1<html> 2<head> 3<title>Debugging</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%"> <font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Debugging</b></font> 14 </td> 15 <td width="112"><a href="http://spirit.sf.net"><img src="theme/spirit.gif" width="112" height="48" align="right" border="0"></a></td> 16 </tr> 17</table> 18<br> 19<table border="0"> 20 <tr> 21 <td width="10"></td> 22 <td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td> 23 <td width="30"><a href="position_iterator.html"><img src="theme/l_arr.gif" border="0"></a></td> 24 <td width="30"><a href="error_handling.html"><img src="theme/r_arr.gif" border="0"></a></td> 25 </tr> 26</table> 27<p>The top-down nature of Spirit makes the generated parser easy to micro- debug 28 using the standard debugger bundled with the C++ compiler we are using. With 29 recursive-descent, the parse traversal utilizes the hardware stack through C++ 30 function call mechanisms. There are no difficult to debug tables or state machines 31 that obscure the parsing logic flow. The stack trace we see in the debugger 32 follows faithfully the hierarchical grammar structure.</p> 33<p> Since any production rule can initiate a parse traversal , it is a lot easier 34 to pinpoint the bugs by focusing on one or a few rules. For relatively complex 35 parsing tasks, the same way we write robust C++ programs, it is advisable to 36 develop a grammar iteratively on a per-module basis where each module is a small 37 subset of the complete grammar. That way, we can stress-test individual modules 38 piecemeal until we reach the top-most module. For instance, when developing 39 a scripting language, we can start with expressions, then move on to statements, 40 then functions, upwards until we have a complete grammar. </p> 41<p> At some point when the grammar gets quite complicated, it is desirable to 42 visualize the parse traversal and see what's happening. There are some facilities 43 in the framework that aid in the visualisation of the parse traversal for the 44 purpose of debugging. The following macros enable these features.</p> 45<a name="debugging_macros"></a> 46<h2>Debugging Macros</h2> 47<a name="spirit_assert_exception"></a> 48<h3>BOOST_SPIRIT_ASSERT_EXCEPTION</h3> 49<p> Spirit contains assertions that may activate when spirit is used incorrectly. 50 By default these assertions use the assert macro from the standard library. 51 If you want spirit to throw an exception instead, define <tt>BOOST_SPIRIT_ASSERT_EXCEPTION</tt> 52 to the name of the class that you want to be thrown. This class's constructor 53 will be passed a <tt>const char*</tt> stringified version of the file, line, 54 and assertion condition, when it is thrown. If you want to totally disable the 55 assertion, <tt>#define NDEBUG</tt>.</p> 56<a name="spirit_debug"></a> 57<h3>BOOST_SPIRIT_DEBUG</h3> 58<p>Define this to enable debugging.</p> 59<p>With debugging enabled, special output is generated at key points of the 60 parse process, using the standard output operator (<tt><span class="keyword">operator</span><span class="special"><<</span></tt>) 61 with <tt>BOOST_SPIRIT_DEBUG_OUT</tt> (default is <tt><span class="identifier">std</span><span class="special">::</span><span class="identifier">cout</span></tt>, 62 see below) as its left operand.</p> 63<table width="80%" border="0" align="center"> 64 <tr> 65 <td class="note_box"><img src="theme/note.gif"> In order to use spirit's 66 debugging support you must ensure that appropriate overloads of 67 <tt><span class="identifier">operator</span><span class="special"><<</span></tt> 68 taking <tt>BOOST_SPIRIT_DEBUG_OUT</tt> as its left operand are available. 69 The expected semantics are those of the standard output operator.<br> 70 <br> 71 These overloads may be provided either within the namespace where the 72 corresponding class is declared (will be found through Argument Dependent Lookup) or [within an 73 anonymous namespace] within <tt><span class="keyword">namespace</span> <span class="identifier">boost</span><span class="special">::</span><span class="identifier">spirit</span></tt>, 74 so it is visible where it is called.<br> 75 <br> 76 <img src="theme/alert.gif"> Note in particular that when <tt>BOOST_SPIRIT_DEBUG_FLAGS_CLOSURES</tt> 77 is set, overloads of <tt><span class="identifier">operator</span><span class="special"><<</span></tt> 78 taking instances of the types used in closures as their right operands are required.<br> 79 <br> 80 You may find an example of overloading the output operator for 81 <tt><span class="identifier">std</span><span class="special">::</span><span class="identifier">pair</span></tt> 82 in a <a href="faq.html#output_operator">related FAQ entry</a>.</td> 83 </tr> 84</table> 85 86<p>By default, if the <tt>BOOST_SPIRIT_DEBUG</tt> macro is defined, all available 87 debug output is generated. To fine tune the amount of generated text you can 88 define the <tt>BOOST_SPIRIT_DEBUG_FLAGS</tt> constant to be equal of a combination 89 of the following flags:</p> 90<table width="90%" border="0" align="center"> 91 <tr> 92 <td colspan="2" class="table_title"><b>Available flags to fine tune debug 93 output </b></td> 94 </tr> 95 <tr> 96 <td width="29%" height="27" class="table_cells"><tt>BOOST_SPIRIT_DEBUG_FLAGS_NODES</tt></td> 97 <td width="71%" class="table_cells"><p>print information about nodes (general 98 for all parsers)</p></td> 99 </tr> 100 <tr> 101 <td height="27" class="table_cells"><tt>BOOST_SPIRIT_DEBUG_FLAGS_TREES</tt></td> 102 <td class="table_cells"><p>print information about parse trees and AST's (general 103 for all tree parsers)</p></td> 104 </tr> 105 <tr> 106 <td height="27" class="table_cells"><tt>BOOST_SPIRIT_DEBUG_FLAGS_CLOSURES</tt></td> 107 <td class="table_cells">print information about closures (general for all 108 parsers with closures)</td> 109 </tr> 110 <tr> 111 <td height="27" class="table_cells"><tt>BOOST_SPIRIT_DEBUG_FLAGS_ESCAPE_CHAR</tt></td> 112 <td class="table_cells"><p>print information out of the <tt>esc_char_parser</tt></p></td> 113 </tr> 114 <tr> 115 <td height="27" class="table_cells"><tt>BOOST_SPIRIT_DEBUG_FLAGS_SLEX</tt></td> 116 <td class="table_cells">print information out of the <tt>SLEX</tt> parser</td> 117 </tr> 118</table> 119<p><a name="spirit_debug_out"></a> </p> 120<h3>BOOST_SPIRIT_DEBUG_OUT</h3> 121<p> Define this to redirect the debugging diagnostics printout to somewhere else 122 (e.g. a file or stream). Defaults to <tt>std::cout</tt>.</p> 123<a name="spirit_debug_token printer"></a> 124<h3>BOOST_SPIRIT_DEBUG_TOKEN_PRINTER</h3> 125<p> The <tt>BOOST_SPIRIT_DEBUG_TOKEN_PRINTER</tt> macro allows you to redefine the way characters are printed on the stream. </p> 126<p>If <tt>BOOST_SPIRIT_DEBUG_OUT</tt> is of type <tt>StreamT</tt>, the character type is <tt>CharT</tt> and <tt>BOOST_SPIRIT_DEBUG_TOKEN_PRINTER</tt> is 127 defined to <tt>foo</tt>, it must be compatible with this usage:</p> 128<pre><code><span class=identifier> foo</span><span class=special>(</span><span class=identifier>StreamT</span><span class=special>, </span><span class=identifier>CharT</span><span class=special>)</span></code></pre> 129<p>The default printer requires <tt>operator<<(StreamT, CharT)</tt> to 130 be defined. Additionally, if <tt>CharT</tt> is convertible to a normal character 131 type (<tt>char</tt>, <tt>wchar_t</tt> or <tt>int</tt>), it prints control 132 characters in a friendly manner (e.g., when it receives <span class=special>'\n'</span> it 133 actually prints the <span class=special>\</span> and <span class=special>n</span> characters, 134instead of a newline).</p> 135<a name="spirit_debug_print_some"></a> 136<h3>BOOST_SPIRIT_DEBUG_PRINT_SOME</h3> 137<p> The <tt>BOOST_SPIRIT_DEBUG_PRINT_SOME</tt> constant defines the number of 138 characters from the stream to be printed for diagnosis. This defaults to the 139 first 20 characters.</p> 140<p><a name="spirit_debug_tracenode"></a> </p> 141<h3>BOOST_SPIRIT_DEBUG_TRACENODE</h3> 142<p> By default all parser nodes are traced. This constant may be used to redefine 143 this default. If this is <tt>1</tt> (<tt>true</tt>), then tracing is enabled 144 by default, if this constant is <tt>0</tt> (<tt>false</tt>), the tracing is 145 disabled by default. This preprocessor constant is set to <tt>1 </tt>(<tt>true</tt>) 146 by default.</p> 147<p>Please note, that the following <tt>BOOST_SPIRIT_DEBUG_...() </tt>macros are 148 to be used at function scope only.</p> 149<a name="spirit_debug_node_p_"></a> 150<h3>BOOST_SPIRIT_DEBUG_NODE(p)</h3> 151<p> Define this to print some debugging diagnostics for parser p. This macro</p> 152<ul> 153 <li>Registers the parser name for debugging</li> 154 <li>Enables/disables the tracing for parser depending on <tt>BOOST_SPIRIT_DEBUG_TRACENODE</tt></li> 155</ul> 156<p> <b>Pre-parse</b>: Before entering the rule, the rule name followed by a peek 157 into the data at the current iterator position is printed.</p> 158<p> <b>Post-parse</b>: After parsing the rule, the rule name followed by a peek 159 into the data at the current iterator position is printed. Here, <tt>'/'</tt> 160 before the rule name flags a successful match while <tt>'#'</tt> before the rule 161 name flags an unsuccessful match.</p> 162<p> The following are synonyms for <tt>BOOST_SPIRIT_DEBUG_NODE</tt></p> 163<ol> 164 <li>BOOST_SPIRIT_DEBUG_RULE</li> 165 <li>BOOST_SPIRIT_DEBUG_GRAMMAR</li> 166</ol> 167<a name="spirit_trace_node_p__flag_"></a> 168<h3>BOOST_SPIRIT_DEBUG_TRACE_NODE(p, flag)</h3> 169<p> Similar to <tt>BOOST_SPIRIT_DEBUG_NODE</tt>. Additionally allows selective debugging. 170 This is useful in situations where we want to debug just a hand picked set of 171 nodes.</p> 172<p> The following are synonyms for <tt>BOOST_SPIRIT_DEBUG_TRACE_NODE</tt></p> 173<ol> 174 <li>BOOST_SPIRIT_DEBUG_TRACE_RULE</li> 175 <li>BOOST_SPIRIT_DEBUG_TRACE_GRAMMAR</li> 176</ol> 177<p><a name="spirit_trace_node_p__flag__name_"></a> </p> 178<h3>BOOST_SPIRIT_DEBUG_TRACE_NODE_NAME(p, name, flag)</h3> 179<p> Similar to <tt>BOOST_SPIRIT_DEBUG_NODE</tt>. Additionally allows selective 180 debugging and allows to specify the name used during debug printout. This is 181 useful in situations where we want to debug just a hand picked set of nodes. 182 The <tt>name</tt> may be redefined in situations, where the parser parameter does not reflect the name of the parser to debug.</p> 183<p> The following are synonyms for <tt>BOOST_SPIRIT_DEBUG_TRACE_NODE</tt></p> 184<ol> 185 <li>BOOST_SPIRIT_DEBUG_TRACE_RULE_NAME</li> 186 <li>BOOST_SPIRIT_DEBUG_TRACE_GRAMMAR_NAME</li> 187</ol> 188<hr> 189<p>Here's the original calculator with debugging features enabled:</p> 190<pre> 191 <code><span class=preprocessor>#define </span><span class=identifier>BOOST_SPIRIT_DEBUG </span><span class=comment>///$$$ DEFINE THIS BEFORE ANYTHING ELSE $$$/// 192 </span><span class=preprocessor>#include </span><span class=string>"boost/spirit/include/classic.hpp" 193 194 </span><span class=comment>/***/ 195 196 /*** CALCULATOR GRAMMAR DEFINITIONS HERE ***/ 197 198 </span><span class=identifier>BOOST_SPIRIT_DEBUG_RULE</span><span class=special>(</span><span class=identifier>integer</span><span class=special>); 199 </span><span class=identifier>BOOST_SPIRIT_DEBUG_RULE</span><span class=special>(</span><span class=identifier>group</span><span class=special>); 200 </span><span class=identifier>BOOST_SPIRIT_DEBUG_RULE</span><span class=special>(</span><span class=identifier>factor</span><span class=special>); 201 </span><span class=identifier>BOOST_SPIRIT_DEBUG_RULE</span><span class=special>(</span><span class=identifier>term</span><span class=special>); 202 </span><span class=identifier>BOOST_SPIRIT_DEBUG_RULE</span><span class=special>(</span><span class=identifier>expr</span><span class=special>); 203</span></code></pre> 204<p> <img src="theme/note.gif" width="16" height="16"> Be sure to add the macros <strong>inside</strong> the grammar definition's constructor. Now here's a sample session with the calculator.</p> 205 206<pre><code> <span class="preprocessor">Type an expression...or [q or Q] to quit</span> 207 208 <span class="preprocessor">1 + 2</span> 209 210 grammar(calc): "1 + 2" 211 rule(expression): "1 + 2" 212 rule(term): "1 + 2" 213 rule(factor): "1 + 2" 214 rule(integer): "1 + 2" 215 <span class="preprocessor">push 1</span> 216 /rule(integer): " + 2" 217 /rule(factor): " + 2" 218 /rule(term): " + 2" 219 rule(term): "2" 220 rule(factor): "2" 221 rule(integer): "2" 222 <span class="preprocessor">push 2</span> 223 /rule(integer): "" 224 /rule(factor): "" 225 /rule(term): "" 226 <span class="preprocessor">popped 1 and 2 from the stack. pushing 3 onto the stack.</span> 227 /rule(expression): "" 228 /grammar(calc): "" 229 <span class="preprocessor">------------------------- 230 Parsing succeeded 231 result = 3 232 -------------------------</span></code></pre> 233 234<p> We typed in "1 + 2". Notice that there are two successful branches 235 from the top rule <tt>expr</tt>. The text in red is generated by the parser's 236 semantic actions while the others are generated by the debug-diagnostics of 237 our rules. Notice how the first <tt>integer</tt> rule took "1", the 238 first <tt>term</tt> rule took "+" and finally the second <tt>integer</tt> 239 rule took "2".</p> 240<p>Please note the special meaning of the first characters appearing on the printed 241 lines:</p> 242<ul> 243 <li>a single <span class="literal">'/'</span> starts a line containing the information 244 about a successfully matched parser node (<tt>rule<></tt>, <tt>grammar<></tt> 245 or <tt>subrule<></tt>)</li> 246 <li>a single <span class="literal">'#'</span> starts a line containing the information 247 about a failed parser node</li> 248 <li>a single <span class="literal">'^'</span> starts a line containing the first member (return value/synthesised 249 attribute) of the closure of a successfully matched parser node.</li> 250</ul> 251<p>Check out <a href="../example/fundamental/calc_debug.cpp">calc_debug.cpp</a> to see debugging in action. </p> 252<table border="0"> 253 <tr> 254 <td width="10"></td> 255 <td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td> 256 <td width="30"><a href="position_iterator.html"><img src="theme/l_arr.gif" border="0"></a></td> 257 <td width="30"><a href="error_handling.html"><img src="theme/r_arr.gif" border="0"></a></td> 258 </tr> 259</table> 260<br> 261<hr size="1"> 262<p class="copyright">Copyright © 1998-2003 Joel de Guzman<br> 263 Copyright © 2003 Hartmut Kaiser<br> 264 <br> 265<font size="2">Use, modification and distribution is subject to the Boost Software 266 License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at 267 http://www.boost.org/LICENSE_1_0.txt)</font></p> 268<p class="copyright"> </p> 269</body> 270</html> 271
