1<html> 2<head> 3<title>The Rule</title> 4<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1"> 5<link rel="stylesheet" href="theme/style.css" type="text/css"> 6</head> 7 8<body> 9<table width="100%" border="0" background="theme/bkd2.gif" cellspacing="2"> 10 <tr> 11 <td width="10"> 12 </td> 13 <td width="85%"> 14 <font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>The Rule</b></font> 15 </td> 16 <td width="112"><a href="http://spirit.sf.net"><img src="theme/spirit.gif" width="112" height="48" align="right" border="0"></a></td> 17 </tr> 18</table> 19<br> 20<table border="0"> 21 <tr> 22 <td width="10"></td> 23 <td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td> 24 <td width="30"><a href="numerics.html"><img src="theme/l_arr.gif" border="0"></a></td> 25 <td width="30"><a href="epsilon.html"><img src="theme/r_arr.gif" border="0"></a></td> 26 </tr> 27</table> 28<p>The <b>rule</b> is a polymorphic parser that acts as a named place-holder capturing 29 the behavior of an EBNF expression assigned to it. Naming an EBNF expression 30 allows it to be referenced later. The <tt>rule</tt> is a template class parameterized 31 by the type of the scanner (<tt>ScannerT</tt>), the rule's <a href="indepth_the_parser_context.html">context</a> 32 and its <a href="#tag">tag</a>. Default template parameters are provided to 33 make it easy to use the rule.</p> 34<pre><code><font color="#000000"><span class=identifier> </span><span class=keyword>template</span><span class=special>< 35 </span><span class=keyword>typename </span><span class=identifier>ScannerT </span><span class=special>= </span><span class=identifier>scanner</span><span class=special><>, 36 </span><span class=keyword>typename </span><span class=identifier>ContextT </span><span class=special>= </span><span class=identifier>parser_context</span><span class=special><></span><span class=identifier>, 37 </span><span class="keyword">typename</span><span class=identifier> TagT </span><span class="special">=</span><span class=identifier> parser_address_tag</span><span class=special>> 38 </span><span class=keyword>class </span><span class=identifier>rule</span><span class=special>;</span></font></code></pre> 39<p>Default template parameters are supplied to handle the most common case. <tt>ScannerT</tt> 40 defaults to <tt>scanner<></tt>, a plain vanilla scanner that acts on <tt>char 41 const<span class="operators">*</span></tt> iterators and does nothing special 42 at all other than iterate through all the chars in the null terminated input 43 a character at a time. The rule tag, <tt>TagT</tt>, typically used with <a href="trees.html">ASTs</a>, 44 is used to identify a rule; it is explained <a href="#tag">here</a>. In trivial 45 cases, declaring a rule as <tt>rule<></tt> is enough. You need not be 46 concerned at all with the <tt>ContextT</tt> template parameter unless you wish 47 to tweak the low level behavior of the rule. Detailed information on the <tt>ContextT</tt> 48 template parameter is provided <a href="indepth_the_parser_context.html">elsewhere</a>. 49</p> 50<h3><a name="order_of_parameters"></a>Order of parameters</h3> 51<p>As of v1.8.0, the <tt>ScannerT</tt>, <tt>ContextT</tt> and <tt>TagT</tt> can 52 be specified in any order. If a template parameter is missing, it will assume 53 the defaults. Examples:</p> 54<pre><span class=identifier> rule</span><span class=special><> </span><span class=identifier>rx1</span><span class=special>; 55 </span><span class=identifier>rule</span><span class=special><</span><span class=identifier>scanner</span><span class=special><> </span><span class=special>> </span><span class=identifier>rx2</span><span class=special>; 56</span> <span class=identifier>rule</span><span class=special><</span><span class=identifier>parser_context<code><font color="#000000"><span class=special><></span></font></code> </span><span class=special>> </span><span class=identifier>rx3</span><span class=special>; 57 </span><span class=identifier>rule</span><span class=special><</span><span class=identifier>parser_context<code><font color="#000000"><span class=special><></span></font></code></span><span class=special>, </span><span class=identifier>parser_address_tag</span><span class=special>> </span><span class=identifier>rx4</span><span class=special>; 58</span> <span class=identifier>rule</span><span class=special><</span><span class=identifier>parser_address_tag</span><span class=special>> </span><span class=identifier>rx5</span><span class=special>; 59</span> <span class=identifier>rule</span><span class=special><</span><span class=identifier>parser_address_tag</span><span class=special>, </span><span class=identifier>scanner</span><span class=special><>, </span><span class=identifier>parser_context<code><font color="#000000"><span class=special><></span></font></code> </span><span class=special>> </span><span class=identifier>rx6</span><span class=special>; 60 </span><span class=identifier>rule</span><span class=special><</span><span class=identifier>parser_context<code><font color="#000000"><span class=special><></span></font></code></span><span class=special>, </span><span class=identifier>scanner</span><span class=special><>, </span><span class=identifier>parser_address_tag</span><span class=special>> </span><span class=identifier>rx7</span><span class=special>;</span></pre> 61<h3><a name="multiple_scanner_support" id="multiple_scanner_support"></a>Multiple scanners</h3> 62<p>As of v1.8.0, rules can use one or more scanner types. There are cases, for 63 instance, where we need a rule that can work on the phrase and character levels. 64 Rule/scanner mismatch has been a source of confusion and is the no. 1 <a href="faq.html#scanner_business">FAQ</a>. 65 To address this issue, we now have multiple scanner support. Example:</p> 66<pre><span class=special> </span><span class=keyword>typedef </span><span class=identifier>scanner_list</span><span class=special><</span><span class=identifier>scanner</span><span class=special><>, </span><span class=identifier>phrase_scanner_t</span><span class=special>> </span><span class=identifier>scanners</span><span class=special>; 67 68 </span><span class=identifier>rule</span><span class=special><</span><span class=identifier>scanners</span><span class=special>> </span><span class=identifier>r </span><span class=special>= </span><span class=special>+</span><span class=identifier>anychar_p</span><span class=special>; 69 </span><span class=identifier>assert</span><span class=special>(</span><span class=identifier>parse</span><span class=special>(</span><span class=string>"abcdefghijk"</span><span class=special>, </span><span class=identifier>r</span><span class=special>).</span><span class=identifier>full</span><span class=special>); 70 </span><span class=identifier>assert</span><span class=special>(</span><span class=identifier>parse</span><span class=special>(</span><span class=string>"a b c d e f g h i j k"</span><span class=special>, </span><span class=identifier>r</span><span class=special>, </span><span class=identifier>space_p</span><span class=special>).</span><span class=identifier>full</span><span class=special>);</span></pre> 71<p>Notice how rule <tt>r</tt> is used in both the phrase and character levels. 72</p> 73<p>By default support for multiple scanners is disabled. The macro 74 <tt>BOOST_SPIRIT_RULE_SCANNERTYPE_LIMIT</tt> must be defined to the 75 maximum number of scanners allowed in a scanner_list. The value must 76 be greater than 1 to enable multiple scanners. Given the 77 example above, to define a limit of two scanners for the list, the 78 following line must be inserted into the source file before the 79 inclusion of Spirit headers: 80</p> 81<pre><span class=special> </span><span class=preprocessor>#define </span><span class=identifier>BOOST_SPIRIT_RULE_SCANNERTYPE_LIMIT</span> <span class=literal>2</span></pre> 82<table width="80%" border="0" align="center"> 83 <tr> 84 <td class="note_box"><img src="theme/bulb.gif" width="13" height="18"> See 85 the techniques section for an <a href="techniques.html#multiple_scanner_support">example</a> 86 of a <a href="grammar.html">grammar</a> using a multiple scanner enabled 87 rule, <a href="scanner.html#lexeme_scanner">lexeme_scanner</a> and <a href="scanner.html#as_lower_scanner">as_lower_scanner.</a></td> 88 </tr> 89</table> 90<h3>Rule Declarations</h3> 91<p>The rule class models EBNF's production rule. Example:</p> 92<pre><code><font color="#000000"> <span class=identifier>rule</span><span class=special><> </span><span class=identifier>a_rule </span><span class=special>= </span><span class=special>*(</span><span class=identifier>a </span><span class=special>| </span><span class=identifier>b</span><span class=special>) </span><span class=special>& </span><span class=special>+(</span><span class=identifier>c </span><span class=special>| </span><span class=identifier>d </span><span class=special>| </span><span class=identifier>e</span><span class=special>);</span></font></code></pre> 93<p>The type and behavior of the right-hand (rhs) EBNF expression, which may be 94 arbitrarily complex, is encoded in the rule named a_rule. a_rule may now be 95 referenced elsewhere in the grammar:</p> 96<pre><code><font color="#000000"> <span class=identifier>rule</span><span class=special><> </span><span class=identifier>another_rule </span><span class=special>= </span><span class=identifier>f </span><span class=special>>> </span><span class=identifier>g </span><span class=special>>> </span><span class=identifier>h </span><span class=special>>> </span><span class=identifier>a_rule</span><span class=special>;</span></font></code></pre> 97<table width="80%" border="0" align="center"> 98 <tr> 99 <td class="note_box"><img src="theme/alert.gif" width="16" height="16"> <b>Referencing 100 rules <br> 101 </b><br> 102 When a rule is referenced anywhere in the right hand side of an EBNF expression, 103 the rule is held by the expression by reference. It is the responsibility 104 of the client to ensure that the referenced rule stays in scope and does 105 not get destructed while it is being referenced. </td> 106 </tr> 107</table> 108<pre><span class=special> </span><span class=identifier>a </span><span class=special>= </span><span class=identifier>int_p</span><span class=special>; 109 </span><span class=identifier>b </span><span class=special>= </span><span class=identifier>a</span><span class=special>; 110 </span><span class=identifier>c </span><span class=special>= </span><span class=identifier>int_p </span><span class=special>>> </span><span class=identifier>b</span><span class=special>;</span></pre> 111<h3>Copying Rules</h3> 112<p>The rule is a weird C++ citizen, unlike any other C++ object. It does not have 113 the proper copy and assignment semantics and cannot be stored and passed around 114 by value. If you need to copy a rule you have to explicitly call its member 115 function <tt>copy()</tt>:</p> 116<pre><span class=special> </span><span class=identifier>r</span><span class="special">.</span><span class=identifier>copy()</span><span class=special>;</span></pre> 117<p>However, be warned that copying a rule will not deep copy other referenced 118 rules of the source rule being copied. This might lead to dangling references. 119 Again, it is the responsibility of the client to ensure that all referenced 120 rules stay in scope and does not get destructed while it is being referenced. 121 Caveat emptor.</p> 122<p>If you copy a rule, then you'll want to place it in a storage somewhere. The 123 problem is how? The storage can't be another rule:</p> 124<pre> <code><font color="#000000"><span class=identifier>rule</span><span class=special><></span></font></code> r2 <span class="special">=</span> <span class=identifier>r</span><span class="special">.</span><span class=identifier>copy()</span><span class=special>; </span><span class="comment">// BAD!</span></pre> 125<p>because rules are weird and does not have the expected C++ copy-constructor 126 and assignment semantics! As a general rule: <strong>Don't put a copied rule 127 into another rule! </strong>Instead, use the <a href="stored_rule.html">stored_rule</a> 128 for that purpose.</p> 129<h3>Forward declarations</h3> 130<p>A <tt>rule</tt> may be declared before being defined to allow cyclic structures 131 typically found in BNF declarations. Example:</p> 132<pre><code><font color="#000000"><span class=special> </span><span class=identifier>rule</span><span class=special><> </span><span class=identifier>a</span><span class=special>, </span><span class=identifier>b</span><span class=special>, </span><span class=identifier>c</span><span class=special>; 133 134 </span><span class=identifier>a </span><span class=special>= </span><span class=identifier>b </span><span class=special>| </span><span class=identifier>a</span><span class=special>; 135 </span><span class=identifier>b </span><span class=special>= </span><span class=identifier>c </span><span class=special>| </span><span class=identifier>a</span><span class=special>;</span></font></code></pre> 136<h3>Recursion</h3> 137<p>The right-hand side of a rule may reference other rules, including itself. 138 The limitation is that direct or indirect left recursion is not allowed (this 139 is an unchecked run-time error that results in an infinite loop). This is typical 140 of top-down parsers. Example:</p> 141<pre><code><font color="#000000"><span class=special> </span><span class=identifier>a </span><span class=special>= </span><span class=identifier>a </span><span class=special>| </span><span class=identifier>b</span><span class=special>; </span><span class=comment>// infinite loop!</span></font></code></pre> 142<table width="80%" border="0" align="center"> 143 <tr> 144 <td class="note_box"><img src="theme/lens.gif" width="15" height="16"> <b>What 145 is left recursion?<br> 146 </b><br> 147 Left recursion happens when you have a rule that calls itself before anything 148 else. A top-down parser will go into an infinite loop when this happens. 149 See the <a href="faq.html#left_recursion">FAQ</a> for details on how to 150 eliminate left recursion.</td> 151 </tr> 152</table> 153<h3>Undefined rules</h3> 154<p>An undefined rule matches nothing and is semantically equivalent to <tt>nothing_p</tt>.</p> 155<h3>Redeclarations</h3> 156<p>Like any other C++ assignment, a second assignment to a rule is destructive 157 and will redefine it. The old definition is lost. Rules are dynamic. A rule 158 can change its definition anytime:</p> 159<pre><code><font color="#000000"><span class=identifier> r </span><span class=special>= </span><span class=identifier>a_definition</span><span class=special>; 160</span><span class=identifier> r </span><span class=special>= </span><span class=identifier>another_definition</span><span class=special>;</span></font></code></pre> 161<p>Rule <tt>r</tt> loses the old definition when the second assignment is made. 162 As mentioned, an undefined rule matches nothing and is semantically equivalent 163 to <tt>nothing_p</tt>. 164<h3>Dynamic Parsers</h3> 165<p>Hosting declarative EBNF in imperative C++ yields an interesting blend. We 166 have the best of both worlds. We have the ability to conveniently modify the 167 grammar at run time using imperative constructs such as <tt>if</tt>, <tt>else</tt> 168 statements. Example:</p> 169<pre><code><font color="#000000"><span class=special> </span><span class=keyword>if </span><span class=special>(</span><span class=identifier>feature_is_available</span><span class=special>) 170 </span><span class=identifier>r </span><span class=special>= </span><span class=identifier>add_this_feature</span><span class=special>;</span></font></code></pre> 171<p>Rules are essentially dynamic parsers. A dynamic parser is characterized by 172 its ability to modify its behavior at run time. Initially, an undefined rule 173 matches nothing. At any time, the rule may be defined and redefined, thus, dynamically 174 altering its behavior.</p> 175<h3>No start rule</h3> 176<p>Typically, parsers have what is called a start symbol, chosen to be the root 177 of the grammar where parsing starts. The Spirit parser framework has no notion 178 of a start symbol. Any rule can be a start symbol. This feature promotes step-wise 179 creation of parsers. We can build parsers from the bottom up while fully testing 180 each level or module up until we get to the top-most level.</p> 181<h3><a name="tag"></a>Parser Tags</h3> 182<p>Rules may be tagged for identification purposes. This is necessary, especially 183 when dealing with <a href="trees.html">parse trees and ASTs</a> to see which 184 rule created a specific AST/parse tree node. Each rule has an ID of type <tt>parser_id</tt>. 185 This ID can be obtained through the rule's <tt>id()</tt> member function:</p> 186<pre><code><font color="#000000"><span class=identifier> my_rule</span><span class=special>.</span><span class=identifier>id</span><span class=special>(); </span><span class=comment>// get my_rule's id</span></font></code></pre> 187<p>The <tt>parser_id</tt> class is declared as:</p> 188<pre> <span class="keyword">class</span> <span class="identifier">parser_id</span><br> <span class="special">{</span><br> <span class="keyword">public</span><span class="special">:</span><br> parser_id<span class="special">();</span><br> <span class="keyword">explicit</span> parser_id<span class="special">(</span><span class="keyword">void const</span><span class="special">*</span> p<span class="special">);</span><br> parser_id<span class="special">(</span><span class="keyword">std::size_t</span> l<span class="special">);</span> 189 190 <span class="keyword">bool</span> <span class="keyword">operator</span><span class="special">==(</span><span class="identifier">parser_id</span> <span class="keyword">const</span><span class="special">&</span> x<span class="special">)</span> const<span class="special">;</span><br> <span class="keyword">bool</span> <span class="keyword">operator</span><span class="special">!=(</span><span class="identifier">parser_id</span> <span class="keyword">const</span><span class="special">&</span> x<span class="special">)</span> const<span class="special">;</span> 191 <span class="keyword">bool</span> <span class="keyword"> operator</span><span class="special"><(</span><span class="identifier">parser_id</span> <span class="keyword">const</span><span class="special">&</span> x<span class="special">)</span> const<span class="special">;</span> 192 <span class="special"></span><span class="keyword">std::size_t</span><span class="identifier"> to_long</span><span class="special">()</span> <span class="keyword">const</span><span class="special">; 193 };</span></pre> 194<h3>parser_address_tag</h3> 195<p>The rule's <tt>TagT</tt> template parameter supplies this ID. This defaults 196 to <tt>parser_address_tag</tt>. The <tt>parser_address_tag</tt> uses the address 197 of the rule as its ID. This is often not the most convenient, since it is not 198 always possible to get the address of a rule to compare against. </p> 199<h3>parser_tag</h3> 200<p>It is possible to have specific constant integers to identify a rule. For this 201 purpose, we can use the <tt>parser_tag<N></tt>, where N is a constant 202 integer:</p> 203<pre><code><font color="#000000"><span class=identifier> rule</span><span class=special><</span><span class=identifier>parser_tag</span><span class="special"><</span><span class=identifier>123</span><span class="special">> > </span><span class="identifier">my_rule</span><span class="special">; </span><span class="comment">// set my_rule's id to 123</span></font></code></pre> 204<h3>dynamic_parser_tag</h3> 205<p>The <tt>parser_tag<N></tt> can only specify a <strong>static ID</strong>, 206 which is defined at compile time. If you need the ID to be <strong>dynamic</strong> 207 (changeable at runtime), you can use the <tt>dynamic_parser_tag</tt> class as 208 the <tt>TagT</tt> template parameter. This template parameter enables the <tt>set_id()</tt> 209 function, which may be used to set the required id at runtime:</p> 210<pre><code><font color="#000000"><span class=identifier> rule</span><span class=special><</span><span class=identifier>dynamic_parser_tag</span><span class="special">> </span><span class="identifier">my_dynrule</span><span class="special">;</span> 211 my_dynrule.set_id(1234); <span class="comment">// set my_dynrule's id to 1234</span></font></code></pre> 212<p>If the <tt>set_id()</tt> function isn't called, the parser id defaults to the 213 address of the rule as its ID, just like the <tt>parser_address_tag</tt> template 214 parameter would do. </p> 215<table border="0"> 216 <tr> 217 <td width="10"></td> 218 <td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td> 219 <td width="30"><a href="numerics.html"><img src="theme/l_arr.gif" border="0"></a></td> 220 <td width="30"><a href="epsilon.html"><img src="theme/r_arr.gif" border="0"></a></td> 221 </tr> 222</table> 223<br> 224<hr size="1"> 225<p class="copyright">Copyright © 1998-2003 Joel de Guzman<br> 226 <br> 227 <font size="2">Use, modification and distribution is subject to the Boost Software 228 License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at 229 http://www.boost.org/LICENSE_1_0.txt)</font></p> 230</body> 231</html> 232
