gcc/libstdc++-v3/docs/26_numerics/howto.html
Phil Edwards 58fdec72e2 configopts.html: Mention new options.
2000-08-28  Phil Edwards  <pme@sources.redhat.com>

	* docs/configopts.html:  Mention new options.
	* docs/install.html:  Formatting changes for platform-specific
	  pre-reqs, previously only Cygwin.
	* docs/mail.html:  Fix new link.
	* docs/thanks.html:  A couple more people.
	* docs/26_numerics/howto.html:  Fix typo.

From-SVN: r36018
2000-08-28 18:41:24 +00:00

144 lines
5.8 KiB
HTML

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<META NAME="AUTHOR" CONTENT="pme@sources.redhat.com (Phil Edwards)">
<META NAME="KEYWORDS" CONTENT="HOWTO, libstdc++, egcs, g++, libg++, STL">
<META NAME="DESCRIPTION" CONTENT="HOWTO for the libstdc++ chapter 26.">
<META NAME="GENERATOR" CONTENT="vi and eight fingers">
<TITLE>libstdc++-v3 HOWTO: Chapter 26</TITLE>
<LINK REL="home" HREF="http://sources.redhat.com/libstdc++/docs/26_numerics/">
<LINK REL=StyleSheet HREF="../lib3styles.css">
<!-- $Id: howto.html,v 1.4 2000/07/11 21:45:08 pme Exp $ -->
</HEAD>
<BODY>
<H1 CLASS="centered"><A NAME="top">Chapter 26: Numerics</A></H1>
<P>Chapter 26 deals with building block abstractions to aid in
numerical computing:
<UL>
<LI>Template data structures such as <TT>valarray&lt;&gt;</TT>
and <TT>complex&lt;&gt;</TT>.
<LI>Template numerical functions such as <TT>accumulate</TT>;
<TT>inner_product</TT>; <TT>partial_sum</TT> and
<TT>adjacent_difference</TT>.
</UL>
All of the Standard C math functions are of course included in C++,
and overloaded versions for <TT>long</TT>, <TT>float</TT>, and
<TT>long double</TT> have been added for all of them.
</P>
<!-- ####################################################### -->
<HR>
<H1>Contents</H1>
<UL>
<LI><A HREF="#1">Complex Number Processing</A>
<LI><A HREF="#2">Array Processing</A>
<LI><A HREF="#3">Numerical Functions</A>
</UL>
<HR>
<!-- ####################################################### -->
<H2><A NAME="1">Complex Number Processing</A></H2>
<P>Using <TT>complex&lt;&gt;</TT> becomes even more comple- er, sorry,
<EM>complicated</EM>, with the not-quite-gratuitously-incompatible
addition of complex types to the C language. David Tribble has
compiled a list of C++98 and C99 conflict points; his description of
C's new type versus those of C++ and how to get them playing together
nicely is
<A HREF="http://home.flash.net/~dtribble/text/cdiffs.htm#C99.complex">here</A>.
</P>
<P><TT>complex&lt;&gt;</TT> is intended to be instantiated with a
floating-point type. As long as you meet that and some other basic
requirements, then the resulting instantiation has all of the usual
math operators defined, as well as definitions of <TT>op&lt;&lt;</TT>
and <TT>op&gt;&gt;</TT> that work with iostreams: <TT>op&lt;&lt;</TT>
prints <TT>(u,v)</TT> and <TT>op&gt;&gt;</TT> can read <TT>u</TT>,
<TT>(u)</TT>, and <TT>(u,v)</TT>.
</P>
<P>Return <A HREF="#top">to top of page</A> or
<A HREF="../faq/index.html">to the FAQ</A>.
</P>
<HR>
<H2><A NAME="2">Array Processing</A></H2>
<P>One of the major reasons why FORTRAN can chew through numbers so well
is that it is defined to be free of pointer aliasing, an assumption
that C89 is not allowed to make, and neither is C++. C99 adds a new
keyword, <TT>restrict</TT>, to apply to individual pointers. The C++
solution is contained in the library rather than the language
(although many vendors can be expected to add this to their compilers
as an extension).
</P>
<P>That library solution is a set of two classes, five template classes,
and &quot;a whole bunch&quot; of functions. The classes are required
to be free of pointer aliasing, so compilers can optimize the
daylights out of them the same way that they have been for FORTRAN.
They are collectively called <TT>valarray</TT>, although strictly
speaking this is only one of the five template classes, and they are
designed to be familiar to people who have worked with the BLAS
libraries before.
</P>
<P>Some more stuff should go here once somebody has time to write it.
</P>
<P>Return <A HREF="#top">to top of page</A> or
<A HREF="../faq/index.html">to the FAQ</A>.
</P>
<HR>
<H2><A NAME="3">Numerical Functions</A></H2>
<P>There are four generalized functions in the &lt;numeric&gt; header
that follow the same conventions as those in &lt;algorithm&gt;. Each
of them is overloaded: one signature for common default operations,
and a second for fully general operations. Their names are
self-explanatory to anyone who works with numerics on a regular basis:
<UL>
<LI><TT>accumulate</TT>
<LI><TT>inner_product</TT>
<LI><TT>partial_sum</TT>
<LI><TT>adjacent_difference</TT>
</UL>
</P>
<P>Here is a simple example of the two forms of <TT>accumulate</TT>.
<PRE>
int ar[50];
int someval = somefunction();
// ...initialize members of ar to something...
int sum = std::accumulate(ar,ar+50,0);
int sum_stuff = std::accumulate(ar,ar+50,someval);
int product = std::accumulate(ar,ar+50,1,std::multiplies&lt;int&gt;());
</PRE>
The first call adds all the members of the array, using zero as an
initial value for <TT>sum</TT>. The second does the same, but uses
<TT>someval</TT> as the starting value (thus, <TT>sum_stuff == sum +
someval</TT>). The final call uses the second of the two signatures,
and multiplies all the members of the array; here we must obviously
use 1 as a starting value instead of 0.
</P>
<P>The other three functions have similar dual-signature forms.
</P>
<P>Return <A HREF="#top">to top of page</A> or
<A HREF="../faq/index.html">to the FAQ</A>.
</P>
<!-- ####################################################### -->
<HR>
<P CLASS="fineprint"><EM>
Comments and suggestions are welcome, and may be sent to
<A HREF="mailto:pme@sources.redhat.com">Phil Edwards</A> or
<A HREF="mailto:gdr@egcs.cygnus.com">Gabriel Dos Reis</A>.
<BR> $Id: howto.html,v 1.4 2000/07/11 21:45:08 pme Exp $
</EM></P>
</BODY>
</HTML>