<a rel=previous href="#R%20and%20Emacs">R and Emacs</a>,
Up:<a rel=up href="#Top">Top</a>
<br>

<h1>7 R Miscellania</h1>

<ul>
<li><a href="#Why%20does%20R%20run%20out%20of%20memory%3f">Why does R run out of memory?</a>: 
<li><a href="#Why%20does%20sourcing%20a%20correct%20file%20fail%3f">Why does sourcing a correct file fail?</a>: 
<li><a href="#How%20can%20I%20set%20components%20of%20a%20list%20to%20NULL%3f">How can I set components of a list to NULL?</a>: 
<li><a href="#How%20can%20I%20save%20my%20workspace%3f">How can I save my workspace?</a>: 
<li><a href="#How%20can%20I%20clean%20up%20my%20workspace%3f">How can I clean up my workspace?</a>: 
<li><a href="#How%20can%20I%20get%20eval()%20and%20D()%20to%20work%3f">How can I get eval() and D() to work?</a>: 
<li><a href="#Why%20do%20my%20matrices%20lose%20dimensions%3f">Why do my matrices lose dimensions?</a>: 
<li><a href="#How%20does%20autoloading%20work%3f">How does autoloading work?</a>: 
<li><a href="#How%20should%20I%20set%20options%3f">How should I set options?</a>: 
<li><a href="#How%20do%20file%20names%20work%20in%20Windows%3f">How do file names work in Windows?</a>: 
<li><a href="#Why%20does%20plotting%20give%20a%20color%20allocation%20error%3f">Why does plotting give a color allocation error?</a>: 
<li><a href="#Is%20R%20Y2K-compliant%3f">Is R Y2K-compliant?</a>: 
</ul>

<p><hr>
Node:<a name="Why%20does%20R%20run%20out%20of%20memory%3f">Why does R run out of memory?</a>,
Next:<a rel=next href="#Why%20does%20sourcing%20a%20correct%20file%20fail%3f">Why does sourcing a correct file fail?</a>,
Previous:<a rel=previous href="#R%20Miscellania">R Miscellania</a>,
Up:<a rel=up href="#R%20Miscellania">R Miscellania</a>
<br>

<h2>7.1 Why does R run out of memory?</h2>

<p>R (currently) uses a <em>static</em> memory model.  This means that when
it starts up, it asks the operating system to reserve a fixed amount of
memory for it.  The size of this chunk cannot be changed subsequently. 
Hence, it can happen that not enough memory was allocated, e.g., when
trying to read large data sets into R.

<p>In these cases, you should restart R with more memory available, using
the command line options <code>--nsize</code> and <code>--vsize</code>.  To
understand these options, one needs to know that R maintains separate
areas for fixed and variable sized objects.  The first of these is
allocated as an array of "cons cells" (Lisp programmers will know what
they are, others may think of them as the building blocks of the
language itself, parse trees, etc.), and the second are thrown on a
"heap".  The <code>--nsize</code> option can be used to specify the number
of cons cells which R is to use (the default is 250000), and the
<code>--vsize</code> option to specify the size of the vector heap in bytes
(the default is 6&nbsp;MB).  Boths options must either be integers or
integers ending with <code>M</code>, <code>K</code>, or <code>k</code> meaning `Mega'
(2^20), (computer) `Kilo' (2^10), or regular `kilo' (1000).

<p>E.g., to read in a table of 5000 observations on 40 numeric variables,
<code>R --vsize 6M</code> should do (which currently is the default).

<p>Note that the information on where to find vectors and strings on the
heap is stored using cons cells.  Thus, it may also be necessary to
allocate more space for cons cells in order to perform computations with
very "large" variable-size objects.

<p>You can find out the current memory consumption (the proportion of heap
and cons cells used) by typing <kbd>gc()</kbd> at the R prompt.  This may
help you in finding out whether to increase <code>--vsize</code> or
<code>--nsize</code>.  Note that following <kbd>gcinfo(TRUE)</kbd>, automatic
garbage collection always prints memory use statistics.

<p>As of version 0.62.3, R will tell you whether you ran out of cons or
heap memory.

<p>The defaults for <code>--nsize</code> and <code>--vsize</code> can be changed by
setting the environment variables <code>R_NSIZE</code> and <code>R_VSIZE</code>
respectively, perhaps most conveniently on Unix in the R environment
file (<code>~/.Renviron</code> by default).

<p>When using <code>read.table()</code>, the memory requirements are in fact
higher than anticipated, because the file is first read in as one long
string which is then split again.  Use <code>scan()</code> if possible in
case you run out of memory when reading in a large table.

<p><hr>
Node:<a name="Why%20does%20sourcing%20a%20correct%20file%20fail%3f">Why does sourcing a correct file fail?</a>,
Next:<a rel=next href="#How%20can%20I%20set%20components%20of%20a%20list%20to%20NULL%3f">How can I set components of a list to NULL?</a>,
Previous:<a rel=previous href="#Why%20does%20R%20run%20out%20of%20memory%3f">Why does R run out of memory?</a>,
Up:<a rel=up href="#R%20Miscellania">R Miscellania</a>
<br>

<h2>7.2 Why does sourcing a correct file fail?</h2>

<p>R sometimes has problems parsing a file which does not end in a newline. 
This can happen for example when Emacs is used for editing the file and
<code>next-line-add-newlines</code> is set to <code>nil</code>.  To avoid the
problem, either set <code>require-final-newline</code> to a non-<code>nil</code>
value in one of your Emacs startup files, or make sure R-mode (see <a href="#Is%20there%20Emacs%20support%20for%20R%3f">Is there Emacs support for R?</a>) is used for editing R source files (which
locally ensures this setting).

<p>Earlier R versions had a similar problem when reading in data files, but
this should have been taken care of now.

<p><hr>
Node:<a name="How%20can%20I%20set%20components%20of%20a%20list%20to%20NULL%3f">How can I set components of a list to NULL?</a>,
Next:<a rel=next href="#How%20can%20I%20save%20my%20workspace%3f">How can I save my workspace?</a>,
Previous:<a rel=previous href="#Why%20does%20sourcing%20a%20correct%20file%20fail%3f">Why does sourcing a correct file fail?</a>,
Up:<a rel=up href="#R%20Miscellania">R Miscellania</a>
<br>

<h2>7.3 How can I set components of a list to NULL?</h2>

<p>You can use

<pre>x[i] &lt;- list(NULL)
</pre>

<p>to set component <code>i</code> of the list <code>x</code> to <code>NULL</code>, similarly
for named components.  Do not set <code>x[i]</code> or <code>x[[i]]</code> to
<code>NULL</code>, because this will remove the corresponding component from
the list.

<p>For dropping the row names of a matrix <code>x</code>, it may be easier to use
<code>rownames(x) &lt;- NULL</code>, similarly for column names.

<p><hr>
Node:<a name="How%20can%20I%20save%20my%20workspace%3f">How can I save my workspace?</a>,
Next:<a rel=next href="#How%20can%20I%20clean%20up%20my%20workspace%3f">How can I clean up my workspace?</a>,
Previous:<a rel=previous href="#How%20can%20I%20set%20components%20of%20a%20list%20to%20NULL%3f">How can I set components of a list to NULL?</a>,
Up:<a rel=up href="#R%20Miscellania">R Miscellania</a>
<br>

<h2>7.4 How can I save my workspace?</h2>

<p><code>save.image()</code> saves the objects in the user's <code>.GlobalEnv</code> to
the file <code>.RData</code> in the R startup directory.  (This is also what
happens after <kbd>q("yes")</kbd>.)  Using <code>save.image(<var>file</var>)</code> one
can save the image under a different name.

<p><hr>
Node:<a name="How%20can%20I%20clean%20up%20my%20workspace%3f">How can I clean up my workspace?</a>,
Next:<a rel=next href="#How%20can%20I%20get%20eval()%20and%20D()%20to%20work%3f">How can I get eval() and D() to work?</a>,
Previous:<a rel=previous href="#How%20can%20I%20save%20my%20workspace%3f">How can I save my workspace?</a>,
Up:<a rel=up href="#R%20Miscellania">R Miscellania</a>
<br>

<h2>7.5 How can I clean up my workspace?</h2>

<p>To remove all objects in the currently active environment (typically
<code>.GlobalEnv</code>), you can do

<pre>rm(list = ls(all = TRUE))
</pre>

<p>(Without <code>all = TRUE</code>, only the objects with names not starting
with a `.' are removed.)

<p><hr>
Node:<a name="How%20can%20I%20get%20eval()%20and%20D()%20to%20work%3f">How can I get eval() and D() to work?</a>,
Next:<a rel=next href="#Why%20do%20my%20matrices%20lose%20dimensions%3f">Why do my matrices lose dimensions?</a>,
Previous:<a rel=previous href="#How%20can%20I%20clean%20up%20my%20workspace%3f">How can I clean up my workspace?</a>,
Up:<a rel=up href="#R%20Miscellania">R Miscellania</a>
<br>

<h2>7.6 How can I get eval() and D() to work?</h2>

<p>Strange things will happen if you use <code>eval(print(x), envir = e)</code>
or <code>D(x^2, "x")</code>.  The first one will either tell you that
"<code>x</code>" is not found, or print the value of the wrong <code>x</code>. 
The other one will likely return zero if <code>x</code> exists, and an error
otherwise.

<p>This is because in both cases, the first argument is evaluated in the
calling environment first.  The result (which should be an object of
mode `expression' or `call') is then evaluated or differentiated.  What
you (most likely) really want is obtained by "quoting" the first
argument upon surrounding it with <code>expression()</code>.  For example,

<pre>R&gt; D(expression(x^2), "x")
2 * x
</pre>

<p>Although this behavior may initially seem to be rather strange, is
perfectly logical.  The "intuitive" behavior could easily be
implemented, but problems would arise whenever the expression is
contained in a variable, passed as a parameter, or is the result of a
function call.  Consider for instance the semantics in cases like

<pre>D2 &lt;- function(e, n) D(D(e, n), n)
</pre>

<p>or

<pre>g &lt;- function(y) eval(substitute(y), sys.frame(sys.parent(n = 2)))
g(a * b)
</pre>

<p>See the help pages for more examples.

<p><hr>
Node:<a name="Why%20do%20my%20matrices%20lose%20dimensions%3f">Why do my matrices lose dimensions?</a>,
Next:<a rel=next href="#How%20does%20autoloading%20work%3f">How does autoloading work?</a>,
Previous:<a rel=previous href="#How%20can%20I%20get%20eval()%20and%20D()%20to%20work%3f">How can I get eval() and D() to work?</a>,
Up:<a rel=up href="#R%20Miscellania">R Miscellania</a>
<br>

<h2>7.7 Why do my matrices lose dimensions?</h2>

<p>When a matrix with a single row or column is created by a subscripting
operation, e.g., <code>row &lt;- mat[2, ]</code>, it is by default turned into a
vector.  In a similar way if an array with dimension, say, 2&nbsp;x&nbsp;3&nbsp;x&nbsp;1&nbsp;x&nbsp;4 is created by subscripting it will be coerced into a 2&nbsp;x&nbsp;3&nbsp;x&nbsp;4
array, losing the unnecessary dimension.  After much discussion this has
been determined to be a <em>feature</em>.

<p>To prevent this happening, add the option <code>drop = FALSE</code> to the
subscripting. For example,

<pre>rowmatrix &lt;- mat[2, , drop = FALSE]  # creates a row matrix
colmatrix &lt;- mat[, 2, drop = FALSE]  # creates a column matrix
a &lt;- b[1, 1, 1, drop = FALSE]        # creates a 1 x 1 x 1 array
</pre>

<p>The <code>drop = FALSE</code> option should be used defensively when
programming.  For example, the statement

<pre>somerows &lt;- mat[index, ]
</pre>

<p>will return a vector rather than a matrix if <code>index</code> happens to
have length 1, causing errors later in the code.  It should probably be
rewritten as

<pre>somerows &lt;- mat[index, , drop = FALSE]
</pre>

<p><hr>
Node:<a name="How%20does%20autoloading%20work%3f">How does autoloading work?</a>,
Next:<a rel=next href="#How%20should%20I%20set%20options%3f">How should I set options?</a>,
Previous:<a rel=previous href="#Why%20do%20my%20matrices%20lose%20dimensions%3f">Why do my matrices lose dimensions?</a>,
Up:<a rel=up href="#R%20Miscellania">R Miscellania</a>
<br>

<h2>7.8 How does autoloading work?</h2>

<p>R has a special environment called <code>.AutoloadEnv</code>.  Using
<kbd>autoload(<var>name</var>, <var>pkg</var>)</kbd>, where <var>name</var> and
<var>pkg</var> are strings giving the names of an object and the package
containing it, stores some information in this environment.  When R
tries to evaluate <var>name</var>, it loads the corresponding package
<var>pkg</var> and reevaluates <var>name</var> in the new package's
environment.

<p>Using this mechanism makes R behave as if the package was loaded, but
does not occupy memory (yet).

<p>See the help page for <code>autoload()</code> for a very nice example.

<p><hr>
Node:<a name="How%20should%20I%20set%20options%3f">How should I set options?</a>,
Next:<a rel=next href="#How%20do%20file%20names%20work%20in%20Windows%3f">How do file names work in Windows?</a>,
Previous:<a rel=previous href="#How%20does%20autoloading%20work%3f">How does autoloading work?</a>,
Up:<a rel=up href="#R%20Miscellania">R Miscellania</a>
<br>

<h2>7.9 How should I set options?</h2>

<p>The function <code>options()</code> allows setting and examining a variety of
global "options" which affect the way in which R computes and displays
its results.  The variable <code>.Options</code> holds the current values of
these options, but should never directly be assigned to unless you want
to drive yourself crazy--simply pretend that it is a "read-only"
variable.

<p>For example, given

<pre>test1 &lt;- function(x = pi, dig = 3) {
  oo &lt;- options(digits = dig); on.exit(options(oo));
  cat(.Options$digits, x, "\n")
}
test2 &lt;- function(x = pi, dig = 3) {
  .Options$digits &lt;- dig
  cat(.Options$digits, x, "\n")
}
</pre>

<p>we obtain:

<pre>R&gt; test1()
3 3.14
R&gt; test2()
3 3.141593
</pre>

<p>What is really used is the <em>global</em> value of <code>.Options</code>, and
using <kbd>options(OPT = VAL)</kbd> correctly updates it.  Local copies of
<code>.Options</code>, either in <code>.GlobalEnv</code> or in a function
environment (frame), are just silently disregarded.

<p><hr>
Node:<a name="How%20do%20file%20names%20work%20in%20Windows%3f">How do file names work in Windows?</a>,
Next:<a rel=next href="#Why%20does%20plotting%20give%20a%20color%20allocation%20error%3f">Why does plotting give a color allocation error?</a>,
Previous:<a rel=previous href="#How%20should%20I%20set%20options%3f">How should I set options?</a>,
Up:<a rel=up href="#R%20Miscellania">R Miscellania</a>
<br>

<h2>7.10 How do file names work in Windows?</h2>

<p>As R uses C-style string handling, <code>\</code> is treated as an escape
character, so that for example one can enter a newline as <code>\n</code>. 
When you really need a <code>\</code>, you have to escape it with another
<code>\</code>.

<p>Thus, in filenames use something like <code>"c:\\data\\money.dat"</code>.  You
can also replace <code>\</code> by <code>/</code> (<code>"c:/data/money.dat"</code>).

<p><hr>
Node:<a name="Why%20does%20plotting%20give%20a%20color%20allocation%20error%3f">Why does plotting give a color allocation error?</a>,
Next:<a rel=next href="#Is%20R%20Y2K-compliant%3f">Is R Y2K-compliant?</a>,
Previous:<a rel=previous href="#How%20do%20file%20names%20work%20in%20Windows%3f">How do file names work in Windows?</a>,
Up:<a rel=up href="#R%20Miscellania">R Miscellania</a>
<br>

<h2>7.11 Why does plotting give a color allocation error?</h2>

<p>Sometimes plotting, e.g., when running <code>demo(image)</code>, results in
"Error: color allocation error".  This is an X problem, and only
indirectly related to R.  It occurs when applications started prior to R
have used all the available colors.  (How many colors are available
depends on the X configuration; sometimes only 256 colors can be used.)

<p>One application which is notorious for "eating" colors is Netscape. 
If the problem occurs when Netscape is running, try (re)starting it with
either the <code>-no-install</code> (to use the default colormap) or the
<code>-install</code> (to install a private colormap) option.

<p>You could also set the <code>colortype</code> of <code>X11()</code> to `pseudo.cube'
rather than the default `pseudo'.  See the help page for <code>X11()</code>
for more information.

<p><hr>
Node:<a name="Is%20R%20Y2K-compliant%3f">Is R Y2K-compliant?</a>,
Previous:<a rel=previous href="#Why%20does%20plotting%20give%20a%20color%20allocation%20error%3f">Why does plotting give a color allocation error?</a>,
Up:<a rel=up href="#R%20Miscellania">R Miscellania</a>
<br>

<h2>7.12 Is R Y2K-compliant?</h2>

<p>We expect R to be Y2K compliant when compiled and run on a Y2K compliant
system.  In particular R does not internally represent or manipulate
dates as two-digit quantities.  However, no guarantee of Y2K compliance
is provided for R.  R is free software and comes with <em>no warranty
whatsover</em>.

<p>R, like any other programming language, can be used to write programs
and manipulate data in ways that are not Y2K compliant.

<p><hr>
Node:<a name="R%20Programming">R Programming</a>,
Next:<a rel=next href="#R%20Bugs">R Bugs</a>,
Previous:<a rel=previous href="#R%20Miscellania">R Miscellania</a>,
Up:<a rel=up href="#Top">Top</a>
<br>

<h1>8 R Programming</h1>

<ul>
<li><a href="#How%20should%20I%20write%20summary%20methods%3f">How should I write summary methods?</a>: 
<li><a href="#How%20can%20I%20debug%20dynamically%20loaded%20code%3f">How can I debug dynamically loaded code?</a>: 
<li><a href="#How%20can%20I%20inspect%20R%20objects%20when%20debugging%3f">How can I inspect R objects when debugging?</a>: 
</ul>

<p><hr>
Node:<a name="How%20should%20I%20write%20summary%20methods%3f">How should I write summary methods?</a>,
Next:<a rel=next href="#How%20can%20I%20debug%20dynamically%20loaded%20code%3f">How can I debug dynamically loaded code?</a>,
Previous:<a rel=previous href="#R%20Programming">R Programming</a>,
Up:<a rel=up href="#R%20Programming">R Programming</a>
<br>

<h2>8.1 How should I write summary methods?</h2>

<p>Suppose you want to provide a summary method for class <code>foo</code>.  Then
<code>summary.foo()</code> should not print anything, but return an object of
class <code>summary.foo</code>, <em>and</em> you should write a method
<code>print.summary.foo()</code> which nicely prints the summary information
and invisibly returns its object.  This approach is preferred over
having <code>summary.foo()</code> print summary information and return
something useful, as sometimes you need to grab something computed by
<code>summary()</code> inside a function or similar.  In such cases you don't
want anything printed.

<p><hr>
Node:<a name="How%20can%20I%20debug%20dynamically%20loaded%20code%3f">How can I debug dynamically loaded code?</a>,
Next:<a rel=next href="#How%20can%20I%20inspect%20R%20objects%20when%20debugging%3f">How can I inspect R objects when debugging?</a>,
Previous:<a rel=previous href="#How%20should%20I%20write%20summary%20methods%3f">How should I write summary methods?</a>,
Up:<a rel=up href="#R%20Programming">R Programming</a>
<br>

<h2>8.2 How can I debug dynamically loaded code?</h2>

<p>According to Doug Bates, the secret of symbolic debugging of dynamically
loaded code is to

<ul>
<li>Call the debugger on the R executable.  In the shell, you can e.g. use
<kbd>R -d gdb</kbd>; see below for debugging from within Emacs. 
<li>Start the R program. 
<li>At the R prompt, use <code>dyn.load()</code> to load your library. 
<li>Send an interrupt signal.  Inside of GUD mode in Emacs, you send
<kbd>C-c C-c</kbd>.  This will put you back to the debugger prompt. 
<li>Set the breakpoints in your code. 
<li>Continue execution of R by typing <kbd>signal 0&lt;RET&gt;</kbd>. 
</ul>

<p>When using GUD mode for debugging from within Emacs, you may find it
most convenient to use the directory with your code in it as the current
working directory and then make a symbolic link from that directory to
the R executable.  That way <code>.gdbinit</code> can stay in the directory
with the code and be used to set up the environment and the search paths
for the source, e.g. as follows:

<pre>set env R_HOME /opt/R
set env R_PAPERSIZE letter
set env R_PRINTCMD lpr
dir /opt/R/src/appl
dir /opt/R/src/main
dir /opt/R/src/nmath
dir /opt/R/src/unix
</pre>

<p><hr>
Node:<a name="How%20can%20I%20inspect%20R%20objects%20when%20debugging%3f">How can I inspect R objects when debugging?</a>,
Previous:<a rel=previous href="#How%20can%20I%20debug%20dynamically%20loaded%20code%3f">How can I debug dynamically loaded code?</a>,
Up:<a rel=up href="#R%20Programming">R Programming</a>
<br>

<h2>8.3 How can I inspect R objects when debugging?</h2>

<p>In the C implementation underlying R, all objects are so-called SEXPs
(from Lisp's "S-expressions" which comes from "symbolic
expression"), which are pointers to data structures called SEXPRECs. 
(See the file <code>src/include/Rinternals.h</code> in the R sources for the
definition of the SEXPREC type.)  For example, let

<pre>R&gt; DF &lt;- data.frame(a = 1:3, b = 4:6)
</pre>

<p>By setting a breakpoint at <code>do_get</code> and typing <kbd>get("DF")</kbd> at
the R prompt, one can find out the address in memory of <code>DF</code>, e.g.

<pre>Value returned is $1 = (SEXPREC *) 0x40583e1c
(gdb) p *$1
$2 = {
  sxpinfo = {type = 19, obj = 1, named = 1, gp = 0,
    mark = 0, debug = 0, trace = 0, = 0},
  attrib = 0x40583e80,
  u = {
    vecsxp = {
      length = 2,
      type = {c = 0x40634700 "0&gt;X@D&gt;X@0&gt;X@", i = 0x40634700,
        f = 0x40634700, z = 0x40634700, s = 0x40634700},
      truelength = 1075851272,
    },
    primsxp = {offset = 2},
    symsxp = {pname = 0x2, value = 0x40634700, internal = 0x40203008},
    listsxp = {carval = 0x2, cdrval = 0x40634700, tagval = 0x40203008},
    envsxp = {frame = 0x2, enclos = 0x40634700},
    closxp = {formals = 0x2, body = 0x40634700, env = 0x40203008},
    promsxp = {value = 0x2, expr = 0x40634700, env = 0x40203008}
  }
}
</pre>

<p>(Debugger output reformatted for better legibility).

<p>Using <code>PrintValue()</code> one can "inspect" the values of the various
elements of the SEXP, e.g.,

<pre>(gdb) p PrintValue($1-&gt;attrib)
$names
[1] "a" "b"

$row.names
[1] "1" "2" "3"

$class
[1] "data.frame"

$3 = void
</pre>

<p>(Make sure, however, to use <code>PrintValue()</code> on SEXPs with the `obj'
bit turned on only.)

<p>To find out where exactly the corresponding information is stored, one
needs to go "deeper":

<pre>(gdb) set $a = $1-&gt;attrib
(gdb) p $a-&gt;u.listsxp.tagval-&gt;u.symsxp.pname-&gt;u.vecsxp.type.c
$4 = 0x405d40e8 "names"
(gdb) p $a-&gt;u.listsxp.carval-&gt;u.vecsxp.type.s[1]-&gt;u.vecsxp.type.c
$5 = 0x40634378 "b"
(gdb) p $1-&gt;u.vecsxp.type.s[0]-&gt;u.vecsxp.type.i[0]
$6 = 1
(gdb) p $1-&gt;u.vecsxp.type.s[1]-&gt;u.vecsxp.type.i[1]
$7 = 5
</pre>

<p><hr>
Node:<a name="R%20Bugs">R Bugs</a>,
Next:<a rel=next href="#Acknowledgments">Acknowledgments</a>,
Previous:<a rel=previous href="#R%20Programming">R Programming</a>,
Up:<a rel=up href="#Top">Top</a>
<br>

<h1>9 R Bugs</h1>

<ul>
<li><a href="#What%20is%20a%20bug%3f">What is a bug?</a>: 
<li><a href="#How%20to%20report%20a%20bug">How to report a bug</a>: 
</ul>

<p><hr>
Node:<a name="What%20is%20a%20bug%3f">What is a bug?</a>,
Next:<a rel=next href="#How%20to%20report%20a%20bug">How to report a bug</a>,
Previous:<a rel=previous href="#R%20Bugs">R Bugs</a>,
Up:<a rel=up href="#R%20Bugs">R Bugs</a>
<br>

<h2>9.1 What is a bug?</h2>

<p>If R executes an illegal instruction, or dies with an operating system
error message that indicates a problem in the program (as opposed to
something like "disk full"), then it is certainly a bug.  If you call
<code>.Internal()</code>, <code>.External()</code>, <code>.C()</code> or <code>.Fortran()</code>
yourself (or in a function you wrote), you can always crash R by using
wrong argument types (modes).  This is not a bug.

<p>Taking forever to complete a command can be a bug, but you must make
certain that it was really R's fault.  Some commands simply take a long
time.  If the input was such that you <em>know</em> it should have been
processed quickly, report a bug.  If you don't know whether the command
should take a long time, find out by looking in the manual or by asking
for assistance.

<p>If a command you are familiar with causes an R error message in a case
where its usual definition ought to be reasonable, it is probably a
bug. If a command does the wrong thing, that is a bug.  But be sure you
know for certain what it ought to have done.  If you aren't familiar
with the command, or don't know for certain how the command is supposed
to work, then it might actually be working right.  Rather than jumping
to conclusions, show the problem to someone who knows for certain.

<p>Finally, a command's intended definition may not be best for statistical
analysis.  This is a very important sort of problem, but it is also a
matter of judgment.  Also, it is easy to come to such a conclusion out
of ignorance of some of the existing features.  It is probably best not
to complain about such a problem until you have checked the
documentation in the usual ways, feel confident that you understand it,
and know for certain that what you want is not available.  If you are
not sure what the command is supposed to do after a careful reading of
the manual this indicates a bug in the manual.  The manual's job is to
make everything clear.  It is just as important to report documentation
bugs as program bugs.  However, we know that the introductory
documentation is seriously inadequate, so you don't need to report this.

<p>If the online argument list of a function disagrees with the manual, one
of them must be wrong, so report the bug.

<p><hr>
Node:<a name="How%20to%20report%20a%20bug">How to report a bug</a>,
Previous:<a rel=previous href="#What%20is%20a%20bug%3f">What is a bug?</a>,
Up:<a rel=up href="#R%20Bugs">R Bugs</a>
<br>

<h2>9.2 How to report a bug</h2>

<p>When you decide that there is a bug, it is important to report it and to
report it in a way which is useful.  What is most useful is an exact
description of what commands you type, starting with the shell command
to run R, until the problem happens.  Always include the version of R,
machine, and operating system that you are using; type <kbd>version</kbd> in
R to print this.

<p>The most important principle in reporting a bug is to report
<em>facts</em>, not hypotheses or categorizations.  It is always easier to
report the facts, but people seem to prefer to strain to posit
explanations and report them instead.  If the explanations are based on
guesses about how R is implemented, they will be useless; others will
have to try to figure out what the facts must have been to lead to such
speculations.  Sometimes this is impossible.  But in any case, it is
unnecessary work for the ones trying to fix the problem.

<p>For example, suppose that on a data set which you know to be quite large
the command

<pre>R&gt; data.frame(x, y, z, monday, tuesday)
</pre>

<p>never returns. Do not report that <code>data.frame()</code> fails for large
data sets.  Perhaps it fails when a variable name is a day of the week. 
If this is so then when others got your report they would try out the
<code>data.frame()</code> command on a large data set, probably with no day of
the week variable name, and not see any problem.  There is no way in the
world that others could guess that they should try a day of the week
variable name.

<p>Or perhaps the command fails because the last command you used was a
method for <code>"["()</code> that had a bug causing R's internal data
structures to be corrupted and making the <code>data.frame()</code> command
fail from then on.  This is why others need to know what other commands
you have typed (or read from your startup file).

<p>It is very useful to try and find simple examples that produce
apparently the same bug, and somewhat useful to find simple examples
that might be expected to produce the bug but actually do not.  If you
want to debug the problem and find exactly what caused it, that is
wonderful. You should still report the facts as well as any explanations
or solutions.  Please include an example that reproduces the problem,
preferably the simplest one you have found.

<p>Invoking R with the <code>--vanilla</code> option may help in isolating a
bug. This ensures that the site profile and saved data files are not
read.

<p>On Unix systems a bug report can be generated using the function
<code>bug.report()</code>. This automatically includes the version information
and sends the bug to the correct address.  Alternatively the bug report
can be emailed to <a href="mailto:r-bugs@lists.r-project.org">r-bugs@lists.r-project.org</a> or submitted to the Web
page at <a href="http://bugs.r-project.org">http://bugs.r-project.org</a>.

<p>Bug reports on contributed packages should perhaps be sent to the
package maintainer rather than to r-bugs.

<p><hr>
Node:<a name="Acknowledgments">Acknowledgments</a>,
Previous:<a rel=previous href="#R%20Bugs">R Bugs</a>,
Up:<a rel=up href="#Top">Top</a>
<br>

<h1>10 Acknowledgments</h1>

<p>Of course, many many thanks to Robert and Ross for the R system, and to
the package writers and porters for adding to it.

<p>Special thanks go to Doug Bates, Peter Dalgaard, Paul Gilbert, Fritz
Leisch, Jim Lindsey, Thomas Lumley, Martin Maechler, Brian D. Ripley,
Anthony Rossini, and Andreas Weingessel for their comments which helped
me improve this FAQ.

<p>More to some soon <small>...</small>

</body></html>
The R Project SVN R

(root)/branches/R-exp-gnome/doc/html/faq.html – Rev 6771
