1 Revision history for Variable-Magic
3 0.39 2009-12-01 00:05 UTC
4 + Add : You can use a function name as a callback by passing a string
5 reference to wizard() instead of a code reference.
6 + Fix : Compatiblity with perl 5.11.1 and 5.11.2.
7 + Fix : Scalars stored into the data slot no longer leak.
8 + Fix : Thread destruction should not segfault anymore.
9 + Opt : As a result of removing the deprecated features, less memory is
10 used for both threaded (a pointer table is used instead of a
11 hash) and non-threaded (no global structure is needed anymore)
13 + Rem : As advertised in the precendent version, all the signature
14 related features were removed. This includes the 'sig' option
15 to wizard(), the getsig() and gensig() functions, and the SIG_*
18 0.38 2009-10-04 16:10 UTC
19 + Dep : All the signature-related features are DEPRECATED and will be
20 removed in december 2009. This includes the 'sig' option to
21 wizard(), the getsig() and gensig() functions, and the SIG_*
23 Afaik those features were never used in real life, and they make
24 the XS code slow and brittle. If you want your wizards to be
25 globally available, you'll have to handle that yourself by
26 storing them in a hash.
27 + Doc : More examples in a new COOKBOOK section.
28 + Fix : Compatiblity with the 5.11.0 release. The new compatibility
29 constant VMG_COMPAT_ARRAY_PUSH_NOLEN_VOID was added to cover
31 + Fix : Work around Kwalitee test misfailures.
33 0.37 2009-08-25 15:25 UTC
34 + Add : The new constant VMG_FORKSAFE can be tested to know whether the
35 module will behave nicely when fork()ing. It's currently always
36 true except on Windows where you need perl 5.10.1 for it to be
38 + Doc : Nits and clarifications.
40 0.36 2009-07-05 16:30 UTC
41 + Chg : getdata() now returns an empty list when no magic is present.
42 It used to return undef but did not croak as stated in the doc.
43 Thanks Matt S. Trout for pointing this out.
44 + Fix : Building and tests with ActiveState Perl 5.8 build >= 822.
46 0.35 2009-05-15 20:50 UTC
47 This is a maintenance release. Users of 0.34 can skip this update.
48 + Chg : uvar magic used to be enabled for 5.9.5. It now requires 5.10 or
50 + Fix : Silence compilation warnings for cxinc().
51 + Rem : The workaround introduced in 0.31 for the "Unknown errors" issue
52 was removed, as regression tests show that 0.34 provides a
54 + Tst : Skip t/40-threads.t on 5.8.x with old versions of threads and
57 0.34 2009-04-19 16:55 UTC
58 + Fix : Destruction of wizards in require.
59 + Fix : "panic: restartop" in 5.10 when dieing at compile time triggers
62 0.33 2009-03-26 00:00 UTC
63 + Chg : cast() and getsig() now croak too when an invalid signature is
65 + Fix : It was possible to generate with gensig() the same signature
66 twice before actually defining the wizards.
67 + Tst : More stash tests.
68 + Tst : Test when the magic takes place in some cases.
69 + Tst : Improved coverage for the op object feature.
71 0.32 2009-03-01 13:45 UTC
72 + Chg : dispell() and getdata() now croak when an invalid signature is
74 + Doc : More examples and a brand new synopsis.
75 + Fix : The signature is no longer stored in the mg_private member of
76 the MAGIC struct. This fixes possible clashes with magics from
78 + Fix : op info objects weren't blessed into the right class.
79 + Fix : Races when initializing the op names cache under threads.
81 0.31 2009-02-19 23:50 UTC
82 + Doc : A somewhat better introduction. The FUNCTION section was moved
83 before CONSTANTS and HISTORY since it's more important. And
84 'assignation' is really 'assignment', says Ricardo. :)
85 + Fix : The op_info features are now thread safe.
86 + Fix : Segfaults that occured when trying to get the op name during
88 + Fix : Segfaults and wrong "Unknown error" exceptions that happened
89 when dieing in require caused a free callback to fire.
90 Thanks Florian Ragwitz for reporting.
92 0.30 2009-02-12 17:05 UTC
93 + Add : You can now get in all the callbacks either the name or a B::OP
94 object representing the current op if you pass the new 'op_info'
95 option to wizard() with respective values VMG_OP_INFO_NAME and
97 + Doc : Document magic on array/hashes values.
98 + Fix : Completely skip free callbacks during global destruction.
99 + Tst : Tests now pass again on 5.8.3 and lower (they were broken since
101 + Tst : Test magic on symbol table and on array/hash values.
103 0.29 2009-02-08 11:10 UTC
104 + Add : Set $_[2] to the default length for len magic on scalars.
105 + Chg : Perl 5.8 is required.
106 + Chg : The CLONE method will no longer be defined for non-threaded
109 + Fix : The logic for turning off the GMAGICAL flag on hashes when no
110 uvar magic is involved could do nasty things on tied hashes.
111 + Upd : More resources in META.yml.
113 0.28 2009-01-24 17:05 UTC
114 + Add : The 'copy_key' option to wizard(), that enables the safe edition
115 of $_[2] in uvar callbacks to redirect actions to another key.
116 + Fix : Exception objects thrown from callbacks at compile time were not
118 + Fix : free callbacks triggered during global destruction could lead to
119 segfaults because the wizard was already freed.
120 + Fix : Refcount of scalars with free callback.
121 + Fix : Segfaults when using get or uvar magic simultaneously with clear
123 + Doc : The documentation about what the callbacks take and should
124 return is now somewhat clearer.
126 0.27 2009-01-18 23:40 UTC
127 + Fix : Offset of 1 with len magic on scalars.
128 + Fix : Segfaults that seem to happen when one croak in a callback at
129 compile time (thanks Ash Berlin for the testcase).
130 + Upd : Resources in META.yml (Florian Ragwitz)
131 + Tst : Large cleanup resulting in a wider set of tested behaviours.
133 0.26 2008-11-29 22:05 UTC
134 + Fix : Leaks of cloned coderefs that access lexicals.
135 + Fix : Building with maint-5.8 and blead (thanks Andreas Koenig).
137 0.25 2008-11-12 19:40 UTC
138 + Fix : Compatibility with 5.8.9.
139 + Fix : Old Pod::Coverage don't ignore CLONE.
140 + Upd : META.yml spec updated to 1.4.
142 0.24 2008-10-12 14:55 UTC
143 + Fix : Really fix it.
145 0.23 2008-10-11 17:25 UTC
146 + Fix : Building on perls with multiplicity but not ithreads.
148 0.22 2008-09-29 19:40 UTC
149 + Doc : Explain how magic differs from overloading.
150 + Fix : Disable thread safety for perl 5.8 on Windows. If you want it
151 enabled, reliable feedback is welcome.
153 0.21_02 2008-09-28 23:10 UTC
154 + Fix : Hide previously deleted magic tokens in vmg_svt_free (Yuval
157 0.21_01 2008-09-22 13:25 UTC
158 + Chg : Shrink the context data.
159 + Fix : Enable thread safety features for perls that just lack
161 + Tst : Output threads and threads::shared versions in threads tests.
163 0.20 2008-09-20 23:35 UTC
164 + Fix : The module is now thread safe. You can test if it had been
165 built with thread safety features enabled through the
166 VMG_THREADSAFE constant.
167 + Fix : Correct backtrace when wizard() croaks.
168 + Fix : 'cast $a, undef' should throw an 'invalid wizard' error instead
169 of an 'invalid signature'.
171 0.19 2008-07-05 15:05 UTC
172 + Add : The patchlevel with which the module has been built is now
173 exported with the VMG_PERL_PATCHLEVEL constant.
174 + Chg : Version checking macros overhaul.
175 + Fix : 'uvar' magic should only be enabled in maint-5.10 and higher
176 since it's never going to be backported to 5.8.
177 + Fix : Likewise, the copy callback API change should only be enabled in
179 + Fix : Instead of those ackward fake refcount increments in the wizard
180 destructor, flag the wizard with SVf_BREAK and force temps
181 cleanup. This should fix the wizard destruction for all perls,
182 and solves the leak introduced by the previous solution.
184 0.18 2008-07-04 12:20 UTC
185 + Fix : Enable uvar magic when patchlevel >= 28419.
186 + Fix : Set VMG_COMPAT_ARRAY_PUSH_NOLEN for Windows ActiveState Perl
189 0.17 2008-05-11 09:05 UTC
190 + Fix : 5.8.4 and lower don't have mPUSHi. Gah. Yeah, I know, I should
193 0.16 2008-05-10 22:05 UTC
194 + Add : The samples/copy.pl script.
195 + Chg : The sv_magical() + vmg_mg_magical() combo was simplified into
197 + Tst : t/33-code.t was testing scalars, not code.
198 + Tst : is() and like() are better than ok().
199 + Tst : 100% coverage reached.
201 0.15 2008-04-11 18:25 UTC
202 + Chg : Factor vmg_cb_call{1,2,3}() into one function.
203 + Fix : len magic is no longer called when taking the length() of a
204 magical scalar since p32969. The VMG_COMPAT_SCALAR_LENGTH_NOLEN
205 constant was added to cover this.
206 + Tst : More tests for t/22-len.t.
208 0.14 2008-03-24 12:35 UTC
209 + Fix : t/16-huf.t failures on Solaris and FreeBSD caused by not
210 updating mg->mg_ptr after Renew-ing it on dispell.
211 + Fix : Undefining MGf_DUP caused the wizard's magic to be wrongly
212 initialized. MGf_DUP now appears as true from the user point of
213 view, but the dup callback isn't actually set.
214 + Fix : Warnings with blead due to copy callbacks now taking an I32 (API
216 + Fix : vmg_svt_val() and vmg_uvar_del() aren't required if no uvar
218 + Tst : Tests now output the version of optional modules loaded.
220 0.13 2008-03-19 14:35 UTC
221 + Doc : Link to coverage report.
222 + Fix : Correct dependencies listing in META.yml.
223 + Tst : Improved test coverage.
224 + Tst : Print the patchlevel as a comment.
225 + Tst : Use is() where it's relevant.
226 + Tst : t/16-huf.t now really tests interaction with H::U::FH.
228 0.12 2008-02-07 18:15 UTC
229 + Fix : POD error. Thanks to Chris Williams (BinGOs) for the quick
232 0.11 2008-02-07 17:55 UTC
233 + Add : Copy callbacks now receive the current key/index in $_[2].
234 The current value/element is now in $_[3].
235 + Chg : The trigger of copy magic on list assignment with perls greater
236 than 5.10.0 was caused by uvar magic. Hence,
237 VMG_COMPAT_HASH_LISTASSIGN_COPY was removed. Use VMG_UVAR
239 + Fix : Build failures on patched perls.
240 + Tst : Added missing exported symbols to 01-import.t.
242 0.10 2008-02-04 11:30 UTC
243 + Add : New script : samples/vm_vs_tie.pl, that benchmarks our uvar
244 magic versus tied hashes.
245 + Add : The VMG_COMPAT_* constants can be used from userspace to check
246 perl magic abilities.
247 + Fix : Callbacks that returned undef made us croak, breaking the
248 variable behaviour (regression test in 14-callbacks.t).
249 + Fix : uvar callbacks weren't tested for non-NULL-ity before being
250 called (regression test in 28-uvar.t).
251 + Tst : Fix typo in 25-copy.t that prevented Tie::Hash tests to be ran.
253 0.09 2008-02-02 11:30 UTC
254 + Doc : Explicitely say that uvar callbacks are safely ignored for
256 + Doc : Document caveats and fix the usual set of typos.
257 + Fix : vmg_dispell() didn't check if the ext magic were ours when
258 counting wizards that have uvar callbacks, resulting in a
259 possible memory misread.
260 + Fix : getdata() now returns directly the data object, and no longer a
261 copy. This caused a leak.
262 + Tst : Prefix author tests by 9*-.
263 + Tst : New optional author test : 95-portability-files.t, that uses
264 Test::Portability::Files when it's present.
265 + Tst : New test : 14-self.t, that tests application of magic on the
267 + Tst : Move Hash::Util::FieldHash tests out of 11-multiple.t to
270 0.08 2008-02-01 16:55 UTC
271 + Add : copy magic for tied arrays/hashes.
273 + Add : uvar magics : fetch, store, exists, delete for hashes.
276 Internal development versions.
278 0.06 2007-11-20 10:10 UTC
279 + Chg : 5.7.3 is now officially required.
280 + Fix : "data" test failures on 5.8.{0,2}.
281 + Fix : Drand01() vs rand().
283 0.05 2007-11-19 09:10 UTC
284 + Fix : 5.10.0_RC1 compatibility fix.
286 0.04 2007-08-28 12:25 UTC
287 + Chg : s/require (XSLoader)/use $1/.
288 + Fix : Tests are now strict.
289 + Fix : Complete dependencies.
291 0.03 2007-08-01 17:20 UTC
292 + Add : Passing the signature of an already defined magic to wizard()
293 now returns the corresponding magic object.
294 + Add : You can pass the numeric signature as the wizard argument of
295 cast(), dispell() and getdata().
296 + Add : Any argument specified after the wizard (or the signature) in a
297 call to cast() is now passed to the private data constructor in
299 + Chg : $_[0] is now always a reference to the magic variable in all
300 callbacks. The reason for changing from the previous behaviour
301 is that one may want to apply the same magic to a plain scalar
302 and to a scalar reference, and needs a way to distinguish
303 between them in the callback (say, ref()).
304 + Fix : Wizard object destruction used not to free the signature.
306 0.02 2007-07-27 13:50 UTC
307 + Fix : In response to test report 548152 :
308 Newx() and SvMAGIC_set() not present on older perls.
309 + Fix : In response to test report 548275 :
310 Since perl 5.9.5, 'clear' magic is invoked when an array is
311 undefined (bug #43357). Moreover, 'len' magic is no longer
312 called by pushing an element since perl 5.9.3.
313 + Fix : Missing glob test in MANIFEST.
315 0.01 2007-07-25 16:15 UTC
316 First version, released on an unsuspecting world.