2495 lines
127 KiB
HTML
2495 lines
127 KiB
HTML
<?xml version="1.0" encoding="utf-8" ?>
|
||
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html
|
||
|
||
lang="en" xmlns="http://www.w3.org/1999/xhtml" xml:lang="en">
|
||
<head>
|
||
<meta http-equiv="content-type" content="application/xhtml+xml; charset=UTF-8" />
|
||
<meta name="generator" content="Docutils 0.3.8: http://docutils.sourceforge.net/" />
|
||
<title>Appendix A - An Introduction to Preprocessor Metaprogramming</title>
|
||
<meta name="copyright" content="From "C++ Template Metaprogramming," by David Abrahams and Aleksey Gurtovoy. Copyright (c) 2005 by Pearson Education, Inc. Reprinted with permission." />
|
||
<style type="text/css">
|
||
|
||
/*
|
||
:Author: David Goodger
|
||
:Contact: goodger@users.sourceforge.net
|
||
:date: $Date: 2004/06/23 13:31:26 $
|
||
:version: $Revision: 1.37 $
|
||
:copyright: This stylesheet has been placed in the public domain.
|
||
|
||
Default cascading style sheet for the HTML output of Docutils.
|
||
*/
|
||
|
||
.first {
|
||
margin-top: 0 }
|
||
|
||
.last {
|
||
margin-bottom: 0 }
|
||
|
||
a.toc-backref {
|
||
text-decoration: none ;
|
||
color: black }
|
||
|
||
blockquote.epigraph {
|
||
margin: 2em 5em ; }
|
||
|
||
dd {
|
||
margin-bottom: 0.5em }
|
||
|
||
/* Uncomment (& remove this text!) to get bold-faced definition list terms
|
||
dt {
|
||
font-weight: bold }
|
||
*/
|
||
|
||
div.abstract {
|
||
margin: 2em 5em }
|
||
|
||
div.abstract p.topic-title {
|
||
font-weight: bold ;
|
||
text-align: center }
|
||
|
||
div.attention, div.caution, div.danger, div.error, div.hint,
|
||
div.important, div.note, div.tip, div.warning, div.admonition {
|
||
margin: 2em ;
|
||
border: medium outset ;
|
||
padding: 1em }
|
||
|
||
div.attention p.admonition-title, div.caution p.admonition-title,
|
||
div.danger p.admonition-title, div.error p.admonition-title,
|
||
div.warning p.admonition-title {
|
||
color: red ;
|
||
font-weight: bold ;
|
||
font-family: sans-serif }
|
||
|
||
div.hint p.admonition-title, div.important p.admonition-title,
|
||
div.note p.admonition-title, div.tip p.admonition-title,
|
||
div.admonition p.admonition-title {
|
||
font-weight: bold ;
|
||
font-family: sans-serif }
|
||
|
||
div.dedication {
|
||
margin: 2em 5em ;
|
||
text-align: center ;
|
||
font-style: italic }
|
||
|
||
div.dedication p.topic-title {
|
||
font-weight: bold ;
|
||
font-style: normal }
|
||
|
||
div.figure {
|
||
margin-left: 2em }
|
||
|
||
div.footer, div.header {
|
||
font-size: smaller }
|
||
|
||
div.sidebar {
|
||
margin-left: 1em ;
|
||
border: medium outset ;
|
||
padding: 0em 1em ;
|
||
background-color: #ffffee ;
|
||
width: 40% ;
|
||
float: right ;
|
||
clear: right }
|
||
|
||
div.sidebar p.rubric {
|
||
font-family: sans-serif ;
|
||
font-size: medium }
|
||
|
||
div.system-messages {
|
||
margin: 5em }
|
||
|
||
div.system-messages h1 {
|
||
color: red }
|
||
|
||
div.system-message {
|
||
border: medium outset ;
|
||
padding: 1em }
|
||
|
||
div.system-message p.system-message-title {
|
||
color: red ;
|
||
font-weight: bold }
|
||
|
||
div.topic {
|
||
margin: 2em }
|
||
|
||
h1.title {
|
||
text-align: center }
|
||
|
||
h2.subtitle {
|
||
text-align: center }
|
||
|
||
hr {
|
||
width: 75% }
|
||
|
||
ol.simple, ul.simple {
|
||
margin-bottom: 1em }
|
||
|
||
ol.arabic {
|
||
list-style: decimal }
|
||
|
||
ol.loweralpha {
|
||
list-style: lower-alpha }
|
||
|
||
ol.upperalpha {
|
||
list-style: upper-alpha }
|
||
|
||
ol.lowerroman {
|
||
list-style: lower-roman }
|
||
|
||
ol.upperroman {
|
||
list-style: upper-roman }
|
||
|
||
p.attribution {
|
||
text-align: right ;
|
||
margin-left: 50% }
|
||
|
||
p.caption {
|
||
font-style: italic }
|
||
|
||
p.credits {
|
||
font-style: italic ;
|
||
font-size: smaller }
|
||
|
||
p.label {
|
||
white-space: nowrap }
|
||
|
||
p.rubric {
|
||
font-weight: bold ;
|
||
font-size: larger ;
|
||
color: maroon ;
|
||
text-align: center }
|
||
|
||
p.sidebar-title {
|
||
font-family: sans-serif ;
|
||
font-weight: bold ;
|
||
font-size: larger }
|
||
|
||
p.sidebar-subtitle {
|
||
font-family: sans-serif ;
|
||
font-weight: bold }
|
||
|
||
p.topic-title {
|
||
font-weight: bold }
|
||
|
||
pre.address {
|
||
margin-bottom: 0 ;
|
||
margin-top: 0 ;
|
||
font-family: serif ;
|
||
font-size: 100% }
|
||
|
||
pre.line-block {
|
||
font-family: serif ;
|
||
font-size: 100% }
|
||
|
||
pre.literal-block, pre.doctest-block {
|
||
margin-left: 2em ;
|
||
margin-right: 2em ;
|
||
background-color: #eeeeee }
|
||
|
||
span.classifier {
|
||
font-family: sans-serif ;
|
||
font-style: oblique }
|
||
|
||
span.classifier-delimiter {
|
||
font-family: sans-serif ;
|
||
font-weight: bold }
|
||
|
||
span.interpreted {
|
||
font-family: sans-serif }
|
||
|
||
span.option {
|
||
white-space: nowrap }
|
||
|
||
span.option-argument {
|
||
font-style: italic }
|
||
|
||
span.pre {
|
||
white-space: pre }
|
||
|
||
span.problematic {
|
||
color: red }
|
||
|
||
table {
|
||
margin-top: 0.5em ;
|
||
margin-bottom: 0.5em }
|
||
|
||
table.citation {
|
||
border-left: solid thin gray ;
|
||
padding-left: 0.5ex }
|
||
|
||
table.docinfo {
|
||
margin: 2em 4em }
|
||
|
||
table.footnote {
|
||
border-left: solid thin black ;
|
||
padding-left: 0.5ex }
|
||
|
||
dt {
|
||
font-weight: bold }
|
||
|
||
td, th {
|
||
padding-left: 0.5em ;
|
||
padding-right: 0.5em ;
|
||
vertical-align: top }
|
||
|
||
th.docinfo-name, th.field-name {
|
||
font-weight: bold ;
|
||
text-align: left ;
|
||
white-space: nowrap }
|
||
|
||
h1 tt, h2 tt, h3 tt, h4 tt, h5 tt, h6 tt {
|
||
font-size: 100% }
|
||
|
||
tt {
|
||
background-color: #eeeeee }
|
||
|
||
ul.auto-toc {
|
||
list-style-type: none }
|
||
|
||
</style> </head>
|
||
<body><br />
|
||
<div class="document" id="preprocessor-title">
|
||
<h1 class="title">Appendix A - An Introduction to Preprocessor
|
||
Metaprogramming</h1>
|
||
<table class="docinfo" frame="void" rules="none">
|
||
<colgroup><col class="docinfo-name" /> <col class="docinfo-content" />
|
||
</colgroup>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<th class="docinfo-name">Copyright:</th>
|
||
<td>From "C++ Template Metaprogramming," by David Abrahams and
|
||
Aleksey Gurtovoy. Copyright (c) 2005 by Pearson Education, Inc.
|
||
Reprinted with permission.</td>
|
||
</tr>
|
||
<tr class="field">
|
||
<th class="docinfo-name">ISBN:</th>
|
||
<td class="field-body">0321227255</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<div class="section" id="motivation">
|
||
<h1><a name="motivation">A.1 Motivation</a></h1>
|
||
<p>Even with the full power of template metaprogramming and the <a class="reference"
|
||
|
||
href="http://www.boost.org/libs/mpl">Boost Metaprogramming library</a>
|
||
at our disposal, some C++ coding jobs still require a great deal of
|
||
boilerplate code repetition. We saw one example in Chapter 5, when we
|
||
implemented <tt class="docutils literal"><span class="pre">tiny_size</span></tt>:</p>
|
||
<pre class="literal-block">template <class T0, class T1, class T2>
|
||
struct tiny_size
|
||
: mpl::int_<3> {};
|
||
</pre>
|
||
<!-- : rst-mode hack -->
|
||
<!-- @prefix.append('struct none {};') -->
|
||
<p>Aside from the repeated pattern in the parameter list of the primary
|
||
template above, there are three partial specializations below, which
|
||
also follow a predictable pattern:</p>
|
||
<pre class="literal-block">template <class T0, class T1>
|
||
struct tiny_size<T0,T1,none>
|
||
: mpl::int_<2> {};
|
||
|
||
template <class T0>
|
||
struct tiny_size<T0,none,none>
|
||
: mpl::int_<1> {};
|
||
|
||
template <>
|
||
struct tiny_size<none,none,none>
|
||
: mpl::int_<0> {};
|
||
</pre>
|
||
<!-- : rst-mode hack -->
|
||
<!-- @compile('all') -->
|
||
<p>In this case there is only a small amount of code with such a
|
||
"mechanical" flavor, but had we been implementing <tt class="docutils literal"><span
|
||
|
||
class="pre">large</span></tt> instead of <tt class="docutils literal"><span
|
||
|
||
class="pre">tiny</span></tt>, there might easily have been a great
|
||
deal more. When the number of instances of a pattern grows beyond two
|
||
or three, writing them by hand tends to become error-prone. Perhaps
|
||
more importantly, the code gets hard to read, because the important
|
||
abstraction in the code is really the pattern, not the individual
|
||
instances.</p>
|
||
<div class="section" id="code-generation">
|
||
<h2><a name="code-generation">A.1.1 Code Generation</a></h2>
|
||
<p>Rather than being written out by hand, mechanical-looking code
|
||
should really be generated mechanically. Having written a program to
|
||
spit out instances of the code pattern, a library author has two
|
||
choices: She can either ship pre-generated source code files, or she
|
||
can ship the generator itself. Either approach has drawbacks. If
|
||
clients only get the generated source, they are stuck with whatever
|
||
the library author generated—and experience shows that if they are
|
||
happy with three instances of a pattern today, someone will need
|
||
four tomorrow. If clients get the generator program, on the other
|
||
hand, they also need the resources to execute it (e.g.,
|
||
interpreters), and they must integrate the generator into their
|
||
build processes...</p>
|
||
</div>
|
||
<div class="section" id="enter-the-preprocessor">
|
||
<h2><a name="enter-the-preprocessor">A.1.2 Enter the Preprocessor</a></h2>
|
||
<p>...unless the generator is a preprocessor metaprogram. Though not
|
||
designed for that purpose, the C and C++ preprocessors can be made
|
||
to execute sophisticated programs during the preprocessing phase of
|
||
compilation. Users can control the code generation process with
|
||
preprocessor <tt class="docutils literal"><span class="pre">#define</span></tt>s
|
||
in code or <tt class="docutils literal"><span class="pre">-D</span></tt>
|
||
options on the compiler's command line, making build integration
|
||
trivial. For example, we might parameterize the primary <tt class="docutils literal"><span
|
||
|
||
class="pre">tiny_size</span></tt> template above as follows:</p>
|
||
<pre class="literal-block">#include <<strong>boost/preprocessor/repetition/enum_params</strong>.hpp>
|
||
|
||
#ifndef TINY_MAX_SIZE
|
||
# define TINY_MAX_SIZE 3 // default maximum size is 3
|
||
#endif
|
||
|
||
template <<strong>BOOST_PP_ENUM_PARAMS(TINY_MAX_SIZE, class T)</strong>>
|
||
struct tiny_size
|
||
: mpl::int_<TINY_MAX_SIZE>
|
||
{};
|
||
</pre>
|
||
<!-- : rst-mode hack -->
|
||
<!-- @compile(pop = None) -->
|
||
<p>To test the metaprogram, run your compiler in its "preprocessing"
|
||
mode (usually the <tt class="docutils literal"><span class="pre">-E</span></tt>
|
||
option), with the Boost root directory in your <tt class="docutils literal"><span
|
||
|
||
class="pre">#include</span></tt> path. For instance:<a class="footnote-reference"
|
||
|
||
href="#minusp" id="id2" name="id2">[1]</a></p>
|
||
<pre class="literal-block">g++ -P -E -Ipath/to/boost_1_32_0 -I. test.cpp
|
||
</pre>
|
||
<!-- @ignore() -->
|
||
<table class="docutils footnote" frame="void" id="minusp" rules="none">
|
||
<colgroup><col class="label" /><col /></colgroup>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td class="label"><a class="fn-backref" href="#id2" name="minusp">[1]</a></td>
|
||
<td>GCC's <tt class="docutils literal"><span class="pre">-P</span></tt>
|
||
option inhibits the generation of source file and line number
|
||
markers in preprocessed output.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<p>Given the appropriate metaprograms, users would be able to adjust
|
||
not only the number of parameters to <tt class="docutils literal"><span
|
||
|
||
class="pre">tiny_size</span></tt>, but the maximum size of the
|
||
entire <tt class="docutils literal"><span class="pre">tiny</span></tt>
|
||
implementation just by <tt class="docutils literal"><span class="pre">#define</span></tt>-ing
|
||
<tt class="docutils literal"><span class="pre">TINY_MAX_SIZE</span></tt>.</p>
|
||
<p>The Boost Preprocessor library <a class="citation-reference" href="#mk04"
|
||
|
||
id="id3" name="id3">[MK04]</a> plays a role in preprocessor
|
||
metaprogramming similar to the one played by the MPL in template
|
||
metaprogramming: It supplies a framework of high-level components
|
||
(like <tt class="docutils literal"><span class="pre">BOOST_PP_ENUM_PARAMS</span></tt>)
|
||
that make otherwise-painful metaprogramming jobs approachable. In
|
||
this appendix we won't attempt to cover nitty-gritty details of how
|
||
the preprocessor works, nor principles of preprocessor
|
||
metaprogramming in general, nor even many details of how the
|
||
Preprocessor <em>library</em> works. We <em>will</em> show you
|
||
enough at a high level that you'll be able to use the library
|
||
productively and learn the rest on your own.</p>
|
||
<table class="docutils citation" frame="void" id="mk04" rules="none">
|
||
<colgroup><col class="label" /><col /></colgroup>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td class="label"><a class="fn-backref" href="#id3" name="mk04">[MK04]</a></td>
|
||
<td>Paul Mensonides and Vesa Karvonen. "The Boost Preprocessor
|
||
Library." <a class="reference" href="http://www.boost.org/libs/preprocessor">http://www.boost.org/libs/preprocessor</a>.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
</div>
|
||
<div class="section" id="fundamental-abstractions-of-the-preprocessor">
|
||
<h1><a name="fundamental-abstractions-of-the-preprocessor">A.2 Fundamental
|
||
Abstractions of the Preprocessor</a></h1>
|
||
<p>We began our discussion of template metaprogramming in Chapter 2 by
|
||
describing its metadata (potential template arguments) and
|
||
metafunctions (class templates). On the basis of those two fundamental
|
||
abstractions, we built up the entire picture of compile-time
|
||
computation covered in the rest of this book. In this section we'll
|
||
lay a similar foundation for the preprocessor metaprogrammer. Some of
|
||
what we cover here may be a review for you, but it's important to
|
||
identify the basic concepts going into detail.</p>
|
||
<div class="section" id="preprocessing-tokens">
|
||
<h2><a name="preprocessing-tokens">A.2.1 Preprocessing Tokens</a></h2>
|
||
<p>The fundamental unit of data in the preprocessor is the <strong>preprocessing
|
||
token</strong>. Preprocessing tokens correspond roughly to the
|
||
tokens you're used to working with in C++, such as identifiers,
|
||
operator symbols, and literals. Technically, there are some
|
||
differences between <em>preprocessing tokens</em> and regular <em>tokens</em>
|
||
(see section 2 of the C++ standard for details), but they can be
|
||
ignored for the purposes of this discussion. In fact, we'll be using
|
||
the terms interchangeably here.</p>
|
||
</div>
|
||
<div class="section" id="macros">
|
||
<h2><a name="macros">A.2.2 Macros</a></h2>
|
||
<p>Preprocessor macros come in two flavors. <strong>Object-like
|
||
macros</strong> can be defined this way:</p>
|
||
<blockquote>
|
||
<div class="line-block">
|
||
<div class="line"><tt class="docutils literal"><span class="pre">#define</span></tt>
|
||
<em>identifier</em> <em>replacement-list</em></div>
|
||
</div>
|
||
</blockquote>
|
||
<!-- @litre_translator.line_offset -= 7 -->
|
||
<p>where the <em>identifier</em> names the macro being defined, and <em>replacement-list</em>
|
||
is a sequence of zero or more tokens. Where the <em>identifier</em>
|
||
appears in subsequent program text, it is <strong>expanded</strong>
|
||
by the preprocessor into its <em>replacement-list</em>.</p>
|
||
<p><strong>Function-like macros</strong>, which act as the
|
||
"metafunctions of the preprocessing phase," are defined as follows:</p>
|
||
<blockquote>
|
||
<div class="line-block">
|
||
<div class="line"><tt class="docutils literal"><span class="pre">#define</span></tt>
|
||
<em>identifier</em>(<em>a</em><sub>1</sub>, <em>a</em><sub>2</sub>,
|
||
... <em>a</em><sub>n</sub>) <em>replacement-list</em></div>
|
||
</div>
|
||
</blockquote>
|
||
<!-- @litre_translator.line_offset -= 7 -->
|
||
<p>where each <em>a</em><sub>i</sub> is an identifier naming a <strong>macro
|
||
parameter</strong>. When the macro name appears in subsequent
|
||
program text followed by a suitable argument list, it is expanded
|
||
into its <em>replacement-list</em>, except that each argument is
|
||
substituted for the corresponding parameter where it appears in the
|
||
<em>replacement-list</em>.<a class="footnote-reference" href="#expansion"
|
||
|
||
id="id4" name="id4">[2]</a></p>
|
||
<table class="docutils footnote" frame="void" id="expansion" rules="none">
|
||
<colgroup><col class="label" /><col /></colgroup>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td class="label"><a class="fn-backref" href="#id4" name="expansion">[2]</a></td>
|
||
<td>We have omitted many details of how macro expansion works.
|
||
We encourage you to take a few minutes to study section 16.3
|
||
of the C++ standard, which describes that process in
|
||
straightforward terms.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
<div class="section" id="macro-arguments">
|
||
<h2><a name="macro-arguments">A.2.3 Macro Arguments</a></h2>
|
||
<div class="admonition-definition admonition">
|
||
<p class="first admonition-title">Definition</p>
|
||
<p>A <strong>macro argument</strong> is a nonempty sequence of:</p>
|
||
<ul class="last simple">
|
||
<li>Preprocessing tokens other than commas or parentheses, <em>and/or</em></li>
|
||
<li>Preprocessing tokens surrounded by matched pairs of
|
||
parentheses.</li>
|
||
</ul>
|
||
</div>
|
||
<p>This definition has consequences for preprocessor metaprogramming
|
||
that must not be underestimated. Note, first of all, that the
|
||
following tokens have special status:</p>
|
||
<blockquote>
|
||
<pre class="literal-block">, ( )
|
||
</pre> </blockquote>
|
||
<!-- @ignore() -->
|
||
<p>As a result, a macro argument can never contain an unmatched
|
||
parenthesis, or a comma that is not surrounded by matched
|
||
parentheses. For example, both lines following the definition of FOO
|
||
below are ill-formed:</p>
|
||
<pre class="literal-block">#define FOO(X) X // Unary identity macro
|
||
FOO(,) // un-parenthesized comma or two empty arguments
|
||
FOO()) // unmatched parenthesis or missing argument
|
||
</pre>
|
||
<!-- @def pp_failure(options = ['-E'], **kw):
|
||
compile( expect_error = not 'mwcc' in config.compiler , options = options, **kw)pp_failure() -->
|
||
<p>Note also that the following tokens do <em>not</em> have special
|
||
status; the preprocessor knows nothing about matched pairs of
|
||
braces, brackets, or angle brackets:</p>
|
||
<blockquote>
|
||
<pre class="literal-block">{ } [ ] < >
|
||
</pre> </blockquote>
|
||
<!-- @ignore() -->
|
||
<p>As a result, these lines are also ill-formed:</p>
|
||
<pre class="literal-block">FOO(std::pair<int<strong>,</strong> long>) // two arguments
|
||
FOO({ int x = 1<strong>,</strong> y = 2; return x+y; }) // two arguments
|
||
</pre>
|
||
<!-- @example.prepend('#define FOO(X) X')
|
||
pp_failure() -->
|
||
<p>It <em>is</em> possible to pass either string of tokens above as
|
||
part of a single macro argument, provided it is parenthesized:</p>
|
||
<pre class="literal-block">FOO(<strong>(</strong>std::pair<int,int><strong>)</strong>) // one argument
|
||
FOO(<strong>(</strong>{ int x = 1, y = 2; return x+y; }<strong>)</strong>) // one argument
|
||
</pre>
|
||
<!-- @example.prepend('#define FOO(X) X')
|
||
compile(options = ['-E']) -->
|
||
<p>However, because of the special status of commas, it is impossible
|
||
to strip parentheses from a macro argument without knowing the
|
||
number of comma-separated token sequences it contains.<a class="footnote-reference"
|
||
|
||
href="#c99" id="id5" name="id5">[3]</a> If you are writing a macro
|
||
that needs to be able to accept an argument containing a variable
|
||
number of commas, your users will either have to parenthesize that
|
||
argument <em>and</em> pass you the number of comma-separated token
|
||
sequences as an additional argument, or they will have to encode the
|
||
same information in one of the preprocessor data structures covered
|
||
later in this appendix.</p>
|
||
<table class="docutils footnote" frame="void" id="c99" rules="none">
|
||
<colgroup><col class="label" /><col /></colgroup>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td class="label"><a name="c99">[3]</a></td>
|
||
<td><em>(<a class="fn-backref" href="#id5">1</a>, <a class="fn-backref"
|
||
|
||
href="#id12">2</a>)</em> The C99 preprocessor, by virtue
|
||
of its variadic macros, can do that and more. The C++
|
||
standardization committee is likely to adopt C99's
|
||
preprocessor extensions for the next version of the C++
|
||
standard.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
</div>
|
||
<div class="section" id="preprocessor-library-structure">
|
||
<h1><a name="preprocessor-library-structure">A.3 Preprocessor Library
|
||
Structure</a></h1>
|
||
<p>Since in-depth coverage of the Boost Preprocessor library is beyond
|
||
the scope of this book, we'll try to give you the <em>tools</em> to
|
||
gain an in-depth understanding of the library here. To do that, you'll
|
||
need to use the electronic Preprocessor library documentation, which
|
||
begins with the index.html file in the <tt class="docutils literal"><span
|
||
|
||
class="pre">libs/preprocessor/</span></tt> subdirectory of your
|
||
Boost installation.</p>
|
||
<p>On the left of your browser window you'll see an index, and if you
|
||
follow the "Headers" link, it will reveal the structure of the <tt class="docutils literal"><span
|
||
|
||
class="pre">boost/preprocessor/</span></tt> directory. Most of the
|
||
library's headers are grouped into subdirectories according to related
|
||
functionality. The top-level directory contains only a few headers
|
||
that provide general-purpose macros, along with a header for each
|
||
subdirectory that simply <tt class="docutils literal"><span class="pre">#include</span></tt>s
|
||
all the headers in that subdirectory. For example, <tt class="docutils literal"><span
|
||
|
||
class="pre">boost/preprocessor/selection.hpp</span></tt> does
|
||
nothing more than to <tt class="docutils literal"><span class="pre">#include</span></tt>
|
||
the <tt class="docutils literal"><span class="pre">min.hpp</span></tt>
|
||
and <tt class="docutils literal"><span class="pre">max.hpp</span></tt>
|
||
headers that comprise the contents of <tt class="docutils literal"><span
|
||
|
||
class="pre">boost/preprocessor/selection/</span></tt>. The headers
|
||
whose names <em>don't</em> correspond to subdirectories generally
|
||
declare a macro whose name is the same as the name of the header,
|
||
without the extension, and with a <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_</span></tt> prefix. For example, <tt class="docutils literal"><span
|
||
|
||
class="pre">boost/preprocessor/selection/max.hpp</span></tt>
|
||
declares <tt class="docutils literal"><span class="pre">BOOST_PP_MAX</span></tt>.</p>
|
||
<p>You'll also notice that often a header will declare an additional
|
||
macro with a <tt class="docutils literal"><span class="pre">_D</span></tt>,
|
||
<tt class="docutils literal"><span class="pre">_R</span></tt>, or <tt
|
||
|
||
class="docutils literal"><span class="pre">_Z</span></tt> suffix.<a
|
||
|
||
class="footnote-reference" href="#suffix" id="id6" name="id6">[4]</a>
|
||
For instance, <tt class="docutils literal"><span class="pre">boost/preprocessor/selection/max.hpp</span></tt>
|
||
also declares <tt class="docutils literal"><span class="pre">BOOST_PP_MAX_D</span></tt>.
|
||
For the purposes of this appendix, you should ignore those macros.
|
||
Eventually you will want to understand how they can be used to
|
||
optimize preprocessing speed; consult the Topics section of the
|
||
library documentation under the subheading "reentrancy" for that
|
||
information.</p>
|
||
<table class="docutils footnote" frame="void" id="suffix" rules="none">
|
||
<colgroup><col class="label" /><col /></colgroup>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td class="label"><a class="fn-backref" href="#id6" name="suffix">[4]</a></td>
|
||
<td>Macros with <tt class="docutils literal"><span class="pre">_1ST</span></tt>,
|
||
<tt class="docutils literal"><span class="pre">_2ND</span></tt>,
|
||
or <tt class="docutils literal"><span class="pre">_3RD</span></tt>
|
||
suffixes, if they appear, should be ignored for a different
|
||
reason: They are deprecated and will be removed from the library
|
||
soon.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
<div class="section" id="preprocessor-library-abstractions">
|
||
<h1><a name="preprocessor-library-abstractions">A.4 Preprocessor
|
||
Library Abstractions</a></h1>
|
||
<p>In this section we'll discuss the basic abstractions of the
|
||
Preprocessor library, and give some simple examples of each.</p>
|
||
<div class="section" id="repetition">
|
||
<h2><a name="repetition">A.4.1 Repetition</a></h2>
|
||
<p>The repeated generation of <tt class="docutils literal"><span class="pre">class</span>
|
||
<span class="pre">T0</span></tt>, <tt class="docutils literal"><span
|
||
|
||
class="pre">class</span> <span class="pre">T1</span></tt>... <tt
|
||
|
||
class="docutils literal"><span class="pre">class</span> <span class="pre">T</span></tt><em>n</em>
|
||
that we achieved using <tt class="docutils literal"><span class="pre">BOOST_PP_ENUM_PARAMS</span></tt>
|
||
was a specific case of the general concept of <strong>horizontal
|
||
repetition</strong>. The library also has a concept of vertical
|
||
repetition, which we'll get to in a moment. Horizontal repetition
|
||
macros are all found in the library's <tt class="docutils literal"><span
|
||
|
||
class="pre">repetition/</span></tt> subdirectory.</p>
|
||
<div class="section" id="horizontal-repetition">
|
||
<h3><a name="horizontal-repetition">A.4.1.1 Horizontal Repetition</a></h3>
|
||
<p>To generate the <tt class="docutils literal"><span class="pre">tiny_size</span></tt>
|
||
specializations using horizontal repetition, we might write the
|
||
following:</p>
|
||
<pre class="literal-block">#include <boost/preprocessor/repetition.hpp>
|
||
#include <boost/preprocessor/arithmetic/sub.hpp>
|
||
#include <boost/preprocessor/punctuation/comma_if.hpp>
|
||
|
||
#define TINY_print(z, n, data) data
|
||
|
||
#define TINY_size(z, n, unused) \
|
||
template <BOOST_PP_ENUM_PARAMS(n, class T)> \
|
||
struct tiny_size< \
|
||
BOOST_PP_ENUM_PARAMS(n,T) \
|
||
BOOST_PP_COMMA_IF(n) \
|
||
BOOST_PP_ENUM( \
|
||
BOOST_PP_SUB(TINY_MAX_SIZE,n), TINY_print, none) \
|
||
> \
|
||
: mpl::int_<n> {};
|
||
|
||
BOOST_PP_REPEAT(TINY_MAX_SIZE, TINY_size, ~)
|
||
|
||
#undef TINY_size
|
||
#undef TINY_print
|
||
</pre>
|
||
<!-- @import re
|
||
compile('all', pop = None)example.sub('BOOST_PP_REPEAT.*', '', flags = re.DOTALL) -->
|
||
<p>The code generation process is kicked off by calling <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_REPEAT</span></tt>, a <strong>higher-order
|
||
macro</strong> that repeatedly invokes the macro named by its
|
||
second argument (<tt class="docutils literal"><span class="pre">TINY_size</span></tt>).
|
||
The first argument specifies the number of repeated invocations,
|
||
and the third one can be any data; it is passed on unchanged to
|
||
the macro being invoked. In this case, <tt class="docutils literal"><span
|
||
|
||
class="pre">TINY_size</span></tt> doesn't use that data, so
|
||
the choice to pass <tt class="docutils literal"><span class="pre">~</span></tt>
|
||
was arbitrary.<a class="footnote-reference" href="#markers" id="id7"
|
||
|
||
name="id7">[5]</a></p>
|
||
<table class="docutils footnote" frame="void" id="markers" rules="none">
|
||
<colgroup><col class="label" /><col /></colgroup>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td class="label"><a class="fn-backref" href="#id7" name="markers">[5]</a></td>
|
||
<td><tt class="docutils literal"><span class="pre">~</span></tt>
|
||
is not an <em>entirely</em> arbitrary choice. Both <tt class="docutils literal"><span
|
||
|
||
class="pre">@</span></tt> and <tt class="docutils literal"><span
|
||
|
||
class="pre">$</span></tt> might have been good choices,
|
||
except that they are technically not part of the basic
|
||
character set that C++ implementations are required to
|
||
support. An identifier like <tt class="docutils literal"><span
|
||
|
||
class="pre">ignored</span></tt> might be subject to
|
||
macro expansion, leading to unexpected results.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<p>Each time the <tt class="docutils literal"><span class="pre">TINY_size</span></tt>
|
||
macro is invoked by <tt class="docutils literal"><span class="pre">BOOST_PP_REPEAT</span></tt>,
|
||
it generates a different specialization of <tt class="docutils literal"><span
|
||
|
||
class="pre">tiny_size</span></tt>. The macro accepts three
|
||
parameters.</p>
|
||
<ul class="simple">
|
||
<li><tt class="docutils literal"><span class="pre">z</span></tt>
|
||
is related to the <tt class="docutils literal"><span class="pre">_Z</span></tt>
|
||
macro suffix we mentioned earlier. You'll never need to use it
|
||
except for optimization purposes, and can safely ignore it for
|
||
now.</li>
|
||
<li><tt class="docutils literal"><span class="pre">n</span></tt>
|
||
is the repetition index. In repeated invocations of <tt class="docutils literal"><span
|
||
|
||
class="pre">TINY_size</span></tt>, <tt class="docutils literal"><span
|
||
|
||
class="pre">n</span></tt> will be <tt class="docutils literal"><span
|
||
|
||
class="pre">0</span></tt>, then <tt class="docutils literal"><span
|
||
|
||
class="pre">1</span></tt>, then <tt class="docutils literal"><span
|
||
|
||
class="pre">2</span></tt>, and so on.</li>
|
||
<li><tt class="docutils literal"><span class="pre">unused</span></tt>,
|
||
in this case, will be <tt class="docutils literal"><span class="pre">~</span></tt>
|
||
on each repetition. In general, the final argument to a macro
|
||
invoked by <tt class="docutils literal"><span class="pre">BOOST_PP_REPEAT</span></tt>
|
||
is always the same as its invoker's final argument.</li>
|
||
</ul>
|
||
<p>Because its <em>replacement-list</em> covers several lines, all
|
||
but the last line of <tt class="docutils literal"><span class="pre">TINY_size</span></tt>
|
||
is continued with a trailing backslash. The first few of those
|
||
lines just invoke <tt class="docutils literal"><span class="pre">BOOST_PP_ENUM_PARAMS</span></tt>
|
||
(which we already used in the primary template) to generate
|
||
comma-separated lists, so each invocation of <tt class="docutils literal"><span
|
||
|
||
class="pre">TINY_size</span></tt> produces something
|
||
equivalent to:<a class="footnote-reference" href="#cont" id="id8"
|
||
|
||
name="id8">[6]</a></p>
|
||
<pre class="literal-block">template <<strong>class T0, class T1, ... class T</strong><em>n-1</em>>
|
||
struct tiny_size<
|
||
<strong>T0, T1, ... T</strong><em>n-1</em>
|
||
<em>...more...</em>
|
||
>
|
||
: mpl::int_<n> {};
|
||
</pre>
|
||
<table class="docutils footnote" frame="void" id="cont" rules="none">
|
||
<colgroup><col class="label" /><col /></colgroup>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td class="label"><a class="fn-backref" href="#id8" name="cont">[6]</a></td>
|
||
<td>Note that the line continuation characters <em>and</em>
|
||
the newlines following them are removed by the preprocessor,
|
||
so the resulting code actually appears on a single line in
|
||
the preprocessed output.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<!-- @ignore() -->
|
||
<p><tt class="docutils literal"><span class="pre">BOOST_PP_COMMA_IF</span></tt>
|
||
generates a comma if its numeric argument is not <tt class="docutils literal"><span
|
||
|
||
class="pre">0</span></tt>. When <tt class="docutils literal"><span
|
||
|
||
class="pre">n</span></tt> is <tt class="docutils literal"><span
|
||
|
||
class="pre">0</span></tt>, the list generated by the preceding
|
||
line will be empty, and a leading comma directly following the <tt
|
||
|
||
class="docutils literal"><span class="pre"><</span></tt>
|
||
character would be ill-formed.</p>
|
||
<p>The next line uses <tt class="docutils literal"><span class="pre">BOOST_PP_ENUM</span></tt>
|
||
to generate <tt class="docutils literal"><span class="pre">TINY_MAX_SIZE-n</span></tt>
|
||
comma-separated copies of <tt class="docutils literal"><span class="pre">none</span></tt>.
|
||
<tt class="docutils literal"><span class="pre">BOOST_PP_ENUM</span></tt>
|
||
is just like <tt class="docutils literal"><span class="pre">BOOST_PP_REPEAT</span></tt>
|
||
except that it generates commas between repetitions, so its second
|
||
argument (<tt class="docutils literal"><span class="pre">TINY_print</span></tt>,
|
||
here) must have the same signature as <tt class="docutils literal"><span
|
||
|
||
class="pre">TINY_size</span></tt>. In this case, <tt class="docutils literal"><span
|
||
|
||
class="pre">TINY_print</span></tt> ignores its repetition
|
||
index <tt class="docutils literal"><span class="pre">n</span></tt>,
|
||
and simply yields its third argument, <tt class="docutils literal"><span
|
||
|
||
class="pre">none</span></tt>.</p>
|
||
<p><tt class="docutils literal"><span class="pre">BOOST_PP_SUB</span></tt>
|
||
implements token subtraction. It's crucial to understand that
|
||
although the preprocessor <em>itself</em> can evaluate ordinary
|
||
arithmetic expressions:</p>
|
||
<pre class="literal-block">#define X 3
|
||
...
|
||
#if <strong>X - 1 > 0</strong> // OK
|
||
<em>whatever</em>
|
||
#endif
|
||
</pre>
|
||
<!-- @compile() -->
|
||
<!-- @litre_translator.line_offset -= 7 -->
|
||
<p>preprocessor <em>metaprograms</em> can only operate on tokens.
|
||
Normally, when a macro in the Preprocessor library expects a
|
||
numeric argument, it must be passed as a single token. If we had
|
||
written <tt class="docutils literal"><span class="pre">TINY_MAX_SIZE-n</span></tt>
|
||
instead of <tt class="docutils literal"><span class="pre">BOOST_PP_SUB(TINY_MAX_SIZE,n)</span></tt>
|
||
above, the first argument to <tt class="docutils literal"><span class="pre">BOOST_PP_ENUM</span></tt>
|
||
would have contained three tokens at each invocation: first <tt class="docutils literal"><span
|
||
|
||
class="pre">3-0</span></tt>, then <tt class="docutils literal"><span
|
||
|
||
class="pre">3-1</span></tt>, and finally <tt class="docutils literal"><span
|
||
|
||
class="pre">3-2</span></tt>. <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_SUB</span></tt>, though, generates
|
||
single-token results: first <tt class="docutils literal"><span class="pre">3</span></tt>,
|
||
then <tt class="docutils literal"><span class="pre">2</span></tt>,
|
||
and finally <tt class="docutils literal"><span class="pre">1</span></tt>,
|
||
in successive repetitions.</p>
|
||
<div class="sidebar">
|
||
<p class="first sidebar-title">Naming Conventions</p>
|
||
<p class="last">Note that <tt class="docutils literal"><span class="pre">TINY_print</span></tt>
|
||
and <tt class="docutils literal"><span class="pre">TINY_size</span></tt>
|
||
are <tt class="docutils literal"><span class="pre">#undef</span></tt>'d
|
||
immediately
|
||
after they're used, with no intervening <tt class="docutils literal"><span
|
||
|
||
class="pre">#include</span></tt>s. They can therefore be
|
||
thought of as "local" macro definitions. Because the
|
||
preprocessor doesn't respect scope boundaries, it's important to
|
||
choose names carefully to prevent clashes. We recommend <tt class="docutils literal"><span
|
||
|
||
class="pre">PREFIXED_lower_case</span></tt> names for local
|
||
macros and <tt class="docutils literal"><span class="pre">PREFIXED_UPPER_CASE</span></tt>
|
||
names for global ones. The only exceptions are one-letter
|
||
lowercase names, which are safe to use for local macros: No
|
||
other header is likely to <tt class="docutils literal"><span class="pre">#define</span></tt>
|
||
a global single-letter lowercase macro—that would be <em>very</em>
|
||
bad manners.</p>
|
||
</div>
|
||
</div>
|
||
<div class="section" id="vertical-repetition">
|
||
<h3><a name="vertical-repetition">A.4.1.2 Vertical Repetition</a></h3>
|
||
<p>If you send the previous example through your preprocessor,
|
||
you'll see one long line containing something like this:</p>
|
||
<pre class="literal-block">template <> struct tiny_size< none , none , none > : mpl::int_<0>
|
||
{}; template < class T0> struct tiny_size< T0 , none , none > :
|
||
mpl::int_<1> {}; template < class T0 , class T1> struct tiny_size
|
||
< T0 , T1 , none > : mpl::int_<2> {};
|
||
</pre>
|
||
<!-- @compile('all', pop = 1) -->
|
||
<p>The distinguishing feature of horizontal repetition is that all
|
||
instances of the repeated pattern are generated on the same line
|
||
of preprocessed output. For some jobs, like generating the primary
|
||
<tt class="docutils literal"><span class="pre">tiny_size</span></tt>
|
||
template, that's perfectly appropriate. In this case, however,
|
||
there are at least two disadvantages.</p>
|
||
<ol class="arabic simple">
|
||
<li>It's hard to verify that our metaprogram is doing the right
|
||
thing without reformatting the resulting code by hand.</li>
|
||
<li>The efficiency of nested horizontal repetitions varies widely
|
||
across preprocessors. Each specialization generated by means of
|
||
horizontal repetition contains three other horizontal
|
||
repetitions: two invocations of <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_ENUM_PARAMS</span></tt> and one
|
||
invocation of <tt class="docutils literal"><span class="pre">BOOST_PP_ENUM</span></tt>.
|
||
When <tt class="docutils literal"><span class="pre">TINY_MAX_SIZE</span></tt>
|
||
is <tt class="docutils literal"><span class="pre">3</span></tt>,
|
||
you'll probably never care, but on at least one preprocessor
|
||
still in use today, compilation begins to slow noticeably when <tt
|
||
|
||
class="docutils literal"><span class="pre">TINY_MAX_SIZE</span></tt>
|
||
reaches <tt class="docutils literal"><span class="pre">8</span></tt>.<a
|
||
|
||
class="footnote-reference" href="#nest" id="id9" name="id9">[7]</a></li>
|
||
</ol>
|
||
<blockquote>
|
||
<table class="docutils footnote" frame="void" id="nest" rules="none">
|
||
<colgroup><col class="label" /><col /></colgroup>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td class="label"><a class="fn-backref" href="#id9" name="nest">[7]</a></td>
|
||
<td>That said, other preprocessors can handle 256 * 256
|
||
nested repetitions without any speed problems whatsoever.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</blockquote>
|
||
<p>The solution to these problems, naturally, is <strong>vertical
|
||
repetition</strong>, which generates instances of a pattern
|
||
across multiple lines. The Preprocessor library provides two means
|
||
of vertical repetition: <strong>local iteration</strong> and <strong>file
|
||
iteration</strong>.</p>
|
||
<div class="section" id="local-iteration">
|
||
<h4><a name="local-iteration">Local Iteration</a></h4>
|
||
<p>The most expedient way to demonstrate local iteration in our
|
||
example is to replace the invocation of <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_REPEAT</span></tt> with the following:</p>
|
||
<pre class="literal-block">#include <boost/preprocessor/<strong>iteration/local.hpp</strong>>
|
||
|
||
#define BOOST_PP_LOCAL_MACRO(n) TINY_size(~, n, ~)
|
||
#define BOOST_PP_LOCAL_LIMITS (0, <strong>TINY_MAX_SIZE - 1</strong>)
|
||
<strong>#include</strong> BOOST_PP_LOCAL_ITERATE()
|
||
</pre>
|
||
<!-- @compile('all', pop = 1) -->
|
||
<p>Local iteration repeatedly invokes the user-defined macro with
|
||
the special name <tt class="docutils literal"><span class="pre">BOOST_PP_LOCAL_MACRO</span></tt>,
|
||
whose argument will be an iteration index. Since we already had
|
||
<tt class="docutils literal"><span class="pre">TINY_size</span></tt>
|
||
lying around, we've just defined <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_LOCAL_MACRO</span></tt> to invoke it.
|
||
The range of iteration indices are given by another user-defined
|
||
macro, <tt class="docutils literal"><span class="pre">BOOST_PP_LOCAL_LIMITS</span></tt>,
|
||
which must expand to a parenthesized pair of integer values
|
||
representing the <em>inclusive</em> range of index values
|
||
passed to <tt class="docutils literal"><span class="pre">BOOST_PP_LOCAL_MACRO</span></tt>.
|
||
Note that this is one of the rare places where the library
|
||
expects a numeric argument that can be an expression consisting
|
||
of multiple tokens.</p>
|
||
<p>Finally, the repetition is initiated by <tt class="docutils literal"><span
|
||
|
||
class="pre">#include</span></tt>-ing the result of invoking
|
||
<tt class="docutils literal"><span class="pre">BOOST_PP_LOCAL_ITERATE</span></tt>,
|
||
which will ultimately be a file in the Preprocessor library
|
||
itself. You may find it surprising that many preprocessors can
|
||
handle repeated file inclusion more quickly than nested
|
||
horizontal repetition, but that is in fact the case.</p>
|
||
<p>If we throw the new example at our preprocessor, we'll see the
|
||
following, on three separate lines in the output:</p>
|
||
<pre class="literal-block">template <> struct tiny_size< none , none , none > : mpl::int_<0>
|
||
{};
|
||
|
||
template < class T0> struct tiny_size< T0 , none , none > : mpl::
|
||
int_<1> {};
|
||
|
||
template < class T0 , class T1> struct tiny_size< T0 , T1 , none
|
||
> : mpl::int_<2> {};
|
||
</pre>
|
||
<!-- @compile('all', pop = 1) -->
|
||
<p>That represents a great improvement in verifiability, but it's
|
||
still not ideal. As <tt class="docutils literal"><span class="pre">TINY_MAX_SIZE</span></tt>
|
||
grows, it gets harder and harder to see that the pattern is
|
||
generating what we'd like. If we could get some more line breaks
|
||
into the output it would retain a more recognizable form.</p>
|
||
<p>Both repetition methods we've used so far have another
|
||
drawback, though it doesn't show up in this example. Consider
|
||
what would happen if <tt class="docutils literal"><span class="pre">tiny_size</span></tt>
|
||
had a member function that we wanted to debug. If you've ever
|
||
tried to use a debugger to step through a function generated by
|
||
a preprocessor macro, you know that it's a frustrating
|
||
experience at best: The debugger shows you the line from which
|
||
the macro was ultimately invoked, which usually looks nothing at
|
||
all like the code that was generated. Worse, as far as the
|
||
debugger is concerned, <em>every</em> statement in that
|
||
generated function occupies that same line.</p>
|
||
</div>
|
||
<div class="section" id="file-iteration">
|
||
<h4><a name="file-iteration">File Iteration</a></h4>
|
||
<p>Clearly, debuggability depends on preserving the association
|
||
between generated code and the lines in the source file that
|
||
describe the code pattern. File iteration generates pattern
|
||
instances by repeatedly <tt class="docutils literal"><span class="pre">#include</span></tt>-ing
|
||
the same source file. The effect of file iteration on
|
||
debuggability is similar to that of templates: Although separate
|
||
instances appear to occupy the same source lines in the
|
||
debugger, we do have the experience of stepping through the
|
||
function's source code.</p>
|
||
<p>To apply file iteration in our example, we can replace our
|
||
earlier local iteration code and the definition of <tt class="docutils literal"><span
|
||
|
||
class="pre">TINY_size</span></tt>, with:</p>
|
||
<pre class="literal-block">#include <boost/preprocessor/iteration/iterate.hpp>
|
||
#define BOOST_PP_ITERATION_LIMITS (0, TINY_MAX_SIZE - 1)
|
||
#define BOOST_PP_FILENAME_1 "tiny_size_spec.hpp"
|
||
#include BOOST_PP_ITERATE()
|
||
</pre>
|
||
<p><tt class="docutils literal"><span class="pre">BOOST_PP_ITERATION_LIMITS</span></tt>
|
||
follows the same pattern as <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_LOCAL_LIMITS</span></tt> did, allowing
|
||
us to specify an inclusive range of iteration indices. <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_FILENAME_1</span></tt> specifies the
|
||
name of the file to repeatedly <tt class="docutils literal"><span
|
||
|
||
class="pre">#include</span></tt> (we'll show you that file
|
||
in a moment). The trailing <tt class="docutils literal"><span class="pre">1</span></tt>
|
||
indicates that this is the first nesting level of file
|
||
iteration—should we need to invoke file iteration again from
|
||
within <tt class="docutils literal"><span class="pre">tiny_size_spec.hpp</span></tt>,
|
||
we'd need to use <tt class="docutils literal"><span class="pre">BOOST_PP_FILENAME_2</span></tt>
|
||
instead.</p>
|
||
<p>The contents of <tt class="docutils literal"><span class="pre">tiny_size_spec.hpp</span></tt>
|
||
should look familiar to you; most of it is the same as <tt class="docutils literal"><span
|
||
|
||
class="pre">TINY_size</span></tt>'s <em>replacement-list</em>,
|
||
without the backslashes:</p>
|
||
<pre class="literal-block">#define n BOOST_PP_ITERATION()
|
||
|
||
template <BOOST_PP_ENUM_PARAMS(n, class T)>
|
||
struct tiny_size<
|
||
BOOST_PP_ENUM_PARAMS(n,T)
|
||
BOOST_PP_COMMA_IF(n)
|
||
BOOST_PP_ENUM(BOOST_PP_SUB(TINY_MAX_SIZE,n), TINY_print, none)
|
||
>
|
||
: mpl::int_<n> {};
|
||
|
||
#undef n
|
||
</pre>
|
||
<!-- @import tempfile, os
|
||
open(os.path.join(tempfile.gettempdir(),'tiny_size_spec.hpp'), 'w' ).write(str(example))ignore()vertical_options = ['-I'+tempfile.gettempdir(), '-c']
|
||
compile('all', options = vertical_options, pop = 1) -->
|
||
<p>The Library transmits the iteration index to us in the result
|
||
of <tt class="docutils literal"><span class="pre">BOOST_PP_ITERATION()</span></tt>;
|
||
<tt class="docutils literal"><span class="pre">n</span></tt> is
|
||
nothing more than a convenient local macro used to reduce
|
||
syntactic noise. Note that we didn't use <tt class="docutils literal"><span
|
||
|
||
class="pre">#include</span></tt> guards because we need <tt
|
||
|
||
class="docutils literal"><span class="pre">tiny_size_spec.hpp</span></tt>
|
||
to be processed multiple times.</p>
|
||
<p>The preprocessed result should now preserve the line structure
|
||
of the pattern and be more verifiable for larger values of <tt
|
||
|
||
class="docutils literal"><span class="pre">TINY_MAX_SIZE</span></tt>.
|
||
For instance, when <tt class="docutils literal"><span class="pre">TINY_MAX_SIZE</span></tt>
|
||
is <tt class="docutils literal"><span class="pre">8</span></tt>,
|
||
the following excerpt appears in the output of GCC's
|
||
preprocessing phase:</p>
|
||
<pre class="literal-block"><em>...</em>
|
||
template < class T0 , class T1 , class T2 , class T3>
|
||
struct tiny_size<
|
||
T0 , T1 , T2 , T3
|
||
,
|
||
none , none , none , none
|
||
>
|
||
: mpl::int_<4> {};
|
||
|
||
template < class T0 , class T1 , class T2 , class T3 , class T4>
|
||
struct tiny_size<
|
||
T0 , T1 , T2 , T3 , T4
|
||
,
|
||
none , none , none
|
||
>
|
||
: mpl::int_<5> {};
|
||
<em>...etc.</em>
|
||
</pre>
|
||
<!-- @compile('all', options = vertical_options + ['-DTINY_MAX_SIZE=8']) -->
|
||
</div>
|
||
<div class="section" id="self-iteration">
|
||
<h4><a name="self-iteration">Self-Iteration</a></h4>
|
||
<p>Creating an entirely new file like <tt class="docutils literal"><span
|
||
|
||
class="pre">tiny_size_spec.hpp</span></tt> each time we want
|
||
to express a trivial code pattern for file repetition can be
|
||
inconvenient. Fortunately, the library provides a macro that
|
||
allows us to place the pattern right in the file that invokes
|
||
the iteration. <tt class="docutils literal"><span class="pre">BOOST_PP_IS_ITERATING</span></tt>
|
||
is defined to a nonzero value whenever we're inside an
|
||
iteration. We can use that value to select between the part of a
|
||
file that invokes the iteration and the part that provides the
|
||
repeated pattern. Here's a complete <tt class="docutils literal"><span
|
||
|
||
class="pre">tiny_size.hpp</span></tt> file that demonstrates
|
||
self-iteration. Note in particular the placement and use of the
|
||
<tt class="docutils literal"><span class="pre">#include</span></tt>
|
||
guard <tt class="docutils literal"><span class="pre">TINY_SIZE_HPP_INCLUDED</span></tt>:</p>
|
||
<pre class="literal-block">#ifndef <strong>BOOST_PP_IS_ITERATING</strong>
|
||
|
||
# ifndef TINY_SIZE_HPP_INCLUDED
|
||
# define TINY_SIZE_HPP_INCLUDED
|
||
|
||
# include <boost/preprocessor/repetition.hpp>
|
||
# include <boost/preprocessor/arithmetic/sub.hpp>
|
||
# include <boost/preprocessor/punctuation/comma_if.hpp>
|
||
# include <boost/preprocessor/iteration/iterate.hpp>
|
||
|
||
# ifndef TINY_MAX_SIZE
|
||
# define TINY_MAX_SIZE 3 // default maximum size is 3
|
||
# endif
|
||
|
||
// primary template
|
||
template <BOOST_PP_ENUM_PARAMS(TINY_MAX_SIZE, class T)>
|
||
struct tiny_size
|
||
: mpl::int_<TINY_MAX_SIZE>
|
||
{};
|
||
|
||
// generate specializations
|
||
# define BOOST_PP_ITERATION_LIMITS (0, TINY_MAX_SIZE - 1)
|
||
# define BOOST_PP_FILENAME_1 "tiny_size.hpp" // this file
|
||
# include BOOST_PP_ITERATE()
|
||
|
||
# endif // TINY_SIZE_HPP_INCLUDED
|
||
|
||
#else // <strong>BOOST_PP_IS_ITERATING</strong>
|
||
|
||
# define n BOOST_PP_ITERATION()
|
||
|
||
# define TINY_print(z, n, data) data
|
||
|
||
// specialization pattern
|
||
template <BOOST_PP_ENUM_PARAMS(n, class T)>
|
||
struct tiny_size<
|
||
BOOST_PP_ENUM_PARAMS(n,T)
|
||
BOOST_PP_COMMA_IF(n)
|
||
BOOST_PP_ENUM(BOOST_PP_SUB(TINY_MAX_SIZE,n), TINY_print, none)
|
||
>
|
||
: mpl::int_<n> {};
|
||
|
||
# undef TINY_print
|
||
# undef n
|
||
|
||
#endif // <strong>BOOST_PP_IS_ITERATING</strong>
|
||
</pre>
|
||
<!-- @compile(source_file = 'tiny_size.hpp') --> </div>
|
||
<div class="section" id="more">
|
||
<h4><a name="more">More</a></h4>
|
||
<p>There's a good deal more to file iteration than what we've been
|
||
able to show you here. For more details, we encourage you to
|
||
delve into the library's electronic documentation of <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_ITERATE</span></tt> and friends. Also,
|
||
it's important to note that no single technique for repetition
|
||
is superior to any other: Your choice may depend on convenience,
|
||
verifiability, debuggability, compilation speed, and your own
|
||
sense of "logical coherence."</p>
|
||
</div>
|
||
</div>
|
||
</div>
|
||
<div class="section" id="arithmetic-logical-and-comparison-operations">
|
||
<h2><a name="arithmetic-logical-and-comparison-operations">A.4.2 Arithmetic,
|
||
Logical, and Comparison Operations</a></h2>
|
||
<p>As we mentioned earlier, many of the Preprocessor library
|
||
interfaces require single-token numeric arguments, and when those
|
||
numbers need to be computed arithmetically, straightforward
|
||
arithmetic expressions are inappropriate. We used <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_SUB</span></tt> to subtract two numeric
|
||
tokens in our <tt class="docutils literal"><span class="pre">tiny_size</span></tt>
|
||
examples. The library contains a suite of operations for
|
||
non-negative integral token arithmetic in its <tt class="docutils literal"><span
|
||
|
||
class="pre">arithmetic/</span></tt> subdirectory, as shown in
|
||
Table A.1</p>
|
||
<table border="1" class="docutils">
|
||
<caption>Preprocessor Library Arithmetic Operations</caption> <colgroup>
|
||
<col width="44%" /> <col width="56%" /> </colgroup>
|
||
<thead valign="bottom">
|
||
<tr>
|
||
<th>Expression</th>
|
||
<th>Value of Single Token Result</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_ADD(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">+</span> <span class="pre">y</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_DEC(x)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">-</span> <span class="pre">1</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_DIV(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">/</span> <span class="pre">y</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_INC(x)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">+</span> <span class="pre">1</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_MOD(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">%</span> <span class="pre">y</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_MUL(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">*</span> <span class="pre">y</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SUB(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">-</span> <span class="pre">y</span></tt></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<p>The <tt class="docutils literal"><span class="pre">logical/</span></tt>
|
||
subdirectory contains the convenient Boolean token operations shown
|
||
in Table A.2 and the more efficient operations shown in Table A.3,
|
||
which require that their operands are either <tt class="docutils literal"><span
|
||
|
||
class="pre">0</span></tt> or <tt class="docutils literal"><span
|
||
|
||
class="pre">1</span></tt> (a single bit).</p>
|
||
<table border="1" class="docutils">
|
||
<caption>Preprocessor Library Integer Logical Operations</caption> <colgroup>
|
||
<col width="44%" /> <col width="56%" /> </colgroup>
|
||
<thead valign="bottom">
|
||
<tr>
|
||
<th>Expression</th>
|
||
<th>Value of Single Token Result</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_AND(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">&&</span> <span class="pre">y</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_NOR(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">!(x</span> <span
|
||
|
||
class="pre">||</span> <span class="pre">y)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_OR(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">||</span> <span class="pre">y</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_XOR(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(bool)x</span>
|
||
<span class="pre">!=</span> <span class="pre">(bool)y</span>
|
||
<span class="pre">?</span> <span class="pre">1</span> <span
|
||
|
||
class="pre">:</span> <span class="pre">0</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_NOT(x)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">?</span> <span class="pre">0</span> <span class="pre">:</span>
|
||
<span class="pre">1</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_BOOL(x)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">?</span> <span class="pre">1</span> <span class="pre">:</span>
|
||
<span class="pre">0</span></tt></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<table border="1" class="docutils">
|
||
<caption>Preprocessor Library Bit Logical Operations</caption> <colgroup>
|
||
<col width="44%" /> <col width="56%" /> </colgroup>
|
||
<thead valign="bottom">
|
||
<tr>
|
||
<th>Expression</th>
|
||
<th>Value of Single Token Result</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_BITAND(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">&&</span> <span class="pre">y</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_BITNOR(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">!(x</span> <span
|
||
|
||
class="pre">||</span> <span class="pre">y)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_BITOR(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">||</span> <span class="pre">y</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_BITXOR(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(bool)x</span>
|
||
<span class="pre">!=</span> <span class="pre">(bool)y</span>
|
||
<span class="pre">?</span> <span class="pre">1</span> <span
|
||
|
||
class="pre">:</span> <span class="pre">0</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_COMPL(x)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">?</span> <span class="pre">0</span> <span class="pre">:</span>
|
||
<span class="pre">1</span></tt></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<p>Finally, the <tt class="docutils literal"><span class="pre">comparison/</span></tt>
|
||
subdirectory provides the token integral comparison operations shown
|
||
in Table A.4.</p>
|
||
<table border="1" class="docutils">
|
||
<caption>Preprocessor Library Comparison Operations</caption> <colgroup>
|
||
<col width="46%" /> <col width="54%" /> </colgroup>
|
||
<thead valign="bottom">
|
||
<tr>
|
||
<th>Expression</th>
|
||
<th>Value of Single Token Result</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_EQUAL(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">==</span> <span class="pre">y</span> <span
|
||
|
||
class="pre">?</span> <span class="pre">1</span> <span class="pre">:</span>
|
||
<span class="pre">0</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_NOT_EQUAL(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">!=</span> <span class="pre">y</span> <span
|
||
|
||
class="pre">?</span> <span class="pre">1</span> <span class="pre">:</span>
|
||
<span class="pre">0</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_LESS(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre"><</span> <span class="pre">y</span> <span
|
||
|
||
class="pre">?</span> <span class="pre">1</span> <span class="pre">:</span>
|
||
<span class="pre">0</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_LESS_EQUAL(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre"><=</span> <span class="pre">y</span> <span
|
||
|
||
class="pre">?</span> <span class="pre">1</span> <span class="pre">:</span>
|
||
<span class="pre">0</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_GREATER(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">></span> <span class="pre">y</span> <span
|
||
|
||
class="pre">?</span> <span class="pre">1</span> <span class="pre">:</span>
|
||
<span class="pre">0</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_GREATER_EQUAL(x,y)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">x</span> <span
|
||
|
||
class="pre">>=</span> <span class="pre">y</span> <span
|
||
|
||
class="pre">?</span> <span class="pre">1</span> <span class="pre">:</span>
|
||
<span class="pre">0</span></tt></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<p>Because it's common to have a choice among several workable
|
||
comparison operators, it may be useful to know that <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_EQUAL</span></tt> and <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_NOT_EQUAL</span></tt> are likely to be O(1)
|
||
while the other comparison operators are generally slower.</p>
|
||
</div>
|
||
<div class="section" id="control-structures">
|
||
<h2><a name="control-structures">A.4.3 Control Structures</a></h2>
|
||
<p>In its <tt class="docutils literal"><span class="pre">control/</span></tt>
|
||
directory, the Preprocessor Library supplies a macro <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_IF(c,t,f)</span></tt> that fulfills a
|
||
similar role to the one filled by <tt class="docutils literal"><span
|
||
|
||
class="pre">mpl::if_</span></tt>. To explore the "control"
|
||
group, we'll generate code for a framework of generic function
|
||
objects: the Boost Function Library.<a class="footnote-reference" href="#function"
|
||
|
||
id="id10" name="id10">[8]</a> <tt class="docutils literal"><span
|
||
|
||
class="pre">boost::function</span></tt> is partially specialized
|
||
to match function type arguments of each arity up to the maximum
|
||
supported by the library:</p>
|
||
<pre class="literal-block">template <class Signature> struct function; // primary template
|
||
|
||
template <class R> // arity = 0
|
||
struct function<R()>
|
||
<em>definition not shown...</em>
|
||
|
||
template <class R, class A0> // arity = 1
|
||
struct function<R(A0)>
|
||
<em>definition not shown...</em>
|
||
|
||
template <class R, class A0, class A1> // arity = 2
|
||
struct function<R(A0,A1)>
|
||
<em>definition not shown...</em>
|
||
|
||
template <class R, class A0, class A1, class A2> // arity = 3
|
||
struct function<R(A0,A1,A2)>
|
||
<em>definition not shown...</em>
|
||
|
||
<em>etc.</em>
|
||
</pre>
|
||
<!-- @example.replace(')>', ')>;')
|
||
compile() -->
|
||
<table class="docutils footnote" frame="void" id="function" rules="none">
|
||
<colgroup><col class="label" /><col /></colgroup>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td class="label"><a class="fn-backref" href="#id10" name="function">[8]</a></td>
|
||
<td>We touched briefly on the design of Boost Function when we
|
||
discussed type erasure in Chapter 9. See the Function library
|
||
documentation at <tt class="docutils literal"><span class="pre">boost_1_32_0/libs/function/index.html</span></tt>
|
||
on the CD that accompanies this book for more information.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<p>We've already covered a few strategies that can be used to generate
|
||
the pattern above, so we won't belabor that part of the problem; the
|
||
file iteration approach we used for <tt class="docutils literal"><span
|
||
|
||
class="pre">tiny_size</span></tt> would be fine:</p>
|
||
<pre class="literal-block">#ifndef BOOST_PP_IS_ITERATING
|
||
|
||
# ifndef BOOST_FUNCTION_HPP_INCLUDED
|
||
# define BOOST_FUNCTION_HPP_INCLUDED
|
||
|
||
# include <boost/preprocessor/repetition.hpp>
|
||
# include <boost/preprocessor/iteration/iterate.hpp>
|
||
|
||
# ifndef FUNCTION_MAX_ARITY
|
||
# define FUNCTION_MAX_ARITY 15
|
||
# endif
|
||
|
||
<strong>template <class Signature> struct function;</strong> // primary template
|
||
|
||
// generate specializations
|
||
# define BOOST_PP_ITERATION_LIMITS (0, FUNCTION_MAX_ARITY)
|
||
# define BOOST_PP_FILENAME_1 "boost/function.hpp" // this file
|
||
# include BOOST_PP_ITERATE()
|
||
|
||
# endif // BOOST_FUNCTION_HPP_INCLUDED
|
||
|
||
#else // BOOST_PP_IS_ITERATING
|
||
|
||
# define n BOOST_PP_ITERATION()
|
||
|
||
// specialization pattern
|
||
<strong>template <class R BOOST_PP_ENUM_TRAILING_PARAMS(n, class A)></strong>
|
||
<strong>struct function<R ( BOOST_PP_ENUM_PARAMS(n,A) )></strong>
|
||
<em>definition not shown...</em>
|
||
|
||
# undef n
|
||
|
||
#endif // BOOST_PP_IS_ITERATING
|
||
</pre>
|
||
<p><tt class="docutils literal"><span class="pre">BOOST_PP_ENUM_TRAILING_PARAMS</span></tt>,
|
||
used above, is just like <tt class="docutils literal"><span class="pre">BOOST_PP_ENUM_PARAMS</span></tt>
|
||
except that when its first argument is not <tt class="docutils literal"><span
|
||
|
||
class="pre">0</span></tt>, it generates a leading comma.</p>
|
||
<!-- @example.replace_emphasis(';//')
|
||
tmpdir = tempfile.gettempdir()tmpboost = os.path.join(tmpdir,'boost')try: os.mkdir(tmpboost)except: pass
|
||
tmp_boost_function = os.path.join(tmpdir, 'boost/function.hpp')compile( options = vertical_options , source_file = tmp_boost_function
|
||
, pop = None) -->
|
||
<div class="section" id="argument-selection">
|
||
<h3><a name="argument-selection">A.4.3.1 Argument Selection</a></h3>
|
||
<p>For the sake of interoperability with C++ standard library
|
||
algorithms, it might be nice if <tt class="docutils literal"><span
|
||
|
||
class="pre">function</span></tt>s of one or two arguments were
|
||
derived from appropriate specializations of <tt class="docutils literal"><span
|
||
|
||
class="pre">std::unary_function</span></tt> or <tt class="docutils literal"><span
|
||
|
||
class="pre">std::binary_function</span></tt>, respectively.<a
|
||
|
||
class="footnote-reference" href="#ebo" id="id11" name="id11">[9]</a>
|
||
<tt class="docutils literal"><span class="pre">BOOST_PP_IF</span></tt>
|
||
is a great tool for dealing with special cases:</p>
|
||
<pre class="literal-block"># include <boost/preprocessor/control/if.hpp>
|
||
# include <boost/preprocessor/comparison/equal.hpp>
|
||
|
||
// specialization pattern
|
||
template <class R BOOST_PP_ENUM_TRAILING_PARAMS(n, class A)>
|
||
struct function<R ( BOOST_PP_ENUM_PARAMS(n,A) )>
|
||
BOOST_PP_IF(
|
||
BOOST_PP_EQUAL(n,2), <strong>: std::binary_function<A0, A1, R></strong>
|
||
, BOOST_PP_IF(
|
||
BOOST_PP_EQUAL(n,1), <strong>: std::unary_function<A0, R></strong>
|
||
, <em>...empty argument...</em>
|
||
)
|
||
)
|
||
{ <em>...class body omitted...</em> };
|
||
</pre>
|
||
<!-- @pp_failure() -->
|
||
<table class="docutils footnote" frame="void" id="ebo" rules="none">
|
||
<colgroup><col class="label" /><col /></colgroup>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td class="label"><a class="fn-backref" href="#id11" name="ebo">[9]</a></td>
|
||
<td>While derivation from <tt class="docutils literal"><span
|
||
|
||
class="pre">std::unary_function</span></tt> or <tt class="docutils literal"><span
|
||
|
||
class="pre">std::binary_function</span></tt> might be
|
||
necessary for interoperability with some older library
|
||
implementations, it may inhibit the Empty Base Optimization
|
||
(EBO) from taking effect when two such derived classes are
|
||
part of the same object. For more information, see section
|
||
9.4. In general, it's better to expose <tt class="docutils literal"><span
|
||
|
||
class="pre">first_argument_type</span></tt>, <tt class="docutils literal"><span
|
||
|
||
class="pre">second_argument_type</span></tt>, and <tt class="docutils literal"><span
|
||
|
||
class="pre">result_type</span></tt> <tt class="docutils literal"><span
|
||
|
||
class="pre">typedef</span></tt>s directly.</td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<p>Well, our first attempt has run into several problems. First off,
|
||
you're not allowed to pass an empty argument to the preprocessor.<a
|
||
|
||
class="footnote-reference" href="#c99" id="id12" name="id12">[3]</a>
|
||
Secondly, because angle brackets don't get special treatment, the
|
||
commas in the <tt class="docutils literal"><span class="pre">std::unary_function</span></tt>
|
||
and <tt class="docutils literal"><span class="pre">std::binary_function</span></tt>
|
||
specializations above are treated as macro argument separators,
|
||
and the preprocessor will complain that we've passed the wrong
|
||
number of arguments to <tt class="docutils literal"><span class="pre">BOOST_PP_IF</span></tt>
|
||
in two places.</p>
|
||
<p>Because it captures all of the issues, let's focus on the inner <tt
|
||
|
||
class="docutils literal"><span class="pre">BOOST_PP_IF</span></tt>
|
||
invocation for a moment. The strategy that <tt class="docutils literal"><span
|
||
|
||
class="pre">mpl::eval_if</span></tt> uses, of selecting a
|
||
nullary function to invoke, could work nicely here. The
|
||
preprocessor doesn't have a direct analogue for <tt class="docutils literal"><span
|
||
|
||
class="pre">mpl::eval_if</span></tt>, but it doesn't really
|
||
need one: We can get the right effect by adding a second set of
|
||
parentheses to <tt class="docutils literal"><span class="pre">BOOST_PP_IF</span></tt>.</p>
|
||
<pre class="literal-block">#define BOOST_FUNCTION_unary() : std::unary_function<A0,R>
|
||
#define BOOST_FUNCTION_empty() // nothing
|
||
|
||
...
|
||
|
||
, BOOST_PP_IF(
|
||
BOOST_PP_EQUAL(n,1), BOOST_FUNCTION_unary
|
||
, BOOST_FUNCTION_empty
|
||
)<strong>()</strong>
|
||
|
||
#undef BOOST_FUNCTION_empty
|
||
#undef BOOST_FUNCTION_unary
|
||
</pre>
|
||
<!-- @ignore() -->
|
||
<p>A nullary macro that generates nothing is so commonly needed that
|
||
the library's "facilities" group provides one: <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_EMPTY</span></tt>. To complete the
|
||
example we'll need to delay evaluation all the way to the outer <tt
|
||
|
||
class="docutils literal"><span class="pre">BOOST_PP_IF</span></tt>
|
||
invocation, because <tt class="docutils literal"><span class="pre">std::binary_function<A0,A1,R></span></tt>
|
||
also has a "comma problem":</p>
|
||
<pre class="literal-block"># include <boost/preprocessor/<strong>facilities/empty.hpp</strong>>
|
||
|
||
# define BOOST_FUNCTION_binary() : std::binary_function<A0,A1,R>
|
||
# define BOOST_FUNCTION_unary() : std::unary_function<A0,R>
|
||
|
||
// specialization pattern
|
||
template <class R BOOST_PP_ENUM_TRAILING_PARAMS(n, class A)>
|
||
struct function<R ( BOOST_PP_ENUM_PARAMS(n,A) )>
|
||
BOOST_PP_IF(
|
||
BOOST_PP_EQUAL(n,2), BOOST_FUNCTION_binary
|
||
, BOOST_PP_IF(
|
||
BOOST_PP_EQUAL(n,1), BOOST_FUNCTION_unary
|
||
, <strong>BOOST_PP_EMPTY</strong>
|
||
)
|
||
)<strong>()</strong>
|
||
{
|
||
<em>...class body omitted...</em>
|
||
};
|
||
|
||
# undef BOOST_FUNCTION_unary
|
||
# undef BOOST_FUNCTION_binary
|
||
# undef n
|
||
</pre>
|
||
<!-- @stack.pop()
|
||
stack[-1].replace('// specialization pattern', '////\n%s\n////' % str(example))compile(source_file = tmp_boost_function, pop = None) -->
|
||
<p>Note that because we happened to be using file iteration, we
|
||
could have also used <tt class="docutils literal"><span class="pre">#if</span></tt>
|
||
on <tt class="docutils literal"><span class="pre">n</span></tt>'s
|
||
value directly:</p>
|
||
<pre class="literal-block"> template <class R BOOST_PP_ENUM_TRAILING_PARAMS(n, class A)>
|
||
struct function<R ( BOOST_PP_ENUM_PARAMS(n,A) )>
|
||
<strong>#if n == 2</strong>
|
||
: std::binary_function<A0, A1, R>
|
||
<strong>#elif n == 1</strong>
|
||
: std::unary_function<A0, R>
|
||
<strong>#endif</strong>
|
||
</pre>
|
||
<!-- @stack.pop()
|
||
stack[-1].sub( r'////.*////', '////\n%s\n////' % str(example), flags = re.DOTALL)compile(source_file = tmp_boost_function, pop = None) -->
|
||
<p><tt class="docutils literal"><span class="pre">BOOST_PP_IF</span></tt>
|
||
has the advantage of enabling us to encapsulate the logic in a
|
||
reusable macro, parameterized on <tt class="docutils literal"><span
|
||
|
||
class="pre">n</span></tt>, that is compatible with all
|
||
repetition constructs:</p>
|
||
<pre class="literal-block">#define BOOST_FUNCTION_BASE(n) \
|
||
BOOST_PP_IF(BOOST_PP_EQUAL(n,2), BOOST_FUNCTION_binary \
|
||
, BOOST_PP_IF(BOOST_PP_EQUAL(n,1), BOOST_FUNCTION_unary \
|
||
, BOOST_PP_EMPTY \
|
||
) \
|
||
)()
|
||
</pre>
|
||
<!-- @compile(options = ['-E']) --> </div>
|
||
<div class="section" id="other-selection-constructs">
|
||
<h3><a name="other-selection-constructs">A.4.3.2 Other Selection
|
||
Constructs</a></h3>
|
||
<p><tt class="docutils literal"><span class="pre">BOOST_PP_IDENTITY</span></tt>,
|
||
also in the "facilities" group, is an interesting cousin of <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_EMPTY</span></tt>:</p>
|
||
<pre class="literal-block">#define BOOST_PP_IDENTITY(tokens) tokens BOOST_PP_EMPTY
|
||
</pre>
|
||
<!-- @ignore() -->
|
||
<p>You can think of it as creating a nullary macro that returns <tt
|
||
|
||
class="docutils literal"><span class="pre">tokens</span></tt>:
|
||
When empty parentheses are appended, the trailing <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_EMPTY</span></tt> is expanded leaving
|
||
just <tt class="docutils literal"><span class="pre">tokens</span></tt>
|
||
behind. If we had wanted inheritance from <tt class="docutils literal"><span
|
||
|
||
class="pre">mpl::empty_base</span></tt> when <tt class="docutils literal"><span
|
||
|
||
class="pre">function</span></tt>'s arity is not one or two, we
|
||
could have used <tt class="docutils literal"><span class="pre">BOOST_PP_IDENTITY</span></tt>:</p>
|
||
<pre class="literal-block">// specialization pattern
|
||
template <class R BOOST_PP_ENUM_TRAILING_PARAMS(n, class A)>
|
||
struct function<R ( BOOST_PP_ENUM_PARAMS(n,A) )>
|
||
BOOST_PP_IF(
|
||
BOOST_PP_EQUAL(n,2), BOOST_FUNCTION_binary
|
||
, BOOST_PP_IF(
|
||
BOOST_PP_EQUAL(n,1), BOOST_FUNCTION_unary
|
||
, <strong>BOOST_PP_IDENTITY(: mpl::empty_base)</strong>
|
||
)
|
||
)<strong>()</strong>
|
||
{
|
||
<em>...class body omitted...</em>
|
||
};
|
||
</pre>
|
||
<!-- @stack.pop()
|
||
stack[-1].sub( r'////.*////', '////\n%s\n////' % str(example), flags = re.DOTALL)compile(source_file = tmp_boost_function, pop = None) -->
|
||
<p>It's also worth knowing about <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_EXPR_IF</span></tt>, which generates its
|
||
second argument or nothing, depending on the Boolean value of its
|
||
first:</p>
|
||
<pre class="literal-block">#define BOOST_PP_EXPR_IF(c,tokens) \
|
||
BOOST_PP_IF(c,BOOST_PP_IDENTITY(tokens),BOOST_PP_EMPTY)()
|
||
</pre>
|
||
<!-- @example.append(
|
||
'int BOOST_PP_EXPR_IF(1,main) BOOST_PP_EXPR_IF(0,quack) () {}')compile() -->
|
||
<p>So <tt class="docutils literal"><span class="pre">BOOST_PP_EXPR_IF(1,foo)</span></tt>
|
||
expands to <tt class="docutils literal"><span class="pre">foo</span></tt>,
|
||
while <tt class="docutils literal"><span class="pre">BOOST_PP_EXPR_IF(0,foo)</span></tt>
|
||
expands to nothing.</p>
|
||
</div>
|
||
</div>
|
||
<div class="section" id="token-pasting">
|
||
<h2><a name="token-pasting">A.4.4 Token Pasting</a></h2>
|
||
<p>It would be nice if there were a generic way to access the return
|
||
and parameter types of <em>all</em> function objects, rather than
|
||
just the unary and binary ones. A metafunction returning the
|
||
signature as an MPL sequence would do the trick. We could just
|
||
specialize <tt class="docutils literal"><span class="pre">signature</span></tt>
|
||
for each <tt class="docutils literal"><span class="pre">function</span></tt>
|
||
arity:</p>
|
||
<pre class="literal-block">template <class F> struct signature; // primary template
|
||
|
||
// partial specializations for boost::function
|
||
template <class R>
|
||
struct signature<function<R()> >
|
||
: mpl::vector1<R> {};
|
||
|
||
template <class R, class A0>
|
||
struct signature<function<R(A0)> >
|
||
: mpl::vector2<R,A0> {};
|
||
|
||
template <class R, class A0, class A1>
|
||
struct signature<function<R(A0,A1)> >
|
||
: mpl::vector3<R,A0,A1> {};
|
||
|
||
...
|
||
</pre>
|
||
<!-- @example.prepend('template <class T> struct function;')
|
||
compile() -->
|
||
<p>To generate these specializations, we might add the following to
|
||
our pattern:</p>
|
||
<pre class="literal-block">template <class R BOOST_PP_ENUM_TRAILING_PARAMS(n, class A)>
|
||
struct signature<function<R( BOOST_PP_ENUM_PARAMS(n,A) )> >
|
||
: mpl::<strong>BOOST_PP_CAT</strong>(vector,n)<
|
||
R BOOST_PP_ENUM_TRAILING_PARAMS(n,A)
|
||
> {};
|
||
</pre>
|
||
<!-- @stack.pop()
|
||
stack[-1].replace( ';//', ''';// template <class T> struct signature; %s''' % example)
|
||
compile(source_file = tmp_boost_function) -->
|
||
<p><tt class="docutils literal"><span class="pre">BOOST_PP_CAT</span></tt>
|
||
implements <strong>token pasting</strong>; its two arguments are
|
||
"glued" together into a single token. Since this is a
|
||
general-purpose macro, it sits in <tt class="docutils literal"><span
|
||
|
||
class="pre">cat.hpp</span></tt> at the top level of the
|
||
library's directory tree.</p>
|
||
<p>Although the preprocessor has a built-in token-pasting operator, <tt
|
||
|
||
class="docutils literal"><span class="pre">##</span></tt>, it only
|
||
works within a macro definition. If we'd used it here, it wouldn't
|
||
have taken effect at all:</p>
|
||
<pre class="literal-block">template <class R>
|
||
struct signature<function<R()> >
|
||
: mpl::<strong>vector##1</strong><R> {};
|
||
|
||
template <class R, class A0>
|
||
struct signature<function<R(A0)> >
|
||
: mpl::<strong>vector##2</strong><R,A0> {};
|
||
|
||
template <class R, class A0, class A1>
|
||
struct signature<function<R(A0,A1)> >
|
||
: mpl::<strong>vector##3</strong><R,A0,A1> {};
|
||
|
||
...
|
||
</pre>
|
||
<!-- @example.replace('##','')
|
||
example.prepend(''' template <class T> struct function; template <class T> struct signature;''')
|
||
compile() -->
|
||
<p>Also, <tt class="docutils literal"><span class="pre">##</span></tt>
|
||
often yields surprising results by taking effect before its
|
||
arguments have been expanded:</p>
|
||
<pre class="literal-block">#define N 10
|
||
#define VEC(i) vector##i
|
||
|
||
VEC(N) // vectorN
|
||
</pre>
|
||
<!-- @example.wrap('typedef int vectorN;', 'x;')
|
||
compile() -->
|
||
<p>By contrast, <tt class="docutils literal"><span class="pre">BOOST_PP_CAT</span></tt>
|
||
delays concatenation until after its arguments have been fully
|
||
evaluated:</p>
|
||
<pre class="literal-block">#define N 10
|
||
#define VEC(i) BOOST_PP_CAT(vector,i)
|
||
|
||
VEC(N) // vector10
|
||
</pre>
|
||
<!-- @example.wrap('''
|
||
#include <boost/preprocessor/cat.hpp> typedef int vector10; ''', 'x;')compile() -->
|
||
</div>
|
||
<div class="section" id="data-types">
|
||
<h2><a name="data-types">A.4.5 Data Types</a></h2>
|
||
<p>The Preprocessor library also provides <strong>data types</strong>,
|
||
which you can think of as being analogous to the MPL's type
|
||
sequences. Preprocessor data types store <em>macro arguments</em>
|
||
instead of C++ types.</p>
|
||
<div class="section" id="sequences">
|
||
<h3><a name="sequences">A.4.5.1 Sequences</a></h3>
|
||
<p>A <strong>sequence</strong> (or <strong>seq</strong> for short)
|
||
is any string of nonempty parenthesized <em>macro arguments</em>.
|
||
For instance, here's a three-element sequence:</p>
|
||
<pre class="literal-block">#define MY_SEQ (f(12))(a + 1)(foo)
|
||
</pre>
|
||
<!-- @ignore() -->
|
||
<p>Here's how we might use a sequence to generate specializations of
|
||
the <tt class="docutils literal"><span class="pre">is_integral</span></tt>
|
||
template from the Boost Type Traits library (see Chapter 2):</p>
|
||
<pre class="literal-block">#include <boost/preprocessor/seq.hpp>
|
||
|
||
template <class T>
|
||
struct is_integral : mpl::false_ {};
|
||
|
||
// a seq of integral types with unsigned counterparts
|
||
#define BOOST_TT_basic_ints (char)(short)(int)(long)
|
||
|
||
// generate a seq containing "signed t" and "unsigned t"
|
||
#define BOOST_TT_int_pair(r,data,t) (signed t)(unsigned t)
|
||
|
||
// a seq of all the integral types
|
||
#define BOOST_TT_ints \
|
||
(bool)(char) \
|
||
BOOST_PP_SEQ_FOR_EACH(BOOST_TT_int_pair, ~, BOOST_TT_basic_ints)
|
||
|
||
// generate an is_integral specialization for type t
|
||
#define BOOST_TT_is_integral_spec(r,data,t) \
|
||
template <> \
|
||
struct is_integral<t> : mpl::true_ {};
|
||
|
||
BOOST_PP_SEQ_FOR_EACH(BOOST_TT_is_integral_spec, ~, BOOST_TT_ints)
|
||
|
||
#undef BOOST_TT_is_integral_spec
|
||
#undef BOOST_TT_ints
|
||
#undef BOOST_TT_int_pair
|
||
#undef BOOST_TT_basic_ints
|
||
</pre>
|
||
<!-- @compile() -->
|
||
<p><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_FOR_EACH</span></tt>
|
||
is a higher-order macro, similar to <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_REPEAT</span></tt>, that invokes its
|
||
first argument on each element of its third argument.</p>
|
||
<p>Sequences are the most efficient, most flexible, and
|
||
easiest-to-use of the library's data structures, provided that you
|
||
never need to make an empty one: An empty sequence would contain
|
||
no tokens, and so couldn't be passed as a macro argument. The
|
||
other data structures covered here all have an empty
|
||
representation.</p>
|
||
<p>The facilities for manipulating sequences are all in the
|
||
library's <tt class="docutils literal"><span class="pre">seq/</span></tt>
|
||
subdirectory. They are summarized in Table A.5, where <tt class="docutils literal"><span
|
||
|
||
class="pre">t</span></tt> is the sequence <tt class="docutils literal"><span
|
||
|
||
class="pre">(</span></tt><em>t</em><sub>0</sub><tt class="docutils literal"><span
|
||
|
||
class="pre">)(</span></tt><em>t</em><sub>1</sub><tt class="docutils literal"><span
|
||
|
||
class="pre">)...(</span></tt><em>t</em><sub>k</sub><tt class="docutils literal"><span
|
||
|
||
class="pre">)</span></tt>. Where <em>s</em>, <em>r</em>, and
|
||
<em>d</em> appear, they have a similar purpose to the <tt class="docutils literal"><span
|
||
|
||
class="pre">z</span></tt> parameters we discussed earlier (and
|
||
suggested you ignore for now).</p>
|
||
<table border="1" class="docutils">
|
||
<caption>Preprocessor Sequence Operations</caption> <colgroup> <col
|
||
|
||
width="51%" /> <col width="49%" /> </colgroup>
|
||
<thead valign="bottom">
|
||
<tr>
|
||
<th>Expression</th>
|
||
<th>Result</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_CAT(t)</span></tt></td>
|
||
<td><em>t</em><sub>0</sub><em>t</em><sub>1</sub>...<em>t</em><sub>k</sub></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_ELEM(n,t)</span></tt></td>
|
||
<td><em>t</em><sub>n</sub></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_ENUM(t)</span></tt></td>
|
||
<td><em>t</em><sub>0</sub>, <em>t</em><sub>1</sub>, ...<em>t</em><sub>k</sub></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_FILTER(pred,data,t)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">t</span></tt>
|
||
without the elements that don't satisfy <tt class="docutils literal"><span
|
||
|
||
class="pre">pred</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_FIRST_N(n,t)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(</span></tt><em>t</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt>...<tt
|
||
|
||
class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>n-1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_FOLD_LEFT(op,</span>
|
||
<span class="pre">x,</span> <span class="pre">t)</span></tt></td>
|
||
<td>...<tt class="docutils literal"><span class="pre">op(</span></tt><em>s</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,op(</span></tt><em>s</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,op(</span></tt><em>s</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,x</span></tt>,<em>t</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">),</span></tt><em>t</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">),</span></tt><em>t</em><sub>2</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt>...</td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_FOLD_RIGHT(op,</span>
|
||
<span class="pre">x,</span> <span class="pre">t)</span></tt></td>
|
||
<td>...<tt class="docutils literal"><span class="pre">op(</span></tt><em>s</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,op(</span></tt><em>s</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,op(</span></tt><em>s</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,x</span></tt>,<em>t</em><sub>k</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">),</span></tt><em>t</em><sub>k-1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">),</span></tt>
|
||
<em>t</em><sub>k-2</sub><tt class="docutils literal"><span class="pre">)</span></tt>...</td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_FOR_EACH(f,</span>
|
||
<span class="pre">x,</span> <span class="pre">t)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">f(</span></tt><em>r</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span> <span
|
||
|
||
class="pre">x,</span></tt><em>t</em><sub>0</sub><tt class="docutils literal"><span
|
||
|
||
class="pre">)</span> <span class="pre">f(</span></tt><em>r</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span> <span
|
||
|
||
class="pre">x,</span></tt><em>t</em><sub>1</sub><tt class="docutils literal"><span
|
||
|
||
class="pre">)</span></tt>...<tt class="docutils literal"><span
|
||
|
||
class="pre">f(</span></tt><em>r</em><tt class="docutils literal"><span
|
||
|
||
class="pre">,</span> <span class="pre">x,</span></tt><em>t</em><sub>k</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_FOR_EACH_I(g,</span>
|
||
<span class="pre">x,</span> <span class="pre">t)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">g(</span></tt><em>r</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span> <span
|
||
|
||
class="pre">x,</span> <span class="pre">0,</span></tt>
|
||
<em>t</em><sub>0</sub><tt class="docutils literal"><span class="pre">)</span>
|
||
<span class="pre">g(</span></tt><em>r</em><tt class="docutils literal"><span
|
||
|
||
class="pre">,</span> <span class="pre">x,</span> <span
|
||
|
||
class="pre">1,</span></tt> <em>t</em><sub>1</sub><tt class="docutils literal"><span
|
||
|
||
class="pre">)</span></tt>... <tt class="docutils literal"><span
|
||
|
||
class="pre">g(</span></tt><em>r</em><tt class="docutils literal"><span
|
||
|
||
class="pre">,</span> <span class="pre">x,</span> <span
|
||
|
||
class="pre">k,</span></tt> <em>t</em><sub>k</sub><tt class="docutils literal"><span
|
||
|
||
class="pre">)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_FOR_EACH_PRODUCT(h,</span>
|
||
<span class="pre">x,</span> <span class="pre">t)</span></tt></td>
|
||
<td>
|
||
<dl class="first last docutils">
|
||
<dt>Cartesian product—</dt>
|
||
<dd>see online docs</dd>
|
||
</dl>
|
||
</td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_INSERT(t,i,tokens)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(</span></tt><em>t</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt>...<tt
|
||
|
||
class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>i-1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(tokens)</span>
|
||
<span class="pre">(</span></tt><em>t</em><sub>i</sub><tt class="docutils literal"><span
|
||
|
||
class="pre">)(</span></tt><em>t</em><sub>i+1</sub><tt class="docutils literal"><span
|
||
|
||
class="pre">)</span></tt>...<tt class="docutils literal"><span
|
||
|
||
class="pre">(</span></tt><em>t</em><sub>k</sub><tt class="docutils literal"><span
|
||
|
||
class="pre">)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_POP_BACK(t)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(</span></tt><em>t</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt>...<tt
|
||
|
||
class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>k-1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_POP_FRONT(t)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(</span></tt><em>t</em><sub>2</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt>...<tt
|
||
|
||
class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>k</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_PUSH_BACK(t,tokens)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(</span></tt><em>t</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt>...<tt
|
||
|
||
class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>k</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(tokens)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_PUSH_FRONT(t,tokens)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(tokens)(</span></tt><em>t</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(</span></tt><em>t</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt>...<tt
|
||
|
||
class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>k</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_REMOVE(t,i)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(</span></tt><em>t</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt>...<tt
|
||
|
||
class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>i-1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(</span></tt><em>t</em><sub>i+1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt>...<tt
|
||
|
||
class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>k</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_REPLACE(t,i,tokens)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(</span></tt><em>t</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt>...<tt
|
||
|
||
class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>i-1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(tokens)(</span></tt><em>t</em><sub>i+1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt>...<tt
|
||
|
||
class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>k</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_REST_N(n,t)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>n</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(</span></tt><em>t</em><sub>n+1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt>...<tt
|
||
|
||
class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>k</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_REVERSE(t)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>k</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(</span></tt><em>t</em><sub>k-1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt>...<tt
|
||
|
||
class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_HEAD(t)</span></tt></td>
|
||
<td><em>t</em><sub>0</sub></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_TAIL(t)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(</span></tt><em>t</em><sub>2</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt>...<tt
|
||
|
||
class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>k</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_SIZE(t)</span></tt></td>
|
||
<td><em>k+1</em></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_SUBSEQ(t,i,m)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>i</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)(</span></tt><em>t</em><sub>i+1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt>...<tt
|
||
|
||
class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>i+m-1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_TO_ARRAY(t)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>k+1</em>
|
||
<tt class="docutils literal"><span class="pre">,(</span></tt><em>t</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt><em>t</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt>...<em>t</em><sub>k</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">))</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_TO_TUPLE(t)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>t</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt> <em>t</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt>...<em>t</em><sub>k</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_TRANSFORM(f,</span>
|
||
<span class="pre">x,</span> <span class="pre">t)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(f(</span></tt><em>r</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,x,</span></tt><em>t</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">))</span> <span
|
||
|
||
class="pre">(f(</span></tt><em>r</em><tt class="docutils literal"><span
|
||
|
||
class="pre">,x,</span></tt><em>t</em><sub>1</sub><tt class="docutils literal"><span
|
||
|
||
class="pre">))</span></tt>...<tt class="docutils literal"><span
|
||
|
||
class="pre">(f(</span></tt><em>r</em><tt class="docutils literal"><span
|
||
|
||
class="pre">,x,</span></tt><em>t</em><sub>k</sub><tt class="docutils literal"><span
|
||
|
||
class="pre">))</span></tt></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
<p>It's worth noting that while there is no upper limit on the
|
||
length of a sequence, operations such as <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_SEQ_ELEM</span></tt> that take numeric
|
||
arguments will only work with values up to 256.</p>
|
||
</div>
|
||
<div class="section" id="tuples">
|
||
<h3><a name="tuples">A.4.5.2 Tuples</a></h3>
|
||
<p>A <strong>tuple</strong> is a very simple data structure for
|
||
which the library provides random access and a few other basic
|
||
operations. A tuple takes the form of a parenthesized,
|
||
comma-separated list of <em>macro arguments</em>. For example,
|
||
this is a three-element tuple:</p>
|
||
<pre class="literal-block">#define TUPLE3 (f(12), a + 1, foo)
|
||
</pre>
|
||
<p>The operations in the library's <tt class="docutils literal"><span
|
||
|
||
class="pre">tuple/</span></tt> subdirectory can handle tuples
|
||
of up to 25 elements. For example, a tuple's <tt class="docutils literal"><span
|
||
|
||
class="pre">N</span></tt>th element can be accessed via <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_TUPLE_ELEM</span></tt>, as follows:</p>
|
||
<pre class="literal-block"> // length index tuple
|
||
BOOST_PP_TUPLE_ELEM( 3 , 1 , TUPLE3) // a + 1
|
||
</pre>
|
||
<!-- @def gen_id(id = 'a', hdr = 'tuple'):
|
||
example.wrap(''' #include <boost/preprocessor/%s.hpp> int const %s = 0; int const x =''' % (hdr,id), ';')
|
||
compile('all', pop = 1)gen_id() -->
|
||
<p>Notice that we had to pass the tuple's length as the second
|
||
argument to <tt class="docutils literal"><span class="pre">BOOST_PP_TUPLE_ELEM</span></tt>;
|
||
in fact, <em>all</em> tuple operations require explicit
|
||
specification of the tuple's length. We're not going to summarize
|
||
the other four operations in the "tuple" group here—you can
|
||
consult the Preprocessor library's electronic documentation for
|
||
more details. We note, however, that sequences can be transformed
|
||
into tuples with <tt class="docutils literal"><span class="pre">BOOST_PP_SEQ_TO_TUPLE</span></tt>,
|
||
and nonempty tuples can be transformed back into sequences with <tt
|
||
|
||
class="docutils literal"><span class="pre">BOOST_PP_TUPLE_TO_SEQ</span></tt>.</p>
|
||
<p>The greatest strength of tuples is that they conveniently take
|
||
the same representation as a macro argument list:</p>
|
||
<pre class="literal-block">#define FIRST_OF_THREE(a1,a2,a3) a1
|
||
#define SECOND_OF_THREE(a1,a2,a3) a2
|
||
#define THIRD_OF_THREE(a1,a2,a3) a3
|
||
|
||
// uses tuple as an argument list
|
||
# define SELECT(selector, tuple) <strong>selector tuple</strong>
|
||
|
||
SELECT(THIRD_OF_THREE, TUPLE3) // foo
|
||
</pre>
|
||
<!-- @gen_id('foo') --> </div>
|
||
<div class="section" id="arrays">
|
||
<h3><a name="arrays">A.4.5.3 Arrays</a></h3>
|
||
<p>An <strong>array</strong> is just a tuple containing a
|
||
non-negative integer and a tuple of that length:</p>
|
||
<pre class="literal-block">#define ARRAY3 ( 3, TUPLE3 )
|
||
</pre>
|
||
<p>Because an array carries its length around with it, the library's
|
||
interface for operating on arrays is much more convenient than the
|
||
one used for tuples:</p>
|
||
<pre class="literal-block">BOOST_PP_ARRAY_ELEM(1, ARRAY3) // a + 1
|
||
</pre>
|
||
<!-- @gen_id(hdr = 'array')
|
||
del stack[-2:] -->
|
||
<p>The facilities for manipulating arrays of up to 25 elements are
|
||
all in the library's <tt class="docutils literal"><span class="pre">array/</span></tt>
|
||
subdirectory. They are summarized in Table A.6, where <tt class="docutils literal"><span
|
||
|
||
class="pre">a</span></tt> is the array <tt class="docutils literal"><span
|
||
|
||
class="pre">(</span></tt><em>k</em><tt class="docutils literal"><span
|
||
|
||
class="pre">,</span> <span class="pre">(</span></tt><em>a</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt><em>a</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,...</span></tt><em>a</em><sub>k-1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">))</span></tt>.</p>
|
||
<table border="1" class="docutils">
|
||
<caption>Preprocessor Array Operations</caption> <colgroup> <col
|
||
|
||
width="52%" /> <col width="48%" /> </colgroup>
|
||
<thead valign="bottom">
|
||
<tr>
|
||
<th>Expression</th>
|
||
<th>Result</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody valign="top">
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_ARRAY_DATA(a)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>a</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt><em>a</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt>...
|
||
<em>a</em><sub>k-1</sub><tt class="docutils literal"><span class="pre">)</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_ARRAY_ELEM(i,a)</span></tt></td>
|
||
<td><em>a</em><sub>i</sub></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_ARRAY_INSERT(a,</span>
|
||
<span class="pre">i,</span> <span class="pre">tokens)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>k+1</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,(</span></tt><em>a</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt><em>a</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt>...<em>a</em><sub>i-1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span> <span
|
||
|
||
class="pre">tokens,</span></tt> <em>a</em><sub>i</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt><em>a</em><sub>i+1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt>...
|
||
<em>a</em><sub>k-1</sub><tt class="docutils literal"><span class="pre">))</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_ARRAY_POP_BACK(a)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>k-1</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,(</span></tt><em>a</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt><em>a</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt>...
|
||
<em>a</em><sub>k-2</sub><tt class="docutils literal"><span class="pre">))</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_ARRAY_POP_FRONT(a)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>k-1</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,(</span></tt><em>a</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt><em>a</em><sub>2</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt>...
|
||
<em>a</em><sub>k-1</sub><tt class="docutils literal"><span class="pre">))</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_ARRAY_PUSH_BACK(a,</span>
|
||
<span class="pre">tokens)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>k+1</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,(</span></tt><em>a</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt><em>a</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt>...
|
||
<em>a</em><sub>k-1</sub><tt class="docutils literal"><span class="pre">,</span>
|
||
<span class="pre">tokens))</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_ARRAY_PUSH_FRONT(a,</span>
|
||
<span class="pre">tokens)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>k+1</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,(tokens,</span></tt>
|
||
<em>a</em><sub>1</sub><tt class="docutils literal"><span class="pre">,</span></tt><em>a</em><sub>2</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt>...
|
||
<em>a</em><sub>k-1</sub><tt class="docutils literal"><span class="pre">))</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_ARRAY_REMOVE(a,</span>
|
||
<span class="pre">i)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>k-1</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,(</span></tt><em>a</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt><em>a</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt>...
|
||
<em>a</em><sub>i-1</sub><tt class="docutils literal"><span class="pre">,</span></tt><em>a</em><sub>i+1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt>...
|
||
<em>a</em><sub>k-1</sub><tt class="docutils literal"><span class="pre">))</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_ARRAY_REPLACE(a,</span>
|
||
<span class="pre">i,</span> <span class="pre">tokens)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>k</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,(</span></tt><em>a</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt><em>a</em><sub>1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt>...
|
||
<em>a</em><sub>i-1</sub><tt class="docutils literal"><span class="pre">,</span>
|
||
<span class="pre">tokens,</span></tt> <em>a</em><sub>i+1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt>...
|
||
<em>a</em><sub>k-1</sub><tt class="docutils literal"><span class="pre">))</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_ARRAY_REVERSE(a)</span></tt></td>
|
||
<td><tt class="docutils literal"><span class="pre">(</span></tt><em>k</em><tt
|
||
|
||
class="docutils literal"><span class="pre">,(</span></tt><em>a</em><sub>k-1</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt><em>a</em><sub>k-2</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">,</span></tt>...
|
||
<em>a</em><sub>1</sub><tt class="docutils literal"><span class="pre">,</span></tt><em>a</em><sub>0</sub><tt
|
||
|
||
class="docutils literal"><span class="pre">))</span></tt></td>
|
||
</tr>
|
||
<tr>
|
||
<td><tt class="docutils literal"><span class="pre">BOOST_PP_ARRAY_SIZE(a)</span></tt></td>
|
||
<td><em>k</em></td>
|
||
</tr>
|
||
</tbody>
|
||
</table>
|
||
</div>
|
||
<div class="section" id="lists">
|
||
<h3><a name="lists">A.4.5.4 Lists</a></h3>
|
||
<p>A <strong>list</strong> is a two-element tuple whose first
|
||
element is the first element of the list, and whose second element
|
||
is a list of the remaining elements, or <tt class="docutils literal"><span
|
||
|
||
class="pre">BOOST_PP_NIL</span></tt> if there are no remaining
|
||
elements. Lists have access characteristics similar to those of a
|
||
runtime linked list. Here is a three-element list:</p>
|
||
<pre class="literal-block">#define LIST3 (<strong>f(12)</strong>, (<strong>a + 1</strong>, (<strong>foo</strong>, BOOST_PP_NIL)))
|
||
</pre>
|
||
<!-- @ignore() -->
|
||
<p>The facilities for manipulating lists are all in the library's <tt
|
||
|
||
class="docutils literal"><span class="pre">list/</span></tt>
|
||
subdirectory. Because the operations are a subset of those
|
||
provided for sequences, we're not going to summarize them here—it
|
||
should be easy to understand the list operations by reading the
|
||
documentation on the basis of our coverage of sequences.</p>
|
||
<p>Like sequences, lists have no fixed upper length bound. Unlike
|
||
sequences, lists can also be empty. It's rare to need more than 25
|
||
elements in a preprocessor data structure, and lists tend to be
|
||
slower to manipulate and harder to read than any of the other
|
||
structures, so they should normally be used only as a last resort.</p>
|
||
</div>
|
||
</div>
|
||
</div>
|
||
<div class="section" id="exercise">
|
||
<h1><a name="exercise">A.5 Exercise</a></h1>
|
||
<dl class="docutils">
|
||
<dt>A-0</dt>
|
||
<dd>Fully preprocessor-ize the <tt class="docutils literal"><span class="pre">tiny</span></tt>
|
||
type sequence implemented in Chapter 5 so that all boilerplate code
|
||
is eliminated and the maximum size of a <tt class="docutils literal"><span
|
||
|
||
class="pre">tiny</span></tt> sequence can be adjusted by
|
||
changing <tt class="docutils literal"><span class="pre">TINY_MAX_SIZE</span></tt>.</dd>
|
||
</dl>
|
||
<!-- on hold:
|
||
It isn't uncommon to need token-wise arithmetic operations forpurposes other than invoking Preprocessor Library repetitionmacros. For example, let's write a metafunction to generate
|
||
function types from "signature" type sequences that specify thefunction's return and parameter types:: template <unsigned Size, class Signature>
|
||
struct to_function_impl; template <class Signature> struct to_function
|
||
: to_function_impl<mpl::size<Signature>::type, Signature> {};The challenge now is to implement ``to_function_impl``. For
|
||
``Size == 3``, an appropriate specialization might look like this:: template <class Signature> struct to_function_impl<3,Signature>
|
||
{ typedef mpl::begin<Signature>::type i0; typedef mpl::deref<i0>::type t0;
|
||
typedef mpl::next<i0>::type i1; typedef mpl::deref<i1>::type t1; typedef mpl::next<i1>::type i2;
|
||
typedef mpl::deref<i2>::type t2; typedef t0 type(t1,t2); };
|
||
A local macro to generate a single ``to_function_impl``specialization would look something like this:
|
||
.. parsed-literal:: #define to_function_impl_spec(size) \\ template <class Signature> \\
|
||
struct to_function_impl<3,Signature> \\ { \\ typedef mpl::begin<Signature>::type i0; \\ typedef mpl::deref<i0>::type t0; \\
|
||
\\ BOOST_PP_REPEAT_FROM_TO(1, size, to_function_t, ~) \\ \\ typedef t0 type(BOOST_PP_ENUM_SHIFTED_PARAMS(size,t)); \\
|
||
}; #define to_function_t(z, n, unused) \\ typedef mpl::next<BOOST_PP_CAT(i,\ **BOOST_PP_DEC(n)**)>::type \\
|
||
BOOST_PP_CAT(i,n); \\ \\ typedef mpl::deref<BOOST_PP_CAT(i,n)>::type BOOST_PP_CAT(t,n);
|
||
We've used some new library macros above; here is a brief rundown:* ``BOOST_PP_REPEAT_FROM_TO`` is just like ``BOOST_PP_REPEAT``, except that it accepts an initial repetition index. Since every
|
||
function has a return type, we don't need to worry about the case where ``Size == 0``.* ``BOOST_PP_ENUM_SHIFTED_PARAMS`` is just like
|
||
``BOOST_PP_ENUM_PARAMS``, except that repetition indices start at ``1`` instead of ``0``.* ``BOOST_PP_CAT`` implements token pasting; its two arguments are
|
||
"glued" together into a single token. Since this is a general-purpose macro, it sits in ``cat.hpp`` at the top level of the library's directory tree. [#paste]_
|
||
.. [#paste] The preprocessor's built-in token-pasting operator, ``##``, often yields surprising results by taking effect before its arguments have been expanded. By contrast, ``BOOST_PP_CAT`` delays concatenation until after its arguments have been fully
|
||
evaluated.* Finally, though it only performs trivial arithmetic, ``BOOST_PP_DEC`` plays a crucial role in generating an
|
||
appropriate prior iterator identifier for our own code in ``to_function_t``.If we didn't have ``BOOST_PP_REPEAT_FROM_TO`` at our disposal in
|
||
the previous example, we might've had to use ``BOOST_PP_REPEAT``,which always starts iterating at ``0``. Consequently``to_function_t`` would've been responsible for producing thedeclarations of ``i0`` and ``t0`` as well as those of the other
|
||
nested types. To manage that, it would need a way to selectdifferent expansions depending on the value of ``n``.In its ``control/`` directory, the Preprocessor Library supplies a
|
||
macro ``BOOST_PP_IF(c,t,f)`` that fulfills a similar role to theone filled by ``mpl::if_``. Rewriting the example accordingly, weget:
|
||
.. parsed-literal:: #define to_function_impl_spec(size) \\ template <class Signature> \\
|
||
struct to_function_impl<3,Signature> \\ { \\ BOOST_PP_REPEAT_FROM_TO(1, size, to_function_t, ~) \\ \\
|
||
typedef t0 type(BOOST_PP_ENUM_SHIFTED_PARAMS(size,t)); \\ }; #define to_function_t(z, n, unused) \\
|
||
typedef BOOST_PP_IF( \\ n, \\ mpl::next<BOOST_PP_CAT(i,BOOST_PP_DEC(n))>::type, \\ typedef mpl::begin<Signature>::type i0; \\
|
||
) \\ BOOST_PP_CAT(i,n); \\ \\ typedef mpl::deref<BOOST_PP_CAT(i,n)>::type BOOST_PP_CAT(t,n);
|
||
Although the formulation above will work, it does unnecessary workwhen ``n == 0``, evaluating the "true" branch of the conditionalonly to discard it. -->
|
||
</div>
|
||
</div>
|
||
<hr class="docutils footer" />
|
||
<div class="footer"> Generated on: 2005-10-17 19:34 UTC. Generated by <a class="reference"
|
||
|
||
href="http://docutils.sourceforge.net/">Docutils</a> from <a class="reference"
|
||
|
||
href="http://docutils.sourceforge.net/rst.html">reStructuredText</a>
|
||
source. </div>
|
||
</body>
|
||
</html>
|
||
<!--
|
||
FILE ARCHIVED ON 19:59:10 Mar 30, 2013 AND RETRIEVED FROM THE INTERNET ARCHIVE ON 21:06:19 May 19, 2015. JAVASCRIPT APPENDED BY WAYBACK MACHINE, COPYRIGHT INTERNET ARCHIVE.
|
||
ALL OTHER CONTENT MAY ALSO BE PROTECTED BY COPYRIGHT (17 U.S.C. SECTION 108(a)(3)).-->
|