1 Revision history for Variable-Magic
3 0.55 2014-10-20 23:45 UTC
4 + Fix : [RT #99319] : t/18-opinfo.t: test failure due to change in
6 The new METHOP op class added in perl 5.21.5 is now supported.
7 Thanks James E Keenan for reporting.
9 0.54 2014-09-22 17:30 UTC
10 + Add : The new constant VMG_COMPAT_CODE_COPY_CLONE evaluates to true
11 if your perl calls 'copy' magic when a magical code prototype
12 is cloned, which is currently the case for perl 5.17.0 and
14 + Fix : [RT #90205] : copy magic on subs puts raw CV in $_[3]
15 $_[3] will now contain a reference to the cloned code when
16 'copy' magic is called for a coderef.
17 Thanks Lukas Mai for reporting.
18 + Fix : t/35-stash.t has been taught about perl 5.21.4.
19 + Fix : Tests using run_perl() in t/17-ctl.t will no longer fail on
22 0.53 2013-09-01 17:50 UTC
23 This is a maintenance release. The code contains no functional change.
24 Satisfied users of version 0.52 can skip this update.
25 + Fix : [RT #86338] : typo fix.
26 Thanks dsteinbrunner@pobox.com for the patch.
27 + Tst : Author tests are no longer bundled with this distribution.
28 They are only made available to authors in the git repository.
30 0.52 2012-11-05 02:30 UTC
31 + Add : The new constant VMG_COMPAT_SCALAR_NOLEN evaluates to true
32 when your perl does not call 'len' magic for scalars, which
33 is the case for perl 5.17.4 and above.
35 + Tst : t/35-stash.t has been taught about perl 5.17.4.
36 + Tst : Author tests overhaul.
38 0.51 2012-08-18 15:00 UTC
39 + Fix : It is now safe to call dispell() from inside 'free', 'copy' and
40 'uvar' callbacks to dispell the magic currently in use.
41 Thanks Clinton Gormley for reporting.
42 + Fix : Exceptions thrown from inside a 'free' callback are now always
43 consistently propagated outside of the callback. They used to
44 be lost when the 'free' callback was invoked at the end of an
46 + Fix : The 'reset RMG flag' workaroundn used to allow wizards with
47 both 'uvar' and 'clear' magics to be cast onto a hash, has been
50 0.50 2012-06-24 23:00 UTC
51 + Fix : Less memory is leaked when a wizard is freed during global
52 destruction, or when an exception is thrown from a 'free'
54 + Fix : [RT #77991] : t/17-ctl.t fails on perl 5.14 and 5.16.
55 This was actually an issue with ActivePerl, and this test
56 has learned to cope with it.
57 Thanks Gisle Aas for reporting.
58 + Tst : t/35-stash.t has been taught about perl 5.17.1.
59 + Doc : Many clarifications.
61 0.49 2012-06-05 21:40 UTC
62 This is a maintenance release. The code contains no functional change.
63 Satisfied users of version 0.48 can skip this update.
64 + Fix : [RT #77644] : t/17_ctl.t fails in test 44 since 5.17.0.
65 This test has been taught about perl 5.17.0.
66 Thanks Reini Urban for reporting.
67 + Tst : t/99-kwalitee.t will be skipped when only problematic versions
68 of its dependencies are available.
70 0.48 2012-02-17 23:40 UTC
71 + Add : You can now pass a reference to undef as the magic callback in
72 order to install a no-op callback.
73 Thanks Florian Ragwitz for the suggestion.
75 0.47 2011-10-27 16:55 UTC
76 + Add : The new constant VMG_COMPAT_HASH_DELETE_NOUVAR_VOID evaluates
77 to true when "delete $hash{key}" does not call 'delete' uvar
78 magic in void context.
79 + Chg : The MAGIC tokens created by this module no longer use the
80 mg_private member for storing a "magical" signature.
81 + Fix : Triggering magic in a thread on a variable cloned from the main
82 interpreter, and when the wizard already went out of scope,
83 will not segfault anymore.
84 + Opt : The pointer table is no longer needed to ensure thread safety.
85 The size of the object code is about 8% smaller for threaded
87 + Tst : Threads tests will not fail anymore if resources constraints
88 prevent the system from creating all the required threads.
90 0.46 2011-01-23 16:45 UTC
91 + Fix : [RT #64866] : Assertion failure with perl 5.13.9.
92 The real problem was that dispell() wasn't properly resetting
93 the magical sv flags, which became visible in the test suite
94 only with perl 5.13.9.
95 Thanks Joshua ben Jore for reporting.
97 0.45 2010-11-21 23:15 UTC
98 This is a maintenance release. The code contains no functional change.
99 Users of 0.44 can skip this update.
100 + Doc : C++ compilers are officially NOT supported.
101 + Doc : The minimum perl 5.10.0 requirement for uvar magic has been
103 Thanks Peter Rabbitson for pointing this out and contributing a
105 + Tst : Tune for perl 5.13.7.
106 + Tst : Capture::Tiny will be used in t/17-ctl.t if and only if it can
107 capture a simple run.
109 0.44 2010-09-24 19:10 UTC
110 + Fix : Broken linkage on Windows with gcc 3.4, which appears in
111 particular when using ActivePerl's default compiler suite.
112 For those setups, the Variable::Magic shared library will now
113 be linked against the perl dll directly (instead of the import
114 library). This (should) fix RT #51483.
115 Thanks Christian Walde for helping to reproduce this failure
117 + Rem : Support for development perls from the 5.11 branch but older
118 than the 5.11.0 release was removed. This could cause more
119 recent setups to fail.
120 + Tst : Threads tests are now only run on perl 5.13.4 and higher.
121 They could segfault randomly because of what seems to be an
122 internal bug of Perl, which has been addressed in 5.13.4.
123 There is also an environment variable that allows you to
124 forcefully run those tests, but it should be set only for
125 author testing and not for end users.
127 0.43 2010-06-25 23:35 UTC
128 + Add : The new constant VMG_COMPAT_GLOB_GET tells you whether get magic
129 is called for globs. It's true starting perl 5.13.2.
130 + Chg : All callbacks are now called within an eval-like context.
131 Only free callbacks used to be called that way.
132 + Fix : Some exceptions thrown from a free callback could be lost.
133 + Fix : Croak messages could sometimes be repeated several times.
134 + Fix : t/41-clone.t segfaulting with perl 5.13.2.
136 0.42 2010-05-19 00:15 UTC
137 This is a maintenance release. The code contains no functional change.
138 Users of 0.41 can skip this update.
139 + Fix : Test failures with perl 5.13.
140 + Tst : Improve coverage.
142 0.41 2010-03-15 17:35 UTC
143 + Doc : Tweaks and fixups.
145 + Fix : Compatibility with the soon-to-be-released perl 5.12.0.
146 + Fix : Correctly propagate the errors thrown when variable destruction
147 happens at compile-time and not from inside eval STRING.
148 Thanks Florian Ragwitz and Ash Berlin for reporting.
150 0.40 2010-01-06 23:20 UTC
151 + Fix : Possible memory miswrites when passing data arguments to cast().
152 + Fix : Minor C portability tweaks.
154 0.39 2009-12-01 00:05 UTC
155 + Add : You can use a function name as a callback by passing a string
156 reference to wizard() instead of a code reference.
157 + Fix : Compatiblity with perl 5.11.1 and 5.11.2.
158 + Fix : Scalars stored into the data slot no longer leak.
159 + Fix : Thread destruction should not segfault anymore.
160 + Opt : As a result of removing the deprecated features, less memory is
161 used for both threaded (a pointer table is used instead of a
162 hash) and non-threaded (no global structure is needed anymore)
164 + Rem : As advertised in the precendent version, all the signature
165 related features were removed. This includes the 'sig' option
166 to wizard(), the getsig() and gensig() functions, and the SIG_*
169 0.38 2009-10-04 16:10 UTC
170 + Dep : All the signature-related features are DEPRECATED and will be
171 removed in december 2009. This includes the 'sig' option to
172 wizard(), the getsig() and gensig() functions, and the SIG_*
174 Afaik those features were never used in real life, and they make
175 the XS code slow and brittle. If you want your wizards to be
176 globally available, you'll have to handle that yourself by
177 storing them in a hash.
178 + Doc : More examples in a new COOKBOOK section.
179 + Fix : Compatiblity with the 5.11.0 release. The new compatibility
180 constant VMG_COMPAT_ARRAY_PUSH_NOLEN_VOID was added to cover
182 + Fix : Work around Kwalitee test misfailures.
184 0.37 2009-08-25 15:25 UTC
185 + Add : The new constant VMG_FORKSAFE can be tested to know whether the
186 module will behave nicely when fork()ing. It's currently always
187 true except on Windows where you need perl 5.10.1 for it to be
189 + Doc : Nits and clarifications.
191 0.36 2009-07-05 16:30 UTC
192 + Chg : getdata() now returns an empty list when no magic is present.
193 It used to return undef but did not croak as stated in the doc.
194 Thanks Matt S. Trout for pointing this out.
195 + Fix : Building and tests with ActiveState Perl 5.8 build >= 822.
197 0.35 2009-05-15 20:50 UTC
198 This is a maintenance release. Users of 0.34 can skip this update.
199 + Chg : uvar magic used to be enabled for 5.9.5. It now requires 5.10 or
201 + Fix : Silence compilation warnings for cxinc().
202 + Rem : The workaround introduced in 0.31 for the "Unknown errors" issue
203 was removed, as regression tests show that 0.34 provides a
205 + Tst : Skip t/40-threads.t on 5.8.x with old versions of threads and
208 0.34 2009-04-19 16:55 UTC
209 + Fix : Destruction of wizards in require.
210 + Fix : "panic: restartop" in 5.10 when dieing at compile time triggers
213 0.33 2009-03-26 00:00 UTC
214 + Chg : cast() and getsig() now croak too when an invalid signature is
216 + Fix : It was possible to generate with gensig() the same signature
217 twice before actually defining the wizards.
218 + Tst : More stash tests.
219 + Tst : Test when the magic takes place in some cases.
220 + Tst : Improved coverage for the op object feature.
222 0.32 2009-03-01 13:45 UTC
223 + Chg : dispell() and getdata() now croak when an invalid signature is
225 + Doc : More examples and a brand new synopsis.
226 + Fix : The signature is no longer stored in the mg_private member of
227 the MAGIC struct. This fixes possible clashes with magics from
229 + Fix : op info objects weren't blessed into the right class.
230 + Fix : Races when initializing the op names cache under threads.
232 0.31 2009-02-19 23:50 UTC
233 + Doc : A somewhat better introduction. The FUNCTION section was moved
234 before CONSTANTS and HISTORY since it's more important. And
235 'assignation' is really 'assignment', says Ricardo. :)
236 + Fix : The op_info features are now thread safe.
237 + Fix : Segfaults that occured when trying to get the op name during
239 + Fix : Segfaults and wrong "Unknown error" exceptions that happened
240 when dieing in require caused a free callback to fire.
241 Thanks Florian Ragwitz for reporting.
243 0.30 2009-02-12 17:05 UTC
244 + Add : You can now get in all the callbacks either the name or a B::OP
245 object representing the current op if you pass the new 'op_info'
246 option to wizard() with respective values VMG_OP_INFO_NAME and
248 + Doc : Document magic on array/hashes values.
249 + Fix : Completely skip free callbacks during global destruction.
250 + Tst : Tests now pass again on 5.8.3 and lower (they were broken since
252 + Tst : Test magic on symbol table and on array/hash values.
254 0.29 2009-02-08 11:10 UTC
255 + Add : Set $_[2] to the default length for len magic on scalars.
256 + Chg : Perl 5.8 is required.
257 + Chg : The CLONE method will no longer be defined for non-threaded
260 + Fix : The logic for turning off the GMAGICAL flag on hashes when no
261 uvar magic is involved could do nasty things on tied hashes.
262 + Upd : More resources in META.yml.
264 0.28 2009-01-24 17:05 UTC
265 + Add : The 'copy_key' option to wizard(), that enables the safe edition
266 of $_[2] in uvar callbacks to redirect actions to another key.
267 + Fix : Exception objects thrown from callbacks at compile time were not
269 + Fix : free callbacks triggered during global destruction could lead to
270 segfaults because the wizard was already freed.
271 + Fix : Refcount of scalars with free callback.
272 + Fix : Segfaults when using get or uvar magic simultaneously with clear
274 + Doc : The documentation about what the callbacks take and should
275 return is now somewhat clearer.
277 0.27 2009-01-18 23:40 UTC
278 + Fix : Offset of 1 with len magic on scalars.
279 + Fix : Segfaults that seem to happen when one croak in a callback at
280 compile time (thanks Ash Berlin for the testcase).
281 + Upd : Resources in META.yml (Florian Ragwitz)
282 + Tst : Large cleanup resulting in a wider set of tested behaviours.
284 0.26 2008-11-29 22:05 UTC
285 + Fix : Leaks of cloned coderefs that access lexicals.
286 + Fix : Building with maint-5.8 and blead (thanks Andreas Koenig).
288 0.25 2008-11-12 19:40 UTC
289 + Fix : Compatibility with 5.8.9.
290 + Fix : Old Pod::Coverage don't ignore CLONE.
291 + Upd : META.yml spec updated to 1.4.
293 0.24 2008-10-12 14:55 UTC
294 + Fix : Really fix it.
296 0.23 2008-10-11 17:25 UTC
297 + Fix : Building on perls with multiplicity but not ithreads.
299 0.22 2008-09-29 19:40 UTC
300 + Doc : Explain how magic differs from overloading.
301 + Fix : Disable thread safety for perl 5.8 on Windows. If you want it
302 enabled, reliable feedback is welcome.
304 0.21_02 2008-09-28 23:10 UTC
305 + Fix : Hide previously deleted magic tokens in vmg_svt_free (Yuval
308 0.21_01 2008-09-22 13:25 UTC
309 + Chg : Shrink the context data.
310 + Fix : Enable thread safety features for perls that just lack
312 + Tst : Output threads and threads::shared versions in threads tests.
314 0.20 2008-09-20 23:35 UTC
315 + Fix : The module is now thread safe. You can test if it had been
316 built with thread safety features enabled through the
317 VMG_THREADSAFE constant.
318 + Fix : Correct backtrace when wizard() croaks.
319 + Fix : 'cast $a, undef' should throw an 'invalid wizard' error instead
320 of an 'invalid signature'.
322 0.19 2008-07-05 15:05 UTC
323 + Add : The patchlevel with which the module has been built is now
324 exported with the VMG_PERL_PATCHLEVEL constant.
325 + Chg : Version checking macros overhaul.
326 + Fix : 'uvar' magic should only be enabled in maint-5.10 and higher
327 since it's never going to be backported to 5.8.
328 + Fix : Likewise, the copy callback API change should only be enabled in
330 + Fix : Instead of those ackward fake refcount increments in the wizard
331 destructor, flag the wizard with SVf_BREAK and force temps
332 cleanup. This should fix the wizard destruction for all perls,
333 and solves the leak introduced by the previous solution.
335 0.18 2008-07-04 12:20 UTC
336 + Fix : Enable uvar magic when patchlevel >= 28419.
337 + Fix : Set VMG_COMPAT_ARRAY_PUSH_NOLEN for Windows ActiveState Perl
340 0.17 2008-05-11 09:05 UTC
341 + Fix : 5.8.4 and lower don't have mPUSHi. Gah. Yeah, I know, I should
344 0.16 2008-05-10 22:05 UTC
345 + Add : The samples/copy.pl script.
346 + Chg : The sv_magical() + vmg_mg_magical() combo was simplified into
348 + Tst : t/33-code.t was testing scalars, not code.
349 + Tst : is() and like() are better than ok().
350 + Tst : 100% coverage reached.
352 0.15 2008-04-11 18:25 UTC
353 + Chg : Factor vmg_cb_call{1,2,3}() into one function.
354 + Fix : len magic is no longer called when taking the length() of a
355 magical scalar since p32969. The VMG_COMPAT_SCALAR_LENGTH_NOLEN
356 constant was added to cover this.
357 + Tst : More tests for t/22-len.t.
359 0.14 2008-03-24 12:35 UTC
360 + Fix : t/16-huf.t failures on Solaris and FreeBSD caused by not
361 updating mg->mg_ptr after Renew-ing it on dispell.
362 + Fix : Undefining MGf_DUP caused the wizard's magic to be wrongly
363 initialized. MGf_DUP now appears as true from the user point of
364 view, but the dup callback isn't actually set.
365 + Fix : Warnings with blead due to copy callbacks now taking an I32 (API
367 + Fix : vmg_svt_val() and vmg_uvar_del() aren't required if no uvar
369 + Tst : Tests now output the version of optional modules loaded.
371 0.13 2008-03-19 14:35 UTC
372 + Doc : Link to coverage report.
373 + Fix : Correct dependencies listing in META.yml.
374 + Tst : Improved test coverage.
375 + Tst : Print the patchlevel as a comment.
376 + Tst : Use is() where it's relevant.
377 + Tst : t/16-huf.t now really tests interaction with H::U::FH.
379 0.12 2008-02-07 18:15 UTC
380 + Fix : POD error. Thanks to Chris Williams (BinGOs) for the quick
383 0.11 2008-02-07 17:55 UTC
384 + Add : Copy callbacks now receive the current key/index in $_[2].
385 The current value/element is now in $_[3].
386 + Chg : The trigger of copy magic on list assignment with perls greater
387 than 5.10.0 was caused by uvar magic. Hence,
388 VMG_COMPAT_HASH_LISTASSIGN_COPY was removed. Use VMG_UVAR
390 + Fix : Build failures on patched perls.
391 + Tst : Added missing exported symbols to 01-import.t.
393 0.10 2008-02-04 11:30 UTC
394 + Add : New script : samples/vm_vs_tie.pl, that benchmarks our uvar
395 magic versus tied hashes.
396 + Add : The VMG_COMPAT_* constants can be used from userspace to check
397 perl magic abilities.
398 + Fix : Callbacks that returned undef made us croak, breaking the
399 variable behaviour (regression test in 14-callbacks.t).
400 + Fix : uvar callbacks weren't tested for non-NULL-ity before being
401 called (regression test in 28-uvar.t).
402 + Tst : Fix typo in 25-copy.t that prevented Tie::Hash tests to be ran.
404 0.09 2008-02-02 11:30 UTC
405 + Doc : Explicitely say that uvar callbacks are safely ignored for
407 + Doc : Document caveats and fix the usual set of typos.
408 + Fix : vmg_dispell() didn't check if the ext magic were ours when
409 counting wizards that have uvar callbacks, resulting in a
410 possible memory misread.
411 + Fix : getdata() now returns directly the data object, and no longer a
412 copy. This caused a leak.
413 + Tst : Prefix author tests by 9*-.
414 + Tst : New optional author test : 95-portability-files.t, that uses
415 Test::Portability::Files when it's present.
416 + Tst : New test : 14-self.t, that tests application of magic on the
418 + Tst : Move Hash::Util::FieldHash tests out of 11-multiple.t to
421 0.08 2008-02-01 16:55 UTC
422 + Add : copy magic for tied arrays/hashes.
424 + Add : uvar magics : fetch, store, exists, delete for hashes.
427 Internal development versions.
429 0.06 2007-11-20 10:10 UTC
430 + Chg : 5.7.3 is now officially required.
431 + Fix : "data" test failures on 5.8.{0,2}.
432 + Fix : Drand01() vs rand().
434 0.05 2007-11-19 09:10 UTC
435 + Fix : 5.10.0_RC1 compatibility fix.
437 0.04 2007-08-28 12:25 UTC
438 + Chg : s/require (XSLoader)/use $1/.
439 + Fix : Tests are now strict.
440 + Fix : Complete dependencies.
442 0.03 2007-08-01 17:20 UTC
443 + Add : Passing the signature of an already defined magic to wizard()
444 now returns the corresponding magic object.
445 + Add : You can pass the numeric signature as the wizard argument of
446 cast(), dispell() and getdata().
447 + Add : Any argument specified after the wizard (or the signature) in a
448 call to cast() is now passed to the private data constructor in
450 + Chg : $_[0] is now always a reference to the magic variable in all
451 callbacks. The reason for changing from the previous behaviour
452 is that one may want to apply the same magic to a plain scalar
453 and to a scalar reference, and needs a way to distinguish
454 between them in the callback (say, ref()).
455 + Fix : Wizard object destruction used not to free the signature.
457 0.02 2007-07-27 13:50 UTC
458 + Fix : In response to test report 548152 :
459 Newx() and SvMAGIC_set() not present on older perls.
460 + Fix : In response to test report 548275 :
461 Since perl 5.9.5, 'clear' magic is invoked when an array is
462 undefined (bug #43357). Moreover, 'len' magic is no longer
463 called by pushing an element since perl 5.9.3.
464 + Fix : Missing glob test in MANIFEST.
466 0.01 2007-07-25 16:15 UTC
467 First version, released on an unsuspecting world.