boost/more
1
0
Fork 0
mirror of https://github.com/boostorg/more.git synced 2024-12-27 23:40:13 +08:00
Code Issues Packages Projects Releases Wiki Activity
021ed44c05
more/lib_guide.htm

411 lines
20 KiB
HTML
Raw Blame History

<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<title>Boost Library Requirements and Guidelines</title>
<meta name="GENERATOR" content="Microsoft FrontPage 5.0">
<meta name="ProgId" content="FrontPage.Editor.Document">
<meta name="Microsoft Border" content="none, default">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<table border="1" bgcolor="#007F7F" cellpadding="2">
<tr>
<td bgcolor="#FFFFFF"><img src="../c++boost.gif" alt="c++boost.gif (8819 bytes)" width="277" height="86"></td>
<td><a href="../index.htm"><font face="Arial" color="#FFFFFF"><big>Home</big></font></a></td>
<td><a href="../libs/libraries.htm"><font face="Arial" color="#FFFFFF"><big>Libraries</big></font></a></td>
<td><a href="../people/people.htm"><font face="Arial" color="#FFFFFF"><big>People</big></font></a></td>
<td><a href="faq.htm"><font face="Arial" color="#FFFFFF"><big>FAQ</big></font></a></td>
<td><a href="index.htm"><font face="Arial" color="#FFFFFF"><big>More</big></font></a></td>
</tr>
</table>
<h1 align="left">Boost Library Requirements and Guidelines</h1>
<p align="left"><a href="#Introduction">Introduction</a><br>
<a href="#Requirements">Requirements</a><br>
&nbsp;&nbsp;&nbsp; <a href="#License">License requirements</a><br>
&nbsp;&nbsp;&nbsp; <a href="#Portability">Portability requirements</a><br>
&nbsp;&nbsp;&nbsp; <a href="#Ownership">Ownership</a><br>
<a href="#Guidelines">Guidelines</a><br>
&nbsp;&nbsp;&nbsp; <a href="#Design and Programming">Design and programming</a><br>
&nbsp;&nbsp;&nbsp; <a href="#Design and Programming">Directory structure and
filenames</a><br>
&nbsp;&nbsp;&nbsp; <a href="#Naming consistency">Naming consistency</a><br>
&nbsp;&nbsp;&nbsp; <a href="#Documentation">Documentation</a><br>
<a href="#Rationale">Rationale</a><br>
&nbsp;&nbsp;&nbsp; <a href="#Exception-specification">Exception-specification
rationale</a><br>
&nbsp;&nbsp;&nbsp; <a href="#Naming">Naming conventions rationale</a><br>
&nbsp;&nbsp;&nbsp; <a href="#code fonts">Source code fonts rationale</a><br>
&nbsp;&nbsp;&nbsp; <a href="#Rationale rationale">Rationale rationale</a><br>
&nbsp;&nbsp;&nbsp; <a href="#Acknowledgements">Acknowledgements rationale</a></p>
<h2 align="left"><a name="Introduction">Introduction</a></h2>
<p align="left">This page describes requirements and guidelines for the content
of a library submitted to Boost.</p>
<p align="left">See the <a href="submission_process.htm">Boost Library
Submission Process</a> page for a description of the process involved.</p>
<h2 align="left"><a name="Requirements">Requirements</a></h2>
<p>To avoid the frustration and wasted time of a proposed library being
rejected, it must meets these requirements:</p>
<ul>
<li>The license must meet the <a href="#License">license requirements</a>
below. Restricted licenses like the GPL and LGPL are not acceptable.
</li>
<li>The
copyright <a href="#Ownership">ownership</a> must be clear.
</li>
<li>The library must be generally useful and not restricted to a narrow
problem domain.
</li>
<li>The library must meet the <a href="#Portability">portability requirements</a>
below.&nbsp;
</li>
<li>The library must come reasonably close to meeting the <a href="#Guidelines">Guidelines</a>
below.
<ul>
<li><a href="#Design and Programming">Design and Programming</a></li>
<li><a href="#Directory structure">Directory Structure</a></li>
<li><a href="#Documentation">Documentation</a></li>
</ul>
</li>
<li>The author must be willing to participate in discussions on the mailing
list, and to refine the library accordingly.</li>
</ul>
<p>There's no requirement that an author read the mailing list for a time before
making a submission. It has been noted, however, that submissions which begin
&quot;I just started to read this mailing list ...&quot; seem to fail, often
embarrassingly.</p>
<h3 align="left"><a name="License">License</a> requirements</h3>
<ul>
<li>Must be simple to read and understand.
</li>
<li>Must grant permission to copy, use and modify the software for any use
(commercial and non-commercial) for no fee.
</li>
<li>Must require that the license appear on all copies of the software source
code.
</li>
<li>Must not require that the license appear with executables or other binary
uses of the library.
</li>
<li>Must not require that the source code be
available for execution or other binary uses of the library.
</li>
<li>May restrict the use of the name and description of the library to the
standard version found on the Boost web site.</li>
</ul>
<h3 align="left"><a name="Portability">Portability</a> requirements</h3>
<ul>
<li>
<p align="left">A library's interface must portable and not restricted to a
particular compiler or operating system.
</li>
<li>
<p align="left">A library's implementation must if possible be portable and
not restricted to a particular compiler or operating system.&nbsp; If a
portable implementation is not possible, non-portable constructions are
acceptable if reasonably easy to port to other environments, and
implementations are provided for at least two popular operating systems
(such as UNIX and Windows).
</li>
<li>
<p align="left">There is no requirement that a library run on C++ compilers
which do not conform to the ISO standard.&nbsp;
</li>
<li>
<p align="left">There is no requirement that a library run on any particular
C++ compiler.&nbsp; Boost contributors often try to ensure their libraries
work with popular compilers.&nbsp; The boost/config.hpp <a href="../libs/config/config.htm">configuration
header</a> is the preferred mechanism for working around compiler
deficiencies.</li>
</ul>
<p align="left">Since there is no absolute way to prove portability, many boost
submissions demonstrate practical portability by compiling and executing
correctly with two different C++ compilers, often under different operating
systems.&nbsp; Otherwise reviewers may disbelieve that porting is in fact
practical.</p>
<h3 align="left"><a name="Ownership">Ownership</a></h3>
<p align="left">Are you sure you own the library you are thinking of
submitting?&nbsp;&nbsp; &quot;How to Copyright Software&quot; by MJ Salone, Nolo
Press, 1990 says:</p>
<blockquote>
<p align="left">Doing work on your own time that is very similar to
programming you do for your employer on company time can raise nasty legal
problems.&nbsp; In this situation, it's best to get a written release from
your employer in advance.</p>
</blockquote>
<p align="left">Place a copyright notice in all the important files you submit.
Boost won't accept libraries without clear copyright information.</p>
<h2 align="left"><a name="Guidelines">Guidelines</a></h2>
<p align="left">Please use these guidelines as a checklist for preparing the
content a library submission.&nbsp; Not every guideline applies to every
library, but a reasonable effort to comply is expected.</p>
<h3><a name="Design and Programming">Design and Programming</a></h3>
<ul>
<li>Aim first for clarity and correctness; optimization should be only a
secondary concern in most Boost libraries.</li>
</ul>
<ul>
<li>Aim for ISO Standard C++. Than means making effective use of the standard
features of the language, and avoiding non-standard compiler extensions. It
also means using the C++ Standard Library where applicable.</li>
</ul>
<ul>
<li>Headers should be good neighbors. See the <a href="header.htm">header
policy</a>. See <a href="#Naming consistency">Naming consistency</a>.</li>
</ul>
<ul>
<li>Follow quality programming practices. See, for example, &quot;Effective
C++&quot; 2nd Edition, and &quot;More Effective C++&quot;, both by Scott
Meyers, published by Addison Wesley.</li>
</ul>
<ul>
<li>Use the C++ Standard Library or other Boost libraries, but only when the
benefits outweigh the costs.&nbsp; Do not use libraries other than the C++
Standard Library or Boost. See <a href="library_reuse.htm">Library reuse</a>.</li>
</ul>
<ul>
<li>Read <a href="imp_vars.htm">Implementation Variation</a> to see how to
supply performance, platform, or other implementation variations.</li>
</ul>
<ul>
<li>Use the lowercase/underscore <a href="#Naming">naming conventions</a> of
the C++ standard library.&nbsp; Template parameter names begin with an
uppercase letter. Macro (gasp!) names should be all uppercase and begin with
BOOST_.</li>
</ul>
<ul>
<li>Choose meaningful names - explicit is better than implicit, and readability counts.
There is a strong preference for clear and descriptive names, even if
lengthy.</li>
</ul>
<ul>
<li>Use exceptions to report errors where appropriate, and write code that is
safe in the face of exceptions.</li>
</ul>
<ul>
<li>Avoid exception-specifications. See <a href="#Exception-specification">exception-specification
rationale</a>.</li>
</ul>
<ul>
<li>Provide sample programs or confidence tests so potential users can see how
to use your library.</li>
</ul>
<ul>
<li>Provide a regression test program or programs which follow the <a href="test_policy.htm">Test
Policies and Protocols</a>.</li>
</ul>
<ul>
<li>Although some boost members use proportional fonts, tabs, and unrestricted
line lengths in their own code, boost's widely distributed source code
should follow more conservative guidelines:
<ul>
<li>Use fixed-width fonts.&nbsp; See <a href="#code fonts">fonts rationale</a>.</li>
<li>Use spaces rather than tabs.</li>
<li>Limit line lengths to 80 characters.</li>
</ul>
</li>
</ul>
<ul>
<li>Begin all source files with:
<ul>
<li>A comment line describing the contents of the file.</li>
<li>Comments describing copyright and licensing.</li>
<li>A comment line referencing the Boost home page in the form:<br>
<code>// See http://www.boost.org for updates, documentation, and
revision history.</code><br>
[Including revision history in source files is no longer recommended;
the publicly available CVS repository better serves that purpose.]</li>
</ul>
</li>
</ul>
<h3><a name="Directory structure">Directory Structure</a> and Filenames</h3>
<ul>
<li>File and directory names must contain only lowercase ASCII letters ,
numbers, underscores, and a period.&nbsp; Leading character must be
alphabetic. Maximum length 31. Only a single period is permitted.&nbsp;
These requirements ensure file and directory names are relatively portable.</li>
<li>All libraries have at their highest level a primary directory named for
the particular library. See <a href="#Naming consistency">Naming consistency</a>. The primary directory may have sub-directories.</li>
<li>For very simple libraries implemented entirely within the library header,
all files go in the primary directory (except headers, which go in the boost
header directory).</li>
</ul>
<blockquote>
<p><b>Boost standard sub-directory names</b></p>
<table border="1" cellpadding="5">
<tr>
<td><b>Sub-directory</b></td>
<td><b>Contents</b></td>
<td><b>Required</b></td>
</tr>
<tr>
<td><code>build</code></td>
<td>Library build files such as a Jamfile.</td>
<td>If any build files.</td>
</tr>
<tr>
<td>doc</td>
<td>Documentation (HTML) files.</td>
<td>If several doc files.</td>
</tr>
<tr>
<td><code>example</code></td>
<td>Sample program files.</td>
<td>If several sample files.</td>
</tr>
<tr>
<td><code>src</code></td>
<td>Source files which must be compiled to build the library.&nbsp;</td>
<td>If any source files.</td>
</tr>
<tr>
<td><code>test</code></td>
<td>Regression or other test programs or scripts.</td>
<td>If several test files.</td>
</tr>
</table>
</blockquote>
<p>If the documentation is in a doc sub-directory, the primary directory should
contain an index.html file which redirects to the doc subdirectory:</p>
<blockquote>
<pre>&lt;html&gt;
&lt;head&gt;
&lt;meta http-equiv=&quot;refresh&quot; content=&quot;0; URL=doc/index.html&quot;&gt;
&lt;/head&gt;
&lt;body&gt;
Automatic redirection failed, please go to
&lt;a href=&quot;doc/index.html&quot;&gt;doc/index.html&lt;/a&gt;
&lt;/body&gt;
&lt;/html&gt;</pre>
</blockquote>
<h3><a name="Naming consistency">Naming consistency</a></h3>
<p>As library developers and users have gained experience with Boost, the
following consistent naming approach has come to be viewed as very helpful,
particularly for larger libraries which need their own header subdirectories and
namespaces.</p>
<p>Here is how it works. The library is given a name which describes the
contents of the library.&nbsp; Cryptic abbreviations are not acceptable.
Following the practice of the C++ Standard Library, names are usually singular
rather than plural.&nbsp; For example, a library dealing with file systems might
chose the name &quot;filesystem&quot;, but not &quot;filesystems&quot;, &quot;fs&quot; or &quot;nicecode&quot;.</p>
<ul>
<li>The library's primary directory (in parent <i>boost-root/libs</i>) is
given that same name.&nbsp; For example, <i>boost-root/libs/filesystem</i>.<br>
&nbsp;</li>
<li>The library's primary header directory (in parent <i>boost-root/boost</i>)
is given that same name. For example, <i>boost-root/boost/filesystem</i>.<br>
&nbsp;</li>
<li>The library's primary namespace (in parent <i>::boost</i>) is given that
same name. For example, <i>::boost::filesystem</i>.</li>
</ul>
<h3><a name="Documentation">Documentation</a></h3>
<p>Even the simplest library needs some documentation; the amount should be
proportional to the need.&nbsp; The documentation should assume the readers have
a basic knowledge of C++, but are not necessarily experts.</p>
<p>The format for documentation should be HTML, and should not require an
advanced browser or server-side extensions. The documentation entry point should
always be a file named index.html or index.htm.</p>
<p>There is no single right way to do documentation. HTML documentation is often
organized quite differently from traditional printed documents. Task-oriented
styles differ from reference oriented styles. In the end, it comes down to the
question: Is the documentation sufficient for the mythical &quot;average&quot;
C++ programmer to use the library successfully?</p>
<p>Appropriate topics for documentation often include:
<ul>
<li>General introduction to the library.</li>
<li>Description of each class.</li>
<li>Relationship between classes.</li>
<li>For each function, as applicable, description, requirements (preconditions),
effects, post-conditions, returns, and throws.</li>
<li>Discussion of error detection and recovery strategy.</li>
<li>How to use including description of typical uses.</li>
<li>How to compile and link.</li>
<li>How to test.</li>
<li>Version or revision history.</li>
<li>Rationale for design decisions.&nbsp; See <a href="#Rationale">Rationale
rationale</a>.</li>
<li>Acknowledgements.&nbsp; See <a href="#Acknowledgements">Acknowledgments
rationale.</a></li>
</ul>
<p>If you need more help with how to write documentation you can check out the
article on <a href="writingdoc/index.html">Writing Documentation for Boost</a>.</p>
<h2><a name="Rationale">Rationale</a></h2>
<p>Rationale for some of the requirements and guidelines follows.</p>
<hr>
<h3><a name="Exception-specification">Exception-specification</a> rationale</h3>
<p>Exception specifications [ISO 15.4] are sometimes coded to indicate what
exceptions may be thrown, or because the programmer hopes they will improved
performance.&nbsp; But consider the following member from a smart pointer:</p>
<pre> T&amp; operator*() const throw() { return *ptr; }</pre>
<p>This function calls no other functions; it only manipulates fundamental data
types like pointers Therefore, no runtime behavior of the
exception-specification can ever be invoked.&nbsp; The function is completely
exposed to the compiler; indeed it is declared inline Therefore, a smart
compiler can easily deduce that the functions are incapable of throwing
exceptions, and make the same optimizations it would have made based on the
empty exception-specification. A &quot;dumb&quot; compiler, however, may make
all kinds of pessimizations.</p>
<p>For example, some compilers turn off inlining if there is an
exception-specification.&nbsp; Some compilers add try/catch blocks. Such
pessimizations can be a performance disaster which makes the code unusable in
practical applications.</p>
<p>Although initially appealing, an exception-specification tends to have
consequences that require <b>very</b> careful thought to understand. The biggest
problem with exception-specifications is that programmers use them as though
they have the effect the programmer would like, instead of the effect they
actually have.</p>
<p>A non-inline function is the one place a &quot;throws nothing&quot;
exception-specification may have some benefit with some compilers.</p>
<hr>
<h3><a name="Naming">Naming</a> conventions rationale</h3>
<p>The C++ standard committee's Library Working Group discussed this issue in
detail, and over a long period of time. The discussion was repeated again in
early boost postings. A short summary:</p>
<ul>
<li>Naming conventions are contentious, and although several are widely used,
no one style predominates.
</li>
<li>Given the intent to propose portions of boost for the next revision of the
C++ standard library, boost decided to follow the standard library's
conventions.
</li>
<li>Once a library settles on a particular convention, a vast majority of
stakeholders want that style to be consistently used.
</li>
</ul>
<hr>
<h3>Source <a name="code fonts">code fonts</a> rationale</h3>
<p>Dave Abrahams comments: An important purpose (I daresay the primary purpose)
of source code is communication: the documentation of intent. This is a doubly
important goal for boost, I think. Using a fixed-width font allows us to
communicate with more people, in more ways (diagrams are possible) right there
in the source. Code written for fixed-width fonts using spaces will read
reasonably well when viewed with a variable-width font, and as far as I can tell
every editor supporting variable-width fonts also supports fixed width. I don't
think the converse is true.</p>
<hr>
<h3><a name="Rationale rationale">Rationale rationale</a></h3>
<p>Rationale is defined as &quot;The fundamental reasons for something;
basis.&quot; by the American Heritage Dictionary.</p>
<p>Beman Dawes comments:&nbsp; Failure to supply contemporaneous rationale for
design decisions is a major defect in many software projects. Lack of accurate
rationale causes issues to revisited endlessly, causes maintenance bugs when a
maintainer changes something without realizing it was done a certain way for
some purpose, and shortens the useful lifetime of software.</p>
<p>Rationale is fairly easy to provide at the time decisions are made, but very
hard to accurately recover even a short time later.</p>
<hr>
<h3><a name="Acknowledgements">Acknowledgements</a> rationale</h3>
<p>As a library matures, it almost always accumulates improvements suggested to
the authors by other boost members.&nbsp; It is a part of the culture of
boost.org to acknowledge such contributions, identifying the person making the
suggestion.&nbsp; Major contributions are usually acknowledged in the
documentation, while minor fixes are often mentioned in comments within the code
itself.</p>
<hr>
<p>Revised <!--webbot bot="Timestamp" s-type="EDITED" s-format="%d %B, %Y" startspan -->15 August, 2002<!--webbot bot="Timestamp" endspan i-checksum="34415" --></p>
</body>
</html>
Reference in New Issue View Git Blame Copy Permalink