| 1 | # -*- mode: perl; perl-indent-level: 2; -*- |
| 2 | # Memoize.pm |
| 3 | # |
| 4 | # Transparent memoization of idempotent functions |
| 5 | # |
| 6 | # Copyright 1998, 1999, 2000, 2001 M-J. Dominus. |
| 7 | # You may copy and distribute this program under the |
| 8 | # same terms as Perl itself. If in doubt, |
| 9 | # write to mjd-perl-memoize+@plover.com for a license. |
| 10 | # |
| 11 | # Version 1.01 $Revision: 1.18 $ $Date: 2001/06/24 17:16:47 $ |
| 12 | |
| 13 | package Memoize; |
| 14 | $VERSION = '1.01_01'; |
| 15 | |
| 16 | # Compile-time constants |
| 17 | sub SCALAR () { 0 } |
| 18 | sub LIST () { 1 } |
| 19 | |
| 20 | |
| 21 | # |
| 22 | # Usage memoize(functionname/ref, |
| 23 | # { NORMALIZER => coderef, INSTALL => name, |
| 24 | # LIST_CACHE => descriptor, SCALAR_CACHE => descriptor } |
| 25 | # |
| 26 | |
| 27 | use Carp; |
| 28 | use Exporter; |
| 29 | use vars qw($DEBUG); |
| 30 | use Config; # Dammit. |
| 31 | @ISA = qw(Exporter); |
| 32 | @EXPORT = qw(memoize); |
| 33 | @EXPORT_OK = qw(unmemoize flush_cache); |
| 34 | use strict; |
| 35 | |
| 36 | my %memotable; |
| 37 | my %revmemotable; |
| 38 | my @CONTEXT_TAGS = qw(MERGE TIE MEMORY FAULT HASH); |
| 39 | my %IS_CACHE_TAG = map {($_ => 1)} @CONTEXT_TAGS; |
| 40 | |
| 41 | # Raise an error if the user tries to specify one of thesepackage as a |
| 42 | # tie for LIST_CACHE |
| 43 | |
| 44 | my %scalar_only = map {($_ => 1)} qw(DB_File GDBM_File SDBM_File ODBM_File NDBM_File); |
| 45 | |
| 46 | sub memoize { |
| 47 | my $fn = shift; |
| 48 | my %options = @_; |
| 49 | my $options = \%options; |
| 50 | |
| 51 | unless (defined($fn) && |
| 52 | (ref $fn eq 'CODE' || ref $fn eq '')) { |
| 53 | croak "Usage: memoize 'functionname'|coderef {OPTIONS}"; |
| 54 | } |
| 55 | |
| 56 | my $uppack = caller; # TCL me Elmo! |
| 57 | my $cref; # Code reference to original function |
| 58 | my $name = (ref $fn ? undef : $fn); |
| 59 | |
| 60 | # Convert function names to code references |
| 61 | $cref = &_make_cref($fn, $uppack); |
| 62 | |
| 63 | # Locate function prototype, if any |
| 64 | my $proto = prototype $cref; |
| 65 | if (defined $proto) { $proto = "($proto)" } |
| 66 | else { $proto = "" } |
| 67 | |
| 68 | # I would like to get rid of the eval, but there seems not to be any |
| 69 | # other way to set the prototype properly. The switch here for |
| 70 | # 'usethreads' works around a bug in threadperl having to do with |
| 71 | # magic goto. It would be better to fix the bug and use the magic |
| 72 | # goto version everywhere. |
| 73 | my $wrapper = |
| 74 | $Config{usethreads} |
| 75 | ? eval "sub $proto { &_memoizer(\$cref, \@_); }" |
| 76 | : eval "sub $proto { unshift \@_, \$cref; goto &_memoizer; }"; |
| 77 | |
| 78 | my $normalizer = $options{NORMALIZER}; |
| 79 | if (defined $normalizer && ! ref $normalizer) { |
| 80 | $normalizer = _make_cref($normalizer, $uppack); |
| 81 | } |
| 82 | |
| 83 | my $install_name; |
| 84 | if (defined $options->{INSTALL}) { |
| 85 | # INSTALL => name |
| 86 | $install_name = $options->{INSTALL}; |
| 87 | } elsif (! exists $options->{INSTALL}) { |
| 88 | # No INSTALL option provided; use original name if possible |
| 89 | $install_name = $name; |
| 90 | } else { |
| 91 | # INSTALL => undef means don't install |
| 92 | } |
| 93 | |
| 94 | if (defined $install_name) { |
| 95 | $install_name = $uppack . '::' . $install_name |
| 96 | unless $install_name =~ /::/; |
| 97 | no strict; |
| 98 | local($^W) = 0; # ``Subroutine $install_name redefined at ...'' |
| 99 | *{$install_name} = $wrapper; # Install memoized version |
| 100 | } |
| 101 | |
| 102 | $revmemotable{$wrapper} = "" . $cref; # Turn code ref into hash key |
| 103 | |
| 104 | # These will be the caches |
| 105 | my %caches; |
| 106 | for my $context (qw(SCALAR LIST)) { |
| 107 | # suppress subsequent 'uninitialized value' warnings |
| 108 | $options{"${context}_CACHE"} ||= ''; |
| 109 | |
| 110 | my $cache_opt = $options{"${context}_CACHE"}; |
| 111 | my @cache_opt_args; |
| 112 | if (ref $cache_opt) { |
| 113 | @cache_opt_args = @$cache_opt; |
| 114 | $cache_opt = shift @cache_opt_args; |
| 115 | } |
| 116 | if ($cache_opt eq 'FAULT') { # no cache |
| 117 | $caches{$context} = undef; |
| 118 | } elsif ($cache_opt eq 'HASH') { # user-supplied hash |
| 119 | my $cache = $cache_opt_args[0]; |
| 120 | my $package = ref(tied %$cache); |
| 121 | if ($context eq 'LIST' && $scalar_only{$package}) { |
| 122 | croak("You can't use $package for LIST_CACHE because it can only store scalars"); |
| 123 | } |
| 124 | $caches{$context} = $cache; |
| 125 | } elsif ($cache_opt eq '' || $IS_CACHE_TAG{$cache_opt}) { |
| 126 | # default is that we make up an in-memory hash |
| 127 | $caches{$context} = {}; |
| 128 | # (this might get tied later, or MERGEd away) |
| 129 | } else { |
| 130 | croak "Unrecognized option to `${context}_CACHE': `$cache_opt' should be one of (@CONTEXT_TAGS); aborting"; |
| 131 | } |
| 132 | } |
| 133 | |
| 134 | # Perhaps I should check here that you didn't supply *both* merge |
| 135 | # options. But if you did, it does do something reasonable: They |
| 136 | # both get merged to the same in-memory hash. |
| 137 | if ($options{SCALAR_CACHE} eq 'MERGE') { |
| 138 | $caches{SCALAR} = $caches{LIST}; |
| 139 | } elsif ($options{LIST_CACHE} eq 'MERGE') { |
| 140 | $caches{LIST} = $caches{SCALAR}; |
| 141 | } |
| 142 | |
| 143 | # Now deal with the TIE options |
| 144 | { |
| 145 | my $context; |
| 146 | foreach $context (qw(SCALAR LIST)) { |
| 147 | # If the relevant option wasn't `TIE', this call does nothing. |
| 148 | _my_tie($context, $caches{$context}, $options); # Croaks on failure |
| 149 | } |
| 150 | } |
| 151 | |
| 152 | # We should put some more stuff in here eventually. |
| 153 | # We've been saying that for serveral versions now. |
| 154 | # And you know what? More stuff keeps going in! |
| 155 | $memotable{$cref} = |
| 156 | { |
| 157 | O => $options, # Short keys here for things we need to access frequently |
| 158 | N => $normalizer, |
| 159 | U => $cref, |
| 160 | MEMOIZED => $wrapper, |
| 161 | PACKAGE => $uppack, |
| 162 | NAME => $install_name, |
| 163 | S => $caches{SCALAR}, |
| 164 | L => $caches{LIST}, |
| 165 | }; |
| 166 | |
| 167 | $wrapper # Return just memoized version |
| 168 | } |
| 169 | |
| 170 | # This function tries to load a tied hash class and tie the hash to it. |
| 171 | sub _my_tie { |
| 172 | my ($context, $hash, $options) = @_; |
| 173 | my $fullopt = $options->{"${context}_CACHE"}; |
| 174 | |
| 175 | # We already checked to make sure that this works. |
| 176 | my $shortopt = (ref $fullopt) ? $fullopt->[0] : $fullopt; |
| 177 | |
| 178 | return unless defined $shortopt && $shortopt eq 'TIE'; |
| 179 | carp("TIE option to memoize() is deprecated; use HASH instead") |
| 180 | if $^W; |
| 181 | |
| 182 | my @args = ref $fullopt ? @$fullopt : (); |
| 183 | shift @args; |
| 184 | my $module = shift @args; |
| 185 | if ($context eq 'LIST' && $scalar_only{$module}) { |
| 186 | croak("You can't use $module for LIST_CACHE because it can only store scalars"); |
| 187 | } |
| 188 | my $modulefile = $module . '.pm'; |
| 189 | $modulefile =~ s{::}{/}g; |
| 190 | eval { require $modulefile }; |
| 191 | if ($@) { |
| 192 | croak "Memoize: Couldn't load hash tie module `$module': $@; aborting"; |
| 193 | } |
| 194 | my $rc = (tie %$hash => $module, @args); |
| 195 | unless ($rc) { |
| 196 | croak "Memoize: Couldn't tie hash to `$module': $!; aborting"; |
| 197 | } |
| 198 | 1; |
| 199 | } |
| 200 | |
| 201 | sub flush_cache { |
| 202 | my $func = _make_cref($_[0], scalar caller); |
| 203 | my $info = $memotable{$revmemotable{$func}}; |
| 204 | die "$func not memoized" unless defined $info; |
| 205 | for my $context (qw(S L)) { |
| 206 | my $cache = $info->{$context}; |
| 207 | if (tied %$cache && ! (tied %$cache)->can('CLEAR')) { |
| 208 | my $funcname = defined($info->{NAME}) ? |
| 209 | "function $info->{NAME}" : "anonymous function $func"; |
| 210 | my $context = {S => 'scalar', L => 'list'}->{$context}; |
| 211 | croak "Tied cache hash for $context-context $funcname does not support flushing"; |
| 212 | } else { |
| 213 | %$cache = (); |
| 214 | } |
| 215 | } |
| 216 | } |
| 217 | |
| 218 | # This is the function that manages the memo tables. |
| 219 | sub _memoizer { |
| 220 | my $orig = shift; # stringized version of ref to original func. |
| 221 | my $info = $memotable{$orig}; |
| 222 | my $normalizer = $info->{N}; |
| 223 | |
| 224 | my $argstr; |
| 225 | my $context = (wantarray() ? LIST : SCALAR); |
| 226 | |
| 227 | if (defined $normalizer) { |
| 228 | no strict; |
| 229 | if ($context == SCALAR) { |
| 230 | $argstr = &{$normalizer}(@_); |
| 231 | } elsif ($context == LIST) { |
| 232 | ($argstr) = &{$normalizer}(@_); |
| 233 | } else { |
| 234 | croak "Internal error \#41; context was neither LIST nor SCALAR\n"; |
| 235 | } |
| 236 | } else { # Default normalizer |
| 237 | local $^W = 0; |
| 238 | $argstr = join chr(28),@_; |
| 239 | } |
| 240 | |
| 241 | if ($context == SCALAR) { |
| 242 | my $cache = $info->{S}; |
| 243 | _crap_out($info->{NAME}, 'scalar') unless $cache; |
| 244 | if (exists $cache->{$argstr}) { |
| 245 | return $cache->{$argstr}; |
| 246 | } else { |
| 247 | my $val = &{$info->{U}}(@_); |
| 248 | # Scalars are considered to be lists; store appropriately |
| 249 | if ($info->{O}{SCALAR_CACHE} eq 'MERGE') { |
| 250 | $cache->{$argstr} = [$val]; |
| 251 | } else { |
| 252 | $cache->{$argstr} = $val; |
| 253 | } |
| 254 | $val; |
| 255 | } |
| 256 | } elsif ($context == LIST) { |
| 257 | my $cache = $info->{L}; |
| 258 | _crap_out($info->{NAME}, 'list') unless $cache; |
| 259 | if (exists $cache->{$argstr}) { |
| 260 | my $val = $cache->{$argstr}; |
| 261 | # If LISTCONTEXT=>MERGE, then the function never returns lists, |
| 262 | # so we have a scalar value cached, so just return it straightaway: |
| 263 | return ($val) if $info->{O}{LIST_CACHE} eq 'MERGE'; |
| 264 | # Maybe in a later version we can use a faster test. |
| 265 | |
| 266 | # Otherwise, we cached an array containing the returned list: |
| 267 | return @$val; |
| 268 | } else { |
| 269 | my $q = $cache->{$argstr} = [&{$info->{U}}(@_)]; |
| 270 | @$q; |
| 271 | } |
| 272 | } else { |
| 273 | croak "Internal error \#42; context was neither LIST nor SCALAR\n"; |
| 274 | } |
| 275 | } |
| 276 | |
| 277 | sub unmemoize { |
| 278 | my $f = shift; |
| 279 | my $uppack = caller; |
| 280 | my $cref = _make_cref($f, $uppack); |
| 281 | |
| 282 | unless (exists $revmemotable{$cref}) { |
| 283 | croak "Could not unmemoize function `$f', because it was not memoized to begin with"; |
| 284 | } |
| 285 | |
| 286 | my $tabent = $memotable{$revmemotable{$cref}}; |
| 287 | unless (defined $tabent) { |
| 288 | croak "Could not figure out how to unmemoize function `$f'"; |
| 289 | } |
| 290 | my $name = $tabent->{NAME}; |
| 291 | if (defined $name) { |
| 292 | no strict; |
| 293 | local($^W) = 0; # ``Subroutine $install_name redefined at ...'' |
| 294 | *{$name} = $tabent->{U}; # Replace with original function |
| 295 | } |
| 296 | undef $memotable{$revmemotable{$cref}}; |
| 297 | undef $revmemotable{$cref}; |
| 298 | |
| 299 | # This removes the last reference to the (possibly tied) memo tables |
| 300 | # my ($old_function, $memotabs) = @{$tabent}{'U','S','L'}; |
| 301 | # undef $tabent; |
| 302 | |
| 303 | # # Untie the memo tables if they were tied. |
| 304 | # my $i; |
| 305 | # for $i (0,1) { |
| 306 | # if (tied %{$memotabs->[$i]}) { |
| 307 | # warn "Untying hash #$i\n"; |
| 308 | # untie %{$memotabs->[$i]}; |
| 309 | # } |
| 310 | # } |
| 311 | |
| 312 | $tabent->{U}; |
| 313 | } |
| 314 | |
| 315 | sub _make_cref { |
| 316 | my $fn = shift; |
| 317 | my $uppack = shift; |
| 318 | my $cref; |
| 319 | my $name; |
| 320 | |
| 321 | if (ref $fn eq 'CODE') { |
| 322 | $cref = $fn; |
| 323 | } elsif (! ref $fn) { |
| 324 | if ($fn =~ /::/) { |
| 325 | $name = $fn; |
| 326 | } else { |
| 327 | $name = $uppack . '::' . $fn; |
| 328 | } |
| 329 | no strict; |
| 330 | if (defined $name and !defined(&$name)) { |
| 331 | croak "Cannot operate on nonexistent function `$fn'"; |
| 332 | } |
| 333 | # $cref = \&$name; |
| 334 | $cref = *{$name}{CODE}; |
| 335 | } else { |
| 336 | my $parent = (caller(1))[3]; # Function that called _make_cref |
| 337 | croak "Usage: argument 1 to `$parent' must be a function name or reference.\n"; |
| 338 | } |
| 339 | $DEBUG and warn "${name}($fn) => $cref in _make_cref\n"; |
| 340 | $cref; |
| 341 | } |
| 342 | |
| 343 | sub _crap_out { |
| 344 | my ($funcname, $context) = @_; |
| 345 | if (defined $funcname) { |
| 346 | croak "Function `$funcname' called in forbidden $context context; faulting"; |
| 347 | } else { |
| 348 | croak "Anonymous function called in forbidden $context context; faulting"; |
| 349 | } |
| 350 | } |
| 351 | |
| 352 | 1; |
| 353 | |
| 354 | |
| 355 | |
| 356 | |
| 357 | |
| 358 | =head1 NAME |
| 359 | |
| 360 | Memoize - Make functions faster by trading space for time |
| 361 | |
| 362 | =head1 SYNOPSIS |
| 363 | |
| 364 | # This is the documentation for Memoize 1.01 |
| 365 | use Memoize; |
| 366 | memoize('slow_function'); |
| 367 | slow_function(arguments); # Is faster than it was before |
| 368 | |
| 369 | |
| 370 | This is normally all you need to know. However, many options are available: |
| 371 | |
| 372 | memoize(function, options...); |
| 373 | |
| 374 | Options include: |
| 375 | |
| 376 | NORMALIZER => function |
| 377 | INSTALL => new_name |
| 378 | |
| 379 | SCALAR_CACHE => 'MEMORY' |
| 380 | SCALAR_CACHE => ['HASH', \%cache_hash ] |
| 381 | SCALAR_CACHE => 'FAULT' |
| 382 | SCALAR_CACHE => 'MERGE' |
| 383 | |
| 384 | LIST_CACHE => 'MEMORY' |
| 385 | LIST_CACHE => ['HASH', \%cache_hash ] |
| 386 | LIST_CACHE => 'FAULT' |
| 387 | LIST_CACHE => 'MERGE' |
| 388 | |
| 389 | =head1 DESCRIPTION |
| 390 | |
| 391 | `Memoizing' a function makes it faster by trading space for time. It |
| 392 | does this by caching the return values of the function in a table. |
| 393 | If you call the function again with the same arguments, C<memoize> |
| 394 | jumps in and gives you the value out of the table, instead of letting |
| 395 | the function compute the value all over again. |
| 396 | |
| 397 | Here is an extreme example. Consider the Fibonacci sequence, defined |
| 398 | by the following function: |
| 399 | |
| 400 | # Compute Fibonacci numbers |
| 401 | sub fib { |
| 402 | my $n = shift; |
| 403 | return $n if $n < 2; |
| 404 | fib($n-1) + fib($n-2); |
| 405 | } |
| 406 | |
| 407 | This function is very slow. Why? To compute fib(14), it first wants |
| 408 | to compute fib(13) and fib(12), and add the results. But to compute |
| 409 | fib(13), it first has to compute fib(12) and fib(11), and then it |
| 410 | comes back and computes fib(12) all over again even though the answer |
| 411 | is the same. And both of the times that it wants to compute fib(12), |
| 412 | it has to compute fib(11) from scratch, and then it has to do it |
| 413 | again each time it wants to compute fib(13). This function does so |
| 414 | much recomputing of old results that it takes a really long time to |
| 415 | run---fib(14) makes 1,200 extra recursive calls to itself, to compute |
| 416 | and recompute things that it already computed. |
| 417 | |
| 418 | This function is a good candidate for memoization. If you memoize the |
| 419 | `fib' function above, it will compute fib(14) exactly once, the first |
| 420 | time it needs to, and then save the result in a table. Then if you |
| 421 | ask for fib(14) again, it gives you the result out of the table. |
| 422 | While computing fib(14), instead of computing fib(12) twice, it does |
| 423 | it once; the second time it needs the value it gets it from the table. |
| 424 | It doesn't compute fib(11) four times; it computes it once, getting it |
| 425 | from the table the next three times. Instead of making 1,200 |
| 426 | recursive calls to `fib', it makes 15. This makes the function about |
| 427 | 150 times faster. |
| 428 | |
| 429 | You could do the memoization yourself, by rewriting the function, like |
| 430 | this: |
| 431 | |
| 432 | # Compute Fibonacci numbers, memoized version |
| 433 | { my @fib; |
| 434 | sub fib { |
| 435 | my $n = shift; |
| 436 | return $fib[$n] if defined $fib[$n]; |
| 437 | return $fib[$n] = $n if $n < 2; |
| 438 | $fib[$n] = fib($n-1) + fib($n-2); |
| 439 | } |
| 440 | } |
| 441 | |
| 442 | Or you could use this module, like this: |
| 443 | |
| 444 | use Memoize; |
| 445 | memoize('fib'); |
| 446 | |
| 447 | # Rest of the fib function just like the original version. |
| 448 | |
| 449 | This makes it easy to turn memoizing on and off. |
| 450 | |
| 451 | Here's an even simpler example: I wrote a simple ray tracer; the |
| 452 | program would look in a certain direction, figure out what it was |
| 453 | looking at, and then convert the `color' value (typically a string |
| 454 | like `red') of that object to a red, green, and blue pixel value, like |
| 455 | this: |
| 456 | |
| 457 | for ($direction = 0; $direction < 300; $direction++) { |
| 458 | # Figure out which object is in direction $direction |
| 459 | $color = $object->{color}; |
| 460 | ($r, $g, $b) = @{&ColorToRGB($color)}; |
| 461 | ... |
| 462 | } |
| 463 | |
| 464 | Since there are relatively few objects in a picture, there are only a |
| 465 | few colors, which get looked up over and over again. Memoizing |
| 466 | C<ColorToRGB> sped up the program by several percent. |
| 467 | |
| 468 | =head1 DETAILS |
| 469 | |
| 470 | This module exports exactly one function, C<memoize>. The rest of the |
| 471 | functions in this package are None of Your Business. |
| 472 | |
| 473 | You should say |
| 474 | |
| 475 | memoize(function) |
| 476 | |
| 477 | where C<function> is the name of the function you want to memoize, or |
| 478 | a reference to it. C<memoize> returns a reference to the new, |
| 479 | memoized version of the function, or C<undef> on a non-fatal error. |
| 480 | At present, there are no non-fatal errors, but there might be some in |
| 481 | the future. |
| 482 | |
| 483 | If C<function> was the name of a function, then C<memoize> hides the |
| 484 | old version and installs the new memoized version under the old name, |
| 485 | so that C<&function(...)> actually invokes the memoized version. |
| 486 | |
| 487 | =head1 OPTIONS |
| 488 | |
| 489 | There are some optional options you can pass to C<memoize> to change |
| 490 | the way it behaves a little. To supply options, invoke C<memoize> |
| 491 | like this: |
| 492 | |
| 493 | memoize(function, NORMALIZER => function, |
| 494 | INSTALL => newname, |
| 495 | SCALAR_CACHE => option, |
| 496 | LIST_CACHE => option |
| 497 | ); |
| 498 | |
| 499 | Each of these options is optional; you can include some, all, or none |
| 500 | of them. |
| 501 | |
| 502 | =head2 INSTALL |
| 503 | |
| 504 | If you supply a function name with C<INSTALL>, memoize will install |
| 505 | the new, memoized version of the function under the name you give. |
| 506 | For example, |
| 507 | |
| 508 | memoize('fib', INSTALL => 'fastfib') |
| 509 | |
| 510 | installs the memoized version of C<fib> as C<fastfib>; without the |
| 511 | C<INSTALL> option it would have replaced the old C<fib> with the |
| 512 | memoized version. |
| 513 | |
| 514 | To prevent C<memoize> from installing the memoized version anywhere, use |
| 515 | C<INSTALL =E<gt> undef>. |
| 516 | |
| 517 | =head2 NORMALIZER |
| 518 | |
| 519 | Suppose your function looks like this: |
| 520 | |
| 521 | # Typical call: f('aha!', A => 11, B => 12); |
| 522 | sub f { |
| 523 | my $a = shift; |
| 524 | my %hash = @_; |
| 525 | $hash{B} ||= 2; # B defaults to 2 |
| 526 | $hash{C} ||= 7; # C defaults to 7 |
| 527 | |
| 528 | # Do something with $a, %hash |
| 529 | } |
| 530 | |
| 531 | Now, the following calls to your function are all completely equivalent: |
| 532 | |
| 533 | f(OUCH); |
| 534 | f(OUCH, B => 2); |
| 535 | f(OUCH, C => 7); |
| 536 | f(OUCH, B => 2, C => 7); |
| 537 | f(OUCH, C => 7, B => 2); |
| 538 | (etc.) |
| 539 | |
| 540 | However, unless you tell C<Memoize> that these calls are equivalent, |
| 541 | it will not know that, and it will compute the values for these |
| 542 | invocations of your function separately, and store them separately. |
| 543 | |
| 544 | To prevent this, supply a C<NORMALIZER> function that turns the |
| 545 | program arguments into a string in a way that equivalent arguments |
| 546 | turn into the same string. A C<NORMALIZER> function for C<f> above |
| 547 | might look like this: |
| 548 | |
| 549 | sub normalize_f { |
| 550 | my $a = shift; |
| 551 | my %hash = @_; |
| 552 | $hash{B} ||= 2; |
| 553 | $hash{C} ||= 7; |
| 554 | |
| 555 | join(',', $a, map ($_ => $hash{$_}) sort keys %hash); |
| 556 | } |
| 557 | |
| 558 | Each of the argument lists above comes out of the C<normalize_f> |
| 559 | function looking exactly the same, like this: |
| 560 | |
| 561 | OUCH,B,2,C,7 |
| 562 | |
| 563 | You would tell C<Memoize> to use this normalizer this way: |
| 564 | |
| 565 | memoize('f', NORMALIZER => 'normalize_f'); |
| 566 | |
| 567 | C<memoize> knows that if the normalized version of the arguments is |
| 568 | the same for two argument lists, then it can safely look up the value |
| 569 | that it computed for one argument list and return it as the result of |
| 570 | calling the function with the other argument list, even if the |
| 571 | argument lists look different. |
| 572 | |
| 573 | The default normalizer just concatenates the arguments with character |
| 574 | 28 in between. (In ASCII, this is called FS or control-\.) This |
| 575 | always works correctly for functions with only one string argument, |
| 576 | and also when the arguments never contain character 28. However, it |
| 577 | can confuse certain argument lists: |
| 578 | |
| 579 | normalizer("a\034", "b") |
| 580 | normalizer("a", "\034b") |
| 581 | normalizer("a\034\034b") |
| 582 | |
| 583 | for example. |
| 584 | |
| 585 | Since hash keys are strings, the default normalizer will not |
| 586 | distinguish between C<undef> and the empty string. It also won't work |
| 587 | when the function's arguments are references. For example, consider a |
| 588 | function C<g> which gets two arguments: A number, and a reference to |
| 589 | an array of numbers: |
| 590 | |
| 591 | g(13, [1,2,3,4,5,6,7]); |
| 592 | |
| 593 | The default normalizer will turn this into something like |
| 594 | C<"13\034ARRAY(0x436c1f)">. That would be all right, except that a |
| 595 | subsequent array of numbers might be stored at a different location |
| 596 | even though it contains the same data. If this happens, C<Memoize> |
| 597 | will think that the arguments are different, even though they are |
| 598 | equivalent. In this case, a normalizer like this is appropriate: |
| 599 | |
| 600 | sub normalize { join ' ', $_[0], @{$_[1]} } |
| 601 | |
| 602 | For the example above, this produces the key "13 1 2 3 4 5 6 7". |
| 603 | |
| 604 | Another use for normalizers is when the function depends on data other |
| 605 | than those in its arguments. Suppose you have a function which |
| 606 | returns a value which depends on the current hour of the day: |
| 607 | |
| 608 | sub on_duty { |
| 609 | my ($problem_type) = @_; |
| 610 | my $hour = (localtime)[2]; |
| 611 | open my $fh, "$DIR/$problem_type" or die...; |
| 612 | my $line; |
| 613 | while ($hour-- > 0) |
| 614 | $line = <$fh>; |
| 615 | } |
| 616 | return $line; |
| 617 | } |
| 618 | |
| 619 | At 10:23, this function generates the 10th line of a data file; at |
| 620 | 3:45 PM it generates the 15th line instead. By default, C<Memoize> |
| 621 | will only see the $problem_type argument. To fix this, include the |
| 622 | current hour in the normalizer: |
| 623 | |
| 624 | sub normalize { join ' ', (localtime)[2], @_ } |
| 625 | |
| 626 | The calling context of the function (scalar or list context) is |
| 627 | propagated to the normalizer. This means that if the memoized |
| 628 | function will treat its arguments differently in list context than it |
| 629 | would in scalar context, you can have the normalizer function select |
| 630 | its behavior based on the results of C<wantarray>. Even if called in |
| 631 | a list context, a normalizer should still return a single string. |
| 632 | |
| 633 | =head2 C<SCALAR_CACHE>, C<LIST_CACHE> |
| 634 | |
| 635 | Normally, C<Memoize> caches your function's return values into an |
| 636 | ordinary Perl hash variable. However, you might like to have the |
| 637 | values cached on the disk, so that they persist from one run of your |
| 638 | program to the next, or you might like to associate some other |
| 639 | interesting semantics with the cached values. |
| 640 | |
| 641 | There's a slight complication under the hood of C<Memoize>: There are |
| 642 | actually I<two> caches, one for scalar values and one for list values. |
| 643 | When your function is called in scalar context, its return value is |
| 644 | cached in one hash, and when your function is called in list context, |
| 645 | its value is cached in the other hash. You can control the caching |
| 646 | behavior of both contexts independently with these options. |
| 647 | |
| 648 | The argument to C<LIST_CACHE> or C<SCALAR_CACHE> must either be one of |
| 649 | the following four strings: |
| 650 | |
| 651 | MEMORY |
| 652 | FAULT |
| 653 | MERGE |
| 654 | HASH |
| 655 | |
| 656 | or else it must be a reference to a list whose first element is one of |
| 657 | these four strings, such as C<[HASH, arguments...]>. |
| 658 | |
| 659 | =over 4 |
| 660 | |
| 661 | =item C<MEMORY> |
| 662 | |
| 663 | C<MEMORY> means that return values from the function will be cached in |
| 664 | an ordinary Perl hash variable. The hash variable will not persist |
| 665 | after the program exits. This is the default. |
| 666 | |
| 667 | =item C<HASH> |
| 668 | |
| 669 | C<HASH> allows you to specify that a particular hash that you supply |
| 670 | will be used as the cache. You can tie this hash beforehand to give |
| 671 | it any behavior you want. |
| 672 | |
| 673 | A tied hash can have any semantics at all. It is typically tied to an |
| 674 | on-disk database, so that cached values are stored in the database and |
| 675 | retrieved from it again when needed, and the disk file typically |
| 676 | persists after your program has exited. See C<perltie> for more |
| 677 | complete details about C<tie>. |
| 678 | |
| 679 | A typical example is: |
| 680 | |
| 681 | use DB_File; |
| 682 | tie my %cache => 'DB_File', $filename, O_RDWR|O_CREAT, 0666; |
| 683 | memoize 'function', SCALAR_CACHE => [HASH => \%cache]; |
| 684 | |
| 685 | This has the effect of storing the cache in a C<DB_File> database |
| 686 | whose name is in C<$filename>. The cache will persist after the |
| 687 | program has exited. Next time the program runs, it will find the |
| 688 | cache already populated from the previous run of the program. Or you |
| 689 | can forcibly populate the cache by constructing a batch program that |
| 690 | runs in the background and populates the cache file. Then when you |
| 691 | come to run your real program the memoized function will be fast |
| 692 | because all its results have been precomputed. |
| 693 | |
| 694 | =item C<TIE> |
| 695 | |
| 696 | This option is no longer supported. It is still documented only to |
| 697 | aid in the debugging of old programs that use it. Old programs should |
| 698 | be converted to use the C<HASH> option instead. |
| 699 | |
| 700 | memoize ... [TIE, PACKAGE, ARGS...] |
| 701 | |
| 702 | is merely a shortcut for |
| 703 | |
| 704 | require PACKAGE; |
| 705 | { my %cache; |
| 706 | tie %cache, PACKAGE, ARGS...; |
| 707 | } |
| 708 | memoize ... [HASH => \%cache]; |
| 709 | |
| 710 | =item C<FAULT> |
| 711 | |
| 712 | C<FAULT> means that you never expect to call the function in scalar |
| 713 | (or list) context, and that if C<Memoize> detects such a call, it |
| 714 | should abort the program. The error message is one of |
| 715 | |
| 716 | `foo' function called in forbidden list context at line ... |
| 717 | `foo' function called in forbidden scalar context at line ... |
| 718 | |
| 719 | =item C<MERGE> |
| 720 | |
| 721 | C<MERGE> normally means the function does not distinguish between list |
| 722 | and sclar context, and that return values in both contexts should be |
| 723 | stored together. C<LIST_CACHE =E<gt> MERGE> means that list context |
| 724 | return values should be stored in the same hash that is used for |
| 725 | scalar context returns, and C<SCALAR_CACHE =E<gt> MERGE> means the |
| 726 | same, mutatis mutandis. It is an error to specify C<MERGE> for both, |
| 727 | but it probably does something useful. |
| 728 | |
| 729 | Consider this function: |
| 730 | |
| 731 | sub pi { 3; } |
| 732 | |
| 733 | Normally, the following code will result in two calls to C<pi>: |
| 734 | |
| 735 | $x = pi(); |
| 736 | ($y) = pi(); |
| 737 | $z = pi(); |
| 738 | |
| 739 | The first call caches the value C<3> in the scalar cache; the second |
| 740 | caches the list C<(3)> in the list cache. The third call doesn't call |
| 741 | the real C<pi> function; it gets the value from the scalar cache. |
| 742 | |
| 743 | Obviously, the second call to C<pi> is a waste of time, and storing |
| 744 | its return value is a waste of space. Specifying C<LIST_CACHE =E<gt> |
| 745 | MERGE> will make C<memoize> use the same cache for scalar and list |
| 746 | context return values, so that the second call uses the scalar cache |
| 747 | that was populated by the first call. C<pi> ends up being called only |
| 748 | once, and both subsequent calls return C<3> from the cache, regardless |
| 749 | of the calling context. |
| 750 | |
| 751 | Another use for C<MERGE> is when you want both kinds of return values |
| 752 | stored in the same disk file; this saves you from having to deal with |
| 753 | two disk files instead of one. You can use a normalizer function to |
| 754 | keep the two sets of return values separate. For example: |
| 755 | |
| 756 | tie my %cache => 'MLDBM', 'DB_File', $filename, ...; |
| 757 | |
| 758 | memoize 'myfunc', |
| 759 | NORMALIZER => 'n', |
| 760 | SCALAR_CACHE => [HASH => \%cache], |
| 761 | LIST_CACHE => MERGE, |
| 762 | ; |
| 763 | |
| 764 | sub n { |
| 765 | my $context = wantarray() ? 'L' : 'S'; |
| 766 | # ... now compute the hash key from the arguments ... |
| 767 | $hashkey = "$context:$hashkey"; |
| 768 | } |
| 769 | |
| 770 | This normalizer function will store scalar context return values in |
| 771 | the disk file under keys that begin with C<S:>, and list context |
| 772 | return values under keys that begin with C<L:>. |
| 773 | |
| 774 | =back |
| 775 | |
| 776 | =head1 OTHER FACILITIES |
| 777 | |
| 778 | =head2 C<unmemoize> |
| 779 | |
| 780 | There's an C<unmemoize> function that you can import if you want to. |
| 781 | Why would you want to? Here's an example: Suppose you have your cache |
| 782 | tied to a DBM file, and you want to make sure that the cache is |
| 783 | written out to disk if someone interrupts the program. If the program |
| 784 | exits normally, this will happen anyway, but if someone types |
| 785 | control-C or something then the program will terminate immediately |
| 786 | without synchronizing the database. So what you can do instead is |
| 787 | |
| 788 | $SIG{INT} = sub { unmemoize 'function' }; |
| 789 | |
| 790 | C<unmemoize> accepts a reference to, or the name of a previously |
| 791 | memoized function, and undoes whatever it did to provide the memoized |
| 792 | version in the first place, including making the name refer to the |
| 793 | unmemoized version if appropriate. It returns a reference to the |
| 794 | unmemoized version of the function. |
| 795 | |
| 796 | If you ask it to unmemoize a function that was never memoized, it |
| 797 | croaks. |
| 798 | |
| 799 | =head2 C<flush_cache> |
| 800 | |
| 801 | C<flush_cache(function)> will flush out the caches, discarding I<all> |
| 802 | the cached data. The argument may be a function name or a reference |
| 803 | to a function. For finer control over when data is discarded or |
| 804 | expired, see the documentation for C<Memoize::Expire>, included in |
| 805 | this package. |
| 806 | |
| 807 | Note that if the cache is a tied hash, C<flush_cache> will attempt to |
| 808 | invoke the C<CLEAR> method on the hash. If there is no C<CLEAR> |
| 809 | method, this will cause a run-time error. |
| 810 | |
| 811 | An alternative approach to cache flushing is to use the C<HASH> option |
| 812 | (see above) to request that C<Memoize> use a particular hash variable |
| 813 | as its cache. Then you can examine or modify the hash at any time in |
| 814 | any way you desire. You may flush the cache by using C<%hash = ()>. |
| 815 | |
| 816 | =head1 CAVEATS |
| 817 | |
| 818 | Memoization is not a cure-all: |
| 819 | |
| 820 | =over 4 |
| 821 | |
| 822 | =item * |
| 823 | |
| 824 | Do not memoize a function whose behavior depends on program |
| 825 | state other than its own arguments, such as global variables, the time |
| 826 | of day, or file input. These functions will not produce correct |
| 827 | results when memoized. For a particularly easy example: |
| 828 | |
| 829 | sub f { |
| 830 | time; |
| 831 | } |
| 832 | |
| 833 | This function takes no arguments, and as far as C<Memoize> is |
| 834 | concerned, it always returns the same result. C<Memoize> is wrong, of |
| 835 | course, and the memoized version of this function will call C<time> once |
| 836 | to get the current time, and it will return that same time |
| 837 | every time you call it after that. |
| 838 | |
| 839 | =item * |
| 840 | |
| 841 | Do not memoize a function with side effects. |
| 842 | |
| 843 | sub f { |
| 844 | my ($a, $b) = @_; |
| 845 | my $s = $a + $b; |
| 846 | print "$a + $b = $s.\n"; |
| 847 | } |
| 848 | |
| 849 | This function accepts two arguments, adds them, and prints their sum. |
| 850 | Its return value is the numuber of characters it printed, but you |
| 851 | probably didn't care about that. But C<Memoize> doesn't understand |
| 852 | that. If you memoize this function, you will get the result you |
| 853 | expect the first time you ask it to print the sum of 2 and 3, but |
| 854 | subsequent calls will return 1 (the return value of |
| 855 | C<print>) without actually printing anything. |
| 856 | |
| 857 | =item * |
| 858 | |
| 859 | Do not memoize a function that returns a data structure that is |
| 860 | modified by its caller. |
| 861 | |
| 862 | Consider these functions: C<getusers> returns a list of users somehow, |
| 863 | and then C<main> throws away the first user on the list and prints the |
| 864 | rest: |
| 865 | |
| 866 | sub main { |
| 867 | my $userlist = getusers(); |
| 868 | shift @$userlist; |
| 869 | foreach $u (@$userlist) { |
| 870 | print "User $u\n"; |
| 871 | } |
| 872 | } |
| 873 | |
| 874 | sub getusers { |
| 875 | my @users; |
| 876 | # Do something to get a list of users; |
| 877 | \@users; # Return reference to list. |
| 878 | } |
| 879 | |
| 880 | If you memoize C<getusers> here, it will work right exactly once. The |
| 881 | reference to the users list will be stored in the memo table. C<main> |
| 882 | will discard the first element from the referenced list. The next |
| 883 | time you invoke C<main>, C<Memoize> will not call C<getusers>; it will |
| 884 | just return the same reference to the same list it got last time. But |
| 885 | this time the list has already had its head removed; C<main> will |
| 886 | erroneously remove another element from it. The list will get shorter |
| 887 | and shorter every time you call C<main>. |
| 888 | |
| 889 | Similarly, this: |
| 890 | |
| 891 | $u1 = getusers(); |
| 892 | $u2 = getusers(); |
| 893 | pop @$u1; |
| 894 | |
| 895 | will modify $u2 as well as $u1, because both variables are references |
| 896 | to the same array. Had C<getusers> not been memoized, $u1 and $u2 |
| 897 | would have referred to different arrays. |
| 898 | |
| 899 | =item * |
| 900 | |
| 901 | Do not memoize a very simple function. |
| 902 | |
| 903 | Recently someone mentioned to me that the Memoize module made his |
| 904 | program run slower instead of faster. It turned out that he was |
| 905 | memoizing the following function: |
| 906 | |
| 907 | sub square { |
| 908 | $_[0] * $_[0]; |
| 909 | } |
| 910 | |
| 911 | I pointed out that C<Memoize> uses a hash, and that looking up a |
| 912 | number in the hash is necessarily going to take a lot longer than a |
| 913 | single multiplication. There really is no way to speed up the |
| 914 | C<square> function. |
| 915 | |
| 916 | Memoization is not magical. |
| 917 | |
| 918 | =back |
| 919 | |
| 920 | =head1 PERSISTENT CACHE SUPPORT |
| 921 | |
| 922 | You can tie the cache tables to any sort of tied hash that you want |
| 923 | to, as long as it supports C<TIEHASH>, C<FETCH>, C<STORE>, and |
| 924 | C<EXISTS>. For example, |
| 925 | |
| 926 | tie my %cache => 'GDBM_File', $filename, O_RDWR|O_CREAT, 0666; |
| 927 | memoize 'function', SCALAR_CACHE => [HASH => \%cache]; |
| 928 | |
| 929 | works just fine. For some storage methods, you need a little glue. |
| 930 | |
| 931 | C<SDBM_File> doesn't supply an C<EXISTS> method, so included in this |
| 932 | package is a glue module called C<Memoize::SDBM_File> which does |
| 933 | provide one. Use this instead of plain C<SDBM_File> to store your |
| 934 | cache table on disk in an C<SDBM_File> database: |
| 935 | |
| 936 | tie my %cache => 'Memoize::SDBM_File', $filename, O_RDWR|O_CREAT, 0666; |
| 937 | memoize 'function', SCALAR_CACHE => [HASH => \%cache]; |
| 938 | |
| 939 | C<NDBM_File> has the same problem and the same solution. (Use |
| 940 | C<Memoize::NDBM_File instead of plain NDBM_File.>) |
| 941 | |
| 942 | C<Storable> isn't a tied hash class at all. You can use it to store a |
| 943 | hash to disk and retrieve it again, but you can't modify the hash while |
| 944 | it's on the disk. So if you want to store your cache table in a |
| 945 | C<Storable> database, use C<Memoize::Storable>, which puts a hashlike |
| 946 | front-end onto C<Storable>. The hash table is actually kept in |
| 947 | memory, and is loaded from your C<Storable> file at the time you |
| 948 | memoize the function, and stored back at the time you unmemoize the |
| 949 | function (or when your program exits): |
| 950 | |
| 951 | tie my %cache => 'Memoize::Storable', $filename; |
| 952 | memoize 'function', SCALAR_CACHE => [HASH => \%cache]; |
| 953 | |
| 954 | tie my %cache => 'Memoize::Storable', $filename, 'nstore'; |
| 955 | memoize 'function', SCALAR_CACHE => [HASH => \%cache]; |
| 956 | |
| 957 | Include the `nstore' option to have the C<Storable> database written |
| 958 | in `network order'. (See L<Storable> for more details about this.) |
| 959 | |
| 960 | The C<flush_cache()> function will raise a run-time error unless the |
| 961 | tied package provides a C<CLEAR> method. |
| 962 | |
| 963 | =head1 EXPIRATION SUPPORT |
| 964 | |
| 965 | See Memoize::Expire, which is a plug-in module that adds expiration |
| 966 | functionality to Memoize. If you don't like the kinds of policies |
| 967 | that Memoize::Expire implements, it is easy to write your own plug-in |
| 968 | module to implement whatever policy you desire. Memoize comes with |
| 969 | several examples. An expiration manager that implements a LRU policy |
| 970 | is available on CPAN as Memoize::ExpireLRU. |
| 971 | |
| 972 | =head1 BUGS |
| 973 | |
| 974 | The test suite is much better, but always needs improvement. |
| 975 | |
| 976 | There is some problem with the way C<goto &f> works under threaded |
| 977 | Perl, perhaps because of the lexical scoping of C<@_>. This is a bug |
| 978 | in Perl, and until it is resolved, memoized functions will see a |
| 979 | slightly different C<caller()> and will perform a little more slowly |
| 980 | on threaded perls than unthreaded perls. |
| 981 | |
| 982 | Some versions of C<DB_File> won't let you store data under a key of |
| 983 | length 0. That means that if you have a function C<f> which you |
| 984 | memoized and the cache is in a C<DB_File> database, then the value of |
| 985 | C<f()> (C<f> called with no arguments) will not be memoized. If this |
| 986 | is a big problem, you can supply a normalizer function that prepends |
| 987 | C<"x"> to every key. |
| 988 | |
| 989 | =head1 MAILING LIST |
| 990 | |
| 991 | To join a very low-traffic mailing list for announcements about |
| 992 | C<Memoize>, send an empty note to C<mjd-perl-memoize-request@plover.com>. |
| 993 | |
| 994 | =head1 AUTHOR |
| 995 | |
| 996 | Mark-Jason Dominus (C<mjd-perl-memoize+@plover.com>), Plover Systems co. |
| 997 | |
| 998 | See the C<Memoize.pm> Page at http://www.plover.com/~mjd/perl/Memoize/ |
| 999 | for news and upgrades. Near this page, at |
| 1000 | http://www.plover.com/~mjd/perl/MiniMemoize/ there is an article about |
| 1001 | memoization and about the internals of Memoize that appeared in The |
| 1002 | Perl Journal, issue #13. (This article is also included in the |
| 1003 | Memoize distribution as `article.html'.) |
| 1004 | |
| 1005 | My upcoming book will discuss memoization (and many other fascinating |
| 1006 | topics) in tremendous detail. It will be published by Morgan Kaufmann |
| 1007 | in 2002, possibly under the title I<Perl Advanced Techniques |
| 1008 | Handbook>. It will also be available on-line for free. For more |
| 1009 | information, visit http://perl.plover.com/book/ . |
| 1010 | |
| 1011 | To join a mailing list for announcements about C<Memoize>, send an |
| 1012 | empty message to C<mjd-perl-memoize-request@plover.com>. This mailing |
| 1013 | list is for announcements only and has extremely low traffic---about |
| 1014 | two messages per year. |
| 1015 | |
| 1016 | =head1 COPYRIGHT AND LICENSE |
| 1017 | |
| 1018 | Copyright 1998, 1999, 2000, 2001 by Mark Jason Dominus |
| 1019 | |
| 1020 | This library is free software; you may redistribute it and/or modify |
| 1021 | it under the same terms as Perl itself. |
| 1022 | |
| 1023 | =head1 THANK YOU |
| 1024 | |
| 1025 | Many thanks to Jonathan Roy for bug reports and suggestions, to |
| 1026 | Michael Schwern for other bug reports and patches, to Mike Cariaso for |
| 1027 | helping me to figure out the Right Thing to Do About Expiration, to |
| 1028 | Joshua Gerth, Joshua Chamas, Jonathan Roy (again), Mark D. Anderson, |
| 1029 | and Andrew Johnson for more suggestions about expiration, to Brent |
| 1030 | Powers for the Memoize::ExpireLRU module, to Ariel Scolnicov for |
| 1031 | delightful messages about the Fibonacci function, to Dion Almaer for |
| 1032 | thought-provoking suggestions about the default normalizer, to Walt |
| 1033 | Mankowski and Kurt Starsinic for much help investigating problems |
| 1034 | under threaded Perl, to Alex Dudkevich for reporting the bug in |
| 1035 | prototyped functions and for checking my patch, to Tony Bass for many |
| 1036 | helpful suggestions, to Jonathan Roy (again) for finding a use for |
| 1037 | C<unmemoize()>, to Philippe Verdret for enlightening discussion of |
| 1038 | C<Hook::PrePostCall>, to Nat Torkington for advice I ignored, to Chris |
| 1039 | Nandor for portability advice, to Randal Schwartz for suggesting the |
| 1040 | 'C<flush_cache> function, and to Jenda Krynicky for being a light in |
| 1041 | the world. |
| 1042 | |
| 1043 | Special thanks to Jarkko Hietaniemi, the 5.8.0 pumpking, for including |
| 1044 | this module in the core and for his patient and helpful guidance |
| 1045 | during the integration process. |
| 1046 | |
| 1047 | =cut |