1 Revision history for Variable-Magic
3 0.44 2010-09-24 19:10 UTC
4 + Fix : Broken linkage on Windows with gcc 3.4, which appears in
5 particular when using ActivePerl's default compiler suite.
6 For those setups, the Variable::Magic shared library will now
7 be linked against the perl dll directly (instead of the import
8 library). This (should) fix RT #51483.
9 Thanks Christian Walde for helping to reproduce this failure
11 + Rem : Support for development perls from the 5.11 branch but older
12 than the 5.11.0 release was removed. This could cause more
13 recent setups to fail.
14 + Tst : Threads tests are now only run on perl 5.13.4 and higher.
15 They could segfault randomly because of what seems to be an
16 internal bug of Perl, which has been addressed in 5.13.4.
17 There is also an environment variable that allows you to
18 forcefully run those tests, but it should be set only for
19 author testing and not for end users.
21 0.43 2010-06-25 23:35 UTC
22 + Add : The new constant VMG_COMPAT_GLOB_GET tells you whether get magic
23 is called for globs. It's true starting perl 5.13.2.
24 + Chg : All callbacks are now called within an eval-like context.
25 Only free callbacks used to be called that way.
26 + Fix : Some exceptions thrown from a free callback could be lost.
27 + Fix : Croak messages could sometimes be repeated several times.
28 + Fix : t/41-clone.t segfaulting with perl 5.13.2.
30 0.42 2010-05-19 00:15 UTC
31 This is a maintenance release. The code contains no functional change.
32 Users of 0.41 can skip this update.
33 + Fix : Test failures with perl 5.13.
34 + Tst : Improve coverage.
36 0.41 2010-03-15 17:35 UTC
37 + Doc : Tweaks and fixups.
39 + Fix : Compatibility with the soon-to-be-released perl 5.12.0.
40 + Fix : Correctly propagate the errors thrown when variable destruction
41 happens at compile-time and not from inside eval STRING.
42 Thanks Florian Ragwitz and Ash Berlin for reporting.
44 0.40 2010-01-06 23:20 UTC
45 + Fix : Possible memory miswrites when passing data arguments to cast().
46 + Fix : Minor C portability tweaks.
48 0.39 2009-12-01 00:05 UTC
49 + Add : You can use a function name as a callback by passing a string
50 reference to wizard() instead of a code reference.
51 + Fix : Compatiblity with perl 5.11.1 and 5.11.2.
52 + Fix : Scalars stored into the data slot no longer leak.
53 + Fix : Thread destruction should not segfault anymore.
54 + Opt : As a result of removing the deprecated features, less memory is
55 used for both threaded (a pointer table is used instead of a
56 hash) and non-threaded (no global structure is needed anymore)
58 + Rem : As advertised in the precendent version, all the signature
59 related features were removed. This includes the 'sig' option
60 to wizard(), the getsig() and gensig() functions, and the SIG_*
63 0.38 2009-10-04 16:10 UTC
64 + Dep : All the signature-related features are DEPRECATED and will be
65 removed in december 2009. This includes the 'sig' option to
66 wizard(), the getsig() and gensig() functions, and the SIG_*
68 Afaik those features were never used in real life, and they make
69 the XS code slow and brittle. If you want your wizards to be
70 globally available, you'll have to handle that yourself by
71 storing them in a hash.
72 + Doc : More examples in a new COOKBOOK section.
73 + Fix : Compatiblity with the 5.11.0 release. The new compatibility
74 constant VMG_COMPAT_ARRAY_PUSH_NOLEN_VOID was added to cover
76 + Fix : Work around Kwalitee test misfailures.
78 0.37 2009-08-25 15:25 UTC
79 + Add : The new constant VMG_FORKSAFE can be tested to know whether the
80 module will behave nicely when fork()ing. It's currently always
81 true except on Windows where you need perl 5.10.1 for it to be
83 + Doc : Nits and clarifications.
85 0.36 2009-07-05 16:30 UTC
86 + Chg : getdata() now returns an empty list when no magic is present.
87 It used to return undef but did not croak as stated in the doc.
88 Thanks Matt S. Trout for pointing this out.
89 + Fix : Building and tests with ActiveState Perl 5.8 build >= 822.
91 0.35 2009-05-15 20:50 UTC
92 This is a maintenance release. Users of 0.34 can skip this update.
93 + Chg : uvar magic used to be enabled for 5.9.5. It now requires 5.10 or
95 + Fix : Silence compilation warnings for cxinc().
96 + Rem : The workaround introduced in 0.31 for the "Unknown errors" issue
97 was removed, as regression tests show that 0.34 provides a
99 + Tst : Skip t/40-threads.t on 5.8.x with old versions of threads and
102 0.34 2009-04-19 16:55 UTC
103 + Fix : Destruction of wizards in require.
104 + Fix : "panic: restartop" in 5.10 when dieing at compile time triggers
107 0.33 2009-03-26 00:00 UTC
108 + Chg : cast() and getsig() now croak too when an invalid signature is
110 + Fix : It was possible to generate with gensig() the same signature
111 twice before actually defining the wizards.
112 + Tst : More stash tests.
113 + Tst : Test when the magic takes place in some cases.
114 + Tst : Improved coverage for the op object feature.
116 0.32 2009-03-01 13:45 UTC
117 + Chg : dispell() and getdata() now croak when an invalid signature is
119 + Doc : More examples and a brand new synopsis.
120 + Fix : The signature is no longer stored in the mg_private member of
121 the MAGIC struct. This fixes possible clashes with magics from
123 + Fix : op info objects weren't blessed into the right class.
124 + Fix : Races when initializing the op names cache under threads.
126 0.31 2009-02-19 23:50 UTC
127 + Doc : A somewhat better introduction. The FUNCTION section was moved
128 before CONSTANTS and HISTORY since it's more important. And
129 'assignation' is really 'assignment', says Ricardo. :)
130 + Fix : The op_info features are now thread safe.
131 + Fix : Segfaults that occured when trying to get the op name during
133 + Fix : Segfaults and wrong "Unknown error" exceptions that happened
134 when dieing in require caused a free callback to fire.
135 Thanks Florian Ragwitz for reporting.
137 0.30 2009-02-12 17:05 UTC
138 + Add : You can now get in all the callbacks either the name or a B::OP
139 object representing the current op if you pass the new 'op_info'
140 option to wizard() with respective values VMG_OP_INFO_NAME and
142 + Doc : Document magic on array/hashes values.
143 + Fix : Completely skip free callbacks during global destruction.
144 + Tst : Tests now pass again on 5.8.3 and lower (they were broken since
146 + Tst : Test magic on symbol table and on array/hash values.
148 0.29 2009-02-08 11:10 UTC
149 + Add : Set $_[2] to the default length for len magic on scalars.
150 + Chg : Perl 5.8 is required.
151 + Chg : The CLONE method will no longer be defined for non-threaded
154 + Fix : The logic for turning off the GMAGICAL flag on hashes when no
155 uvar magic is involved could do nasty things on tied hashes.
156 + Upd : More resources in META.yml.
158 0.28 2009-01-24 17:05 UTC
159 + Add : The 'copy_key' option to wizard(), that enables the safe edition
160 of $_[2] in uvar callbacks to redirect actions to another key.
161 + Fix : Exception objects thrown from callbacks at compile time were not
163 + Fix : free callbacks triggered during global destruction could lead to
164 segfaults because the wizard was already freed.
165 + Fix : Refcount of scalars with free callback.
166 + Fix : Segfaults when using get or uvar magic simultaneously with clear
168 + Doc : The documentation about what the callbacks take and should
169 return is now somewhat clearer.
171 0.27 2009-01-18 23:40 UTC
172 + Fix : Offset of 1 with len magic on scalars.
173 + Fix : Segfaults that seem to happen when one croak in a callback at
174 compile time (thanks Ash Berlin for the testcase).
175 + Upd : Resources in META.yml (Florian Ragwitz)
176 + Tst : Large cleanup resulting in a wider set of tested behaviours.
178 0.26 2008-11-29 22:05 UTC
179 + Fix : Leaks of cloned coderefs that access lexicals.
180 + Fix : Building with maint-5.8 and blead (thanks Andreas Koenig).
182 0.25 2008-11-12 19:40 UTC
183 + Fix : Compatibility with 5.8.9.
184 + Fix : Old Pod::Coverage don't ignore CLONE.
185 + Upd : META.yml spec updated to 1.4.
187 0.24 2008-10-12 14:55 UTC
188 + Fix : Really fix it.
190 0.23 2008-10-11 17:25 UTC
191 + Fix : Building on perls with multiplicity but not ithreads.
193 0.22 2008-09-29 19:40 UTC
194 + Doc : Explain how magic differs from overloading.
195 + Fix : Disable thread safety for perl 5.8 on Windows. If you want it
196 enabled, reliable feedback is welcome.
198 0.21_02 2008-09-28 23:10 UTC
199 + Fix : Hide previously deleted magic tokens in vmg_svt_free (Yuval
202 0.21_01 2008-09-22 13:25 UTC
203 + Chg : Shrink the context data.
204 + Fix : Enable thread safety features for perls that just lack
206 + Tst : Output threads and threads::shared versions in threads tests.
208 0.20 2008-09-20 23:35 UTC
209 + Fix : The module is now thread safe. You can test if it had been
210 built with thread safety features enabled through the
211 VMG_THREADSAFE constant.
212 + Fix : Correct backtrace when wizard() croaks.
213 + Fix : 'cast $a, undef' should throw an 'invalid wizard' error instead
214 of an 'invalid signature'.
216 0.19 2008-07-05 15:05 UTC
217 + Add : The patchlevel with which the module has been built is now
218 exported with the VMG_PERL_PATCHLEVEL constant.
219 + Chg : Version checking macros overhaul.
220 + Fix : 'uvar' magic should only be enabled in maint-5.10 and higher
221 since it's never going to be backported to 5.8.
222 + Fix : Likewise, the copy callback API change should only be enabled in
224 + Fix : Instead of those ackward fake refcount increments in the wizard
225 destructor, flag the wizard with SVf_BREAK and force temps
226 cleanup. This should fix the wizard destruction for all perls,
227 and solves the leak introduced by the previous solution.
229 0.18 2008-07-04 12:20 UTC
230 + Fix : Enable uvar magic when patchlevel >= 28419.
231 + Fix : Set VMG_COMPAT_ARRAY_PUSH_NOLEN for Windows ActiveState Perl
234 0.17 2008-05-11 09:05 UTC
235 + Fix : 5.8.4 and lower don't have mPUSHi. Gah. Yeah, I know, I should
238 0.16 2008-05-10 22:05 UTC
239 + Add : The samples/copy.pl script.
240 + Chg : The sv_magical() + vmg_mg_magical() combo was simplified into
242 + Tst : t/33-code.t was testing scalars, not code.
243 + Tst : is() and like() are better than ok().
244 + Tst : 100% coverage reached.
246 0.15 2008-04-11 18:25 UTC
247 + Chg : Factor vmg_cb_call{1,2,3}() into one function.
248 + Fix : len magic is no longer called when taking the length() of a
249 magical scalar since p32969. The VMG_COMPAT_SCALAR_LENGTH_NOLEN
250 constant was added to cover this.
251 + Tst : More tests for t/22-len.t.
253 0.14 2008-03-24 12:35 UTC
254 + Fix : t/16-huf.t failures on Solaris and FreeBSD caused by not
255 updating mg->mg_ptr after Renew-ing it on dispell.
256 + Fix : Undefining MGf_DUP caused the wizard's magic to be wrongly
257 initialized. MGf_DUP now appears as true from the user point of
258 view, but the dup callback isn't actually set.
259 + Fix : Warnings with blead due to copy callbacks now taking an I32 (API
261 + Fix : vmg_svt_val() and vmg_uvar_del() aren't required if no uvar
263 + Tst : Tests now output the version of optional modules loaded.
265 0.13 2008-03-19 14:35 UTC
266 + Doc : Link to coverage report.
267 + Fix : Correct dependencies listing in META.yml.
268 + Tst : Improved test coverage.
269 + Tst : Print the patchlevel as a comment.
270 + Tst : Use is() where it's relevant.
271 + Tst : t/16-huf.t now really tests interaction with H::U::FH.
273 0.12 2008-02-07 18:15 UTC
274 + Fix : POD error. Thanks to Chris Williams (BinGOs) for the quick
277 0.11 2008-02-07 17:55 UTC
278 + Add : Copy callbacks now receive the current key/index in $_[2].
279 The current value/element is now in $_[3].
280 + Chg : The trigger of copy magic on list assignment with perls greater
281 than 5.10.0 was caused by uvar magic. Hence,
282 VMG_COMPAT_HASH_LISTASSIGN_COPY was removed. Use VMG_UVAR
284 + Fix : Build failures on patched perls.
285 + Tst : Added missing exported symbols to 01-import.t.
287 0.10 2008-02-04 11:30 UTC
288 + Add : New script : samples/vm_vs_tie.pl, that benchmarks our uvar
289 magic versus tied hashes.
290 + Add : The VMG_COMPAT_* constants can be used from userspace to check
291 perl magic abilities.
292 + Fix : Callbacks that returned undef made us croak, breaking the
293 variable behaviour (regression test in 14-callbacks.t).
294 + Fix : uvar callbacks weren't tested for non-NULL-ity before being
295 called (regression test in 28-uvar.t).
296 + Tst : Fix typo in 25-copy.t that prevented Tie::Hash tests to be ran.
298 0.09 2008-02-02 11:30 UTC
299 + Doc : Explicitely say that uvar callbacks are safely ignored for
301 + Doc : Document caveats and fix the usual set of typos.
302 + Fix : vmg_dispell() didn't check if the ext magic were ours when
303 counting wizards that have uvar callbacks, resulting in a
304 possible memory misread.
305 + Fix : getdata() now returns directly the data object, and no longer a
306 copy. This caused a leak.
307 + Tst : Prefix author tests by 9*-.
308 + Tst : New optional author test : 95-portability-files.t, that uses
309 Test::Portability::Files when it's present.
310 + Tst : New test : 14-self.t, that tests application of magic on the
312 + Tst : Move Hash::Util::FieldHash tests out of 11-multiple.t to
315 0.08 2008-02-01 16:55 UTC
316 + Add : copy magic for tied arrays/hashes.
318 + Add : uvar magics : fetch, store, exists, delete for hashes.
321 Internal development versions.
323 0.06 2007-11-20 10:10 UTC
324 + Chg : 5.7.3 is now officially required.
325 + Fix : "data" test failures on 5.8.{0,2}.
326 + Fix : Drand01() vs rand().
328 0.05 2007-11-19 09:10 UTC
329 + Fix : 5.10.0_RC1 compatibility fix.
331 0.04 2007-08-28 12:25 UTC
332 + Chg : s/require (XSLoader)/use $1/.
333 + Fix : Tests are now strict.
334 + Fix : Complete dependencies.
336 0.03 2007-08-01 17:20 UTC
337 + Add : Passing the signature of an already defined magic to wizard()
338 now returns the corresponding magic object.
339 + Add : You can pass the numeric signature as the wizard argument of
340 cast(), dispell() and getdata().
341 + Add : Any argument specified after the wizard (or the signature) in a
342 call to cast() is now passed to the private data constructor in
344 + Chg : $_[0] is now always a reference to the magic variable in all
345 callbacks. The reason for changing from the previous behaviour
346 is that one may want to apply the same magic to a plain scalar
347 and to a scalar reference, and needs a way to distinguish
348 between them in the callback (say, ref()).
349 + Fix : Wizard object destruction used not to free the signature.
351 0.02 2007-07-27 13:50 UTC
352 + Fix : In response to test report 548152 :
353 Newx() and SvMAGIC_set() not present on older perls.
354 + Fix : In response to test report 548275 :
355 Since perl 5.9.5, 'clear' magic is invoked when an array is
356 undefined (bug #43357). Moreover, 'len' magic is no longer
357 called by pushing an element since perl 5.9.3.
358 + Fix : Missing glob test in MANIFEST.
360 0.01 2007-07-25 16:15 UTC
361 First version, released on an unsuspecting world.