ceph/src/boost/tools/build/test/test_system.html

   1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
   2 "http://www.w3.org/TR/html4/strict.dtd">
   3
   4 <html>
   5   <head>
   6     <meta name="generator" content=
   7     "HTML Tidy for Linux/x86 (vers 1st March 2002), see www.w3.org">
   8     <!--tidy options: -i -wrap 78 -->
   9     <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
  10
  11     <title>A testing system for B2</title>
  12 <style type="text/css">
  13         hr { color: black }
  14         p.revision { text-align: right; font-style: italic }
  15         pre.code { margin-left: 2em }
  16         pre.example { margin-left: 2em; border: solid black thin }
  17         pre.output { margin-left: 2em }
  18         img.banner { border: 0; float: left }
  19         h1 { text-align: right }
  20         br.clear { clear: left }
  21         div.attention { color: red }
  22
  23 </style>
  24   </head>
  25
  26   <body>
  27     <p><a href="../../../../index.htm"><img class="banner" height="86" width=
  28     "277" alt="C++ Boost" src="../../../../boost.png"></a></p>
  29
  30     <h1>A testing system for B2<br class="clear">
  31     </h1>
  32     <hr>
  33
  34     <dl class="page-index">
  35       <dt><a href="#sec-intro">Introduction for users</a></dt>
  36
  37       <dd>
  38         <dl class="page-index">
  39           <dt><a href="#sec-command-line-options">Command line options</a></dt>
  40         </dl>
  41       </dd>
  42
  43       <dt><a href="#sec-developers">Introduction for developers</a></dt>
  44
  45       <dd>
  46         <dl class="page-index">
  47           <dt><a href="#sec-intro-changing">Changing the working
  48           directory</a></dt>
  49
  50           <dt><a href="#sec-intro-examining">Examining the working directory and
  51           changing it</a></dt>
  52
  53           <dt><a href="#sec-intro-results">Test result</a></dt>
  54         </dl>
  55       </dd>
  56
  57       <dt><a href="#sec-reference">Reference documentation</a></dt>
  58
  59       <dd>
  60         <dl class="page-index">
  61           <dt><a href="#method-__init__">Method __init__</a></dt>
  62
  63           <dt><a href="#method-set_tree">Method <tt>set_tree</tt></a></dt>
  64
  65           <dt><a href="#method-write">Method <tt>write</tt></a></dt>
  66
  67           <dt><a href="#method-copy">Method <tt>copy</tt></a></dt>
  68
  69           <dt><a href="#method-touch">Method <tt>touch</tt></a></dt>
  70
  71           <dt><a href="#method-run_build_system">Method
  72           <tt>run_build_system</tt></a></dt>
  73
  74           <dt><a href="#method-read">Method <tt>read</tt></a></dt>
  75
  76           <dt><a href="#method-read_and_strip">Method
  77           <tt>read_and_strip</tt></a></dt>
  78
  79           <dt><a href="#methods-expectations">Methods for declaring
  80           expectations</a></dt>
  81
  82           <dt><a href="#methods-ignoring">Methods for ignoring
  83           changes</a></dt>
  84
  85           <dt><a href="#methods-result">Methods for explicitly specifying
  86           results</a></dt>
  87
  88           <dt><a href="#class-list">Helper class <tt>List</tt></a></dt>
  89         </dl>
  90       </dd>
  91     </dl>
  92     <hr>
  93
  94     <h2><a name="sec-intro">Introduction for users</a></h2>
  95
  96     <p>The testing system for B2 is a small set of Python modules and
  97     scripts for automatically testing user-obversable behaviour. It uses
  98     components from testing systems of <a href="http://www.scons.org">Scons</a>
  99     and <a href="http://subversion.tigris.org">Subversion</a>, together with
 100     some additional functionality.</p>
 101
 102     <p>To run the tests you need to:</p>
 103
 104     <ol>
 105       <li>Get the source tree of B2 (located at <tt>tools/build</tt>
 106       in Boost)</li>
 107
 108       <li>Have <a href="http://www.python.org">Python</a> installed. Version
 109       2.1 is known to work.</li>
 110
 111       <li>Build Boost.Jam. See <a href=
 112       "../engine/index.html">$boost_build_root/engine/index.html</a> for
 113       instructions.</li>
 114
 115       <li>Configure at least one toolset. You can edit <tt>site-config.jam</tt>
 116       or <tt>user-config.jam</tt> to add new toolsets. Or you can create file
 117       <tt>test-config.jam</tt> in <tt>$boost_build_root/test</tt> directory. In
 118       this case, <tt>site-config.jam</tt> and <tt>user-config.jam</tt> will be
 119       ignored for testing.</li>
 120
 121       <li>Make sure that in the configuration jam file(s) that you use you generate
 122       no console output, ie. with the Boost Build 'ECHO' rule. Such console output
 123       in the configuration jam file(s) will cause a number of tests to automatically
 124       fail which would otherwise succeed.</li>
 125     </ol>
 126
 127     <p>When all is set, you can run all the tests using the <tt>test_all.py</tt>
 128      script or you can run a specific test by starting its Python script
 129      directly.</p>
 130
 131     <p>Examples:</p>
 132
 133 <pre class="code">
 134 python test_all.py
 135 python generators_test.py
 136 </pre>
 137
 138     <p>If everything is OK, you will see a list of passed tests. Otherwise, a
 139     failure will be reported.</p>
 140
 141     <h3><a name="sec-command-line-options">Command line options</a></h3>
 142
 143     <p>Test scripts will use the toolset you configured to be the default or
 144     you can specify a specific one on the command line:</p>
 145
 146 <pre class="code">
 147 python test_all.py borland
 148 python generators_test.py msvc-7.1
 149 </pre>
 150
 151     <p>Other test script flags you can specify on the command line are:</p>
 152
 153     <ul>
 154         <li><tt>--default-bjam</tt> -- By default the test system will use the
 155         Boost Jam executable found built in its default development build
 156         location. This option makes it use the default one available on your
 157         system, i.e. the one found in the system path.</li>
 158
 159         <li><tt>--preserve</tt> -- In case of a failed test its working
 160         directory will be copied to the "failed_test" directory under the
 161         current directory.</li>
 162
 163         <li><tt>--verbose</tt> -- Makes the test system and the run build system
 164         display additional output. Note though that this may cause tests that
 165         check the build system output to fail.</li>
 166     </ul>
 167
 168     <h2><a name="sec-developers">Introduction for developers</a></h2>
 169
 170     <p>It is suggested that every new functionality come together with tests,
 171     and that bugfixes are accompanied by tests. There's no need to say that
 172     tests are good, but two points are extremely important:</p>
 173
 174     <ul>
 175       <li>For an interpreted language like Jam, without any static checks,
 176       testing is simply the only sefeguard we can have.</li>
 177
 178       <li>Good tests allow us to change internal design much more safely, and we
 179       have not gotten everything nailed down yet.</li>
 180     </ul>
 181
 182     <p>Adding a new test is simple:</p>
 183
 184     <ol>
 185       <li>Go to <tt>$boost_build_root/test/test_all.py</tt> and add new test
 186       name to the list at the end of the file. Suppose the test name is "hello".
 187       </li>
 188
 189       <li>Add a new python module, in this example "hello.py", to do the actual
 190       testing.</li>
 191     </ol>
 192
 193     <p>The module, in general will perform these basic actions:</p>
 194
 195     <ol>
 196       <li>Set up the initial working directory state</li>
 197
 198       <li>
 199         Run the build system and check the results:
 200
 201         <ol>
 202           <li>generated output,</li>
 203
 204           <li>changes made to the working directory,</li>
 205
 206           <li>new content of the working directory.</li>
 207         </ol>
 208       </li>
 209
 210       <li>Add, remove or touch files or change their content and then repeat
 211       the previous step until satisfied.</li>
 212
 213       <li>Clean up</li>
 214     </ol>
 215
 216     <p>The "hello.py" module might contain:</p>
 217 <pre class="example">
 218 from BoostBuild import List
 219
 220 # Create a temporary working directory
 221 t = BoostBuild.Tester()
 222
 223 # Create the needed files
 224 t.write("jamroot.jam", "")
 225 t.write("jamfile.jam", """
 226 exe hello : hello.cpp ;
 227 """)
 228 t.write("hello.cpp", """
 229 int main()
 230 {
 231     return 0;
 232 }
 233
 234 """)
 235
 236 t.run_build_system()
 237
 238 # First, create a list of three pathnames.
 239 file_list = List("bin/$toolset/debug/") * List("hello.exe hello.obj")
 240 # Second, assert that those files were added as result of the last build system invocation.
 241 t.expect_addition(file_list)
 242
 243 # Invoke the build system once again.
 244 t.run_build_system("clean")
 245 # Check if the files added previously were removed.
 246 t.expect_removal(file_list)
 247
 248 # Remove temporary directories
 249 t.cleanup()
 250 </pre>
 251
 252     <p>The <tt>test</tt> directory contains a file "template.py" which can be
 253     used as a start for your own tests.</p>
 254
 255     <p>Overview of the most important methods of class <tt>Tester</tt> follows.
 256     </p>
 257
 258     <h3><a name="sec-intro-changing">Changing the working directory</a></h3>
 259
 260     <p>The class <tt>Tester</tt> creates a temporary directory in its
 261     constructor and changes to that directory. It can be modified by calling
 262     these methods:</p>
 263
 264     <ul>
 265       <li><tt>set_tree</tt> -- sets the content of the working directory to be
 266       equal to the content of the specified directory. This method is
 267       preferable when directory tree for testing is large.</li>
 268
 269       <li><tt>write</tt> -- sets the content of file in a working directory.
 270       This is optimal if you want to create a directory tree with 3-4 small
 271       files.</li>
 272
 273       <li><tt>touch</tt> -- changes the modification times of a file</li>
 274     </ul>
 275
 276     <h3><a name="sec-intro-examining">Examining the working directory and
 277     changing it</a></h3>
 278
 279     <p>The method <tt>read</tt>, inherited from the <tt>TestCmd</tt> class, can
 280     be used to read any file in the working directory and check its content.
 281     <tt>Tester</tt> adds another method for tracking changes. Whenever the build
 282     system is run (using <a href="#method-run_build_system"><tt>run_build_system
 283     </tt></a>), the working dir state before and after running is recorded. In
 284     addition, difference between the two states -- i.e. lists of files that were
 285     added, removed, modified or touched -- are stored in two member variables -
 286     <tt>tree_difference</tt> and <tt>unexpected_difference</tt>.</p>
 287
 288     <p>After than, the test author may specify that some change is expected, for
 289     example, by calling <tt>expect_addition("foo")</tt>. This call will check if
 290     the file was indeed added, and if so, will remove its name from the list of
 291     added files in <tt>unexpected_difference</tt>. Likewise, it is possible to
 292     specify that some changes are not interesting, for example a call to
 293     <tt>ignore("*.obj")</tt> will just remove every file with the ".obj"
 294     extension from <tt>unexpected_difference</tt>.</p>
 295
 296     <p>When test has finished with expectations and ignoring, the member
 297     <tt>unexpected_difference</tt> will contain the list of all changes not yet
 298     accounted for. It is possible to assure that this list is empty by calling
 299     the <tt>expect_nothing_more</tt> member function.</p>
 300
 301     <h3><a name="sec-intro-results">Test result</a></h3>
 302
 303     <p>Any of the <tt>expect*</tt> methods below will fail the test if the
 304     expectation is not met. It is also possible to perform manually arbitrary
 305     test and explicitly cause the test to either pass or fail. Ordinary
 306     filesystem functions can be used to work with the directory tree. Methods
 307     <tt>pass_test</tt> and <tt>fail_test</tt> are used to explicitly give the
 308     test outcome.</p>
 309
 310     <p>Typically, after test termination, the working directory is erased. See
 311     the <a href="#sec-command-line-options">"--preserve" command line option</a>
 312     for information on how to preserve the working directory content for failed
 313     tests for debugging purposes.</p>
 314
 315     <h2 id="sec-reference">Reference documentation</h2>
 316
 317     <p>The test system is composed of class <tt>Tester</tt>, derived form
 318     <tt>TestCmd.TestCmd</tt>, and helper class <tt>List</tt>. <tt>Tester</tt>
 319     and <tt>List</tt> methods are described below.</p>
 320
 321     <p>The documentation frequently refers to <tt>filename</tt>. In all cases,
 322     files are specified in unix style: a sequence of components, separated by
 323     "/". This is true on all platforms. In some contexts a list of files is
 324     allowed. In those cases any object with a sequence interface is allowed.</p>
 325
 326     <h3><a name="method-__init__">Method <tt>__init__(self, arguments="",
 327     executable="bjam", match=TestCmd.match_exact, boost_build_path=None,
 328     translate_suffixes=True, pass_toolset=True, use_test_config=True,
 329     ignore_toolset_requirements=True, workdir="", **keywords)</tt></a></h3>
 330
 331     <p><b>Optional arguments:</b></p>
 332
 333     <ul>
 334         <li><tt>arguments</tt>
 335             -- Arguments passed to the run executable.</li>
 336         <li><tt>executable</tt>
 337             -- Name of the executable to invoke.</li>
 338         <li><tt>match</tt>
 339             -- Function to use for compating actual and expected file contents.
 340             </li>
 341         <li><tt>boost_build_path</tt>
 342             -- Boost build path to be passed to the run executable.</li>
 343         <li><tt>translate_suffixes</tt>
 344             -- Whether to update suffixes on the the file names passed from the
 345             test script so they match those actually created by the current
 346             toolset. For example, static library files are specified by using
 347             the .lib suffix but when the 'gcc' toolset is used it actually
 348             creates them using the .a suffix.</li>
 349         <li><tt>pass_toolset</tt>
 350             -- Whether the test system should pass the specified toolset to the
 351             run executable.</li>
 352         <li><tt>use_test_config</tt>
 353             -- Whether the test system should tell the run executable to read in
 354             the test_config.jam configuration file.</li>
 355         <li><tt>ignore_toolset_requirements</tt>
 356             -- Whether the test system should tell the run executable to ignore
 357             toolset requirements.</li>
 358         <li><tt>workdir</tt>
 359             -- Indicates an absolute directory where the test will be run from.
 360             </li>
 361     </ul>
 362
 363     <p><b>Optional arguments inherited from the base class:</b></p>
 364
 365     <ul>
 366         <li><tt>description</tt>
 367             -- Test description string displayed in case of a failed test.</li>
 368         <li><tt>subdir</tt>
 369             -- List of subdirectories to automatically create under the working
 370               directory. Each subdirectory needs to be specified separately
 371               parent coming before its child.</li>
 372         <li><tt>verbose</tt>
 373             -- Flag that may be used to enable more verbose test system output.
 374             Note that it does not also enable more verbose build system output
 375             like the <a href="#sec-command-line-options">"--verbose" command
 376             line option</a> does.</li>
 377     </ul>
 378
 379     <p><b>Effects:</b></p>
 380
 381     <ol>
 382       <li>Remembers the current working directory in member
 383       <tt>original_workdir</tt>.</li>
 384
 385       <li>Determines the location of the executable (<code>bjam</code> by
 386       default) and build system files, assuming that the current directory is
 387       <tt>tools/build/test</tt>. Formulates jam invocation command, which
 388       will include explicit setting for the <tt>BOOST_BUILD_PATH</tt> variable
 389       and arguments passed to this methods, if any. This command will be used
 390       by subsequent invocation of <a href="#method-run_build_system"><tt>
 391       run_build_system</tt></a>. Finally, initializes the base class.</li>
 392
 393       <li>Changes the current working directory to the temporary working
 394       directory created by the base constructor.</li>
 395
 396       <li>If you want to run a test in an existing directory, pass it as
 397       <tt>workdir</tt>.</li>
 398
 399       <li> Most parameters passed to this constructor function may be overruled
 400       for each specific test system run using <a href=
 401       "#method-run_build_system"><tt>run_build_system</tt></a> parameters.
 402     </ol>
 403
 404     <h3><a name="method-set_tree">Method <tt>set_tree(self,
 405     tree_location)</tt></a></h3>
 406
 407     <p><b>Effects:</b></p>
 408
 409     <p>Replaces the content of the current working directory with the content
 410     of directory at <tt>tree_location</tt>. If <tt>tree_location</tt> is not
 411     absolute pathname, it will be treated as relative to
 412     <tt>self.original_workdir</tt>. This methods also explicitly makes the
 413     copied files writeable.</p>
 414
 415     <h3><a name="method-write">Method <tt>write(self, name,
 416     content)</tt></a></h3>
 417
 418     <p><b>Effects:</b></p>
 419
 420     <p>Writes the specified content to the file given by <tt>name</tt> under
 421     the temporary working directory. If the file already exists, it is
 422     overwritten. Any required directories are automatically created.</p>
 423
 424     <h3><a name="method-copy">Method <tt>copy(self, src, dst)</tt></a></h3>
 425
 426     <p><b>Effects:</b></p>
 427
 428     <p>Equvivalent to <tt>self.write(self.read(src), dst)</tt>.</p>
 429
 430     <h3><a name="method-touch">Method <tt>touch(self, names)</tt></a></h3>
 431
 432     <p><b>Effects:</b></p>
 433
 434     <p>Sets the access and modification times for all files in <tt>names</tt> to
 435     the current time. All the elements in <tt>names</tt> should be relative
 436     paths.</p>
 437
 438     <h3><a name="method-run_build_system">Method <tt>run_build_system(self,
 439     extra_args="", subdir="", stdout=None, stderr="", status=0, match=None,
 440     pass_toolset=None, use_test_config=None, ignore_toolset_requirements=None,
 441     expected_duration=None, **kw)</tt></a></h3>
 442
 443     <p><b>Effects:</b></p>
 444
 445     <ol>
 446       <li>Stores the state of the working directory in
 447       <tt>self.previous_tree</tt>.</li>
 448
 449       <li>Changes to <tt>subdir</tt>, if it is specified. It is relative to
 450       the <tt>original_workdir</tt> or the workdir specified in
 451       <tt>__init</tt>.</li>
 452
 453       <li>Invokes the <tt>bjam</tt> executable, passing <tt>extra_args</tt>
 454       to it. The binary should be located under
 455       <tt>&lt;test_invocation_dir&gt;/../jam/src/bin.&lt;platform&gt;</tt>.
 456       This is to make sure tests use the version of jam build from CVS.</li>
 457
 458       <li>Compares the stdout, stderr and exit status of build system
 459       invocation with values to appropriate parameters, if they are not
 460       <tt>None</tt>. If any difference is found, the test fails.</li>
 461
 462       <li>If the <tt>expected_duration</tt> parameter is specified then it
 463       represents the maximal allowed time in seconds for the test to run. The
 464       test will be marked as failed if its duration is greater than the given
 465       <tt>expected_duration</tt> parameter value.</li>
 466
 467       <li>Stores the new state of the working directory in <tt>self.tree</tt>.
 468       Computes the difference between previous and current trees and stores them
 469       in variables <tt>self.tree_difference</tt> and
 470       <tt>self.unexpected_difference</tt>. Both variables are instances of class
 471       <tt>tree.Trees_different</tt>, which have four attributes:
 472       <tt>added_files</tt>, <tt>removed_files</tt>, <tt>modified_files</tt> and
 473       <tt>touched_files</tt>. Each is a list of strings.</p></li>
 474     </ol>
 475
 476     <h3><a name="method-read">Method <tt>read(self, name)</tt></a></h3>
 477
 478     <p><b>Effects:</b></p>
 479
 480     <p>Read the specified file and returns it content. Raises an exception is
 481     the file is absent.</p>
 482
 483     <h3><a name="method-read_and_strip">Method <tt>read_and_strip(self, name)
 484     </tt></a></h3>
 485
 486     <p><b>Effects:</b></p>
 487
 488     <p>Read the specified file and returns it content, after removing trailing
 489     whitespace from every line. Raises an exception is the file is absent.</p>
 490
 491     <p><b>Rationale:</b></p>
 492
 493     <p>Although this method is questionable, there are a lot of cases when jam
 494     or shells it uses insert spaces. It seems that introducing this method is
 495     much simpler than dealing with all those cases.</p>
 496
 497     <h3><a name="methods-expectations">Methods for declaring expectations</a>
 498     </h3>
 499
 500     <p>Accordingly to the number of changes kinds that are detected, there are
 501     four methods that specify that test author expects a specific change to
 502     occur. They check <tt>self.unexpected_difference</tt>, and if the change is
 503     present there, it is removed. Otherwise, test fails.</p>
 504
 505     <p>Each method accepts a list of names. Those names use <tt>/</tt> path
 506     separator on all systems. Additionally, the test system translates suffixes
 507     appropriately. For the test to be portable, suffixes should use Windows
 508     convention: <tt>exe</tt> for executables, <tt>dll</tt> for dynamic libraries
 509     and <tt>lib</tt> for static libraries. Lastly, the string "$toolset" in file
 510     names is replaced by the name of tested toolset.</p>
 511
 512     <p><b>Note:</b> The <tt>List</tt> helper class might be useful to create
 513     lists of names.</p>
 514
 515     <p><b>Note:</b> The file content can be examined using the
 516     <tt>TestCmd.read</tt> function.</p>
 517
 518     <p>The members are:</p>
 519
 520     <ul>
 521       <li>expect_addition</li>
 522       <li>expect_removal</li>
 523       <li>expect_modification</li>
 524       <li>expect_nothing</li>
 525     </ul>
 526
 527     <p>Note that <tt>expect_modification</tt> is used to check that a either
 528     file content or timestamp has changed. The rationale is that some compilers
 529     change content even if sources does not change, and it's easier to have a
 530     method which checks for both content and time changes.</p>
 531
 532     <p>There's also a member <tt>expect_nothing_more</tt>, which checks that all
 533     the changes are either expected or ignored, in other words that
 534     <tt>unexpected_difference</tt> is empty by now.</p>
 535
 536     <p>Lastly, there's a method to compare file content with expected content:
 537     </p>
 538     <tt>expect_content(self, name, content, exact=0)</tt>
 539
 540     <p>The method fails the test if the content of file identified by 'name' is
 541     different from 'content'. If 'exact' is true, the file content is used
 542     as-is, otherwise, two transformations are applied:</p>
 543
 544     <ul>
 545       <li>The <tt>read_and_strip</tt> method is used to read the file, which
 546       removes trailing whitespace</li>
 547
 548       <li>Each backslash in the file content is converted to forward slash.</li>
 549     </ul>
 550
 551     <h3><a name="methods-ignoring">Methods for ignoring changes</a></h3>
 552
 553     <p>There are five methods which ignore changes made to the working tree.
 554     They silently remove elements from <tt>self.unexpected_difference</tt>, and
 555     don't generate error if element is not found. They accept shell style
 556     wildcard.</p>
 557
 558     <p>The following methods correspond to four kinds of changes:</p>
 559
 560     <ul>
 561       <li>ignore_addition(self, wildcard)</li>
 562       <li>ignore_removal(self, wildcard)</li>
 563       <li>ignore_modification(self, wildcard)</li>
 564       <li>ignore_touch(self, wildcard)</li>
 565     </ul>
 566
 567     <p>The method <tt>ignore(self, wildcard)</tt> ignores all the changes made
 568     to files that match a wildcard.</p>
 569
 570     <h3><a name="methods-result">Methods for explicitly specifying results</a>
 571     </h3>
 572
 573     <h4>Method <tt>pass_test(self, condition=1)</tt></h4>
 574
 575     <div class="attention">
 576       At this moment, the method should not be used.
 577     </div>
 578
 579     <h4>Method <tt>fail_test(self, condition=1)</tt></h4>
 580
 581     <p><b>Effects:</b> Cause the test to fail if <tt>condition</tt> is true.</p>
 582
 583     <h3><a name="class-list">Helper class <tt>List</tt></a></h3>
 584     The class has sequence interface and two additional methods.
 585
 586     <h4>Method <tt>__init__(self, string)</tt></h4>
 587
 588     <p><b>Effects:</b> Splits the string on unescaped spaces and tabs. The split
 589     components can further be retrieved using standard sequence access.</p>
 590
 591     <h4>Method <tt>__mul__(self, other)</tt></h4>
 592
 593     <p><b>Effects:</b> Returns an <tt>List</tt> instance, which elements are all
 594     possible concatenations of two string, first of which is from <tt>self</tt>,
 595     and second of which is from <tt>other</tt>.</p>
 596
 597     <p>The class also defines <tt>__str__</tt> and <tt>__repr__</tt> methods.
 598     Finally, there's <tt>__coerce__</tt> method which allows to convert strings
 599     to instances of <tt>List</tt>.</p>
 600
 601     <p><b>Example:</b></p>
 602 <pre>
 603     l = "a b" * List("c d")
 604     for e in l:
 605         print e
 606 </pre>
 607
 608     <p>will output:</p>
 609 <pre>
 610     ac
 611     ad
 612     bc
 613     bd
 614
 615 </pre>
 616     <hr>
 617     <p class="revision">Last modified: May 02, 2008</p>
 618     <p>&copy; Copyright Vladimir Prus 2002, 2003, 2004, 2005.<br>
 619     &copy; Copyright Jurko Gospodnetic 2008.<br>
 620     Distributed under the Boost Software License, Version 1.0.
 621     (See accompanying file LICENSE.txt or https://www.bfgroup.xyz/b2/LICENSE.txt)</p>
 622   </body>
 623 </html>