You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

182 lines
7.1 KiB
HTML

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual describes how to install and use the GNU multiple precision
arithmetic library, version 6.1.0.
Copyright 1991, 1993-2015 Free Software Foundation, Inc.
Permission is granted to copy, distribute and/or modify this document under
the terms of the GNU Free Documentation License, Version 1.3 or any later
version published by the Free Software Foundation; with no Invariant Sections,
with the Front-Cover Texts being "A GNU Manual", and with the Back-Cover
Texts being "You have freedom to copy and modify this GNU Manual, like GNU
software". A copy of the license is included in
GNU Free Documentation License. -->
<!-- Created by GNU Texinfo 6.4, http://www.gnu.org/software/texinfo/ -->
<head>
<title>C++ Interface Internals (GNU MP 6.1.0)</title>
<meta name="description" content="How to install and use the GNU multiple precision arithmetic library, version 6.1.0.">
<meta name="keywords" content="C++ Interface Internals (GNU MP 6.1.0)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link href="index.html#Top" rel="start" title="Top">
<link href="Concept-Index.html#Concept-Index" rel="index" title="Concept Index">
<link href="Internals.html#Internals" rel="up" title="Internals">
<link href="Contributors.html#Contributors" rel="next" title="Contributors">
<link href="Raw-Output-Internals.html#Raw-Output-Internals" rel="prev" title="Raw Output Internals">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
blockquote.smallindentedblock {margin-right: 0em; font-size: smaller}
blockquote.smallquotation {font-size: smaller}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
div.smalldisplay {margin-left: 3.2em}
div.smallexample {margin-left: 3.2em}
div.smalllisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
pre.smalldisplay {font-family: inherit; font-size: smaller}
pre.smallexample {font-size: smaller}
pre.smallformat {font-family: inherit; font-size: smaller}
pre.smalllisp {font-size: smaller}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<a name="C_002b_002b-Interface-Internals"></a>
<div class="header">
<p>
Previous: <a href="Raw-Output-Internals.html#Raw-Output-Internals" accesskey="p" rel="prev">Raw Output Internals</a>, Up: <a href="Internals.html#Internals" accesskey="u" rel="up">Internals</a> &nbsp; [<a href="Concept-Index.html#Concept-Index" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<a name="C_002b_002b-Interface-Internals-1"></a>
<h3 class="section">16.5 C++ Interface Internals</h3>
<a name="index-C_002b_002b-interface-internals"></a>
<p>A system of expression templates is used to ensure something like <code>a=b+c</code>
turns into a simple call to <code>mpz_add</code> etc. For <code>mpf_class</code>
the scheme also ensures the precision of the final
destination is used for any temporaries within a statement like
<code>f=w*x+y*z</code>. These are important features which a naive implementation
cannot provide.
</p>
<p>A simplified description of the scheme follows. The true scheme is
complicated by the fact that expressions have different return types. For
detailed information, refer to the source code.
</p>
<p>To perform an operation, say, addition, we first define a &ldquo;function object&rdquo;
evaluating it,
</p>
<div class="example">
<pre class="example">struct __gmp_binary_plus
{
static void eval(mpf_t f, const mpf_t g, const mpf_t h)
{
mpf_add(f, g, h);
}
};
</pre></div>
<p>And an &ldquo;additive expression&rdquo; object,
</p>
<div class="example">
<pre class="example">__gmp_expr&lt;__gmp_binary_expr&lt;mpf_class, mpf_class, __gmp_binary_plus&gt; &gt;
operator+(const mpf_class &amp;f, const mpf_class &amp;g)
{
return __gmp_expr
&lt;__gmp_binary_expr&lt;mpf_class, mpf_class, __gmp_binary_plus&gt; &gt;(f, g);
}
</pre></div>
<p>The seemingly redundant <code>__gmp_expr&lt;__gmp_binary_expr&lt;&hellip;&gt;&gt;</code> is used to
encapsulate any possible kind of expression into a single template type. In
fact even <code>mpf_class</code> etc are <code>typedef</code> specializations of
<code>__gmp_expr</code>.
</p>
<p>Next we define assignment of <code>__gmp_expr</code> to <code>mpf_class</code>.
</p>
<div class="example">
<pre class="example">template &lt;class T&gt;
mpf_class &amp; mpf_class::operator=(const __gmp_expr&lt;T&gt; &amp;expr)
{
expr.eval(this-&gt;get_mpf_t(), this-&gt;precision());
return *this;
}
template &lt;class Op&gt;
void __gmp_expr&lt;__gmp_binary_expr&lt;mpf_class, mpf_class, Op&gt; &gt;::eval
(mpf_t f, mp_bitcnt_t precision)
{
Op::eval(f, expr.val1.get_mpf_t(), expr.val2.get_mpf_t());
}
</pre></div>
<p>where <code>expr.val1</code> and <code>expr.val2</code> are references to the expression&rsquo;s
operands (here <code>expr</code> is the <code>__gmp_binary_expr</code> stored within the
<code>__gmp_expr</code>).
</p>
<p>This way, the expression is actually evaluated only at the time of assignment,
when the required precision (that of <code>f</code>) is known. Furthermore the
target <code>mpf_t</code> is now available, thus we can call <code>mpf_add</code> directly
with <code>f</code> as the output argument.
</p>
<p>Compound expressions are handled by defining operators taking subexpressions
as their arguments, like this:
</p>
<div class="example">
<pre class="example">template &lt;class T, class U&gt;
__gmp_expr
&lt;__gmp_binary_expr&lt;__gmp_expr&lt;T&gt;, __gmp_expr&lt;U&gt;, __gmp_binary_plus&gt; &gt;
operator+(const __gmp_expr&lt;T&gt; &amp;expr1, const __gmp_expr&lt;U&gt; &amp;expr2)
{
return __gmp_expr
&lt;__gmp_binary_expr&lt;__gmp_expr&lt;T&gt;, __gmp_expr&lt;U&gt;, __gmp_binary_plus&gt; &gt;
(expr1, expr2);
}
</pre></div>
<p>And the corresponding specializations of <code>__gmp_expr::eval</code>:
</p>
<div class="example">
<pre class="example">template &lt;class T, class U, class Op&gt;
void __gmp_expr
&lt;__gmp_binary_expr&lt;__gmp_expr&lt;T&gt;, __gmp_expr&lt;U&gt;, Op&gt; &gt;::eval
(mpf_t f, mp_bitcnt_t precision)
{
// declare two temporaries
mpf_class temp1(expr.val1, precision), temp2(expr.val2, precision);
Op::eval(f, temp1.get_mpf_t(), temp2.get_mpf_t());
}
</pre></div>
<p>The expression is thus recursively evaluated to any level of complexity and
all subexpressions are evaluated to the precision of <code>f</code>.
</p>
<hr>
<div class="header">
<p>
Previous: <a href="Raw-Output-Internals.html#Raw-Output-Internals" accesskey="p" rel="prev">Raw Output Internals</a>, Up: <a href="Internals.html#Internals" accesskey="u" rel="up">Internals</a> &nbsp; [<a href="Concept-Index.html#Concept-Index" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>