mingw/html/lib/B/Deparse.html

   1 <?xml version="1.0" ?>
   2 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
   3 <html xmlns="http://www.w3.org/1999/xhtml">
   4 <head>
   5 <title>B::Deparse - Perl compiler backend to produce perl code</title>
   6 <meta http-equiv="content-type" content="text/html; charset=utf-8" />
   7 <link rev="made" href="mailto:" />
   8 </head>
   9
  10 <body style="background-color: white">
  11 <table border="0" width="100%" cellspacing="0" cellpadding="3">
  12 <tr><td class="block" style="background-color: #cccccc" valign="middle">
  13 <big><strong><span class="block">&nbsp;B::Deparse - Perl compiler backend to produce perl code</span></strong></big>
  14 </td></tr>
  15 </table>
  16
  17 <p><a name="__index__"></a></p>
  18 <!-- INDEX BEGIN -->
  19
  20 <ul>
  21
  22         <li><a href="#name">NAME</a></li>
  23         <li><a href="#synopsis">SYNOPSIS</a></li>
  24         <li><a href="#description">DESCRIPTION</a></li>
  25         <li><a href="#options">OPTIONS</a></li>
  26         <li><a href="#using_b__deparse_as_a_module">USING B::Deparse AS A MODULE</a></li>
  27         <ul>
  28
  29                 <li><a href="#synopsis">Synopsis</a></li>
  30                 <li><a href="#description">Description</a></li>
  31                 <li><a href="#new">new</a></li>
  32                 <li><a href="#ambient_pragmas">ambient_pragmas</a></li>
  33                 <li><a href="#coderef2text">coderef2text</a></li>
  34         </ul>
  35
  36         <li><a href="#bugs">BUGS</a></li>
  37         <li><a href="#author">AUTHOR</a></li>
  38 </ul>
  39 <!-- INDEX END -->
  40
  41 <hr />
  42 <p>
  43 </p>
  44 <h1><a name="name">NAME</a></h1>
  45 <p>B::Deparse - Perl compiler backend to produce perl code</p>
  46 <p>
  47 </p>
  48 <hr />
  49 <h1><a name="synopsis">SYNOPSIS</a></h1>
  50 <p><strong>perl</strong> <strong>-MO=Deparse</strong>[<strong>,-d</strong>][<strong>,-f</strong><em>FILE</em>][<strong>,-p</strong>][<strong>,-q</strong>][<strong>,-l</strong>]
  51         [<strong>,-s</strong><em>LETTERS</em>][<strong>,-x</strong><em>LEVEL</em>] <em>prog.pl</em></p>
  52 <p>
  53 </p>
  54 <hr />
  55 <h1><a name="description">DESCRIPTION</a></h1>
  56 <p>B::Deparse is a backend module for the Perl compiler that generates
  57 perl source code, based on the internal compiled structure that perl
  58 itself creates after parsing a program. The output of B::Deparse won't
  59 be exactly the same as the original source, since perl doesn't keep
  60 track of comments or whitespace, and there isn't a one-to-one
  61 correspondence between perl's syntactical constructions and their
  62 compiled form, but it will often be close. When you use the <strong>-p</strong>
  63 option, the output also includes parentheses even when they are not
  64 required by precedence, which can make it easy to see if perl is
  65 parsing your expressions the way you intended.</p>
  66 <p>While B::Deparse goes to some lengths to try to figure out what your
  67 original program was doing, some parts of the language can still trip
  68 it up; it still fails even on some parts of Perl's own test suite. If
  69 you encounter a failure other than the most common ones described in
  70 the BUGS section below, you can help contribute to B::Deparse's
  71 ongoing development by submitting a bug report with a small
  72 example.</p>
  73 <p>
  74 </p>
  75 <hr />
  76 <h1><a name="options">OPTIONS</a></h1>
  77 <p>As with all compiler backend options, these must follow directly after
  78 the '-MO=Deparse', separated by a comma but not any white space.</p>
  79 <dl>
  80 <dt><strong><a name="item__2dd"><strong>-d</strong></a></strong>
  81
  82 <dd>
  83 <p>Output data values (when they appear as constants) using Data::Dumper.
  84 Without this option, B::Deparse will use some simple routines of its
  85 own for the same purpose. Currently, Data::Dumper is better for some
  86 kinds of data (such as complex structures with sharing and
  87 self-reference) while the built-in routines are better for others
  88 (such as odd floating-point values).</p>
  89 </dd>
  90 </li>
  91 <dt><strong><a name="item__2dffile"><strong>-f</strong><em>FILE</em></a></strong>
  92
  93 <dd>
  94 <p>Normally, B::Deparse deparses the main code of a program, and all the subs
  95 defined in the same file. To include subs defined in other files, pass the
  96 <strong>-f</strong> option with the filename. You can pass the <strong>-f</strong> option several times, to
  97 include more than one secondary file.  (Most of the time you don't want to
  98 use it at all.)  You can also use this option to include subs which are
  99 defined in the scope of a <strong>#line</strong> directive with two parameters.</p>
 100 </dd>
 101 </li>
 102 <dt><strong><a name="item__2dl"><strong>-l</strong></a></strong>
 103
 104 <dd>
 105 <p>Add '#line' declarations to the output based on the line and file
 106 locations of the original code.</p>
 107 </dd>
 108 </li>
 109 <dt><strong><a name="item__2dp"><strong>-p</strong></a></strong>
 110
 111 <dd>
 112 <p>Print extra parentheses. Without this option, B::Deparse includes
 113 parentheses in its output only when they are needed, based on the
 114 structure of your program. With <strong>-p</strong>, it uses parentheses (almost)
 115 whenever they would be legal. This can be useful if you are used to
 116 LISP, or if you want to see how perl parses your input. If you say</p>
 117 </dd>
 118 <dd>
 119 <pre>
 120     if ($var &amp; 0x7f == 65) {print &quot;Gimme an A!&quot;}
 121     print ($which ? $a : $b), &quot;\n&quot;;
 122     $name = $ENV{USER} or &quot;Bob&quot;;</pre>
 123 </dd>
 124 <dd>
 125 <p><code>B::Deparse,-p</code> will print</p>
 126 </dd>
 127 <dd>
 128 <pre>
 129     if (($var &amp; 0)) {
 130         print('Gimme an A!')
 131     };
 132     (print(($which ? $a : $b)), '???');
 133     (($name = $ENV{'USER'}) or '???')</pre>
 134 </dd>
 135 <dd>
 136 <p>which probably isn't what you intended (the <code>'???'</code> is a sign that
 137 perl optimized away a constant value).</p>
 138 </dd>
 139 </li>
 140 <dt><strong><a name="item__2dp"><strong>-P</strong></a></strong>
 141
 142 <dd>
 143 <p>Disable prototype checking. With this option, all function calls are
 144 deparsed as if no prototype was defined for them. In other words,</p>
 145 </dd>
 146 <dd>
 147 <pre>
 148     perl -MO=Deparse,-P -e 'sub foo (\@) { 1 } foo @x'</pre>
 149 </dd>
 150 <dd>
 151 <p>will print</p>
 152 </dd>
 153 <dd>
 154 <pre>
 155     sub foo (\@) {
 156         1;
 157     }
 158     &amp;foo(\@x);</pre>
 159 </dd>
 160 <dd>
 161 <p>making clear how the parameters are actually passed to <code>foo</code>.</p>
 162 </dd>
 163 </li>
 164 <dt><strong><a name="item__2dq"><strong>-q</strong></a></strong>
 165
 166 <dd>
 167 <p>Expand double-quoted strings into the corresponding combinations of
 168 concatenation, uc, ucfirst, lc, lcfirst, quotemeta, and join. For
 169 instance, print</p>
 170 </dd>
 171 <dd>
 172 <pre>
 173     print &quot;Hello, $world, @ladies, \u$gentlemen\E, \u\L$me!&quot;;</pre>
 174 </dd>
 175 <dd>
 176 <p>as</p>
 177 </dd>
 178 <dd>
 179 <pre>
 180     print 'Hello, ' . $world . ', ' . join($&quot;, @ladies) . ', '
 181           . ucfirst($gentlemen) . ', ' . ucfirst(lc $me . '!');</pre>
 182 </dd>
 183 <dd>
 184 <p>Note that the expanded form represents the way perl handles such
 185 constructions internally -- this option actually turns off the reverse
 186 translation that B::Deparse usually does. On the other hand, note that
 187 <code>$x = &quot;$y&quot;</code> is not the same as <code>$x = $y</code>: the former makes the value
 188 of $y into a string before doing the assignment.</p>
 189 </dd>
 190 </li>
 191 <dt><strong><a name="item__2dsletters"><strong>-s</strong><em>LETTERS</em></a></strong>
 192
 193 <dd>
 194 <p>Tweak the style of B::Deparse's output. The letters should follow
 195 directly after the 's', with no space or punctuation. The following
 196 options are available:</p>
 197 </dd>
 198 <dl>
 199 <dt><strong><a name="item_c"><strong>C</strong></a></strong>
 200
 201 <dd>
 202 <p>Cuddle <code>elsif</code>, <code>else</code>, and <code>continue</code> blocks. For example, print</p>
 203 </dd>
 204 <dd>
 205 <pre>
 206     if (...) {
 207          ...
 208     } else {
 209          ...
 210     }</pre>
 211 </dd>
 212 <dd>
 213 <p>instead of</p>
 214 </dd>
 215 <dd>
 216 <pre>
 217     if (...) {
 218          ...
 219     }
 220     else {
 221          ...
 222     }</pre>
 223 </dd>
 224 <dd>
 225 <p>The default is not to cuddle.</p>
 226 </dd>
 227 </li>
 228 <dt><strong><a name="item_inumber"><strong>i</strong><em>NUMBER</em></a></strong>
 229
 230 <dd>
 231 <p>Indent lines by multiples of <em>NUMBER</em> columns. The default is 4 columns.</p>
 232 </dd>
 233 </li>
 234 <dt><strong><a name="item_t"><strong>T</strong></a></strong>
 235
 236 <dd>
 237 <p>Use tabs for each 8 columns of indent. The default is to use only spaces.
 238 For instance, if the style options are <strong>-si4T</strong>, a line that's indented
 239 3 times will be preceded by one tab and four spaces; if the options were
 240 <strong>-si8T</strong>, the same line would be preceded by three tabs.</p>
 241 </dd>
 242 </li>
 243 <dt><strong><a name="item_vstring_2e"><strong>v</strong><em>STRING</em><strong>.</strong></a></strong>
 244
 245 <dd>
 246 <p>Print <em>STRING</em> for the value of a constant that can't be determined
 247 because it was optimized away (mnemonic: this happens when a constant
 248 is used in <strong>v</strong>oid context). The end of the string is marked by a period.
 249 The string should be a valid perl expression, generally a constant.
 250 Note that unless it's a number, it probably needs to be quoted, and on
 251 a command line quotes need to be protected from the shell. Some
 252 conventional values include 0, 1, 42, '', 'foo', and
 253 'Useless use of constant omitted' (which may need to be
 254 <strong>-sv``'Useless use of constant omitted'.''</strong>
 255 or something similar depending on your shell). The default is '???'.
 256 If you're using B::Deparse on a module or other file that's require'd,
 257 you shouldn't use a value that evaluates to false, since the customary
 258 true constant at the end of a module will be in void context when the
 259 file is compiled as a main program.</p>
 260 </dd>
 261 </li>
 262 </dl>
 263 <dt><strong><a name="item__2dxlevel"><strong>-x</strong><em>LEVEL</em></a></strong>
 264
 265 <dd>
 266 <p>Expand conventional syntax constructions into equivalent ones that expose
 267 their internal operation. <em>LEVEL</em> should be a digit, with higher values
 268 meaning more expansion. As with <strong>-q</strong>, this actually involves turning off
 269 special cases in B::Deparse's normal operations.</p>
 270 </dd>
 271 <dd>
 272 <p>If <em>LEVEL</em> is at least 3, <code>for</code> loops will be translated into equivalent
 273 while loops with continue blocks; for instance</p>
 274 </dd>
 275 <dd>
 276 <pre>
 277     for ($i = 0; $i &lt; 10; ++$i) {
 278         print $i;
 279     }</pre>
 280 </dd>
 281 <dd>
 282 <p>turns into</p>
 283 </dd>
 284 <dd>
 285 <pre>
 286     $i = 0;
 287     while ($i &lt; 10) {
 288         print $i;
 289     } continue {
 290         ++$i
 291     }</pre>
 292 </dd>
 293 <dd>
 294 <p>Note that in a few cases this translation can't be perfectly carried back
 295 into the source code -- if the loop's initializer declares a my variable,
 296 for instance, it won't have the correct scope outside of the loop.</p>
 297 </dd>
 298 <dd>
 299 <p>If <em>LEVEL</em> is at least 5, <a href="file://C|\msysgit\mingw\html/pod/perlfunc.html#item_use"><code>use</code></a> declarations will be translated into
 300 <code>BEGIN</code> blocks containing calls to <a href="file://C|\msysgit\mingw\html/pod/perlfunc.html#item_require"><code>require</code></a> and <code>import</code>; for
 301 instance,</p>
 302 </dd>
 303 <dd>
 304 <pre>
 305     use strict 'refs';</pre>
 306 </dd>
 307 <dd>
 308 <p>turns into</p>
 309 </dd>
 310 <dd>
 311 <pre>
 312     sub BEGIN {
 313         require strict;
 314         do {
 315             'strict'-&gt;import('refs')
 316         };
 317     }</pre>
 318 </dd>
 319 <dd>
 320 <p>If <em>LEVEL</em> is at least 7, <code>if</code> statements will be translated into
 321 equivalent expressions using <code>&amp;&amp;</code>, <code>?:</code> and <code>do {}</code>; for instance</p>
 322 </dd>
 323 <dd>
 324 <pre>
 325     print 'hi' if $nice;
 326     if ($nice) {
 327         print 'hi';
 328     }
 329     if ($nice) {
 330         print 'hi';
 331     } else {
 332         print 'bye';
 333     }</pre>
 334 </dd>
 335 <dd>
 336 <p>turns into</p>
 337 </dd>
 338 <dd>
 339 <pre>
 340     $nice and print 'hi';
 341     $nice and do { print 'hi' };
 342     $nice ? do { print 'hi' } : do { print 'bye' };</pre>
 343 </dd>
 344 <dd>
 345 <p>Long sequences of elsifs will turn into nested ternary operators, which
 346 B::Deparse doesn't know how to indent nicely.</p>
 347 </dd>
 348 </li>
 349 </dl>
 350 <p>
 351 </p>
 352 <hr />
 353 <h1><a name="using_b__deparse_as_a_module">USING B::Deparse AS A MODULE</a></h1>
 354 <p>
 355 </p>
 356 <h2><a name="synopsis">Synopsis</a></h2>
 357 <pre>
 358     use B::Deparse;
 359     $deparse = B::Deparse-&gt;new(&quot;-p&quot;, &quot;-sC&quot;);
 360     $body = $deparse-&gt;coderef2text(\&amp;func);
 361     eval &quot;sub func $body&quot;; # the inverse operation</pre>
 362 <p>
 363 </p>
 364 <h2><a name="description">Description</a></h2>
 365 <p>B::Deparse can also be used on a sub-by-sub basis from other perl
 366 programs.</p>
 367 <p>
 368 </p>
 369 <h2><a name="new">new</a></h2>
 370 <pre>
 371     $deparse = B::Deparse-&gt;new(OPTIONS)</pre>
 372 <p>Create an object to store the state of a deparsing operation and any
 373 options. The options are the same as those that can be given on the
 374 command line (see <a href="#options">OPTIONS</a>); options that are separated by commas
 375 after <strong>-MO=Deparse</strong> should be given as separate strings. Some
 376 options, like <strong>-u</strong>, don't make sense for a single subroutine, so
 377 don't pass them.</p>
 378 <p>
 379 </p>
 380 <h2><a name="ambient_pragmas">ambient_pragmas</a></h2>
 381 <pre>
 382     $deparse-&gt;ambient_pragmas(strict =&gt; 'all', '$[' =&gt; $[);</pre>
 383 <p>The compilation of a subroutine can be affected by a few compiler
 384 directives, <strong>pragmas</strong>. These are:</p>
 385 <ul>
 386 <li>
 387 <p>use strict;</p>
 388 </li>
 389 <li>
 390 <p>use warnings;</p>
 391 </li>
 392 <li>
 393 <p>Assigning to the special variable $[</p>
 394 </li>
 395 <li>
 396 <p>use integer;</p>
 397 </li>
 398 <li>
 399 <p>use bytes;</p>
 400 </li>
 401 <li>
 402 <p>use utf8;</p>
 403 </li>
 404 <li>
 405 <p>use re;</p>
 406 </li>
 407 </ul>
 408 <p>Ordinarily, if you use B::Deparse on a subroutine which has
 409 been compiled in the presence of one or more of these pragmas,
 410 the output will include statements to turn on the appropriate
 411 directives. So if you then compile the code returned by coderef2text,
 412 it will behave the same way as the subroutine which you deparsed.</p>
 413 <p>However, you may know that you intend to use the results in a
 414 particular context, where some pragmas are already in scope. In
 415 this case, you use the <strong>ambient_pragmas</strong> method to describe the
 416 assumptions you wish to make.</p>
 417 <p>Not all of the options currently have any useful effect. See
 418 <a href="#bugs">BUGS</a> for more details.</p>
 419 <p>The parameters it accepts are:</p>
 420 <dl>
 421 <dt><strong><a name="item_strict">strict</a></strong>
 422
 423 <dd>
 424 <p>Takes a string, possibly containing several values separated
 425 by whitespace. The special values ``all'' and ``none'' mean what you'd
 426 expect.</p>
 427 </dd>
 428 <dd>
 429 <pre>
 430     $deparse-&gt;ambient_pragmas(strict =&gt; 'subs refs');</pre>
 431 </dd>
 432 </li>
 433 <dt><strong><a name="item___">$[</a></strong>
 434
 435 <dd>
 436 <p>Takes a number, the value of the array base $[.</p>
 437 </dd>
 438 </li>
 439 <dt><strong><a name="item_bytes">bytes</a></strong>
 440
 441 <dt><strong><a name="item_utf8">utf8</a></strong>
 442
 443 <dt><strong><a name="item_integer">integer</a></strong>
 444
 445 <dd>
 446 <p>If the value is true, then the appropriate pragma is assumed to
 447 be in the ambient scope, otherwise not.</p>
 448 </dd>
 449 </li>
 450 <dt><strong><a name="item_re">re</a></strong>
 451
 452 <dd>
 453 <p>Takes a string, possibly containing a whitespace-separated list of
 454 values. The values ``all'' and ``none'' are special. It's also permissible
 455 to pass an array reference here.</p>
 456 </dd>
 457 <dd>
 458 <pre>
 459     $deparser-&gt;ambient_pragmas(re =&gt; 'eval');</pre>
 460 </dd>
 461 </li>
 462 <dt><strong><a name="item_warnings">warnings</a></strong>
 463
 464 <dd>
 465 <p>Takes a string, possibly containing a whitespace-separated list of
 466 values. The values ``all'' and ``none'' are special, again. It's also
 467 permissible to pass an array reference here.</p>
 468 </dd>
 469 <dd>
 470 <pre>
 471     $deparser-&gt;ambient_pragmas(warnings =&gt; [qw[void io]]);</pre>
 472 </dd>
 473 <dd>
 474 <p>If one of the values is the string ``FATAL'', then all the warnings
 475 in that list will be considered fatal, just as with the <strong>warnings</strong>
 476 pragma itself. Should you need to specify that some warnings are
 477 fatal, and others are merely enabled, you can pass the <strong>warnings</strong>
 478 parameter twice:</p>
 479 </dd>
 480 <dd>
 481 <pre>
 482     $deparser-&gt;ambient_pragmas(
 483         warnings =&gt; 'all',
 484         warnings =&gt; [FATAL =&gt; qw/void io/],
 485     );</pre>
 486 </dd>
 487 <dd>
 488 <p>See <a href="file://C|\msysgit\mingw\html/pod/perllexwarn.html">the perllexwarn manpage</a> for more information about lexical warnings.</p>
 489 </dd>
 490 </li>
 491 <dt><strong><a name="item_hint_bits">hint_bits</a></strong>
 492
 493 <dt><strong><a name="item_warning_bits">warning_bits</a></strong>
 494
 495 <dd>
 496 <p>These two parameters are used to specify the ambient pragmas in
 497 the format used by the special variables $^H and ${^WARNING_BITS}.</p>
 498 </dd>
 499 <dd>
 500 <p>They exist principally so that you can write code like:</p>
 501 </dd>
 502 <dd>
 503 <pre>
 504     { my ($hint_bits, $warning_bits);
 505     BEGIN {($hint_bits, $warning_bits) = ($^H, ${^WARNING_BITS})}
 506     $deparser-&gt;ambient_pragmas (
 507         hint_bits    =&gt; $hint_bits,
 508         warning_bits =&gt; $warning_bits,
 509         '$['         =&gt; 0 + $[
 510     ); }</pre>
 511 </dd>
 512 <dd>
 513 <p>which specifies that the ambient pragmas are exactly those which
 514 are in scope at the point of calling.</p>
 515 </dd>
 516 </li>
 517 </dl>
 518 <p>
 519 </p>
 520 <h2><a name="coderef2text">coderef2text</a></h2>
 521 <pre>
 522     $body = $deparse-&gt;coderef2text(\&amp;func)
 523     $body = $deparse-&gt;coderef2text(sub ($$) { ... })</pre>
 524 <p>Return source code for the body of a subroutine (a block, optionally
 525 preceded by a prototype in parens), given a reference to the
 526 sub. Because a subroutine can have no names, or more than one name,
 527 this method doesn't return a complete subroutine definition -- if you
 528 want to eval the result, you should prepend ``sub subname '', or ``sub ''
 529 for an anonymous function constructor. Unless the sub was defined in
 530 the main:: package, the code will include a package declaration.</p>
 531 <p>
 532 </p>
 533 <hr />
 534 <h1><a name="bugs">BUGS</a></h1>
 535 <ul>
 536 <li>
 537 <p>The only pragmas to be completely supported are: <code>use warnings</code>,
 538 <code>use strict 'refs'</code>, <code>use bytes</code>, and <code>use integer</code>. (<a href="#item___"><code>$[</code></a>, which
 539 behaves like a pragma, is also supported.)</p>
 540 <p>Excepting those listed above, we're currently unable to guarantee that
 541 B::Deparse will produce a pragma at the correct point in the program.
 542 (Specifically, pragmas at the beginning of a block often appear right
 543 before the start of the block instead.)
 544 Since the effects of pragmas are often lexically scoped, this can mean
 545 that the pragma holds sway over a different portion of the program
 546 than in the input file.</p>
 547 </li>
 548 <li>
 549 <p>In fact, the above is a specific instance of a more general problem:
 550 we can't guarantee to produce BEGIN blocks or <a href="file://C|\msysgit\mingw\html/pod/perlfunc.html#item_use"><code>use</code></a> declarations in
 551 exactly the right place. So if you use a module which affects compilation
 552 (such as by over-riding keywords, overloading constants or whatever)
 553 then the output code might not work as intended.</p>
 554 <p>This is the most serious outstanding problem, and will require some help
 555 from the Perl core to fix.</p>
 556 </li>
 557 <li>
 558 <p>If a keyword is over-ridden, and your program explicitly calls
 559 the built-in version by using CORE::keyword, the output of B::Deparse
 560 will not reflect this. If you run the resulting code, it will call
 561 the over-ridden version rather than the built-in one. (Maybe there
 562 should be an option to <strong>always</strong> print keyword calls as <code>CORE::name</code>.)</p>
 563 </li>
 564 <li>
 565 <p>Some constants don't print correctly either with or without <strong>-d</strong>.
 566 For instance, neither B::Deparse nor Data::Dumper know how to print
 567 dual-valued scalars correctly, as in:</p>
 568 <pre>
 569     use constant E2BIG =&gt; ($!=7); $y = E2BIG; print $y, 0+$y;</pre>
 570 </li>
 571 <li>
 572 <p>An input file that uses source filtering probably won't be deparsed into
 573 runnable code, because it will still include the <strong>use</strong> declaration
 574 for the source filtering module, even though the code that is
 575 produced is already ordinary Perl which shouldn't be filtered again.</p>
 576 </li>
 577 <li>
 578 <p>Optimised away statements are rendered as '???'. This includes statements that
 579 have a compile-time side-effect, such as the obscure</p>
 580 <pre>
 581     my $x if 0;</pre>
 582 <p>which is not, consequently, deparsed correctly.</p>
 583 </li>
 584 <li>
 585 <p>There are probably many more bugs on non-ASCII platforms (EBCDIC).</p>
 586 </li>
 587 </ul>
 588 <p>
 589 </p>
 590 <hr />
 591 <h1><a name="author">AUTHOR</a></h1>
 592 <p>Stephen McCamant &lt;<a href="mailto:smcc@CSUA.Berkeley.EDU">smcc@CSUA.Berkeley.EDU</a>&gt;, based on an earlier version
 593 by Malcolm Beattie &lt;<a href="mailto:mbeattie@sable.ox.ac.uk">mbeattie@sable.ox.ac.uk</a>&gt;, with contributions from
 594 Gisle Aas, James Duncan, Albert Dvornik, Robin Houston, Dave Mitchell,
 595 Hugo van der Sanden, Gurusamy Sarathy, Nick Ing-Simmons, and Rafael
 596 Garcia-Suarez.</p>
 597 <table border="0" width="100%" cellspacing="0" cellpadding="3">
 598 <tr><td class="block" style="background-color: #cccccc" valign="middle">
 599 <big><strong><span class="block">&nbsp;B::Deparse - Perl compiler backend to produce perl code</span></strong></big>
 600 </td></tr>
 601 </table>
 602
 603 </body>
 604
 605 </html>