1 Revision history for Variable-Magic
3 0.37 2009-08-25 15:25 UTC
4 + Add : The new constant VMG_FORKSAFE can be tested to know whether the
5 module will behave nicely when fork()ing. It's currently always
6 true except on Windows where you need perl 5.10.1 for it to be
8 + Doc : Nits and clarifications.
10 0.36 2009-07-05 16:30 UTC
11 + Chg : getdata() now returns an empty list when no magic is present.
12 It used to return undef but did not croak as stated in the doc.
13 Thanks Matt S. Trout for pointing this out.
14 + Fix : Building and tests with ActiveState Perl 5.8 build >= 822.
16 0.35 2009-05-15 20:50 UTC
17 This is a maintenance release. Users of 0.34 can skip this update.
18 + Chg : uvar magic used to be enabled for 5.9.5. It now requires 5.10 or
20 + Fix : Silence compilation warnings for cxinc().
21 + Rem : The workaround introduced in 0.31 for the "Unknown errors" issue
22 was removed, as regression tests show that 0.34 provides a
24 + Tst : Skip t/40-threads.t on 5.8.x with old versions of threads and
27 0.34 2009-04-19 16:55 UTC
28 + Fix : Destruction of wizards in require.
29 + Fix : "panic: restartop" in 5.10 when dieing at compile time triggers
32 0.33 2009-03-26 00:00 UTC
33 + Chg : cast() and getsig() now croak too when an invalid signature is
35 + Fix : It was possible to generate with gensig() the same signature
36 twice before actually defining the wizards.
37 + Tst : More stash tests.
38 + Tst : Test when the magic takes place in some cases.
39 + Tst : Improved coverage for the op object feature.
41 0.32 2009-03-01 13:45 UTC
42 + Chg : dispell() and getdata() now croak when an invalid signature is
44 + Doc : More examples and a brand new synopsis.
45 + Fix : The signature is no longer stored in the mg_private member of
46 the MAGIC struct. This fixes possible clashes with magics from
48 + Fix : op info objects weren't blessed into the right class.
49 + Fix : Races when initializing the op names cache under threads.
51 0.31 2009-02-19 23:50 UTC
52 + Doc : A somewhat better introduction. The FUNCTION section was moved
53 before CONSTANTS and HISTORY since it's more important. And
54 'assignation' is really 'assignment', says Ricardo. :)
55 + Fix : The op_info features are now thread safe.
56 + Fix : Segfaults that occured when trying to get the op name during
58 + Fix : Segfaults and wrong "Unknown error" exceptions that happened
59 when dieing in require caused a free callback to fire.
60 Thanks Florian Ragwitz for reporting.
62 0.30 2009-02-12 17:05 UTC
63 + Add : You can now get in all the callbacks either the name or a B::OP
64 object representing the current op if you pass the new 'op_info'
65 option to wizard() with respective values VMG_OP_INFO_NAME and
67 + Doc : Document magic on array/hashes values.
68 + Fix : Completely skip free callbacks during global destruction.
69 + Tst : Tests now pass again on 5.8.3 and lower (they were broken since
71 + Tst : Test magic on symbol table and on array/hash values.
73 0.29 2009-02-08 11:10 UTC
74 + Add : Set $_[2] to the default length for len magic on scalars.
75 + Chg : Perl 5.8 is required.
76 + Chg : The CLONE method will no longer be defined for non-threaded
79 + Fix : The logic for turning off the GMAGICAL flag on hashes when no
80 uvar magic is involved could do nasty things on tied hashes.
81 + Upd : More resources in META.yml.
83 0.28 2009-01-24 17:05 UTC
84 + Add : The 'copy_key' option to wizard(), that enables the safe edition
85 of $_[2] in uvar callbacks to redirect actions to another key.
86 + Fix : Exception objects thrown from callbacks at compile time were not
88 + Fix : free callbacks triggered during global destruction could lead to
89 segfaults because the wizard was already freed.
90 + Fix : Refcount of scalars with free callback.
91 + Fix : Segfaults when using get or uvar magic simultaneously with clear
93 + Doc : The documentation about what the callbacks take and should
94 return is now somewhat clearer.
96 0.27 2009-01-18 23:40 UTC
97 + Fix : Offset of 1 with len magic on scalars.
98 + Fix : Segfaults that seem to happen when one croak in a callback at
99 compile time (thanks Ash Berlin for the testcase).
100 + Upd : Resources in META.yml (Florian Ragwitz)
101 + Tst : Large cleanup resulting in a wider set of tested behaviours.
103 0.26 2008-11-29 22:05 UTC
104 + Fix : Leaks of cloned coderefs that access lexicals.
105 + Fix : Building with maint-5.8 and blead (thanks Andreas Koenig).
107 0.25 2008-11-12 19:40 UTC
108 + Fix : Compatibility with 5.8.9.
109 + Fix : Old Pod::Coverage don't ignore CLONE.
110 + Upd : META.yml spec updated to 1.4.
112 0.24 2008-10-12 14:55 UTC
113 + Fix : Really fix it.
115 0.23 2008-10-11 17:25 UTC
116 + Fix : Building on perls with multiplicity but not ithreads.
118 0.22 2008-09-29 19:40 UTC
119 + Doc : Explain how magic differs from overloading.
120 + Fix : Disable thread safety for perl 5.8 on Windows. If you want it
121 enabled, reliable feedback is welcome.
123 0.21_02 2008-09-28 23:10 UTC
124 + Fix : Hide previously deleted magic tokens in vmg_svt_free (Yuval
127 0.21_01 2008-09-22 13:25 UTC
128 + Chg : Shrink the context data.
129 + Fix : Enable thread safety features for perls that just lack
131 + Tst : Output threads and threads::shared versions in threads tests.
133 0.20 2008-09-20 23:35 UTC
134 + Fix : The module is now thread safe. You can test if it had been
135 built with thread safety features enabled through the
136 VMG_THREADSAFE constant.
137 + Fix : Correct backtrace when wizard() croaks.
138 + Fix : 'cast $a, undef' should throw an 'invalid wizard' error instead
139 of an 'invalid signature'.
141 0.19 2008-07-05 15:05 UTC
142 + Add : The patchlevel with which the module has been built is now
143 exported with the VMG_PERL_PATCHLEVEL constant.
144 + Chg : Version checking macros overhaul.
145 + Fix : 'uvar' magic should only be enabled in maint-5.10 and higher
146 since it's never going to be backported to 5.8.
147 + Fix : Likewise, the copy callback API change should only be enabled in
149 + Fix : Instead of those ackward fake refcount increments in the wizard
150 destructor, flag the wizard with SVf_BREAK and force temps
151 cleanup. This should fix the wizard destruction for all perls,
152 and solves the leak introduced by the previous solution.
154 0.18 2008-07-04 12:20 UTC
155 + Fix : Enable uvar magic when patchlevel >= 28419.
156 + Fix : Set VMG_COMPAT_ARRAY_PUSH_NOLEN for Windows ActiveState Perl
159 0.17 2008-05-11 09:05 UTC
160 + Fix : 5.8.4 and lower don't have mPUSHi. Gah. Yeah, I know, I should
163 0.16 2008-05-10 22:05 UTC
164 + Add : The samples/copy.pl script.
165 + Chg : The sv_magical() + vmg_mg_magical() combo was simplified into
167 + Tst : t/33-code.t was testing scalars, not code.
168 + Tst : is() and like() are better than ok().
169 + Tst : 100% coverage reached.
171 0.15 2008-04-11 18:25 UTC
172 + Chg : Factor vmg_cb_call{1,2,3}() into one function.
173 + Fix : len magic is no longer called when taking the length() of a
174 magical scalar since p32969. The VMG_COMPAT_SCALAR_LENGTH_NOLEN
175 constant was added to cover this.
176 + Tst : More tests for t/22-len.t.
178 0.14 2008-03-24 12:35 UTC
179 + Fix : t/16-huf.t failures on Solaris and FreeBSD caused by not
180 updating mg->mg_ptr after Renew-ing it on dispell.
181 + Fix : Undefining MGf_DUP caused the wizard's magic to be wrongly
182 initialized. MGf_DUP now appears as true from the user point of
183 view, but the dup callback isn't actually set.
184 + Fix : Warnings with blead due to copy callbacks now taking an I32 (API
186 + Fix : vmg_svt_val() and vmg_uvar_del() aren't required if no uvar
188 + Tst : Tests now output the version of optional modules loaded.
190 0.13 2008-03-19 14:35 UTC
191 + Doc : Link to coverage report.
192 + Fix : Correct dependencies listing in META.yml.
193 + Tst : Improved test coverage.
194 + Tst : Print the patchlevel as a comment.
195 + Tst : Use is() where it's relevant.
196 + Tst : t/16-huf.t now really tests interaction with H::U::FH.
198 0.12 2008-02-07 18:15 UTC
199 + Fix : POD error. Thanks to Chris Williams (BinGOs) for the quick
202 0.11 2008-02-07 17:55 UTC
203 + Add : Copy callbacks now receive the current key/index in $_[2].
204 The current value/element is now in $_[3].
205 + Chg : The trigger of copy magic on list assignment with perls greater
206 than 5.10.0 was caused by uvar magic. Hence,
207 VMG_COMPAT_HASH_LISTASSIGN_COPY was removed. Use VMG_UVAR
209 + Fix : Build failures on patched perls.
210 + Tst : Added missing exported symbols to 01-import.t.
212 0.10 2008-02-04 11:30 UTC
213 + Add : New script : samples/vm_vs_tie.pl, that benchmarks our uvar
214 magic versus tied hashes.
215 + Add : The VMG_COMPAT_* constants can be used from userspace to check
216 perl magic abilities.
217 + Fix : Callbacks that returned undef made us croak, breaking the
218 variable behaviour (regression test in 14-callbacks.t).
219 + Fix : uvar callbacks weren't tested for non-NULL-ity before being
220 called (regression test in 28-uvar.t).
221 + Tst : Fix typo in 25-copy.t that prevented Tie::Hash tests to be ran.
223 0.09 2008-02-02 11:30 UTC
224 + Doc : Explicitely say that uvar callbacks are safely ignored for
226 + Doc : Document caveats and fix the usual set of typos.
227 + Fix : vmg_dispell() didn't check if the ext magic were ours when
228 counting wizards that have uvar callbacks, resulting in a
229 possible memory misread.
230 + Fix : getdata() now returns directly the data object, and no longer a
231 copy. This caused a leak.
232 + Tst : Prefix author tests by 9*-.
233 + Tst : New optional author test : 95-portability-files.t, that uses
234 Test::Portability::Files when it's present.
235 + Tst : New test : 14-self.t, that tests application of magic on the
237 + Tst : Move Hash::Util::FieldHash tests out of 11-multiple.t to
240 0.08 2008-02-01 16:55 UTC
241 + Add : copy magic for tied arrays/hashes.
243 + Add : uvar magics : fetch, store, exists, delete for hashes.
246 Internal development versions.
248 0.06 2007-11-20 10:10 UTC
249 + Chg : 5.7.3 is now officially required.
250 + Fix : "data" test failures on 5.8.{0,2}.
251 + Fix : Drand01() vs rand().
253 0.05 2007-11-19 09:10 UTC
254 + Fix : 5.10.0_RC1 compatibility fix.
256 0.04 2007-08-28 12:25 UTC
257 + Chg : s/require (XSLoader)/use $1/.
258 + Fix : Tests are now strict.
259 + Fix : Complete dependencies.
261 0.03 2007-08-01 17:20 UTC
262 + Add : Passing the signature of an already defined magic to wizard()
263 now returns the corresponding magic object.
264 + Add : You can pass the numeric signature as the wizard argument of
265 cast(), dispell() and getdata().
266 + Add : Any argument specified after the wizard (or the signature) in a
267 call to cast() is now passed to the private data constructor in
269 + Chg : $_[0] is now always a reference to the magic variable in all
270 callbacks. The reason for changing from the previous behaviour
271 is that one may want to apply the same magic to a plain scalar
272 and to a scalar reference, and needs a way to distinguish
273 between them in the callback (say, ref()).
274 + Fix : Wizard object destruction used not to free the signature.
276 0.02 2007-07-27 13:50 UTC
277 + Fix : In response to test report 548152 :
278 Newx() and SvMAGIC_set() not present on older perls.
279 + Fix : In response to test report 548275 :
280 Since perl 5.9.5, 'clear' magic is invoked when an array is
281 undefined (bug #43357). Moreover, 'len' magic is no longer
282 called by pushing an element since perl 5.9.3.
283 + Fix : Missing glob test in MANIFEST.
285 0.01 2007-07-25 16:15 UTC
286 First version, released on an unsuspecting world.