1 Revision history for Variable-Magic
3 0.46 2011-01-23 16:45 UTC
4 + Fix : [RT #64866] : Assertion failure with perl 5.13.9.
5 The real problem was that dispell() wasn't properly resetting
6 the magical sv flags, which became visible in the test suite
8 Thanks Joshua ben Jore for reporting.
10 0.45 2010-11-21 23:15 UTC
11 This is a maintenance release. The code contains no functional change.
12 Users of 0.44 can skip this update.
13 + Doc : C++ compilers are officially NOT supported.
14 + Doc : The minimum perl 5.10.0 requirement for uvar magic has been
16 Thanks Peter Rabbitson for pointing this out and contributing a
18 + Tst : Tune for perl 5.13.7.
19 + Tst : Capture::Tiny will be used in t/17-ctl.t if and only if it can
22 0.44 2010-09-24 19:10 UTC
23 + Fix : Broken linkage on Windows with gcc 3.4, which appears in
24 particular when using ActivePerl's default compiler suite.
25 For those setups, the Variable::Magic shared library will now
26 be linked against the perl dll directly (instead of the import
27 library). This (should) fix RT #51483.
28 Thanks Christian Walde for helping to reproduce this failure
30 + Rem : Support for development perls from the 5.11 branch but older
31 than the 5.11.0 release was removed. This could cause more
32 recent setups to fail.
33 + Tst : Threads tests are now only run on perl 5.13.4 and higher.
34 They could segfault randomly because of what seems to be an
35 internal bug of Perl, which has been addressed in 5.13.4.
36 There is also an environment variable that allows you to
37 forcefully run those tests, but it should be set only for
38 author testing and not for end users.
40 0.43 2010-06-25 23:35 UTC
41 + Add : The new constant VMG_COMPAT_GLOB_GET tells you whether get magic
42 is called for globs. It's true starting perl 5.13.2.
43 + Chg : All callbacks are now called within an eval-like context.
44 Only free callbacks used to be called that way.
45 + Fix : Some exceptions thrown from a free callback could be lost.
46 + Fix : Croak messages could sometimes be repeated several times.
47 + Fix : t/41-clone.t segfaulting with perl 5.13.2.
49 0.42 2010-05-19 00:15 UTC
50 This is a maintenance release. The code contains no functional change.
51 Users of 0.41 can skip this update.
52 + Fix : Test failures with perl 5.13.
53 + Tst : Improve coverage.
55 0.41 2010-03-15 17:35 UTC
56 + Doc : Tweaks and fixups.
58 + Fix : Compatibility with the soon-to-be-released perl 5.12.0.
59 + Fix : Correctly propagate the errors thrown when variable destruction
60 happens at compile-time and not from inside eval STRING.
61 Thanks Florian Ragwitz and Ash Berlin for reporting.
63 0.40 2010-01-06 23:20 UTC
64 + Fix : Possible memory miswrites when passing data arguments to cast().
65 + Fix : Minor C portability tweaks.
67 0.39 2009-12-01 00:05 UTC
68 + Add : You can use a function name as a callback by passing a string
69 reference to wizard() instead of a code reference.
70 + Fix : Compatiblity with perl 5.11.1 and 5.11.2.
71 + Fix : Scalars stored into the data slot no longer leak.
72 + Fix : Thread destruction should not segfault anymore.
73 + Opt : As a result of removing the deprecated features, less memory is
74 used for both threaded (a pointer table is used instead of a
75 hash) and non-threaded (no global structure is needed anymore)
77 + Rem : As advertised in the precendent version, all the signature
78 related features were removed. This includes the 'sig' option
79 to wizard(), the getsig() and gensig() functions, and the SIG_*
82 0.38 2009-10-04 16:10 UTC
83 + Dep : All the signature-related features are DEPRECATED and will be
84 removed in december 2009. This includes the 'sig' option to
85 wizard(), the getsig() and gensig() functions, and the SIG_*
87 Afaik those features were never used in real life, and they make
88 the XS code slow and brittle. If you want your wizards to be
89 globally available, you'll have to handle that yourself by
90 storing them in a hash.
91 + Doc : More examples in a new COOKBOOK section.
92 + Fix : Compatiblity with the 5.11.0 release. The new compatibility
93 constant VMG_COMPAT_ARRAY_PUSH_NOLEN_VOID was added to cover
95 + Fix : Work around Kwalitee test misfailures.
97 0.37 2009-08-25 15:25 UTC
98 + Add : The new constant VMG_FORKSAFE can be tested to know whether the
99 module will behave nicely when fork()ing. It's currently always
100 true except on Windows where you need perl 5.10.1 for it to be
102 + Doc : Nits and clarifications.
104 0.36 2009-07-05 16:30 UTC
105 + Chg : getdata() now returns an empty list when no magic is present.
106 It used to return undef but did not croak as stated in the doc.
107 Thanks Matt S. Trout for pointing this out.
108 + Fix : Building and tests with ActiveState Perl 5.8 build >= 822.
110 0.35 2009-05-15 20:50 UTC
111 This is a maintenance release. Users of 0.34 can skip this update.
112 + Chg : uvar magic used to be enabled for 5.9.5. It now requires 5.10 or
114 + Fix : Silence compilation warnings for cxinc().
115 + Rem : The workaround introduced in 0.31 for the "Unknown errors" issue
116 was removed, as regression tests show that 0.34 provides a
118 + Tst : Skip t/40-threads.t on 5.8.x with old versions of threads and
121 0.34 2009-04-19 16:55 UTC
122 + Fix : Destruction of wizards in require.
123 + Fix : "panic: restartop" in 5.10 when dieing at compile time triggers
126 0.33 2009-03-26 00:00 UTC
127 + Chg : cast() and getsig() now croak too when an invalid signature is
129 + Fix : It was possible to generate with gensig() the same signature
130 twice before actually defining the wizards.
131 + Tst : More stash tests.
132 + Tst : Test when the magic takes place in some cases.
133 + Tst : Improved coverage for the op object feature.
135 0.32 2009-03-01 13:45 UTC
136 + Chg : dispell() and getdata() now croak when an invalid signature is
138 + Doc : More examples and a brand new synopsis.
139 + Fix : The signature is no longer stored in the mg_private member of
140 the MAGIC struct. This fixes possible clashes with magics from
142 + Fix : op info objects weren't blessed into the right class.
143 + Fix : Races when initializing the op names cache under threads.
145 0.31 2009-02-19 23:50 UTC
146 + Doc : A somewhat better introduction. The FUNCTION section was moved
147 before CONSTANTS and HISTORY since it's more important. And
148 'assignation' is really 'assignment', says Ricardo. :)
149 + Fix : The op_info features are now thread safe.
150 + Fix : Segfaults that occured when trying to get the op name during
152 + Fix : Segfaults and wrong "Unknown error" exceptions that happened
153 when dieing in require caused a free callback to fire.
154 Thanks Florian Ragwitz for reporting.
156 0.30 2009-02-12 17:05 UTC
157 + Add : You can now get in all the callbacks either the name or a B::OP
158 object representing the current op if you pass the new 'op_info'
159 option to wizard() with respective values VMG_OP_INFO_NAME and
161 + Doc : Document magic on array/hashes values.
162 + Fix : Completely skip free callbacks during global destruction.
163 + Tst : Tests now pass again on 5.8.3 and lower (they were broken since
165 + Tst : Test magic on symbol table and on array/hash values.
167 0.29 2009-02-08 11:10 UTC
168 + Add : Set $_[2] to the default length for len magic on scalars.
169 + Chg : Perl 5.8 is required.
170 + Chg : The CLONE method will no longer be defined for non-threaded
173 + Fix : The logic for turning off the GMAGICAL flag on hashes when no
174 uvar magic is involved could do nasty things on tied hashes.
175 + Upd : More resources in META.yml.
177 0.28 2009-01-24 17:05 UTC
178 + Add : The 'copy_key' option to wizard(), that enables the safe edition
179 of $_[2] in uvar callbacks to redirect actions to another key.
180 + Fix : Exception objects thrown from callbacks at compile time were not
182 + Fix : free callbacks triggered during global destruction could lead to
183 segfaults because the wizard was already freed.
184 + Fix : Refcount of scalars with free callback.
185 + Fix : Segfaults when using get or uvar magic simultaneously with clear
187 + Doc : The documentation about what the callbacks take and should
188 return is now somewhat clearer.
190 0.27 2009-01-18 23:40 UTC
191 + Fix : Offset of 1 with len magic on scalars.
192 + Fix : Segfaults that seem to happen when one croak in a callback at
193 compile time (thanks Ash Berlin for the testcase).
194 + Upd : Resources in META.yml (Florian Ragwitz)
195 + Tst : Large cleanup resulting in a wider set of tested behaviours.
197 0.26 2008-11-29 22:05 UTC
198 + Fix : Leaks of cloned coderefs that access lexicals.
199 + Fix : Building with maint-5.8 and blead (thanks Andreas Koenig).
201 0.25 2008-11-12 19:40 UTC
202 + Fix : Compatibility with 5.8.9.
203 + Fix : Old Pod::Coverage don't ignore CLONE.
204 + Upd : META.yml spec updated to 1.4.
206 0.24 2008-10-12 14:55 UTC
207 + Fix : Really fix it.
209 0.23 2008-10-11 17:25 UTC
210 + Fix : Building on perls with multiplicity but not ithreads.
212 0.22 2008-09-29 19:40 UTC
213 + Doc : Explain how magic differs from overloading.
214 + Fix : Disable thread safety for perl 5.8 on Windows. If you want it
215 enabled, reliable feedback is welcome.
217 0.21_02 2008-09-28 23:10 UTC
218 + Fix : Hide previously deleted magic tokens in vmg_svt_free (Yuval
221 0.21_01 2008-09-22 13:25 UTC
222 + Chg : Shrink the context data.
223 + Fix : Enable thread safety features for perls that just lack
225 + Tst : Output threads and threads::shared versions in threads tests.
227 0.20 2008-09-20 23:35 UTC
228 + Fix : The module is now thread safe. You can test if it had been
229 built with thread safety features enabled through the
230 VMG_THREADSAFE constant.
231 + Fix : Correct backtrace when wizard() croaks.
232 + Fix : 'cast $a, undef' should throw an 'invalid wizard' error instead
233 of an 'invalid signature'.
235 0.19 2008-07-05 15:05 UTC
236 + Add : The patchlevel with which the module has been built is now
237 exported with the VMG_PERL_PATCHLEVEL constant.
238 + Chg : Version checking macros overhaul.
239 + Fix : 'uvar' magic should only be enabled in maint-5.10 and higher
240 since it's never going to be backported to 5.8.
241 + Fix : Likewise, the copy callback API change should only be enabled in
243 + Fix : Instead of those ackward fake refcount increments in the wizard
244 destructor, flag the wizard with SVf_BREAK and force temps
245 cleanup. This should fix the wizard destruction for all perls,
246 and solves the leak introduced by the previous solution.
248 0.18 2008-07-04 12:20 UTC
249 + Fix : Enable uvar magic when patchlevel >= 28419.
250 + Fix : Set VMG_COMPAT_ARRAY_PUSH_NOLEN for Windows ActiveState Perl
253 0.17 2008-05-11 09:05 UTC
254 + Fix : 5.8.4 and lower don't have mPUSHi. Gah. Yeah, I know, I should
257 0.16 2008-05-10 22:05 UTC
258 + Add : The samples/copy.pl script.
259 + Chg : The sv_magical() + vmg_mg_magical() combo was simplified into
261 + Tst : t/33-code.t was testing scalars, not code.
262 + Tst : is() and like() are better than ok().
263 + Tst : 100% coverage reached.
265 0.15 2008-04-11 18:25 UTC
266 + Chg : Factor vmg_cb_call{1,2,3}() into one function.
267 + Fix : len magic is no longer called when taking the length() of a
268 magical scalar since p32969. The VMG_COMPAT_SCALAR_LENGTH_NOLEN
269 constant was added to cover this.
270 + Tst : More tests for t/22-len.t.
272 0.14 2008-03-24 12:35 UTC
273 + Fix : t/16-huf.t failures on Solaris and FreeBSD caused by not
274 updating mg->mg_ptr after Renew-ing it on dispell.
275 + Fix : Undefining MGf_DUP caused the wizard's magic to be wrongly
276 initialized. MGf_DUP now appears as true from the user point of
277 view, but the dup callback isn't actually set.
278 + Fix : Warnings with blead due to copy callbacks now taking an I32 (API
280 + Fix : vmg_svt_val() and vmg_uvar_del() aren't required if no uvar
282 + Tst : Tests now output the version of optional modules loaded.
284 0.13 2008-03-19 14:35 UTC
285 + Doc : Link to coverage report.
286 + Fix : Correct dependencies listing in META.yml.
287 + Tst : Improved test coverage.
288 + Tst : Print the patchlevel as a comment.
289 + Tst : Use is() where it's relevant.
290 + Tst : t/16-huf.t now really tests interaction with H::U::FH.
292 0.12 2008-02-07 18:15 UTC
293 + Fix : POD error. Thanks to Chris Williams (BinGOs) for the quick
296 0.11 2008-02-07 17:55 UTC
297 + Add : Copy callbacks now receive the current key/index in $_[2].
298 The current value/element is now in $_[3].
299 + Chg : The trigger of copy magic on list assignment with perls greater
300 than 5.10.0 was caused by uvar magic. Hence,
301 VMG_COMPAT_HASH_LISTASSIGN_COPY was removed. Use VMG_UVAR
303 + Fix : Build failures on patched perls.
304 + Tst : Added missing exported symbols to 01-import.t.
306 0.10 2008-02-04 11:30 UTC
307 + Add : New script : samples/vm_vs_tie.pl, that benchmarks our uvar
308 magic versus tied hashes.
309 + Add : The VMG_COMPAT_* constants can be used from userspace to check
310 perl magic abilities.
311 + Fix : Callbacks that returned undef made us croak, breaking the
312 variable behaviour (regression test in 14-callbacks.t).
313 + Fix : uvar callbacks weren't tested for non-NULL-ity before being
314 called (regression test in 28-uvar.t).
315 + Tst : Fix typo in 25-copy.t that prevented Tie::Hash tests to be ran.
317 0.09 2008-02-02 11:30 UTC
318 + Doc : Explicitely say that uvar callbacks are safely ignored for
320 + Doc : Document caveats and fix the usual set of typos.
321 + Fix : vmg_dispell() didn't check if the ext magic were ours when
322 counting wizards that have uvar callbacks, resulting in a
323 possible memory misread.
324 + Fix : getdata() now returns directly the data object, and no longer a
325 copy. This caused a leak.
326 + Tst : Prefix author tests by 9*-.
327 + Tst : New optional author test : 95-portability-files.t, that uses
328 Test::Portability::Files when it's present.
329 + Tst : New test : 14-self.t, that tests application of magic on the
331 + Tst : Move Hash::Util::FieldHash tests out of 11-multiple.t to
334 0.08 2008-02-01 16:55 UTC
335 + Add : copy magic for tied arrays/hashes.
337 + Add : uvar magics : fetch, store, exists, delete for hashes.
340 Internal development versions.
342 0.06 2007-11-20 10:10 UTC
343 + Chg : 5.7.3 is now officially required.
344 + Fix : "data" test failures on 5.8.{0,2}.
345 + Fix : Drand01() vs rand().
347 0.05 2007-11-19 09:10 UTC
348 + Fix : 5.10.0_RC1 compatibility fix.
350 0.04 2007-08-28 12:25 UTC
351 + Chg : s/require (XSLoader)/use $1/.
352 + Fix : Tests are now strict.
353 + Fix : Complete dependencies.
355 0.03 2007-08-01 17:20 UTC
356 + Add : Passing the signature of an already defined magic to wizard()
357 now returns the corresponding magic object.
358 + Add : You can pass the numeric signature as the wizard argument of
359 cast(), dispell() and getdata().
360 + Add : Any argument specified after the wizard (or the signature) in a
361 call to cast() is now passed to the private data constructor in
363 + Chg : $_[0] is now always a reference to the magic variable in all
364 callbacks. The reason for changing from the previous behaviour
365 is that one may want to apply the same magic to a plain scalar
366 and to a scalar reference, and needs a way to distinguish
367 between them in the callback (say, ref()).
368 + Fix : Wizard object destruction used not to free the signature.
370 0.02 2007-07-27 13:50 UTC
371 + Fix : In response to test report 548152 :
372 Newx() and SvMAGIC_set() not present on older perls.
373 + Fix : In response to test report 548275 :
374 Since perl 5.9.5, 'clear' magic is invoked when an array is
375 undefined (bug #43357). Moreover, 'len' magic is no longer
376 called by pushing an element since perl 5.9.3.
377 + Fix : Missing glob test in MANIFEST.
379 0.01 2007-07-25 16:15 UTC
380 First version, released on an unsuspecting world.