1 Revision history for Variable-Magic
3 0.40 2010-01-06 23:20 UTC
4 + Fix : Possible memory miswrites when passing data arguments to cast().
5 + Fix : Minor C portability tweaks.
7 0.39 2009-12-01 00:05 UTC
8 + Add : You can use a function name as a callback by passing a string
9 reference to wizard() instead of a code reference.
10 + Fix : Compatiblity with perl 5.11.1 and 5.11.2.
11 + Fix : Scalars stored into the data slot no longer leak.
12 + Fix : Thread destruction should not segfault anymore.
13 + Opt : As a result of removing the deprecated features, less memory is
14 used for both threaded (a pointer table is used instead of a
15 hash) and non-threaded (no global structure is needed anymore)
17 + Rem : As advertised in the precendent version, all the signature
18 related features were removed. This includes the 'sig' option
19 to wizard(), the getsig() and gensig() functions, and the SIG_*
22 0.38 2009-10-04 16:10 UTC
23 + Dep : All the signature-related features are DEPRECATED and will be
24 removed in december 2009. This includes the 'sig' option to
25 wizard(), the getsig() and gensig() functions, and the SIG_*
27 Afaik those features were never used in real life, and they make
28 the XS code slow and brittle. If you want your wizards to be
29 globally available, you'll have to handle that yourself by
30 storing them in a hash.
31 + Doc : More examples in a new COOKBOOK section.
32 + Fix : Compatiblity with the 5.11.0 release. The new compatibility
33 constant VMG_COMPAT_ARRAY_PUSH_NOLEN_VOID was added to cover
35 + Fix : Work around Kwalitee test misfailures.
37 0.37 2009-08-25 15:25 UTC
38 + Add : The new constant VMG_FORKSAFE can be tested to know whether the
39 module will behave nicely when fork()ing. It's currently always
40 true except on Windows where you need perl 5.10.1 for it to be
42 + Doc : Nits and clarifications.
44 0.36 2009-07-05 16:30 UTC
45 + Chg : getdata() now returns an empty list when no magic is present.
46 It used to return undef but did not croak as stated in the doc.
47 Thanks Matt S. Trout for pointing this out.
48 + Fix : Building and tests with ActiveState Perl 5.8 build >= 822.
50 0.35 2009-05-15 20:50 UTC
51 This is a maintenance release. Users of 0.34 can skip this update.
52 + Chg : uvar magic used to be enabled for 5.9.5. It now requires 5.10 or
54 + Fix : Silence compilation warnings for cxinc().
55 + Rem : The workaround introduced in 0.31 for the "Unknown errors" issue
56 was removed, as regression tests show that 0.34 provides a
58 + Tst : Skip t/40-threads.t on 5.8.x with old versions of threads and
61 0.34 2009-04-19 16:55 UTC
62 + Fix : Destruction of wizards in require.
63 + Fix : "panic: restartop" in 5.10 when dieing at compile time triggers
66 0.33 2009-03-26 00:00 UTC
67 + Chg : cast() and getsig() now croak too when an invalid signature is
69 + Fix : It was possible to generate with gensig() the same signature
70 twice before actually defining the wizards.
71 + Tst : More stash tests.
72 + Tst : Test when the magic takes place in some cases.
73 + Tst : Improved coverage for the op object feature.
75 0.32 2009-03-01 13:45 UTC
76 + Chg : dispell() and getdata() now croak when an invalid signature is
78 + Doc : More examples and a brand new synopsis.
79 + Fix : The signature is no longer stored in the mg_private member of
80 the MAGIC struct. This fixes possible clashes with magics from
82 + Fix : op info objects weren't blessed into the right class.
83 + Fix : Races when initializing the op names cache under threads.
85 0.31 2009-02-19 23:50 UTC
86 + Doc : A somewhat better introduction. The FUNCTION section was moved
87 before CONSTANTS and HISTORY since it's more important. And
88 'assignation' is really 'assignment', says Ricardo. :)
89 + Fix : The op_info features are now thread safe.
90 + Fix : Segfaults that occured when trying to get the op name during
92 + Fix : Segfaults and wrong "Unknown error" exceptions that happened
93 when dieing in require caused a free callback to fire.
94 Thanks Florian Ragwitz for reporting.
96 0.30 2009-02-12 17:05 UTC
97 + Add : You can now get in all the callbacks either the name or a B::OP
98 object representing the current op if you pass the new 'op_info'
99 option to wizard() with respective values VMG_OP_INFO_NAME and
101 + Doc : Document magic on array/hashes values.
102 + Fix : Completely skip free callbacks during global destruction.
103 + Tst : Tests now pass again on 5.8.3 and lower (they were broken since
105 + Tst : Test magic on symbol table and on array/hash values.
107 0.29 2009-02-08 11:10 UTC
108 + Add : Set $_[2] to the default length for len magic on scalars.
109 + Chg : Perl 5.8 is required.
110 + Chg : The CLONE method will no longer be defined for non-threaded
113 + Fix : The logic for turning off the GMAGICAL flag on hashes when no
114 uvar magic is involved could do nasty things on tied hashes.
115 + Upd : More resources in META.yml.
117 0.28 2009-01-24 17:05 UTC
118 + Add : The 'copy_key' option to wizard(), that enables the safe edition
119 of $_[2] in uvar callbacks to redirect actions to another key.
120 + Fix : Exception objects thrown from callbacks at compile time were not
122 + Fix : free callbacks triggered during global destruction could lead to
123 segfaults because the wizard was already freed.
124 + Fix : Refcount of scalars with free callback.
125 + Fix : Segfaults when using get or uvar magic simultaneously with clear
127 + Doc : The documentation about what the callbacks take and should
128 return is now somewhat clearer.
130 0.27 2009-01-18 23:40 UTC
131 + Fix : Offset of 1 with len magic on scalars.
132 + Fix : Segfaults that seem to happen when one croak in a callback at
133 compile time (thanks Ash Berlin for the testcase).
134 + Upd : Resources in META.yml (Florian Ragwitz)
135 + Tst : Large cleanup resulting in a wider set of tested behaviours.
137 0.26 2008-11-29 22:05 UTC
138 + Fix : Leaks of cloned coderefs that access lexicals.
139 + Fix : Building with maint-5.8 and blead (thanks Andreas Koenig).
141 0.25 2008-11-12 19:40 UTC
142 + Fix : Compatibility with 5.8.9.
143 + Fix : Old Pod::Coverage don't ignore CLONE.
144 + Upd : META.yml spec updated to 1.4.
146 0.24 2008-10-12 14:55 UTC
147 + Fix : Really fix it.
149 0.23 2008-10-11 17:25 UTC
150 + Fix : Building on perls with multiplicity but not ithreads.
152 0.22 2008-09-29 19:40 UTC
153 + Doc : Explain how magic differs from overloading.
154 + Fix : Disable thread safety for perl 5.8 on Windows. If you want it
155 enabled, reliable feedback is welcome.
157 0.21_02 2008-09-28 23:10 UTC
158 + Fix : Hide previously deleted magic tokens in vmg_svt_free (Yuval
161 0.21_01 2008-09-22 13:25 UTC
162 + Chg : Shrink the context data.
163 + Fix : Enable thread safety features for perls that just lack
165 + Tst : Output threads and threads::shared versions in threads tests.
167 0.20 2008-09-20 23:35 UTC
168 + Fix : The module is now thread safe. You can test if it had been
169 built with thread safety features enabled through the
170 VMG_THREADSAFE constant.
171 + Fix : Correct backtrace when wizard() croaks.
172 + Fix : 'cast $a, undef' should throw an 'invalid wizard' error instead
173 of an 'invalid signature'.
175 0.19 2008-07-05 15:05 UTC
176 + Add : The patchlevel with which the module has been built is now
177 exported with the VMG_PERL_PATCHLEVEL constant.
178 + Chg : Version checking macros overhaul.
179 + Fix : 'uvar' magic should only be enabled in maint-5.10 and higher
180 since it's never going to be backported to 5.8.
181 + Fix : Likewise, the copy callback API change should only be enabled in
183 + Fix : Instead of those ackward fake refcount increments in the wizard
184 destructor, flag the wizard with SVf_BREAK and force temps
185 cleanup. This should fix the wizard destruction for all perls,
186 and solves the leak introduced by the previous solution.
188 0.18 2008-07-04 12:20 UTC
189 + Fix : Enable uvar magic when patchlevel >= 28419.
190 + Fix : Set VMG_COMPAT_ARRAY_PUSH_NOLEN for Windows ActiveState Perl
193 0.17 2008-05-11 09:05 UTC
194 + Fix : 5.8.4 and lower don't have mPUSHi. Gah. Yeah, I know, I should
197 0.16 2008-05-10 22:05 UTC
198 + Add : The samples/copy.pl script.
199 + Chg : The sv_magical() + vmg_mg_magical() combo was simplified into
201 + Tst : t/33-code.t was testing scalars, not code.
202 + Tst : is() and like() are better than ok().
203 + Tst : 100% coverage reached.
205 0.15 2008-04-11 18:25 UTC
206 + Chg : Factor vmg_cb_call{1,2,3}() into one function.
207 + Fix : len magic is no longer called when taking the length() of a
208 magical scalar since p32969. The VMG_COMPAT_SCALAR_LENGTH_NOLEN
209 constant was added to cover this.
210 + Tst : More tests for t/22-len.t.
212 0.14 2008-03-24 12:35 UTC
213 + Fix : t/16-huf.t failures on Solaris and FreeBSD caused by not
214 updating mg->mg_ptr after Renew-ing it on dispell.
215 + Fix : Undefining MGf_DUP caused the wizard's magic to be wrongly
216 initialized. MGf_DUP now appears as true from the user point of
217 view, but the dup callback isn't actually set.
218 + Fix : Warnings with blead due to copy callbacks now taking an I32 (API
220 + Fix : vmg_svt_val() and vmg_uvar_del() aren't required if no uvar
222 + Tst : Tests now output the version of optional modules loaded.
224 0.13 2008-03-19 14:35 UTC
225 + Doc : Link to coverage report.
226 + Fix : Correct dependencies listing in META.yml.
227 + Tst : Improved test coverage.
228 + Tst : Print the patchlevel as a comment.
229 + Tst : Use is() where it's relevant.
230 + Tst : t/16-huf.t now really tests interaction with H::U::FH.
232 0.12 2008-02-07 18:15 UTC
233 + Fix : POD error. Thanks to Chris Williams (BinGOs) for the quick
236 0.11 2008-02-07 17:55 UTC
237 + Add : Copy callbacks now receive the current key/index in $_[2].
238 The current value/element is now in $_[3].
239 + Chg : The trigger of copy magic on list assignment with perls greater
240 than 5.10.0 was caused by uvar magic. Hence,
241 VMG_COMPAT_HASH_LISTASSIGN_COPY was removed. Use VMG_UVAR
243 + Fix : Build failures on patched perls.
244 + Tst : Added missing exported symbols to 01-import.t.
246 0.10 2008-02-04 11:30 UTC
247 + Add : New script : samples/vm_vs_tie.pl, that benchmarks our uvar
248 magic versus tied hashes.
249 + Add : The VMG_COMPAT_* constants can be used from userspace to check
250 perl magic abilities.
251 + Fix : Callbacks that returned undef made us croak, breaking the
252 variable behaviour (regression test in 14-callbacks.t).
253 + Fix : uvar callbacks weren't tested for non-NULL-ity before being
254 called (regression test in 28-uvar.t).
255 + Tst : Fix typo in 25-copy.t that prevented Tie::Hash tests to be ran.
257 0.09 2008-02-02 11:30 UTC
258 + Doc : Explicitely say that uvar callbacks are safely ignored for
260 + Doc : Document caveats and fix the usual set of typos.
261 + Fix : vmg_dispell() didn't check if the ext magic were ours when
262 counting wizards that have uvar callbacks, resulting in a
263 possible memory misread.
264 + Fix : getdata() now returns directly the data object, and no longer a
265 copy. This caused a leak.
266 + Tst : Prefix author tests by 9*-.
267 + Tst : New optional author test : 95-portability-files.t, that uses
268 Test::Portability::Files when it's present.
269 + Tst : New test : 14-self.t, that tests application of magic on the
271 + Tst : Move Hash::Util::FieldHash tests out of 11-multiple.t to
274 0.08 2008-02-01 16:55 UTC
275 + Add : copy magic for tied arrays/hashes.
277 + Add : uvar magics : fetch, store, exists, delete for hashes.
280 Internal development versions.
282 0.06 2007-11-20 10:10 UTC
283 + Chg : 5.7.3 is now officially required.
284 + Fix : "data" test failures on 5.8.{0,2}.
285 + Fix : Drand01() vs rand().
287 0.05 2007-11-19 09:10 UTC
288 + Fix : 5.10.0_RC1 compatibility fix.
290 0.04 2007-08-28 12:25 UTC
291 + Chg : s/require (XSLoader)/use $1/.
292 + Fix : Tests are now strict.
293 + Fix : Complete dependencies.
295 0.03 2007-08-01 17:20 UTC
296 + Add : Passing the signature of an already defined magic to wizard()
297 now returns the corresponding magic object.
298 + Add : You can pass the numeric signature as the wizard argument of
299 cast(), dispell() and getdata().
300 + Add : Any argument specified after the wizard (or the signature) in a
301 call to cast() is now passed to the private data constructor in
303 + Chg : $_[0] is now always a reference to the magic variable in all
304 callbacks. The reason for changing from the previous behaviour
305 is that one may want to apply the same magic to a plain scalar
306 and to a scalar reference, and needs a way to distinguish
307 between them in the callback (say, ref()).
308 + Fix : Wizard object destruction used not to free the signature.
310 0.02 2007-07-27 13:50 UTC
311 + Fix : In response to test report 548152 :
312 Newx() and SvMAGIC_set() not present on older perls.
313 + Fix : In response to test report 548275 :
314 Since perl 5.9.5, 'clear' magic is invoked when an array is
315 undefined (bug #43357). Moreover, 'len' magic is no longer
316 called by pushing an element since perl 5.9.3.
317 + Fix : Missing glob test in MANIFEST.
319 0.01 2007-07-25 16:15 UTC
320 First version, released on an unsuspecting world.