1 Revision history for Variable-Magic
3 0.62 2017-11-04 16:35 UTC
4 + Fix : [RT #123314] : Compatibility with CV-in-stash optimisation
5 Thanks Father Chrysostomos for reporting and contributing a
8 0.61 2017-01-24 15:30 UTC
9 + Fix : [RT #115792] : Latest blead stack unwind work causes infloop
11 A change in how the core frees temporary values after eval
12 interacted badly with Variable::Magic and caused infinite loops
13 when a free callback was attached to a wizard. This change was
14 temporarily reverted, but is going to be reinstated, and
15 Variable::Magic is now able to play nicely when this happens.
16 Thanks Peter Rabbitson for reporting and David Mitchell for
20 0.60 2016-09-06 22:30 UTC
21 + Chg : A large chunk of boilerplate XS code, which is also used in
22 other XS modules, has been factored out of the main .xs file
23 to a collection of .h files in the xsh subdirectory.
24 + Fix : [RT #128989] : Bleadperl v5.25.3-266-g1d7e644 breaks V::M.
25 Len magic is no longer called for negative array indices
26 fetches, so the relevant tests are now skipped.
27 Thanks Andreas Koenig for reporting.
29 0.59 2015-10-05 19:05 UTC
30 + Chg : The magic callbacks are now always executed in a new stack.
31 This should have no negative impact (performance-wise or
32 behaviour-wise) on your code, but may actually fix some obscure
34 + Fix : [RT #107294] : Crash during install of Variable-Magic-0.58
35 running "t\18-opinfo.t".
36 A recent change in the core implementation of transliterations
37 caused an undefined behaviour when it interacted with this
38 module, which resulted in visible stack corruption on Win32
39 perls compiled with MSVC when the op_info feature was enabled.
40 This was fixed by using a separate stack for the magic
42 Thanks Peter Stokes for reporting and providing greatly
44 + Fix : The object op_info feature will now handle correctly Unicode
45 transliteration, multideref and custom ops.
47 0.58 2015-07-21 16:00 UTC
48 + Add : If a non-len magic callback returns a reference, it will now
49 only be freed at the end of the statement that caused the
50 magic to trigger. This allows the user to attach free magic
51 (or a plain destructor) to a token returned from the callbacks
52 in order to defer an action after the magic is processed by
54 + Fix : Test failures of threads tests on systems with harsh resource
55 constraints causing the threads to exit() during run.
57 0.57 2015-04-17 15:20 UTC
58 + Chg : The new environment variable to enable thread tests on older
59 perls is PERL_FORCE_TEST_THREADS. Note that this variable
60 should only be turned on by authors.
61 + Fix : Segfaults when the module is loaded by several threads (or
62 Windows emulated processes) ran in parallel.
63 + Fix : Segfaults when the module is loaded in a thread, which spawns
64 itself a new thread, and that child thread outlives its parent.
65 + Fix : Small memory leaks of structures required for thread safety.
66 + Fix : Update the Windows ActivePerl + gcc 3.4 workaround for
67 ExtUtils::MakeMaker 7.04. Thanks Christian Walde for reporting
68 and feedback on this issue.
69 + Tst : The global destruction test will now be exercised on any perl
70 that has DEBUGGING set.
71 + Tst : Optional capturing tests in t/17-ctl.t that were only run when
72 Capture::Tiny was present were converted to an IPC::Open3
73 based helper and will now be run everywhere.
75 0.56 2015-03-11 15:15 UTC
76 + Fix : [RT #101410] : Install fails in blead
77 Even though the change that caused this error was reverted from
78 blead, lvalue uses of ERRSV have been removed from this module
79 so that there will be no breakage when this change is possibly
80 reintroduced in the following months.
81 Thanks Dave Rolsky for reporting.
82 + Tst : $ENV{$Config{ldlibpthname}} is now preserved on all platforms,
83 which will address failures of t/17-ctl.t with unusual
84 compilers (like icc) that link all their compiled objects to
86 + Tst : The global destruction test is now only run on perl 5.13.4 and
87 higher, and only if either Perl::Destruct::Level is installed
88 or PERL_DESTRUCT_LEVEL is set and the perl is a debugging perl.
89 This will solve rare crashes of t/15-self.t on perl 5.13.3 and
92 0.55 2014-10-20 23:45 UTC
93 + Fix : [RT #99319] : t/18-opinfo.t: test failure due to change in
95 The new METHOP op class added in perl 5.21.5 is now supported.
96 Thanks James E Keenan for reporting.
98 0.54 2014-09-22 17:30 UTC
99 + Add : The new constant VMG_COMPAT_CODE_COPY_CLONE evaluates to true
100 if your perl calls 'copy' magic when a magical code prototype
101 is cloned, which is currently the case for perl 5.17.0 and
103 + Fix : [RT #90205] : copy magic on subs puts raw CV in $_[3]
104 $_[3] will now contain a reference to the cloned code when
105 'copy' magic is called for a coderef.
106 Thanks Lukas Mai for reporting.
107 + Fix : t/35-stash.t has been taught about perl 5.21.4.
108 + Fix : Tests using run_perl() in t/17-ctl.t will no longer fail on
111 0.53 2013-09-01 17:50 UTC
112 This is a maintenance release. The code contains no functional change.
113 Satisfied users of version 0.52 can skip this update.
114 + Fix : [RT #86338] : typo fix.
115 Thanks dsteinbrunner@pobox.com for the patch.
116 + Tst : Author tests are no longer bundled with this distribution.
117 They are only made available to authors in the git repository.
119 0.52 2012-11-05 02:30 UTC
120 + Add : The new constant VMG_COMPAT_SCALAR_NOLEN evaluates to true
121 when your perl does not call 'len' magic for scalars, which
122 is the case for perl 5.17.4 and above.
123 This fixes RT #80388.
124 + Tst : t/35-stash.t has been taught about perl 5.17.4.
125 + Tst : Author tests overhaul.
127 0.51 2012-08-18 15:00 UTC
128 + Fix : It is now safe to call dispell() from inside 'free', 'copy' and
129 'uvar' callbacks to dispell the magic currently in use.
130 Thanks Clinton Gormley for reporting.
131 + Fix : Exceptions thrown from inside a 'free' callback are now always
132 consistently propagated outside of the callback. They used to
133 be lost when the 'free' callback was invoked at the end of an
134 eval block or string.
135 + Fix : The 'reset RMG flag' workaroundn used to allow wizards with
136 both 'uvar' and 'clear' magics to be cast onto a hash, has been
139 0.50 2012-06-24 23:00 UTC
140 + Fix : Less memory is leaked when a wizard is freed during global
141 destruction, or when an exception is thrown from a 'free'
143 + Fix : [RT #77991] : t/17-ctl.t fails on perl 5.14 and 5.16.
144 This was actually an issue with ActivePerl, and this test
145 has learned to cope with it.
146 Thanks Gisle Aas for reporting.
147 + Tst : t/35-stash.t has been taught about perl 5.17.1.
148 + Doc : Many clarifications.
150 0.49 2012-06-05 21:40 UTC
151 This is a maintenance release. The code contains no functional change.
152 Satisfied users of version 0.48 can skip this update.
153 + Fix : [RT #77644] : t/17_ctl.t fails in test 44 since 5.17.0.
154 This test has been taught about perl 5.17.0.
155 Thanks Reini Urban for reporting.
156 + Tst : t/99-kwalitee.t will be skipped when only problematic versions
157 of its dependencies are available.
159 0.48 2012-02-17 23:40 UTC
160 + Add : You can now pass a reference to undef as the magic callback in
161 order to install a no-op callback.
162 Thanks Florian Ragwitz for the suggestion.
164 0.47 2011-10-27 16:55 UTC
165 + Add : The new constant VMG_COMPAT_HASH_DELETE_NOUVAR_VOID evaluates
166 to true when "delete $hash{key}" does not call 'delete' uvar
167 magic in void context.
168 + Chg : The MAGIC tokens created by this module no longer use the
169 mg_private member for storing a "magical" signature.
170 + Fix : Triggering magic in a thread on a variable cloned from the main
171 interpreter, and when the wizard already went out of scope,
172 will not segfault anymore.
173 + Opt : The pointer table is no longer needed to ensure thread safety.
174 The size of the object code is about 8% smaller for threaded
176 + Tst : Threads tests will not fail anymore if resources constraints
177 prevent the system from creating all the required threads.
179 0.46 2011-01-23 16:45 UTC
180 + Fix : [RT #64866] : Assertion failure with perl 5.13.9.
181 The real problem was that dispell() wasn't properly resetting
182 the magical sv flags, which became visible in the test suite
183 only with perl 5.13.9.
184 Thanks Joshua ben Jore for reporting.
186 0.45 2010-11-21 23:15 UTC
187 This is a maintenance release. The code contains no functional change.
188 Users of 0.44 can skip this update.
189 + Doc : C++ compilers are officially NOT supported.
190 + Doc : The minimum perl 5.10.0 requirement for uvar magic has been
192 Thanks Peter Rabbitson for pointing this out and contributing a
194 + Tst : Tune for perl 5.13.7.
195 + Tst : Capture::Tiny will be used in t/17-ctl.t if and only if it can
196 capture a simple run.
198 0.44 2010-09-24 19:10 UTC
199 + Fix : Broken linkage on Windows with gcc 3.4, which appears in
200 particular when using ActivePerl's default compiler suite.
201 For those setups, the Variable::Magic shared library will now
202 be linked against the perl dll directly (instead of the import
203 library). This (should) fix RT #51483.
204 Thanks Christian Walde for helping to reproduce this failure
206 + Rem : Support for development perls from the 5.11 branch but older
207 than the 5.11.0 release was removed. This could cause more
208 recent setups to fail.
209 + Tst : Threads tests are now only run on perl 5.13.4 and higher.
210 They could segfault randomly because of what seems to be an
211 internal bug of Perl, which has been addressed in 5.13.4.
212 There is also an environment variable that allows you to
213 forcefully run those tests, but it should be set only for
214 author testing and not for end users.
216 0.43 2010-06-25 23:35 UTC
217 + Add : The new constant VMG_COMPAT_GLOB_GET tells you whether get magic
218 is called for globs. It's true starting perl 5.13.2.
219 + Chg : All callbacks are now called within an eval-like context.
220 Only free callbacks used to be called that way.
221 + Fix : Some exceptions thrown from a free callback could be lost.
222 + Fix : Croak messages could sometimes be repeated several times.
223 + Fix : t/41-clone.t segfaulting with perl 5.13.2.
225 0.42 2010-05-19 00:15 UTC
226 This is a maintenance release. The code contains no functional change.
227 Users of 0.41 can skip this update.
228 + Fix : Test failures with perl 5.13.
229 + Tst : Improve coverage.
231 0.41 2010-03-15 17:35 UTC
232 + Doc : Tweaks and fixups.
234 + Fix : Compatibility with the soon-to-be-released perl 5.12.0.
235 + Fix : Correctly propagate the errors thrown when variable destruction
236 happens at compile-time and not from inside eval STRING.
237 Thanks Florian Ragwitz and Ash Berlin for reporting.
239 0.40 2010-01-06 23:20 UTC
240 + Fix : Possible memory miswrites when passing data arguments to cast().
241 + Fix : Minor C portability tweaks.
243 0.39 2009-12-01 00:05 UTC
244 + Add : You can use a function name as a callback by passing a string
245 reference to wizard() instead of a code reference.
246 + Fix : Compatiblity with perl 5.11.1 and 5.11.2.
247 + Fix : Scalars stored into the data slot no longer leak.
248 + Fix : Thread destruction should not segfault anymore.
249 + Opt : As a result of removing the deprecated features, less memory is
250 used for both threaded (a pointer table is used instead of a
251 hash) and non-threaded (no global structure is needed anymore)
253 + Rem : As advertised in the precendent version, all the signature
254 related features were removed. This includes the 'sig' option
255 to wizard(), the getsig() and gensig() functions, and the SIG_*
258 0.38 2009-10-04 16:10 UTC
259 + Dep : All the signature-related features are DEPRECATED and will be
260 removed in december 2009. This includes the 'sig' option to
261 wizard(), the getsig() and gensig() functions, and the SIG_*
263 Afaik those features were never used in real life, and they make
264 the XS code slow and brittle. If you want your wizards to be
265 globally available, you'll have to handle that yourself by
266 storing them in a hash.
267 + Doc : More examples in a new COOKBOOK section.
268 + Fix : Compatiblity with the 5.11.0 release. The new compatibility
269 constant VMG_COMPAT_ARRAY_PUSH_NOLEN_VOID was added to cover
271 + Fix : Work around Kwalitee test misfailures.
273 0.37 2009-08-25 15:25 UTC
274 + Add : The new constant VMG_FORKSAFE can be tested to know whether the
275 module will behave nicely when fork()ing. It's currently always
276 true except on Windows where you need perl 5.10.1 for it to be
278 + Doc : Nits and clarifications.
280 0.36 2009-07-05 16:30 UTC
281 + Chg : getdata() now returns an empty list when no magic is present.
282 It used to return undef but did not croak as stated in the doc.
283 Thanks Matt S. Trout for pointing this out.
284 + Fix : Building and tests with ActiveState Perl 5.8 build >= 822.
286 0.35 2009-05-15 20:50 UTC
287 This is a maintenance release. Users of 0.34 can skip this update.
288 + Chg : uvar magic used to be enabled for 5.9.5. It now requires 5.10 or
290 + Fix : Silence compilation warnings for cxinc().
291 + Rem : The workaround introduced in 0.31 for the "Unknown errors" issue
292 was removed, as regression tests show that 0.34 provides a
294 + Tst : Skip t/40-threads.t on 5.8.x with old versions of threads and
297 0.34 2009-04-19 16:55 UTC
298 + Fix : Destruction of wizards in require.
299 + Fix : "panic: restartop" in 5.10 when dieing at compile time triggers
302 0.33 2009-03-26 00:00 UTC
303 + Chg : cast() and getsig() now croak too when an invalid signature is
305 + Fix : It was possible to generate with gensig() the same signature
306 twice before actually defining the wizards.
307 + Tst : More stash tests.
308 + Tst : Test when the magic takes place in some cases.
309 + Tst : Improved coverage for the op object feature.
311 0.32 2009-03-01 13:45 UTC
312 + Chg : dispell() and getdata() now croak when an invalid signature is
314 + Doc : More examples and a brand new synopsis.
315 + Fix : The signature is no longer stored in the mg_private member of
316 the MAGIC struct. This fixes possible clashes with magics from
318 + Fix : op info objects weren't blessed into the right class.
319 + Fix : Races when initializing the op names cache under threads.
321 0.31 2009-02-19 23:50 UTC
322 + Doc : A somewhat better introduction. The FUNCTION section was moved
323 before CONSTANTS and HISTORY since it's more important. And
324 'assignation' is really 'assignment', says Ricardo. :)
325 + Fix : The op_info features are now thread safe.
326 + Fix : Segfaults that occured when trying to get the op name during
328 + Fix : Segfaults and wrong "Unknown error" exceptions that happened
329 when dieing in require caused a free callback to fire.
330 Thanks Florian Ragwitz for reporting.
332 0.30 2009-02-12 17:05 UTC
333 + Add : You can now get in all the callbacks either the name or a B::OP
334 object representing the current op if you pass the new 'op_info'
335 option to wizard() with respective values VMG_OP_INFO_NAME and
337 + Doc : Document magic on array/hashes values.
338 + Fix : Completely skip free callbacks during global destruction.
339 + Tst : Tests now pass again on 5.8.3 and lower (they were broken since
341 + Tst : Test magic on symbol table and on array/hash values.
343 0.29 2009-02-08 11:10 UTC
344 + Add : Set $_[2] to the default length for len magic on scalars.
345 + Chg : Perl 5.8 is required.
346 + Chg : The CLONE method will no longer be defined for non-threaded
349 + Fix : The logic for turning off the GMAGICAL flag on hashes when no
350 uvar magic is involved could do nasty things on tied hashes.
351 + Upd : More resources in META.yml.
353 0.28 2009-01-24 17:05 UTC
354 + Add : The 'copy_key' option to wizard(), that enables the safe edition
355 of $_[2] in uvar callbacks to redirect actions to another key.
356 + Fix : Exception objects thrown from callbacks at compile time were not
358 + Fix : free callbacks triggered during global destruction could lead to
359 segfaults because the wizard was already freed.
360 + Fix : Refcount of scalars with free callback.
361 + Fix : Segfaults when using get or uvar magic simultaneously with clear
363 + Doc : The documentation about what the callbacks take and should
364 return is now somewhat clearer.
366 0.27 2009-01-18 23:40 UTC
367 + Fix : Offset of 1 with len magic on scalars.
368 + Fix : Segfaults that seem to happen when one croak in a callback at
369 compile time (thanks Ash Berlin for the testcase).
370 + Upd : Resources in META.yml (Florian Ragwitz)
371 + Tst : Large cleanup resulting in a wider set of tested behaviours.
373 0.26 2008-11-29 22:05 UTC
374 + Fix : Leaks of cloned coderefs that access lexicals.
375 + Fix : Building with maint-5.8 and blead (thanks Andreas Koenig).
377 0.25 2008-11-12 19:40 UTC
378 + Fix : Compatibility with 5.8.9.
379 + Fix : Old Pod::Coverage don't ignore CLONE.
380 + Upd : META.yml spec updated to 1.4.
382 0.24 2008-10-12 14:55 UTC
383 + Fix : Really fix it.
385 0.23 2008-10-11 17:25 UTC
386 + Fix : Building on perls with multiplicity but not ithreads.
388 0.22 2008-09-29 19:40 UTC
389 + Doc : Explain how magic differs from overloading.
390 + Fix : Disable thread safety for perl 5.8 on Windows. If you want it
391 enabled, reliable feedback is welcome.
393 0.21_02 2008-09-28 23:10 UTC
394 + Fix : Hide previously deleted magic tokens in vmg_svt_free (Yuval
397 0.21_01 2008-09-22 13:25 UTC
398 + Chg : Shrink the context data.
399 + Fix : Enable thread safety features for perls that just lack
401 + Tst : Output threads and threads::shared versions in threads tests.
403 0.20 2008-09-20 23:35 UTC
404 + Fix : The module is now thread safe. You can test if it had been
405 built with thread safety features enabled through the
406 VMG_THREADSAFE constant.
407 + Fix : Correct backtrace when wizard() croaks.
408 + Fix : 'cast $a, undef' should throw an 'invalid wizard' error instead
409 of an 'invalid signature'.
411 0.19 2008-07-05 15:05 UTC
412 + Add : The patchlevel with which the module has been built is now
413 exported with the VMG_PERL_PATCHLEVEL constant.
414 + Chg : Version checking macros overhaul.
415 + Fix : 'uvar' magic should only be enabled in maint-5.10 and higher
416 since it's never going to be backported to 5.8.
417 + Fix : Likewise, the copy callback API change should only be enabled in
419 + Fix : Instead of those ackward fake refcount increments in the wizard
420 destructor, flag the wizard with SVf_BREAK and force temps
421 cleanup. This should fix the wizard destruction for all perls,
422 and solves the leak introduced by the previous solution.
424 0.18 2008-07-04 12:20 UTC
425 + Fix : Enable uvar magic when patchlevel >= 28419.
426 + Fix : Set VMG_COMPAT_ARRAY_PUSH_NOLEN for Windows ActiveState Perl
429 0.17 2008-05-11 09:05 UTC
430 + Fix : 5.8.4 and lower don't have mPUSHi. Gah. Yeah, I know, I should
433 0.16 2008-05-10 22:05 UTC
434 + Add : The samples/copy.pl script.
435 + Chg : The sv_magical() + vmg_mg_magical() combo was simplified into
437 + Tst : t/33-code.t was testing scalars, not code.
438 + Tst : is() and like() are better than ok().
439 + Tst : 100% coverage reached.
441 0.15 2008-04-11 18:25 UTC
442 + Chg : Factor vmg_cb_call{1,2,3}() into one function.
443 + Fix : len magic is no longer called when taking the length() of a
444 magical scalar since p32969. The VMG_COMPAT_SCALAR_LENGTH_NOLEN
445 constant was added to cover this.
446 + Tst : More tests for t/22-len.t.
448 0.14 2008-03-24 12:35 UTC
449 + Fix : t/16-huf.t failures on Solaris and FreeBSD caused by not
450 updating mg->mg_ptr after Renew-ing it on dispell.
451 + Fix : Undefining MGf_DUP caused the wizard's magic to be wrongly
452 initialized. MGf_DUP now appears as true from the user point of
453 view, but the dup callback isn't actually set.
454 + Fix : Warnings with blead due to copy callbacks now taking an I32 (API
456 + Fix : vmg_svt_val() and vmg_uvar_del() aren't required if no uvar
458 + Tst : Tests now output the version of optional modules loaded.
460 0.13 2008-03-19 14:35 UTC
461 + Doc : Link to coverage report.
462 + Fix : Correct dependencies listing in META.yml.
463 + Tst : Improved test coverage.
464 + Tst : Print the patchlevel as a comment.
465 + Tst : Use is() where it's relevant.
466 + Tst : t/16-huf.t now really tests interaction with H::U::FH.
468 0.12 2008-02-07 18:15 UTC
469 + Fix : POD error. Thanks to Chris Williams (BinGOs) for the quick
472 0.11 2008-02-07 17:55 UTC
473 + Add : Copy callbacks now receive the current key/index in $_[2].
474 The current value/element is now in $_[3].
475 + Chg : The trigger of copy magic on list assignment with perls greater
476 than 5.10.0 was caused by uvar magic. Hence,
477 VMG_COMPAT_HASH_LISTASSIGN_COPY was removed. Use VMG_UVAR
479 + Fix : Build failures on patched perls.
480 + Tst : Added missing exported symbols to 01-import.t.
482 0.10 2008-02-04 11:30 UTC
483 + Add : New script : samples/vm_vs_tie.pl, that benchmarks our uvar
484 magic versus tied hashes.
485 + Add : The VMG_COMPAT_* constants can be used from userspace to check
486 perl magic abilities.
487 + Fix : Callbacks that returned undef made us croak, breaking the
488 variable behaviour (regression test in 14-callbacks.t).
489 + Fix : uvar callbacks weren't tested for non-NULL-ity before being
490 called (regression test in 28-uvar.t).
491 + Tst : Fix typo in 25-copy.t that prevented Tie::Hash tests to be ran.
493 0.09 2008-02-02 11:30 UTC
494 + Doc : Explicitely say that uvar callbacks are safely ignored for
496 + Doc : Document caveats and fix the usual set of typos.
497 + Fix : vmg_dispell() didn't check if the ext magic were ours when
498 counting wizards that have uvar callbacks, resulting in a
499 possible memory misread.
500 + Fix : getdata() now returns directly the data object, and no longer a
501 copy. This caused a leak.
502 + Tst : Prefix author tests by 9*-.
503 + Tst : New optional author test : 95-portability-files.t, that uses
504 Test::Portability::Files when it's present.
505 + Tst : New test : 14-self.t, that tests application of magic on the
507 + Tst : Move Hash::Util::FieldHash tests out of 11-multiple.t to
510 0.08 2008-02-01 16:55 UTC
511 + Add : copy magic for tied arrays/hashes.
513 + Add : uvar magics : fetch, store, exists, delete for hashes.
516 Internal development versions.
518 0.06 2007-11-20 10:10 UTC
519 + Chg : 5.7.3 is now officially required.
520 + Fix : "data" test failures on 5.8.{0,2}.
521 + Fix : Drand01() vs rand().
523 0.05 2007-11-19 09:10 UTC
524 + Fix : 5.10.0_RC1 compatibility fix.
526 0.04 2007-08-28 12:25 UTC
527 + Chg : s/require (XSLoader)/use $1/.
528 + Fix : Tests are now strict.
529 + Fix : Complete dependencies.
531 0.03 2007-08-01 17:20 UTC
532 + Add : Passing the signature of an already defined magic to wizard()
533 now returns the corresponding magic object.
534 + Add : You can pass the numeric signature as the wizard argument of
535 cast(), dispell() and getdata().
536 + Add : Any argument specified after the wizard (or the signature) in a
537 call to cast() is now passed to the private data constructor in
539 + Chg : $_[0] is now always a reference to the magic variable in all
540 callbacks. The reason for changing from the previous behaviour
541 is that one may want to apply the same magic to a plain scalar
542 and to a scalar reference, and needs a way to distinguish
543 between them in the callback (say, ref()).
544 + Fix : Wizard object destruction used not to free the signature.
546 0.02 2007-07-27 13:50 UTC
547 + Fix : In response to test report 548152 :
548 Newx() and SvMAGIC_set() not present on older perls.
549 + Fix : In response to test report 548275 :
550 Since perl 5.9.5, 'clear' magic is invoked when an array is
551 undefined (bug #43357). Moreover, 'len' magic is no longer
552 called by pushing an element since perl 5.9.3.
553 + Fix : Missing glob test in MANIFEST.
555 0.01 2007-07-25 16:15 UTC
556 First version, released on an unsuspecting world.
projects / perl / modules / Variable-Magic.git / blob