Gitiles
Code Review Sign In
gerrit-public.fairphone.software / toolchain / llvm-project / ca46f5a3b5d75995b2a2807bc885f308e7009372 / . / llvm / docs / CodingStandards.html
blob: 11655e01bacebc069742d36993d7e154c18ea18d [file] [log] [blame]
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
2 "http://www.w3.org/TR/html4/strict.dtd">
3<html>
4<head>
5 <link rel="stylesheet" href="llvm.css" type="text/css">
Misha Brukman6c2522a2008-12-10 23:07:02 +0000[diff] [blame]6 <title>LLVM Coding Standards</title>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]7</head>
8<body>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]9
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]10<div class="doc_title">
Misha Brukman6c2522a2008-12-10 23:07:02 +0000[diff] [blame]11 LLVM Coding Standards
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]12</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]13
14<ol>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]15 <li><a href="#introduction">Introduction</a></li>
Chris Lattner85014f42001-07-23 20:40:41 +0000[diff] [blame]16 <li><a href="#mechanicalissues">Mechanical Source Issues</a>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]17 <ol>
18 <li><a href="#sourceformating">Source Code Formatting</a>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]19 <ol>
20 <li><a href="#scf_commenting">Commenting</a></li>
21 <li><a href="#scf_commentformat">Comment Formatting</a></li>
Misha Brukman03f87d52004-10-26 16:18:43 +0000[diff] [blame]22 <li><a href="#scf_includes"><tt>#include</tt> Style</a></li>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]23 <li><a href="#scf_codewidth">Source Code Width</a></li>
24 <li><a href="#scf_spacestabs">Use Spaces Instead of Tabs</a></li>
25 <li><a href="#scf_indentation">Indent Code Consistently</a></li>
26 </ol></li>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]27 <li><a href="#compilerissues">Compiler Issues</a>
28 <ol>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]29 <li><a href="#ci_warningerrors">Treat Compiler Warnings Like
30 Errors</a></li>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]31 <li><a href="#ci_portable_code">Write Portable Code</a></li>
Chris Lattner9db8e152010-11-30 19:20:40 +0000[diff] [blame]32 <li><a href="#ci_rtti_exceptions">Do not use RTTI or Exceptions</a></li>
Zhanyong Wan19b96fd2010-11-23 05:03:07 +0000[diff] [blame]33 <li><a href="#ci_class_struct">Use of <tt>class</tt>/<tt>struct</tt> Keywords</a></li>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]34 </ol></li>
35 </ol></li>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]36 <li><a href="#styleissues">Style Issues</a>
37 <ol>
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]38 <li><a href="#macro">The High-Level Issues</a>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]39 <ol>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]40 <li><a href="#hl_module">A Public Header File <b>is</b> a
41 Module</a></li>
Zhanyong Wan19b96fd2010-11-23 05:03:07 +0000[diff] [blame]42 <li><a href="#hl_dontinclude"><tt>#include</tt> as Little as Possible</a></li>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]43 <li><a href="#hl_privateheaders">Keep "internal" Headers
44 Private</a></li>
Zhanyong Wan19b96fd2010-11-23 05:03:07 +0000[diff] [blame]45 <li><a href="#hl_earlyexit">Use Early Exits and <tt>continue</tt> to Simplify
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]46 Code</a></li>
Zhanyong Wan19b96fd2010-11-23 05:03:07 +0000[diff] [blame]47 <li><a href="#hl_else_after_return">Don't use <tt>else</tt> after a
48 <tt>return</tt></a></li>
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]49 <li><a href="#hl_predicateloops">Turn Predicate Loops into Predicate
50 Functions</a></li>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]51 </ol></li>
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]52 <li><a href="#micro">The Low-Level Issues</a>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]53 <ol>
Zhanyong Wan4f3b26d2010-12-02 05:10:07 +0000[diff] [blame]54 <li><a href="#ll_naming">Name Types, Functions, Variables, and Enumerators Properly</a></li>
Chris Lattner85ea83e2006-07-27 04:24:14 +0000[diff] [blame]55 <li><a href="#ll_assert">Assert Liberally</a></li>
Zhanyong Wan19b96fd2010-11-23 05:03:07 +0000[diff] [blame]56 <li><a href="#ll_ns_std">Do not use '<tt>using namespace std</tt>'</a></li>
Bill Wendling2b52dc12006-12-09 01:20:34 +0000[diff] [blame]57 <li><a href="#ll_virtual_anch">Provide a virtual method anchor for
Bill Wendlingb0629482006-12-09 01:35:43 +0000[diff] [blame]58 classes in headers</a></li>
Zhanyong Wan19b96fd2010-11-23 05:03:07 +0000[diff] [blame]59 <li><a href="#ll_end">Don't evaluate <tt>end()</tt> every time through a
Chris Lattner56c10942009-06-30 06:13:23 +0000[diff] [blame]60 loop</a></li>
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]61 <li><a href="#ll_iostream"><tt>#include &lt;iostream&gt;</tt> is
62 <em>forbidden</em></a></li>
Chris Lattner02093e12010-12-10 00:54:03 +0000[diff] [blame]63 <li><a href="#ll_raw_ostream">Use <tt>raw_ostream</tt></a></li>
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]64 <li><a href="#ll_avoidendl">Avoid <tt>std::endl</tt></a></li>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]65 </ol></li>
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]66
67 <li><a href="#nano">Microscopic Details</a>
68 <ol>
69 <li><a href="#micro_spaceparen">Spaces Before Parentheses</a></li>
70 <li><a href="#micro_preincrement">Prefer Preincrement</a></li>
71 <li><a href="#micro_namespaceindent">Namespace Indentation</a></li>
72 <li><a href="#micro_anonns">Anonymous Namespaces</a></li>
73 </ol></li>
74
75
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]76 </ol></li>
77 <li><a href="#seealso">See Also</a></li>
78</ol>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]79
Chris Lattner1dab1922004-05-23 21:05:07 +0000[diff] [blame]80<div class="doc_author">
Chris Lattner9567fff2009-07-22 05:43:01 +0000[diff] [blame]81 <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a></p>
Chris Lattner1dab1922004-05-23 21:05:07 +0000[diff] [blame]82</div>
83
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]84
85<!-- *********************************************************************** -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]86<div class="doc_section">
87 <a name="introduction">Introduction</a>
88</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]89<!-- *********************************************************************** -->
90
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]91<div class="doc_text">
92
93<p>This document attempts to describe a few coding standards that are being used
94in the LLVM source tree. Although no coding standards should be regarded as
Misha Brukman0cedb1f2003-10-06 19:26:00 +0000[diff] [blame]95absolute requirements to be followed in all instances, coding standards can be
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]96useful.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]97
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]98<p>This document intentionally does not prescribe fixed standards for religious
Misha Brukman0cedb1f2003-10-06 19:26:00 +0000[diff] [blame]99issues such as brace placement and space usage. For issues like this, follow
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]100the golden rule:</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]101
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]102<blockquote>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]103
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]104<p><b><a name="goldenrule">If you are adding a significant body of source to a
105project, feel free to use whatever style you are most comfortable with. If you
106are extending, enhancing, or bug fixing already implemented code, use the style
107that is already being used so that the source is uniform and easy to
108follow.</a></b></p>
109
110</blockquote>
111
112<p>The ultimate goal of these guidelines is the increase readability and
Misha Brukman0cedb1f2003-10-06 19:26:00 +0000[diff] [blame]113maintainability of our common source base. If you have suggestions for topics to
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]114be included, please mail them to <a
115href="mailto:sabre@nondot.org">Chris</a>.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]116
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]117</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]118
119<!-- *********************************************************************** -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]120<div class="doc_section">
121 <a name="mechanicalissues">Mechanical Source Issues</a>
122</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]123<!-- *********************************************************************** -->
124
125<!-- ======================================================================= -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]126<div class="doc_subsection">
127 <a name="sourceformating">Source Code Formatting</a>
128</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]129
130<!-- _______________________________________________________________________ -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]131<div class="doc_subsubsection">
132 <a name="scf_commenting">Commenting</a>
133</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]134
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]135<div class="doc_text">
136
137<p>Comments are one critical part of readability and maintainability. Everyone
Chris Lattnerbbe228b2009-07-12 00:10:24 +0000[diff] [blame]138knows they should comment, so should you. When writing comments, write them as
139English prose, which means they should use proper capitalization, punctuation,
140etc. Although we all should probably
Misha Brukman0cedb1f2003-10-06 19:26:00 +0000[diff] [blame]141comment our code more than we do, there are a few very critical places that
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]142documentation is very useful:</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]143
Chris Lattner1dab1922004-05-23 21:05:07 +0000[diff] [blame]144<b>File Headers</b>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]145
Reid Spencerc7f87f22007-07-09 08:04:31 +0000[diff] [blame]146<p>Every source file should have a header on it that describes the basic
147purpose of the file. If a file does not have a header, it should not be
148checked into Subversion. Most source trees will probably have a standard
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]149file header format. The standard format for the LLVM source tree looks like
150this:</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]151
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]152<div class="doc_code">
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]153<pre>
Chris Lattnere6f4e072003-10-13 14:58:11 +0000[diff] [blame]154//===-- llvm/Instruction.h - Instruction class definition -------*- C++ -*-===//
Misha Brukman02805a62009-01-02 16:58:42 +0000[diff] [blame]155//
Chris Lattner1dab1922004-05-23 21:05:07 +0000[diff] [blame]156// The LLVM Compiler Infrastructure
157//
Chris Lattnerebf56662007-12-29 19:56:08 +0000[diff] [blame]158// This file is distributed under the University of Illinois Open Source
159// License. See LICENSE.TXT for details.
Misha Brukman02805a62009-01-02 16:58:42 +0000[diff] [blame]160//
Chris Lattner1dab1922004-05-23 21:05:07 +0000[diff] [blame]161//===----------------------------------------------------------------------===//
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]162//
163// This file contains the declaration of the Instruction class, which is the
164// base class for all of the VM instructions.
165//
166//===----------------------------------------------------------------------===//
167</pre>
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]168</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]169
Chris Lattnerebf56662007-12-29 19:56:08 +0000[diff] [blame]170<p>A few things to note about this particular format: The "<tt>-*- C++
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]171-*-</tt>" string on the first line is there to tell Emacs that the source file
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]172is a C++ file, not a C file (Emacs assumes <tt>.h</tt> files are C files by default).
173Note that this tag is not necessary in <tt>.cpp</tt> files. The name of the file is also
Misha Brukman0d640e82004-07-28 22:37:57 +0000[diff] [blame]174on the first line, along with a very short description of the purpose of the
175file. This is important when printing out code and flipping though lots of
176pages.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]177
Chris Lattnerebf56662007-12-29 19:56:08 +0000[diff] [blame]178<p>The next section in the file is a concise note that defines the license
179that the file is released under. This makes it perfectly clear what terms the
180source code can be distributed under and should not be modified in any way.</p>
Chris Lattner1dab1922004-05-23 21:05:07 +0000[diff] [blame]181
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]182<p>The main body of the description does not have to be very long in most cases.
Misha Brukman0cedb1f2003-10-06 19:26:00 +0000[diff] [blame]183Here it's only two lines. If an algorithm is being implemented or something
184tricky is going on, a reference to the paper where it is published should be
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]185included, as well as any notes or "gotchas" in the code to watch out for.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]186
Chris Lattner1dab1922004-05-23 21:05:07 +0000[diff] [blame]187<b>Class overviews</b>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]188
Jim Laskey88b5e792006-07-31 20:18:49 +0000[diff] [blame]189<p>Classes are one fundamental part of a good object oriented design. As such,
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]190a class definition should have a comment block that explains what the class is
Misha Brukman0cedb1f2003-10-06 19:26:00 +0000[diff] [blame]191used for... if it's not obvious. If it's so completely obvious your grandma
192could figure it out, it's probably safe to leave it out. Naming classes
Misha Brukman6d1686c2004-12-04 00:32:12 +0000[diff] [blame]193something sane goes a long ways towards avoiding writing documentation.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]194
195
Chris Lattner1dab1922004-05-23 21:05:07 +0000[diff] [blame]196<b>Method information</b>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]197
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]198<p>Methods defined in a class (as well as any global functions) should also be
Dan Gohman42f56912010-02-26 20:18:32 +0000[diff] [blame]199documented properly. A quick note about what it does and a description of the
Misha Brukman0cedb1f2003-10-06 19:26:00 +0000[diff] [blame]200borderline behaviour is all that is necessary here (unless something
Dan Gohman42f56912010-02-26 20:18:32 +0000[diff] [blame]201particularly tricky or insidious is going on). The hope is that people can
Misha Brukman0cedb1f2003-10-06 19:26:00 +0000[diff] [blame]202figure out how to use your interfaces without reading the code itself... that is
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]203the goal metric.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]204
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]205<p>Good things to talk about here are what happens when something unexpected
206happens: does the method return null? Abort? Format your hard disk?</p>
207
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]208</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]209
210<!-- _______________________________________________________________________ -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]211<div class="doc_subsubsection">
212 <a name="scf_commentformat">Comment Formatting</a>
213</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]214
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]215<div class="doc_text">
216
217<p>In general, prefer C++ style (<tt>//</tt>) comments. They take less space,
Misha Brukman0cedb1f2003-10-06 19:26:00 +0000[diff] [blame]218require less typing, don't have nesting problems, etc. There are a few cases
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]219when it is useful to use C style (<tt>/* */</tt>) comments however:</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]220
221<ol>
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]222 <li>When writing C code: Obviously if you are writing C code, use C style
Misha Brukman6d1686c2004-12-04 00:32:12 +0000[diff] [blame]223 comments.</li>
Misha Brukman03f87d52004-10-26 16:18:43 +0000[diff] [blame]224 <li>When writing a header file that may be <tt>#include</tt>d by a C source
225 file.</li>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]226 <li>When writing a source file that is used by a tool that only accepts C
227 style comments.</li>
228</ol>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]229
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]230<p>To comment out a large block of code, use <tt>#if 0</tt> and <tt>#endif</tt>.
231These nest properly and are better behaved in general than C style comments.</p>
232
233</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]234
Chris Lattner245b5252003-08-07 21:45:47 +0000[diff] [blame]235<!-- _______________________________________________________________________ -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]236<div class="doc_subsubsection">
Misha Brukman03f87d52004-10-26 16:18:43 +0000[diff] [blame]237 <a name="scf_includes"><tt>#include</tt> Style</a>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]238</div>
Chris Lattner245b5252003-08-07 21:45:47 +0000[diff] [blame]239
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]240<div class="doc_text">
241
242<p>Immediately after the <a href="#scf_commenting">header file comment</a> (and
Chris Lattner245b5252003-08-07 21:45:47 +0000[diff] [blame]243include guards if working on a header file), the <a
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]244href="#hl_dontinclude">minimal</a> list of <tt>#include</tt>s required by the
245file should be listed. We prefer these <tt>#include</tt>s to be listed in this
246order:</p>
Chris Lattner245b5252003-08-07 21:45:47 +0000[diff] [blame]247
248<ol>
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]249 <li><a href="#mmheader">Main Module Header</a></li>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]250 <li><a href="#hl_privateheaders">Local/Private Headers</a></li>
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]251 <li><tt>llvm/*</tt></li>
252 <li><tt>llvm/Analysis/*</tt></li>
253 <li><tt>llvm/Assembly/*</tt></li>
Duncan Sandse9da6db2010-09-29 20:09:55 +0000[diff] [blame]254 <li><tt>llvm/Bitcode/*</tt></li>
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]255 <li><tt>llvm/CodeGen/*</tt></li>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]256 <li>...</li>
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]257 <li><tt>Support/*</tt></li>
258 <li><tt>Config/*</tt></li>
259 <li>System <tt>#includes</tt></li>
Chris Lattner245b5252003-08-07 21:45:47 +0000[diff] [blame]260</ol>
261
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]262<p>and each category should be sorted by name.</p>
Chris Lattner245b5252003-08-07 21:45:47 +0000[diff] [blame]263
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]264<p><a name="mmheader">The "Main Module Header"</a> file applies to <tt>.cpp</tt> files
265which implement an interface defined by a <tt>.h</tt> file. This <tt>#include</tt>
Misha Brukman03f87d52004-10-26 16:18:43 +0000[diff] [blame]266should always be included <b>first</b> regardless of where it lives on the file
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]267system. By including a header file first in the <tt>.cpp</tt> files that implement the
Misha Brukman03f87d52004-10-26 16:18:43 +0000[diff] [blame]268interfaces, we ensure that the header does not have any hidden dependencies
269which are not explicitly #included in the header, but should be. It is also a
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]270form of documentation in the <tt>.cpp</tt> file to indicate where the interfaces it
Misha Brukman03f87d52004-10-26 16:18:43 +0000[diff] [blame]271implements are defined.</p>
Chris Lattner245b5252003-08-07 21:45:47 +0000[diff] [blame]272
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]273</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]274
275<!-- _______________________________________________________________________ -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]276<div class="doc_subsubsection">
277 <a name="scf_codewidth">Source Code Width</a>
278</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]279
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]280<div class="doc_text">
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]281
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]282<p>Write your code to fit within 80 columns of text. This helps those of us who
283like to print out code and look at your code in an xterm without resizing
284it.</p>
285
Chris Lattnerfc12d2e2008-07-08 05:12:37 +0000[diff] [blame]286<p>The longer answer is that there must be some limit to the width of the code
287in order to reasonably allow developers to have multiple files side-by-side in
288windows on a modest display. If you are going to pick a width limit, it is
289somewhat arbitrary but you might as well pick something standard. Going with
29090 columns (for example) instead of 80 columns wouldn't add any significant
291value and would be detrimental to printing out code. Also many other projects
292have standardized on 80 columns, so some people have already configured their
293editors for it (vs something else, like 90 columns).</p>
294
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]295<p>This is one of many contentious issues in coding standards, but it is not up
Chris Lattnerfc12d2e2008-07-08 05:12:37 +0000[diff] [blame]296for debate.</p>
297
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]298</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]299
300<!-- _______________________________________________________________________ -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]301<div class="doc_subsubsection">
302 <a name="scf_spacestabs">Use Spaces Instead of Tabs</a>
303</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]304
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]305<div class="doc_text">
306
307<p>In all cases, prefer spaces to tabs in source files. People have different
Benjamin Kramer0f420382009-10-12 14:46:08 +0000[diff] [blame]308preferred indentation levels, and different styles of indentation that they
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]309like; this is fine. What isn't fine is that different editors/viewers expand
310tabs out to different tab stops. This can cause your code to look completely
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]311unreadable, and it is not worth dealing with.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]312
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]313<p>As always, follow the <a href="#goldenrule">Golden Rule</a> above: follow the
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]314style of existing code if you are modifying and extending it. If you like four
Misha Brukman0cedb1f2003-10-06 19:26:00 +0000[diff] [blame]315spaces of indentation, <b>DO NOT</b> do that in the middle of a chunk of code
316with two spaces of indentation. Also, do not reindent a whole source file: it
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]317makes for incredible diffs that are absolutely worthless.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]318
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]319</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]320
321<!-- _______________________________________________________________________ -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]322<div class="doc_subsubsection">
323 <a name="scf_indentation">Indent Code Consistently</a>
324</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]325
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]326<div class="doc_text">
327
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]328<p>Okay, in your first year of programming you were told that indentation is
Misha Brukman0cedb1f2003-10-06 19:26:00 +0000[diff] [blame]329important. If you didn't believe and internalize this then, now is the time.
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]330Just do it.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]331
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]332</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]333
334
335<!-- ======================================================================= -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]336<div class="doc_subsection">
337 <a name="compilerissues">Compiler Issues</a>
338</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]339
340
341<!-- _______________________________________________________________________ -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]342<div class="doc_subsubsection">
343 <a name="ci_warningerrors">Treat Compiler Warnings Like Errors</a>
344</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]345
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]346<div class="doc_text">
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]347
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]348<p>If your code has compiler warnings in it, something is wrong &mdash; you
349aren't casting values correctly, your have "questionable" constructs in your
350code, or you are doing something legitimately wrong. Compiler warnings can
351cover up legitimate errors in output and make dealing with a translation unit
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]352difficult.</p>
353
354<p>It is not possible to prevent all warnings from all compilers, nor is it
Misha Brukman0cedb1f2003-10-06 19:26:00 +0000[diff] [blame]355desirable. Instead, pick a standard compiler (like <tt>gcc</tt>) that provides
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]356a good thorough set of warnings, and stick to it. At least in the case of
Misha Brukman0cedb1f2003-10-06 19:26:00 +0000[diff] [blame]357<tt>gcc</tt>, it is possible to work around any spurious errors by changing the
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]358syntax of the code slightly. For example, a warning that annoys me occurs when
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]359I write code like this:</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]360
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]361<div class="doc_code">
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]362<pre>
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]363if (V = getValue()) {
364 ...
365}
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]366</pre>
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]367</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]368
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]369<p><tt>gcc</tt> will warn me that I probably want to use the <tt>==</tt>
370operator, and that I probably mistyped it. In most cases, I haven't, and I
371really don't want the spurious errors. To fix this particular problem, I
372rewrite the code like this:</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]373
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]374<div class="doc_code">
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]375<pre>
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]376if ((V = getValue())) {
377 ...
378}
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]379</pre>
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]380</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]381
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]382<p>which shuts <tt>gcc</tt> up. Any <tt>gcc</tt> warning that annoys you can
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]383be fixed by massaging the code appropriately.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]384
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]385<p>These are the <tt>gcc</tt> warnings that I prefer to enable:</p>
386
387<div class="doc_code">
388<pre>
389-Wall -Winline -W -Wwrite-strings -Wno-unused
390</pre>
391</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]392
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]393</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]394
395<!-- _______________________________________________________________________ -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]396<div class="doc_subsubsection">
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]397 <a name="ci_portable_code">Write Portable Code</a>
398</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]399
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]400<div class="doc_text">
401
402<p>In almost all cases, it is possible and within reason to write completely
Misha Brukmanc3e78932003-07-28 21:57:18 +0000[diff] [blame]403portable code. If there are cases where it isn't possible to write portable
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]404code, isolate it behind a well defined (and well documented) interface.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]405
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]406<p>In practice, this means that you shouldn't assume much about the host
Chris Lattner44c22412010-11-16 22:19:06 +0000[diff] [blame]407compiler, and Visual Studio tends to be the lowest common denominator.
408If advanced features are used, they should only be an implementation detail of
409a library which has a simple exposed API, and preferably be buried in
410libSystem.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]411
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]412</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]413
Reid Spencerbf6439f2004-09-23 16:03:48 +0000[diff] [blame]414<!-- _______________________________________________________________________ -->
415<div class="doc_subsubsection">
Chris Lattner9db8e152010-11-30 19:20:40 +0000[diff] [blame]416<a name="ci_rtti_exceptions">Do not use RTTI or Exceptions</a>
417</div>
418<div class="doc_text">
419
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]420<p>In an effort to reduce code and executable size, LLVM does not use RTTI
421(e.g. <tt>dynamic_cast&lt;&gt;</tt>) or exceptions. These two language features
422violate the general C++ principle of <i>"you only pay for what you use"</i>,
423causing executable bloat even if exceptions are never used in the code base, or
424if RTTI is never used for a class. Because of this, we turn them off globally
425in the code.</p>
Chris Lattner9db8e152010-11-30 19:20:40 +0000[diff] [blame]426
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]427<p>That said, LLVM does make extensive use of a hand-rolled form of RTTI that
428use templates like <a href="ProgrammersManual.html#isa"><tt>isa&lt;&gt;</tt>,
429<tt>cast&lt;&gt;</tt>, and <tt>dyn_cast&lt;&gt;</tt></a>. This form of RTTI is
430opt-in and can be added to any class. It is also substantially more efficient
431than <tt>dynamic_cast&lt;&gt;</tt>.</p>
Chris Lattner9db8e152010-11-30 19:20:40 +0000[diff] [blame]432
433</div>
434
435<!-- _______________________________________________________________________ -->
436<div class="doc_subsubsection">
Reid Spencerbf6439f2004-09-23 16:03:48 +0000[diff] [blame]437<a name="ci_class_struct">Use of <tt>class</tt> and <tt>struct</tt> Keywords</a>
438</div>
439<div class="doc_text">
Misha Brukmanf2499132004-10-26 15:45:13 +0000[diff] [blame]440
441<p>In C++, the <tt>class</tt> and <tt>struct</tt> keywords can be used almost
442interchangeably. The only difference is when they are used to declare a class:
443<tt>class</tt> makes all members private by default while <tt>struct</tt> makes
444all members public by default.</p>
445
446<p>Unfortunately, not all compilers follow the rules and some will generate
447different symbols based on whether <tt>class</tt> or <tt>struct</tt> was used to
448declare the symbol. This can lead to problems at link time.</p>
449
450<p>So, the rule for LLVM is to always use the <tt>class</tt> keyword, unless
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]451<b>all</b> members are public and the type is a C++
452<a href="http://en.wikipedia.org/wiki/Plain_old_data_structure">POD</a> type, in
453which case <tt>struct</tt> is allowed.</p>
Misha Brukmanf2499132004-10-26 15:45:13 +0000[diff] [blame]454
Reid Spencerbf6439f2004-09-23 16:03:48 +0000[diff] [blame]455</div>
456
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]457<!-- *********************************************************************** -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]458<div class="doc_section">
459 <a name="styleissues">Style Issues</a>
460</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]461<!-- *********************************************************************** -->
462
463
464<!-- ======================================================================= -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]465<div class="doc_subsection">
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]466 <a name="macro">The High-Level Issues</a>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]467</div>
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]468<!-- ======================================================================= -->
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]469
470
471<!-- _______________________________________________________________________ -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]472<div class="doc_subsubsection">
473 <a name="hl_module">A Public Header File <b>is</b> a Module</a>
474</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]475
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]476<div class="doc_text">
477
478<p>C++ doesn't do too well in the modularity department. There is no real
Misha Brukmanc3e78932003-07-28 21:57:18 +0000[diff] [blame]479encapsulation or data hiding (unless you use expensive protocol classes), but it
480is what we have to work with. When you write a public header file (in the LLVM
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]481source tree, they live in the top level "<tt>include</tt>" directory), you are
482defining a module of functionality.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]483
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]484<p>Ideally, modules should be completely independent of each other, and their
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]485header files should only <tt>#include</tt> the absolute minimum number of
486headers possible. A module is not just a class, a function, or a
487namespace: <a href="http://www.cuj.com/articles/2000/0002/0002c/0002c.htm">it's
488a collection of these</a> that defines an interface. This interface may be
489several functions, classes, or data structures, but the important issue is how
490they work together.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]491
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]492<p>In general, a module should be implemented by one or more <tt>.cpp</tt>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]493files. Each of these <tt>.cpp</tt> files should include the header that defines
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]494their interface first. This ensures that all of the dependences of the module
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]495header have been properly added to the module header itself, and are not
496implicit. System headers should be included after user headers for a
497translation unit.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]498
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]499</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]500
501<!-- _______________________________________________________________________ -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]502<div class="doc_subsubsection">
Misha Brukman03f87d52004-10-26 16:18:43 +0000[diff] [blame]503 <a name="hl_dontinclude"><tt>#include</tt> as Little as Possible</a>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]504</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]505
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]506<div class="doc_text">
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]507
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]508<p><tt>#include</tt> hurts compile time performance. Don't do it unless you
509have to, especially in header files.</p>
510
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]511<p>But wait! Sometimes you need to have the definition of a class to use it, or
Misha Brukman03f87d52004-10-26 16:18:43 +0000[diff] [blame]512to inherit from it. In these cases go ahead and <tt>#include</tt> that header
513file. Be aware however that there are many cases where you don't need to have
514the full definition of a class. If you are using a pointer or reference to a
515class, you don't need the header file. If you are simply returning a class
516instance from a prototyped function or method, you don't need it. In fact, for
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]517most cases, you simply don't need the definition of a class. And not
Misha Brukman03f87d52004-10-26 16:18:43 +0000[diff] [blame]518<tt>#include</tt>'ing speeds up compilation.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]519
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]520<p>It is easy to try to go too overboard on this recommendation, however. You
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]521<b>must</b> include all of the header files that you are using &mdash; you can
522include them either directly or indirectly (through another header file). To
523make sure that you don't accidentally forget to include a header file in your
524module header, make sure to include your module header <b>first</b> in the
525implementation file (as mentioned above). This way there won't be any hidden
526dependencies that you'll find out about later.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]527
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]528</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]529
530<!-- _______________________________________________________________________ -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]531<div class="doc_subsubsection">
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]532 <a name="hl_privateheaders">Keep "Internal" Headers Private</a>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]533</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]534
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]535<div class="doc_text">
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]536
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]537<p>Many modules have a complex implementation that causes them to use more than
538one implementation (<tt>.cpp</tt>) file. It is often tempting to put the
539internal communication interface (helper classes, extra functions, etc) in the
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]540public module header file. Don't do this!</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]541
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]542<p>If you really need to do something like this, put a private header file in
543the same directory as the source files, and include it locally. This ensures
544that your private interface remains private and undisturbed by outsiders.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]545
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]546<p>Note however, that it's okay to put extra implementation methods in a public
547class itself. Just make them private (or protected) and all is well.</p>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]548
549</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]550
Bill Wendling2b52dc12006-12-09 01:20:34 +0000[diff] [blame]551<!-- _______________________________________________________________________ -->
552<div class="doc_subsubsection">
Zhanyong Wan19b96fd2010-11-23 05:03:07 +0000[diff] [blame]553 <a name="hl_earlyexit">Use Early Exits and <tt>continue</tt> to Simplify Code</a>
Bill Wendling2b52dc12006-12-09 01:20:34 +0000[diff] [blame]554</div>
555
556<div class="doc_text">
557
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]558<p>When reading code, keep in mind how much state and how many previous
559decisions have to be remembered by the reader to understand a block of code.
560Aim to reduce indentation where possible when it doesn't make it more difficult
561to understand the code. One great way to do this is by making use of early
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]562exits and the <tt>continue</tt> keyword in long loops. As an example of using
563an early exit from a function, consider this "bad" code:</p>
Bill Wendling2b52dc12006-12-09 01:20:34 +0000[diff] [blame]564
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]565<div class="doc_code">
566<pre>
567Value *DoSomething(Instruction *I) {
568 if (!isa&lt;TerminatorInst&gt;(I) &amp;&amp;
569 I-&gt;hasOneUse() &amp;&amp; SomeOtherThing(I)) {
570 ... some long code ....
571 }
572
573 return 0;
574}
575</pre>
576</div>
Bill Wendling2b52dc12006-12-09 01:20:34 +0000[diff] [blame]577
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]578<p>This code has several problems if the body of the '<tt>if</tt>' is large.
579When you're looking at the top of the function, it isn't immediately clear that
580this <em>only</em> does interesting things with non-terminator instructions, and
581only applies to things with the other predicates. Second, it is relatively
582difficult to describe (in comments) why these predicates are important because
583the <tt>if</tt> statement makes it difficult to lay out the comments. Third,
584when you're deep within the body of the code, it is indented an extra level.
585Finally, when reading the top of the function, it isn't clear what the result is
586if the predicate isn't true; you have to read to the end of the function to know
587that it returns null.</p>
Matthijs Kooijmanca4ffaa2008-07-30 12:14:10 +0000[diff] [blame]588
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]589<p>It is much preferred to format the code like this:</p>
590
591<div class="doc_code">
592<pre>
593Value *DoSomething(Instruction *I) {
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]594 // Terminators never need 'something' done to them because ...
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]595 if (isa&lt;TerminatorInst&gt;(I))
596 return 0;
597
598 // We conservatively avoid transforming instructions with multiple uses
599 // because goats like cheese.
600 if (!I-&gt;hasOneUse())
601 return 0;
602
603 // This is really just here for example.
604 if (!SomeOtherThing(I))
605 return 0;
606
607 ... some long code ....
608}
609</pre>
610</div>
611
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]612<p>This fixes these problems. A similar problem frequently happens in <tt>for</tt>
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]613loops. A silly example is something like this:</p>
614
615<div class="doc_code">
616<pre>
617 for (BasicBlock::iterator II = BB-&gt;begin(), E = BB-&gt;end(); II != E; ++II) {
618 if (BinaryOperator *BO = dyn_cast&lt;BinaryOperator&gt;(II)) {
619 Value *LHS = BO-&gt;getOperand(0);
620 Value *RHS = BO-&gt;getOperand(1);
621 if (LHS != RHS) {
622 ...
623 }
624 }
625 }
626</pre>
627</div>
628
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]629<p>When you have very, very small loops, this sort of structure is fine. But if
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]630it exceeds more than 10-15 lines, it becomes difficult for people to read and
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]631understand at a glance. The problem with this sort of code is that it gets very
632nested very quickly. Meaning that the reader of the code has to keep a lot of
633context in their brain to remember what is going immediately on in the loop,
634because they don't know if/when the <tt>if</tt> conditions will have elses etc.
635It is strongly preferred to structure the loop like this:</p>
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]636
637<div class="doc_code">
638<pre>
639 for (BasicBlock::iterator II = BB-&gt;begin(), E = BB-&gt;end(); II != E; ++II) {
640 BinaryOperator *BO = dyn_cast&lt;BinaryOperator&gt;(II);
641 if (!BO) continue;
642
643 Value *LHS = BO-&gt;getOperand(0);
644 Value *RHS = BO-&gt;getOperand(1);
645 if (LHS == RHS) continue;
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]646
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]647 ...
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]648 }
649</pre>
650</div>
651
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]652<p>This has all the benefits of using early exits for functions: it reduces
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]653nesting of the loop, it makes it easier to describe why the conditions are true,
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]654and it makes it obvious to the reader that there is no <tt>else</tt> coming up
655that they have to push context into their brain for. If a loop is large, this
656can be a big understandability win.</p>
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]657
658</div>
659
Chris Lattner14abb832009-07-28 22:54:04 +0000[diff] [blame]660<!-- _______________________________________________________________________ -->
661<div class="doc_subsubsection">
Zhanyong Wan19b96fd2010-11-23 05:03:07 +0000[diff] [blame]662 <a name="hl_else_after_return">Don't use <tt>else</tt> after a <tt>return</tt></a>
Chris Lattner14abb832009-07-28 22:54:04 +0000[diff] [blame]663</div>
664
665<div class="doc_text">
666
667<p>For similar reasons above (reduction of indentation and easier reading),
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]668please do not use '<tt>else</tt>' or '<tt>else if</tt>' after something that
669interrupts control flow &mdash; like <tt>return</tt>, <tt>break</tt>,
670<tt>continue</tt>, <tt>goto</tt>, etc. For example, this is <em>bad</em>:</p>
671
Chris Lattner14abb832009-07-28 22:54:04 +0000[diff] [blame]672<div class="doc_code">
673<pre>
674 case 'J': {
675 if (Signed) {
676 Type = Context.getsigjmp_bufType();
677 if (Type.isNull()) {
678 Error = ASTContext::GE_Missing_sigjmp_buf;
679 return QualType();
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]680 <b>} else {
Chris Lattner14abb832009-07-28 22:54:04 +0000[diff] [blame]681 break;
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]682 }</b>
Chris Lattner14abb832009-07-28 22:54:04 +0000[diff] [blame]683 } else {
684 Type = Context.getjmp_bufType();
685 if (Type.isNull()) {
686 Error = ASTContext::GE_Missing_jmp_buf;
687 return QualType();
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]688 <b>} else {
Chris Lattner14abb832009-07-28 22:54:04 +0000[diff] [blame]689 break;
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]690 }</b>
Chris Lattner14abb832009-07-28 22:54:04 +0000[diff] [blame]691 }
692 }
693 }
694</pre>
695</div>
696
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]697<p>It is better to write it like this:</p>
Chris Lattner14abb832009-07-28 22:54:04 +0000[diff] [blame]698
699<div class="doc_code">
700<pre>
701 case 'J':
702 if (Signed) {
703 Type = Context.getsigjmp_bufType();
704 if (Type.isNull()) {
705 Error = ASTContext::GE_Missing_sigjmp_buf;
706 return QualType();
707 }
708 } else {
709 Type = Context.getjmp_bufType();
710 if (Type.isNull()) {
711 Error = ASTContext::GE_Missing_jmp_buf;
712 return QualType();
713 }
714 }
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]715 <b>break;</b>
Chris Lattner14abb832009-07-28 22:54:04 +0000[diff] [blame]716</pre>
717</div>
718
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]719<p>Or better yet (in this case) as:</p>
Chris Lattner14abb832009-07-28 22:54:04 +0000[diff] [blame]720
721<div class="doc_code">
722<pre>
723 case 'J':
724 if (Signed)
725 Type = Context.getsigjmp_bufType();
726 else
727 Type = Context.getjmp_bufType();
728
729 if (Type.isNull()) {
730 Error = Signed ? ASTContext::GE_Missing_sigjmp_buf :
731 ASTContext::GE_Missing_jmp_buf;
732 return QualType();
733 }
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]734 <b>break;</b>
Chris Lattner14abb832009-07-28 22:54:04 +0000[diff] [blame]735</pre>
736</div>
737
738<p>The idea is to reduce indentation and the amount of code you have to keep
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]739track of when reading the code.</p>
Chris Lattner14abb832009-07-28 22:54:04 +0000[diff] [blame]740
741</div>
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]742
743<!-- _______________________________________________________________________ -->
744<div class="doc_subsubsection">
745 <a name="hl_predicateloops">Turn Predicate Loops into Predicate Functions</a>
746</div>
747
748<div class="doc_text">
749
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]750<p>It is very common to write small loops that just compute a boolean value.
751There are a number of ways that people commonly write these, but an example of
752this sort of thing is:</p>
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]753
754<div class="doc_code">
755<pre>
756 <b>bool FoundFoo = false;</b>
757 for (unsigned i = 0, e = BarList.size(); i != e; ++i)
758 if (BarList[i]-&gt;isFoo()) {
759 <b>FoundFoo = true;</b>
760 break;
761 }
762
763 <b>if (FoundFoo) {</b>
764 ...
765 }
766</pre>
767</div>
768
769<p>This sort of code is awkward to write, and is almost always a bad sign.
770Instead of this sort of loop, we strongly prefer to use a predicate function
771(which may be <a href="#micro_anonns">static</a>) that uses
Chris Lattnerd63f12372009-07-22 16:54:14 +0000[diff] [blame]772<a href="#hl_earlyexit">early exits</a> to compute the predicate. We prefer
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]773the code to be structured like this:</p>
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]774
775<div class="doc_code">
776<pre>
777/// ListContainsFoo - Return true if the specified list has an element that is
778/// a foo.
779static bool ListContainsFoo(const std::vector&lt;Bar*&gt; &amp;List) {
780 for (unsigned i = 0, e = List.size(); i != e; ++i)
781 if (List[i]-&gt;isFoo())
782 return true;
783 return false;
784}
785...
786
787 <b>if (ListContainsFoo(BarList)) {</b>
788 ...
789 }
790</pre>
791</div>
792
793<p>There are many reasons for doing this: it reduces indentation and factors out
794code which can often be shared by other code that checks for the same predicate.
795More importantly, it <em>forces you to pick a name</em> for the function, and
796forces you to write a comment for it. In this silly example, this doesn't add
797much value. However, if the condition is complex, this can make it a lot easier
798for the reader to understand the code that queries for this predicate. Instead
Chris Lattner611a1882009-07-22 16:30:39 +0000[diff] [blame]799of being faced with the in-line details of how we check to see if the BarList
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]800contains a foo, we can trust the function name and continue reading with better
801locality.</p>
Bill Wendling2b52dc12006-12-09 01:20:34 +0000[diff] [blame]802
803</div>
804
805
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]806<!-- ======================================================================= -->
Chris Lattner1dab1922004-05-23 21:05:07 +0000[diff] [blame]807<div class="doc_subsection">
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]808 <a name="micro">The Low-Level Issues</a>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]809</div>
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]810<!-- ======================================================================= -->
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]811
812
813<!-- _______________________________________________________________________ -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]814<div class="doc_subsubsection">
Zhanyong Wan4f3b26d2010-12-02 05:10:07 +0000[diff] [blame]815 <a name="ll_naming">Name Types, Functions, Variables, and Enumerators Properly</a>
816</div>
817
818<div class="doc_text">
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]819
820<p>Poorly-chosen names can mislead the reader and cause bugs. We cannot stress
821enough how important it is to use <em>descriptive</em> names. Pick names that
822match the semantics and role of the underlying entities, within reason. Avoid
823abbreviations unless they are well known. After picking a good name, make sure
824to use consistent capitalization for the name, as inconsistency requires clients
825to either memorize the APIs or to look it up to find the exact spelling.</p>
Zhanyong Wan4f3b26d2010-12-02 05:10:07 +0000[diff] [blame]826
Chris Lattner77c8cc82010-12-10 00:52:35 +0000[diff] [blame]827<p>In general, names should be in camel case (e.g. <tt>TextFileReader</tt>
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]828and <tt>isLValue()</tt>). Different kinds of declarations have different
829rules:</p>
Zhanyong Wan4f3b26d2010-12-02 05:10:07 +0000[diff] [blame]830
Chris Lattner77c8cc82010-12-10 00:52:35 +0000[diff] [blame]831<ul>
Chris Lattner02093e12010-12-10 00:54:03 +0000[diff] [blame]832<li><p><b>Type names</b> (including classes, structs, enums, typedefs, etc)
833 should be nouns and start with an upper-case letter (e.g.
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]834 <tt>TextFileReader</tt>).</p></li>
Chris Lattner77c8cc82010-12-10 00:52:35 +0000[diff] [blame]835
Chris Lattner02093e12010-12-10 00:54:03 +0000[diff] [blame]836<li><p><b>Function names</b> should be verb phrases (as they represent
Chris Lattner77c8cc82010-12-10 00:52:35 +0000[diff] [blame]837 actions), and command-like function should be imperative. The name should
838 be camel case, and start with a lower case letter (e.g. <tt>openFile()</tt>
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]839 or <tt>isFoo()</tt>).</p></li>
Chris Lattner77c8cc82010-12-10 00:52:35 +0000[diff] [blame]840
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]841<li><p><b>Enum declarations</b> (e.g. <tt>enum Foo {...}</tt>) are types, so
842 they should follow the naming conventions for types. A common use for enums
843 is as a discriminator for a union, or an indicator of a subclass. When an
844 enum is used for something like this, it should have a <tt>Kind</tt> suffix
845 (e.g. <tt>ValueKind</tt>).</p></li>
Chris Lattner77c8cc82010-12-10 00:52:35 +0000[diff] [blame]846
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]847<li><p><b>Enumerators</b> (e.g. <tt>enum { Foo, Bar }</tt>) and <b>public member
848 variables</b> should start with an upper-case letter, just like types.
849 Unless the enumerators are defined in their own small namespace or inside a
850 class, enumerators should have a prefix corresponding to the enum
851 declaration name. For example, <tt>enum ValueKind { ... };</tt> may contain
852 enumerators like <tt>VK_Argument</tt>, <tt>VK_BasicBlock</tt>, etc.
853 Enumerators that are just convenience constants are exempt from the
854 requirement for a prefix. For instance:</p>
855
Zhanyong Wan4f3b26d2010-12-02 05:10:07 +0000[diff] [blame]856<div class="doc_code">
857<pre>
858enum {
859 MaxSize = 42,
860 Density = 12
861};
862</pre>
863</div>
Chris Lattner77c8cc82010-12-10 00:52:35 +0000[diff] [blame]864</li>
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]865
Chris Lattner77c8cc82010-12-10 00:52:35 +0000[diff] [blame]866</ul>
867
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]868<p>As an exception, classes that mimic STL classes can have member names in
869STL's style of lower-case words separated by underscores (e.g. <tt>begin()</tt>,
870<tt>push_back()</tt>, and <tt>empty()</tt>).</p>
Chris Lattner77c8cc82010-12-10 00:52:35 +0000[diff] [blame]871
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]872<p>Here are some examples of good and bad names:</p>
Zhanyong Wan4f3b26d2010-12-02 05:10:07 +0000[diff] [blame]873
Zhanyong Wan4f3b26d2010-12-02 05:10:07 +0000[diff] [blame]874<div class="doc_code">
875<pre>
876class VehicleMaker {
877 ...
Chris Lattner77c8cc82010-12-10 00:52:35 +0000[diff] [blame]878 Factory&lt;Tire&gt; F; // Bad -- abbreviation and non-descriptive.
879 Factory&lt;Tire&gt; Factory; // Better.
880 Factory&lt;Tire&gt; TireFactory; // Even better -- if VehicleMaker has more than one
Zhanyong Wan4f3b26d2010-12-02 05:10:07 +0000[diff] [blame]881 // kind of factories.
882};
883
884Vehicle MakeVehicle(VehicleType Type) {
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]885 VehicleMaker M; // Might be OK if having a short life-span.
886 Tire tmp1 = M.makeTire(); // Bad -- 'tmp1' provides no information.
Chris Lattner77c8cc82010-12-10 00:52:35 +0000[diff] [blame]887 Light headlight = M.makeLight("head"); // Good -- descriptive.
Zhanyong Wan4f3b26d2010-12-02 05:10:07 +0000[diff] [blame]888 ...
889}
890</pre>
891</div>
892
893</div>
894
895
896<!-- _______________________________________________________________________ -->
897<div class="doc_subsubsection">
Chris Lattner85ea83e2006-07-27 04:24:14 +0000[diff] [blame]898 <a name="ll_assert">Assert Liberally</a>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]899</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]900
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]901<div class="doc_text">
902
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]903<p>Use the "<tt>assert</tt>" macro to its fullest. Check all of your
Benjamin Kramer0f420382009-10-12 14:46:08 +0000[diff] [blame]904preconditions and assumptions, you never know when a bug (not necessarily even
Misha Brukmanc3e78932003-07-28 21:57:18 +0000[diff] [blame]905yours) might be caught early by an assertion, which reduces debugging time
906dramatically. The "<tt>&lt;cassert&gt;</tt>" header file is probably already
907included by the header files you are using, so it doesn't cost anything to use
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]908it.</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]909
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]910<p>To further assist with debugging, make sure to put some kind of error message
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]911in the assertion statement, which is printed if the assertion is tripped. This
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]912helps the poor debugger make sense of why an assertion is being made and
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]913enforced, and hopefully what to do about it. Here is one complete example:</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]914
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]915<div class="doc_code">
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]916<pre>
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]917inline Value *getOperand(unsigned i) {
918 assert(i &lt; Operands.size() &amp;&amp; "getOperand() out of range!");
919 return Operands[i];
920}
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]921</pre>
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]922</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]923
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]924<p>Here are more examples:</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]925
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]926<div class="doc_code">
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]927<pre>
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]928assert(Ty-&gt;isPointerType() &amp;&amp; "Can't allocate a non pointer type!");
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]929
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]930assert((Opcode == Shl || Opcode == Shr) &amp;&amp; "ShiftInst Opcode invalid!");
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]931
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]932assert(idx &lt; getNumSuccessors() &amp;&amp; "Successor # out of range!");
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]933
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]934assert(V1.getType() == V2.getType() &amp;&amp; "Constant types must be identical!");
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]935
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]936assert(isa&lt;PHINode&gt;(Succ-&gt;front()) &amp;&amp; "Only works on PHId BBs!");
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]937</pre>
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]938</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]939
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]940<p>You get the idea.</p>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]941
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]942<p>Please be aware that, when adding assert statements, not all compilers are aware of
Nick Lewyckyfb75d422008-05-31 23:54:55 +0000[diff] [blame]943the semantics of the assert. In some places, asserts are used to indicate a piece of
944code that should not be reached. These are typically of the form:</p>
945
946<div class="doc_code">
947<pre>
Dan Gohman8ef44982008-11-24 17:18:39 +0000[diff] [blame]948assert(0 &amp;&amp; "Some helpful error message");
Nick Lewyckyfb75d422008-05-31 23:54:55 +0000[diff] [blame]949</pre>
950</div>
951
952<p>When used in a function that returns a value, they should be followed with a return
953statement and a comment indicating that this line is never reached. This will prevent
954a compiler which is unable to deduce that the assert statement never returns from
955generating a warning.</p>
956
957<div class="doc_code">
958<pre>
Dan Gohman8ef44982008-11-24 17:18:39 +0000[diff] [blame]959assert(0 &amp;&amp; "Some helpful error message");
Nick Lewyckyfb75d422008-05-31 23:54:55 +0000[diff] [blame]960// Not reached
961return 0;
962</pre>
963</div>
964
Chris Lattner8a7f4da2010-11-12 00:19:41 +0000[diff] [blame]965<p>Another issue is that values used only by assertions will produce an "unused
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]966value" warning when assertions are disabled. For example, this code will
967warn:</p>
Chris Lattner8a7f4da2010-11-12 00:19:41 +0000[diff] [blame]968
969<div class="doc_code">
970<pre>
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]971unsigned Size = V.size();
972assert(Size &gt; 42 &amp;&amp; "Vector smaller than it should be");
Chris Lattner8a7f4da2010-11-12 00:19:41 +0000[diff] [blame]973
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]974bool NewToSet = Myset.insert(Value);
975assert(NewToSet &amp;&amp; "The value shouldn't be in the set yet");
Chris Lattner8a7f4da2010-11-12 00:19:41 +0000[diff] [blame]976</pre>
977</div>
978
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]979<p>These are two interesting different cases. In the first case, the call to
Chris Lattner8a7f4da2010-11-12 00:19:41 +0000[diff] [blame]980V.size() is only useful for the assert, and we don't want it executed when
981assertions are disabled. Code like this should move the call into the assert
982itself. In the second case, the side effects of the call must happen whether
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]983the assert is enabled or not. In this case, the value should be cast to void to
984disable the warning. To be specific, it is preferred to write the code like
985this:</p>
Chris Lattner8a7f4da2010-11-12 00:19:41 +0000[diff] [blame]986
987<div class="doc_code">
988<pre>
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]989assert(V.size() &gt; 42 &amp;&amp; "Vector smaller than it should be");
Chris Lattner8a7f4da2010-11-12 00:19:41 +0000[diff] [blame]990
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]991bool NewToSet = Myset.insert(Value); (void)NewToSet;
992assert(NewToSet &amp;&amp; "The value shouldn't be in the set yet");
Chris Lattner8a7f4da2010-11-12 00:19:41 +0000[diff] [blame]993</pre>
994</div>
995
996
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]997</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]998
Chris Lattner01e81e62006-01-01 21:59:22 +0000[diff] [blame]999<!-- _______________________________________________________________________ -->
1000<div class="doc_subsubsection">
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1001 <a name="ll_ns_std">Do Not Use '<tt>using namespace std</tt>'</a>
Chris Lattner01e81e62006-01-01 21:59:22 +0000[diff] [blame]1002</div>
1003
1004<div class="doc_text">
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1005
Chris Lattner01e81e62006-01-01 21:59:22 +0000[diff] [blame]1006<p>In LLVM, we prefer to explicitly prefix all identifiers from the standard
Bill Wendlingb0629482006-12-09 01:35:43 +0000[diff] [blame]1007namespace with an "<tt>std::</tt>" prefix, rather than rely on
1008"<tt>using namespace std;</tt>".</p>
Chris Lattner01e81e62006-01-01 21:59:22 +0000[diff] [blame]1009
Bill Wendlingb0629482006-12-09 01:35:43 +0000[diff] [blame]1010<p> In header files, adding a '<tt>using namespace XXX</tt>' directive pollutes
Misha Brukman586a15c2008-12-11 19:37:04 +0000[diff] [blame]1011the namespace of any source file that <tt>#include</tt>s the header. This is
1012clearly a bad thing.</p>
Chris Lattner01e81e62006-01-01 21:59:22 +0000[diff] [blame]1013
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]1014<p>In implementation files (e.g. <tt>.cpp</tt> files), the rule is more of a stylistic
Bill Wendlingb0629482006-12-09 01:35:43 +0000[diff] [blame]1015rule, but is still important. Basically, using explicit namespace prefixes
1016makes the code <b>clearer</b>, because it is immediately obvious what facilities
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1017are being used and where they are coming from. And <b>more portable</b>, because
Bill Wendlingb0629482006-12-09 01:35:43 +0000[diff] [blame]1018namespace clashes cannot occur between LLVM code and other namespaces. The
1019portability rule is important because different standard library implementations
1020expose different symbols (potentially ones they shouldn't), and future revisions
1021to the C++ standard will add more symbols to the <tt>std</tt> namespace. As
1022such, we never use '<tt>using namespace std;</tt>' in LLVM.</p>
Chris Lattner01e81e62006-01-01 21:59:22 +0000[diff] [blame]1023
Bill Wendlingb0629482006-12-09 01:35:43 +0000[diff] [blame]1024<p>The exception to the general rule (i.e. it's not an exception for
1025the <tt>std</tt> namespace) is for implementation files. For example, all of
1026the code in the LLVM project implements code that lives in the 'llvm' namespace.
Chris Lattnereebbbeb2010-11-17 19:47:20 +0000[diff] [blame]1027As such, it is ok, and actually clearer, for the <tt>.cpp</tt> files to have a
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1028'<tt>using namespace llvm;</tt>' directive at the top, after the
Chris Lattnereebbbeb2010-11-17 19:47:20 +0000[diff] [blame]1029<tt>#include</tt>s. This reduces indentation in the body of the file for source
1030editors that indent based on braces, and keeps the conceptual context cleaner.
1031The general form of this rule is that any <tt>.cpp</tt> file that implements
1032code in any namespace may use that namespace (and its parents'), but should not
1033use any others.</p>
Chris Lattner01e81e62006-01-01 21:59:22 +0000[diff] [blame]1034
1035</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]1036
1037<!-- _______________________________________________________________________ -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1038<div class="doc_subsubsection">
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1039 <a name="ll_virtual_anch">Provide a Virtual Method Anchor for Classes
1040 in Headers</a>
Chris Lattner85ea83e2006-07-27 04:24:14 +0000[diff] [blame]1041</div>
1042
1043<div class="doc_text">
1044
1045<p>If a class is defined in a header file and has a v-table (either it has
1046virtual methods or it derives from classes with virtual methods), it must
1047always have at least one out-of-line virtual method in the class. Without
Misha Brukman586a15c2008-12-11 19:37:04 +0000[diff] [blame]1048this, the compiler will copy the vtable and RTTI into every <tt>.o</tt> file
1049that <tt>#include</tt>s the header, bloating <tt>.o</tt> file sizes and
1050increasing link times.</p>
Chris Lattner85ea83e2006-07-27 04:24:14 +0000[diff] [blame]1051
1052</div>
1053
Chris Lattner56c10942009-06-30 06:13:23 +0000[diff] [blame]1054<!-- _______________________________________________________________________ -->
1055<div class="doc_subsubsection">
Nick Lewycky9d0e8b62011-02-20 02:03:04 +0000[diff] [blame]1056 <a name="ll_end">Don't evaluate <tt>end()</tt> every time through a loop</a>
Chris Lattner56c10942009-06-30 06:13:23 +0000[diff] [blame]1057</div>
1058
1059<div class="doc_text">
1060
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1061<p>Because C++ doesn't have a standard "<tt>foreach</tt>" loop (though it can be
1062emulated with macros and may be coming in C++'0x) we end up writing a lot of
1063loops that manually iterate from begin to end on a variety of containers or
1064through other data structures. One common mistake is to write a loop in this
1065style:</p>
Chris Lattner56c10942009-06-30 06:13:23 +0000[diff] [blame]1066
1067<div class="doc_code">
1068<pre>
1069 BasicBlock *BB = ...
1070 for (BasicBlock::iterator I = BB->begin(); I != <b>BB->end()</b>; ++I)
1071 ... use I ...
1072</pre>
1073</div>
1074
1075<p>The problem with this construct is that it evaluates "<tt>BB->end()</tt>"
1076every time through the loop. Instead of writing the loop like this, we strongly
1077prefer loops to be written so that they evaluate it once before the loop starts.
1078A convenient way to do this is like so:</p>
1079
1080<div class="doc_code">
1081<pre>
1082 BasicBlock *BB = ...
1083 for (BasicBlock::iterator I = BB->begin(), E = <b>BB->end()</b>; I != E; ++I)
1084 ... use I ...
1085</pre>
1086</div>
1087
1088<p>The observant may quickly point out that these two loops may have different
1089semantics: if the container (a basic block in this case) is being mutated, then
1090"<tt>BB->end()</tt>" may change its value every time through the loop and the
1091second loop may not in fact be correct. If you actually do depend on this
Chris Lattner6ad089c2009-06-30 06:20:03 +0000[diff] [blame]1092behavior, please write the loop in the first form and add a comment indicating
Chris Lattner56c10942009-06-30 06:13:23 +0000[diff] [blame]1093that you did it intentionally.</p>
1094
1095<p>Why do we prefer the second form (when correct)? Writing the loop in the
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1096first form has two problems. First it may be less efficient than evaluating it
1097at the start of the loop. In this case, the cost is probably minor &mdash; a
1098few extra loads every time through the loop. However, if the base expression is
1099more complex, then the cost can rise quickly. I've seen loops where the end
Chris Lattner56c10942009-06-30 06:13:23 +0000[diff] [blame]1100expression was actually something like: "<tt>SomeMap[x]->end()</tt>" and map
Chris Lattner67f2a4ac2009-06-30 06:27:54 +0000[diff] [blame]1101lookups really aren't cheap. By writing it in the second form consistently, you
Chris Lattner56c10942009-06-30 06:13:23 +0000[diff] [blame]1102eliminate the issue entirely and don't even have to think about it.</p>
1103
Chris Lattner6ad089c2009-06-30 06:20:03 +0000[diff] [blame]1104<p>The second (even bigger) issue is that writing the loop in the first form
Chris Lattner56c10942009-06-30 06:13:23 +0000[diff] [blame]1105hints to the reader that the loop is mutating the container (a fact that a
1106comment would handily confirm!). If you write the loop in the second form, it
1107is immediately obvious without even looking at the body of the loop that the
1108container isn't being modified, which makes it easier to read the code and
1109understand what it does.</p>
1110
1111<p>While the second form of the loop is a few extra keystrokes, we do strongly
1112prefer it.</p>
1113
1114</div>
1115
Chris Lattner85ea83e2006-07-27 04:24:14 +0000[diff] [blame]1116<!-- _______________________________________________________________________ -->
1117<div class="doc_subsubsection">
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1118 <a name="ll_iostream"><tt>#include &lt;iostream&gt;</tt> is Forbidden</a>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1119</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]1120
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1121<div class="doc_text">
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]1122
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1123<p>The use of <tt>#include &lt;iostream&gt;</tt> in library files is
1124hereby <b><em>forbidden</em></b>. The primary reason for doing this is to
1125support clients using LLVM libraries as part of larger systems. In particular,
1126we statically link LLVM into some dynamic libraries. Even if LLVM isn't used,
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1127the static constructors are run whenever an application starts up that uses the
1128dynamic library. There are two problems with this:</p>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1129
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1130<ol>
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1131 <li>The time to run the static c'tors impacts startup time of applications
1132 &mdash; a critical time for GUI apps.</li>
1133
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1134 <li>The static c'tors cause the app to pull many extra pages of memory off the
1135 disk: both the code for the static c'tors in each <tt>.o</tt> file and the
1136 small amount of data that gets touched. In addition, touched/dirty pages
1137 put more pressure on the VM system on low-memory machines.</li>
1138</ol>
1139
1140<p>Note that using the other stream headers (<tt>&lt;sstream&gt;</tt> for
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1141example) is not problematic in this regard &mdash;
1142just <tt>&lt;iostream&gt;</tt>. However, <tt>raw_ostream</tt> provides various
1143APIs that are better performing for almost every use than <tt>std::ostream</tt>
1144style APIs. <b>Therefore new code should always
Chris Lattner5db2ff52009-08-23 21:53:47 +0000[diff] [blame]1145use <a href="#ll_raw_ostream"><tt>raw_ostream</tt></a> for writing, or
1146the <tt>llvm::MemoryBuffer</tt> API for reading files.</b></p>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1147
1148</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]1149
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1150
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]1151<!-- _______________________________________________________________________ -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1152<div class="doc_subsubsection">
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]1153 <a name="ll_raw_ostream">Use <tt>raw_ostream</tt></a>
1154</div>
1155
1156<div class="doc_text">
1157
1158<p>LLVM includes a lightweight, simple, and efficient stream implementation
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1159in <tt>llvm/Support/raw_ostream.h</tt>, which provides all of the common
1160features of <tt>std::ostream</tt>. All new code should use <tt>raw_ostream</tt>
1161instead of <tt>ostream</tt>.</p>
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]1162
1163<p>Unlike <tt>std::ostream</tt>, <tt>raw_ostream</tt> is not a template and can
1164be forward declared as <tt>class raw_ostream</tt>. Public headers should
1165generally not include the <tt>raw_ostream</tt> header, but use forward
1166declarations and constant references to <tt>raw_ostream</tt> instances.</p>
1167
1168</div>
1169
1170
1171<!-- _______________________________________________________________________ -->
1172<div class="doc_subsubsection">
Chris Lattner85ea83e2006-07-27 04:24:14 +0000[diff] [blame]1173 <a name="ll_avoidendl">Avoid <tt>std::endl</tt></a>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1174</div>
Chris Lattner850d4f62002-01-20 19:01:26 +0000[diff] [blame]1175
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1176<div class="doc_text">
1177
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1178<p>The <tt>std::endl</tt> modifier, when used with <tt>iostreams</tt> outputs a
1179newline to the output stream specified. In addition to doing this, however, it
1180also flushes the output stream. In other words, these are equivalent:</p>
Chris Lattner850d4f62002-01-20 19:01:26 +0000[diff] [blame]1181
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]1182<div class="doc_code">
Chris Lattner850d4f62002-01-20 19:01:26 +0000[diff] [blame]1183<pre>
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]1184std::cout &lt;&lt; std::endl;
1185std::cout &lt;&lt; '\n' &lt;&lt; std::flush;
Chris Lattner850d4f62002-01-20 19:01:26 +0000[diff] [blame]1186</pre>
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]1187</div>
Chris Lattner850d4f62002-01-20 19:01:26 +0000[diff] [blame]1188
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1189<p>Most of the time, you probably have no reason to flush the output stream, so
Misha Brukmanb2246152004-07-28 22:31:54 +0000[diff] [blame]1190it's better to use a literal <tt>'\n'</tt>.</p>
Chris Lattner850d4f62002-01-20 19:01:26 +0000[diff] [blame]1191
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1192</div>
Chris Lattner850d4f62002-01-20 19:01:26 +0000[diff] [blame]1193
Bill Wendling2b52dc12006-12-09 01:20:34 +0000[diff] [blame]1194
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1195<!-- ======================================================================= -->
1196<div class="doc_subsection">
1197 <a name="nano">Microscopic Details</a>
1198</div>
1199<!-- ======================================================================= -->
1200
1201<p>This section describes preferred low-level formatting guidelines along with
1202reasoning on why we prefer them.</p>
1203
1204<!-- _______________________________________________________________________ -->
1205<div class="doc_subsubsection">
1206 <a name="micro_spaceparen">Spaces Before Parentheses</a>
1207</div>
1208
1209<div class="doc_text">
1210
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]1211<p>We prefer to put a space before an open parenthesis only in control flow
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1212statements, but not in normal function call expressions and function-like
1213macros. For example, this is good:</p>
1214
1215<div class="doc_code">
1216<pre>
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1217<b>if (</b>x) ...
1218<b>for (</b>i = 0; i != 100; ++i) ...
1219<b>while (</b>llvm_rocks) ...
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1220
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1221<b>somefunc(</b>42);
1222<b><a href="#ll_assert">assert</a>(</b>3 != 4 &amp;&amp; "laws of math are failing me");
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1223
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1224a = <b>foo(</b>42, 92) + <b>bar(</b>x);
1225</pre>
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1226</div>
1227
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1228<p>and this is bad:</p>
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1229
1230<div class="doc_code">
1231<pre>
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1232<b>if(</b>x) ...
1233<b>for(</b>i = 0; i != 100; ++i) ...
1234<b>while(</b>llvm_rocks) ...
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1235
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1236<b>somefunc (</b>42);
1237<b><a href="#ll_assert">assert</a> (</b>3 != 4 &amp;&amp; "laws of math are failing me");
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1238
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1239a = <b>foo (</b>42, 92) + <b>bar (</b>x);
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1240</pre>
1241</div>
1242
1243<p>The reason for doing this is not completely arbitrary. This style makes
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1244control flow operators stand out more, and makes expressions flow better. The
1245function call operator binds very tightly as a postfix operator. Putting a
1246space after a function name (as in the last example) makes it appear that the
1247code might bind the arguments of the left-hand-side of a binary operator with
1248the argument list of a function and the name of the right side. More
1249specifically, it is easy to misread the "a" example as:</p>
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1250
1251<div class="doc_code">
1252<pre>
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1253a = foo <b>(</b>(42, 92) + bar<b>)</b> (x);
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1254</pre>
1255</div>
1256
Bill Wendling510504a2010-12-21 03:31:05 +0000[diff] [blame]1257<p>when skimming through the code. By avoiding a space in a function, we avoid
1258this misinterpretation.</p>
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1259
1260</div>
1261
1262<!-- _______________________________________________________________________ -->
1263<div class="doc_subsubsection">
1264 <a name="micro_preincrement">Prefer Preincrement</a>
1265</div>
1266
1267<div class="doc_text">
1268
1269<p>Hard fast rule: Preincrement (<tt>++X</tt>) may be no slower than
1270postincrement (<tt>X++</tt>) and could very well be a lot faster than it. Use
1271preincrementation whenever possible.</p>
1272
1273<p>The semantics of postincrement include making a copy of the value being
1274incremented, returning it, and then preincrementing the "work value". For
1275primitive types, this isn't a big deal... but for iterators, it can be a huge
1276issue (for example, some iterators contains stack and set objects in them...
1277copying an iterator could invoke the copy ctor's of these as well). In general,
1278get in the habit of always using preincrement, and you won't have a problem.</p>
1279
1280</div>
1281
1282<!-- _______________________________________________________________________ -->
1283<div class="doc_subsubsection">
1284 <a name="micro_namespaceindent">Namespace Indentation</a>
1285</div>
1286
1287<div class="doc_text">
1288
1289<p>
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]1290In general, we strive to reduce indentation wherever possible. This is useful
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1291because we want code to <a href="#scf_codewidth">fit into 80 columns</a> without
1292wrapping horribly, but also because it makes it easier to understand the code.
1293Namespaces are a funny thing: they are often large, and we often desire to put
1294lots of stuff into them (so they can be large). Other times they are tiny,
1295because they just hold an enum or something similar. In order to balance this,
1296we use different approaches for small versus large namespaces.
1297</p>
1298
1299<p>
1300If a namespace definition is small and <em>easily</em> fits on a screen (say,
1301less than 35 lines of code), then you should indent its body. Here's an
1302example:
1303</p>
1304
1305<div class="doc_code">
1306<pre>
Chris Lattnerd63f12372009-07-22 16:54:14 +0000[diff] [blame]1307namespace llvm {
1308 namespace X86 {
1309 /// RelocationType - An enum for the x86 relocation codes. Note that
1310 /// the terminology here doesn't follow x86 convention - word means
1311 /// 32-bit and dword means 64-bit.
1312 enum RelocationType {
1313 /// reloc_pcrel_word - PC relative relocation, add the relocated value to
1314 /// the value already in memory, after we adjust it for where the PC is.
1315 reloc_pcrel_word = 0,
1316
1317 /// reloc_picrel_word - PIC base relative relocation, add the relocated
1318 /// value to the value already in memory, after we adjust it for where the
1319 /// PIC base is.
1320 reloc_picrel_word = 1,
1321
1322 /// reloc_absolute_word, reloc_absolute_dword - Absolute relocation, just
1323 /// add the relocated value to the value already in memory.
1324 reloc_absolute_word = 2,
1325 reloc_absolute_dword = 3
1326 };
1327 }
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1328}
1329</pre>
1330</div>
1331
1332<p>Since the body is small, indenting adds value because it makes it very clear
1333where the namespace starts and ends, and it is easy to take the whole thing in
1334in one "gulp" when reading the code. If the blob of code in the namespace is
Chris Lattner8defde02010-11-17 17:14:55 +0000[diff] [blame]1335larger (as it typically is in a header in the <tt>llvm</tt> or <tt>clang</tt> namespaces), do not
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1336indent the code, and add a comment indicating what namespace is being closed.
1337For example:</p>
1338
1339<div class="doc_code">
1340<pre>
1341namespace llvm {
1342namespace knowledge {
1343
1344/// Grokable - This class represents things that Smith can have an intimate
1345/// understanding of and contains the data associated with it.
1346class Grokable {
1347...
1348public:
1349 explicit Grokable() { ... }
1350 virtual ~Grokable() = 0;
1351
1352 ...
1353
1354};
1355
1356} // end namespace knowledge
1357} // end namespace llvm
1358</pre>
1359</div>
1360
1361<p>Because the class is large, we don't expect that the reader can easily
1362understand the entire concept in a glance, and the end of the file (where the
1363namespaces end) may be a long ways away from the place they open. As such,
1364indenting the contents of the namespace doesn't add any value, and detracts from
1365the readability of the class. In these cases it is best to <em>not</em> indent
1366the contents of the namespace.</p>
1367
1368</div>
1369
1370<!-- _______________________________________________________________________ -->
1371<div class="doc_subsubsection">
1372 <a name="micro_anonns">Anonymous Namespaces</a>
1373</div>
1374
1375<div class="doc_text">
1376
Chris Lattnerd63f12372009-07-22 16:54:14 +0000[diff] [blame]1377<p>After talking about namespaces in general, you may be wondering about
1378anonymous namespaces in particular.
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1379Anonymous namespaces are a great language feature that tells the C++ compiler
1380that the contents of the namespace are only visible within the current
1381translation unit, allowing more aggressive optimization and eliminating the
1382possibility of symbol name collisions. Anonymous namespaces are to C++ as
1383"static" is to C functions and global variables. While "static" is available
1384in C++, anonymous namespaces are more general: they can make entire classes
1385private to a file.</p>
1386
1387<p>The problem with anonymous namespaces is that they naturally want to
1388encourage indentation of their body, and they reduce locality of reference: if
1389you see a random function definition in a C++ file, it is easy to see if it is
1390marked static, but seeing if it is in an anonymous namespace requires scanning
1391a big chunk of the file.</p>
1392
1393<p>Because of this, we have a simple guideline: make anonymous namespaces as
1394small as possible, and only use them for class declarations. For example, this
1395is good:</p>
1396
1397<div class="doc_code">
1398<pre>
1399<b>namespace {</b>
1400 class StringSort {
1401 ...
1402 public:
1403 StringSort(...)
1404 bool operator&lt;(const char *RHS) const;
1405 };
1406<b>} // end anonymous namespace</b>
1407
1408static void Helper() {
1409 ...
1410}
1411
1412bool StringSort::operator&lt;(const char *RHS) const {
1413 ...
1414}
1415
1416</pre>
1417</div>
1418
1419<p>This is bad:</p>
1420
1421
1422<div class="doc_code">
1423<pre>
1424<b>namespace {</b>
1425class StringSort {
1426...
1427public:
1428 StringSort(...)
1429 bool operator&lt;(const char *RHS) const;
1430};
1431
1432void Helper() {
1433 ...
1434}
1435
1436bool StringSort::operator&lt;(const char *RHS) const {
1437 ...
1438}
1439
1440<b>} // end anonymous namespace</b>
1441
1442</pre>
1443</div>
1444
1445
1446<p>This is bad specifically because if you're looking at "Helper" in the middle
1447of a large C++ file, that you have no immediate way to tell if it is local to
1448the file. When it is marked static explicitly, this is immediately obvious.
1449Also, there is no reason to enclose the definition of "operator&lt;" in the
Chris Lattnerd63f12372009-07-22 16:54:14 +0000[diff] [blame]1450namespace just because it was declared there.
Chris Lattner6720d7e2009-07-22 05:40:54 +0000[diff] [blame]1451</p>
1452
1453</div>
1454
1455
1456
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]1457<!-- *********************************************************************** -->
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1458<div class="doc_section">
1459 <a name="seealso">See Also</a>
1460</div>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]1461<!-- *********************************************************************** -->
1462
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1463<div class="doc_text">
1464
1465<p>A lot of these comments and recommendations have been culled for other
1466sources. Two particularly important books for our work are:</p>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]1467
1468<ol>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]1469
Chris Lattnerf9f08bf2007-11-09 21:49:08 +0000[diff] [blame]1470<li><a href="http://www.amazon.com/Effective-Specific-Addison-Wesley-Professional-Computing/dp/0321334876">Effective
1471C++</a> by Scott Meyers. Also
Chris Lattner1dab1922004-05-23 21:05:07 +0000[diff] [blame]1472interesting and useful are "More Effective C++" and "Effective STL" by the same
1473author.</li>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]1474
Chris Lattnerf9f08bf2007-11-09 21:49:08 +0000[diff] [blame]1475<li>Large-Scale C++ Software Design by John Lakos</li>
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]1476
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1477</ol>
1478
1479<p>If you get some free time, and you haven't read them: do so, you might learn
Misha Brukman6d1686c2004-12-04 00:32:12 +0000[diff] [blame]1480something.</p>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1481
1482</div>
1483
Chris Lattnerac457c42001-07-09 03:27:08 +0000[diff] [blame]1484<!-- *********************************************************************** -->
1485
1486<hr>
Misha Brukmand4f290a2004-05-12 18:37:22 +0000[diff] [blame]1487<address>
1488 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
Misha Brukman86242e12008-12-11 17:34:48 +0000[diff] [blame]1489 src="http://jigsaw.w3.org/css-validator/images/vcss-blue" alt="Valid CSS"></a>
Misha Brukmand4f290a2004-05-12 18:37:22 +0000[diff] [blame]1490 <a href="http://validator.w3.org/check/referer"><img
Misha Brukman86242e12008-12-11 17:34:48 +0000[diff] [blame]1491 src="http://www.w3.org/Icons/valid-html401-blue" alt="Valid HTML 4.01"></a>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1492
Misha Brukmand4f290a2004-05-12 18:37:22 +0000[diff] [blame]1493 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
NAKAMURA Takumica46f5a2011-04-09 02:13:37 +0000[diff] [blame^]1494 <a href="http://llvm.org/">LLVM Compiler Infrastructure</a><br>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1495 Last modified: $Date$
Misha Brukmand4f290a2004-05-12 18:37:22 +0000[diff] [blame]1496</address>
Misha Brukmanf196dbb2003-10-24 17:57:33 +0000[diff] [blame]1497
1498</body>
1499</html>