1 Revision history for Variable-Magic
3 0.42 2010-05-19 00:15 UTC
4 This is a maintenance release. The code contains no functional change.
5 Users of 0.41 can skip this update.
6 + Fix : Test failures with perl 5.13.
7 + Tst : Improve coverage.
9 0.41 2010-03-15 17:35 UTC
10 + Doc : Tweaks and fixups.
12 + Fix : Compatibility with the soon-to-be-released perl 5.12.0.
13 + Fix : Correctly propagate the errors thrown when variable destruction
14 happens at compile-time and not from inside eval STRING.
15 Thanks Florian Ragwitz and Ash Berlin for reporting.
17 0.40 2010-01-06 23:20 UTC
18 + Fix : Possible memory miswrites when passing data arguments to cast().
19 + Fix : Minor C portability tweaks.
21 0.39 2009-12-01 00:05 UTC
22 + Add : You can use a function name as a callback by passing a string
23 reference to wizard() instead of a code reference.
24 + Fix : Compatiblity with perl 5.11.1 and 5.11.2.
25 + Fix : Scalars stored into the data slot no longer leak.
26 + Fix : Thread destruction should not segfault anymore.
27 + Opt : As a result of removing the deprecated features, less memory is
28 used for both threaded (a pointer table is used instead of a
29 hash) and non-threaded (no global structure is needed anymore)
31 + Rem : As advertised in the precendent version, all the signature
32 related features were removed. This includes the 'sig' option
33 to wizard(), the getsig() and gensig() functions, and the SIG_*
36 0.38 2009-10-04 16:10 UTC
37 + Dep : All the signature-related features are DEPRECATED and will be
38 removed in december 2009. This includes the 'sig' option to
39 wizard(), the getsig() and gensig() functions, and the SIG_*
41 Afaik those features were never used in real life, and they make
42 the XS code slow and brittle. If you want your wizards to be
43 globally available, you'll have to handle that yourself by
44 storing them in a hash.
45 + Doc : More examples in a new COOKBOOK section.
46 + Fix : Compatiblity with the 5.11.0 release. The new compatibility
47 constant VMG_COMPAT_ARRAY_PUSH_NOLEN_VOID was added to cover
49 + Fix : Work around Kwalitee test misfailures.
51 0.37 2009-08-25 15:25 UTC
52 + Add : The new constant VMG_FORKSAFE can be tested to know whether the
53 module will behave nicely when fork()ing. It's currently always
54 true except on Windows where you need perl 5.10.1 for it to be
56 + Doc : Nits and clarifications.
58 0.36 2009-07-05 16:30 UTC
59 + Chg : getdata() now returns an empty list when no magic is present.
60 It used to return undef but did not croak as stated in the doc.
61 Thanks Matt S. Trout for pointing this out.
62 + Fix : Building and tests with ActiveState Perl 5.8 build >= 822.
64 0.35 2009-05-15 20:50 UTC
65 This is a maintenance release. Users of 0.34 can skip this update.
66 + Chg : uvar magic used to be enabled for 5.9.5. It now requires 5.10 or
68 + Fix : Silence compilation warnings for cxinc().
69 + Rem : The workaround introduced in 0.31 for the "Unknown errors" issue
70 was removed, as regression tests show that 0.34 provides a
72 + Tst : Skip t/40-threads.t on 5.8.x with old versions of threads and
75 0.34 2009-04-19 16:55 UTC
76 + Fix : Destruction of wizards in require.
77 + Fix : "panic: restartop" in 5.10 when dieing at compile time triggers
80 0.33 2009-03-26 00:00 UTC
81 + Chg : cast() and getsig() now croak too when an invalid signature is
83 + Fix : It was possible to generate with gensig() the same signature
84 twice before actually defining the wizards.
85 + Tst : More stash tests.
86 + Tst : Test when the magic takes place in some cases.
87 + Tst : Improved coverage for the op object feature.
89 0.32 2009-03-01 13:45 UTC
90 + Chg : dispell() and getdata() now croak when an invalid signature is
92 + Doc : More examples and a brand new synopsis.
93 + Fix : The signature is no longer stored in the mg_private member of
94 the MAGIC struct. This fixes possible clashes with magics from
96 + Fix : op info objects weren't blessed into the right class.
97 + Fix : Races when initializing the op names cache under threads.
99 0.31 2009-02-19 23:50 UTC
100 + Doc : A somewhat better introduction. The FUNCTION section was moved
101 before CONSTANTS and HISTORY since it's more important. And
102 'assignation' is really 'assignment', says Ricardo. :)
103 + Fix : The op_info features are now thread safe.
104 + Fix : Segfaults that occured when trying to get the op name during
106 + Fix : Segfaults and wrong "Unknown error" exceptions that happened
107 when dieing in require caused a free callback to fire.
108 Thanks Florian Ragwitz for reporting.
110 0.30 2009-02-12 17:05 UTC
111 + Add : You can now get in all the callbacks either the name or a B::OP
112 object representing the current op if you pass the new 'op_info'
113 option to wizard() with respective values VMG_OP_INFO_NAME and
115 + Doc : Document magic on array/hashes values.
116 + Fix : Completely skip free callbacks during global destruction.
117 + Tst : Tests now pass again on 5.8.3 and lower (they were broken since
119 + Tst : Test magic on symbol table and on array/hash values.
121 0.29 2009-02-08 11:10 UTC
122 + Add : Set $_[2] to the default length for len magic on scalars.
123 + Chg : Perl 5.8 is required.
124 + Chg : The CLONE method will no longer be defined for non-threaded
127 + Fix : The logic for turning off the GMAGICAL flag on hashes when no
128 uvar magic is involved could do nasty things on tied hashes.
129 + Upd : More resources in META.yml.
131 0.28 2009-01-24 17:05 UTC
132 + Add : The 'copy_key' option to wizard(), that enables the safe edition
133 of $_[2] in uvar callbacks to redirect actions to another key.
134 + Fix : Exception objects thrown from callbacks at compile time were not
136 + Fix : free callbacks triggered during global destruction could lead to
137 segfaults because the wizard was already freed.
138 + Fix : Refcount of scalars with free callback.
139 + Fix : Segfaults when using get or uvar magic simultaneously with clear
141 + Doc : The documentation about what the callbacks take and should
142 return is now somewhat clearer.
144 0.27 2009-01-18 23:40 UTC
145 + Fix : Offset of 1 with len magic on scalars.
146 + Fix : Segfaults that seem to happen when one croak in a callback at
147 compile time (thanks Ash Berlin for the testcase).
148 + Upd : Resources in META.yml (Florian Ragwitz)
149 + Tst : Large cleanup resulting in a wider set of tested behaviours.
151 0.26 2008-11-29 22:05 UTC
152 + Fix : Leaks of cloned coderefs that access lexicals.
153 + Fix : Building with maint-5.8 and blead (thanks Andreas Koenig).
155 0.25 2008-11-12 19:40 UTC
156 + Fix : Compatibility with 5.8.9.
157 + Fix : Old Pod::Coverage don't ignore CLONE.
158 + Upd : META.yml spec updated to 1.4.
160 0.24 2008-10-12 14:55 UTC
161 + Fix : Really fix it.
163 0.23 2008-10-11 17:25 UTC
164 + Fix : Building on perls with multiplicity but not ithreads.
166 0.22 2008-09-29 19:40 UTC
167 + Doc : Explain how magic differs from overloading.
168 + Fix : Disable thread safety for perl 5.8 on Windows. If you want it
169 enabled, reliable feedback is welcome.
171 0.21_02 2008-09-28 23:10 UTC
172 + Fix : Hide previously deleted magic tokens in vmg_svt_free (Yuval
175 0.21_01 2008-09-22 13:25 UTC
176 + Chg : Shrink the context data.
177 + Fix : Enable thread safety features for perls that just lack
179 + Tst : Output threads and threads::shared versions in threads tests.
181 0.20 2008-09-20 23:35 UTC
182 + Fix : The module is now thread safe. You can test if it had been
183 built with thread safety features enabled through the
184 VMG_THREADSAFE constant.
185 + Fix : Correct backtrace when wizard() croaks.
186 + Fix : 'cast $a, undef' should throw an 'invalid wizard' error instead
187 of an 'invalid signature'.
189 0.19 2008-07-05 15:05 UTC
190 + Add : The patchlevel with which the module has been built is now
191 exported with the VMG_PERL_PATCHLEVEL constant.
192 + Chg : Version checking macros overhaul.
193 + Fix : 'uvar' magic should only be enabled in maint-5.10 and higher
194 since it's never going to be backported to 5.8.
195 + Fix : Likewise, the copy callback API change should only be enabled in
197 + Fix : Instead of those ackward fake refcount increments in the wizard
198 destructor, flag the wizard with SVf_BREAK and force temps
199 cleanup. This should fix the wizard destruction for all perls,
200 and solves the leak introduced by the previous solution.
202 0.18 2008-07-04 12:20 UTC
203 + Fix : Enable uvar magic when patchlevel >= 28419.
204 + Fix : Set VMG_COMPAT_ARRAY_PUSH_NOLEN for Windows ActiveState Perl
207 0.17 2008-05-11 09:05 UTC
208 + Fix : 5.8.4 and lower don't have mPUSHi. Gah. Yeah, I know, I should
211 0.16 2008-05-10 22:05 UTC
212 + Add : The samples/copy.pl script.
213 + Chg : The sv_magical() + vmg_mg_magical() combo was simplified into
215 + Tst : t/33-code.t was testing scalars, not code.
216 + Tst : is() and like() are better than ok().
217 + Tst : 100% coverage reached.
219 0.15 2008-04-11 18:25 UTC
220 + Chg : Factor vmg_cb_call{1,2,3}() into one function.
221 + Fix : len magic is no longer called when taking the length() of a
222 magical scalar since p32969. The VMG_COMPAT_SCALAR_LENGTH_NOLEN
223 constant was added to cover this.
224 + Tst : More tests for t/22-len.t.
226 0.14 2008-03-24 12:35 UTC
227 + Fix : t/16-huf.t failures on Solaris and FreeBSD caused by not
228 updating mg->mg_ptr after Renew-ing it on dispell.
229 + Fix : Undefining MGf_DUP caused the wizard's magic to be wrongly
230 initialized. MGf_DUP now appears as true from the user point of
231 view, but the dup callback isn't actually set.
232 + Fix : Warnings with blead due to copy callbacks now taking an I32 (API
234 + Fix : vmg_svt_val() and vmg_uvar_del() aren't required if no uvar
236 + Tst : Tests now output the version of optional modules loaded.
238 0.13 2008-03-19 14:35 UTC
239 + Doc : Link to coverage report.
240 + Fix : Correct dependencies listing in META.yml.
241 + Tst : Improved test coverage.
242 + Tst : Print the patchlevel as a comment.
243 + Tst : Use is() where it's relevant.
244 + Tst : t/16-huf.t now really tests interaction with H::U::FH.
246 0.12 2008-02-07 18:15 UTC
247 + Fix : POD error. Thanks to Chris Williams (BinGOs) for the quick
250 0.11 2008-02-07 17:55 UTC
251 + Add : Copy callbacks now receive the current key/index in $_[2].
252 The current value/element is now in $_[3].
253 + Chg : The trigger of copy magic on list assignment with perls greater
254 than 5.10.0 was caused by uvar magic. Hence,
255 VMG_COMPAT_HASH_LISTASSIGN_COPY was removed. Use VMG_UVAR
257 + Fix : Build failures on patched perls.
258 + Tst : Added missing exported symbols to 01-import.t.
260 0.10 2008-02-04 11:30 UTC
261 + Add : New script : samples/vm_vs_tie.pl, that benchmarks our uvar
262 magic versus tied hashes.
263 + Add : The VMG_COMPAT_* constants can be used from userspace to check
264 perl magic abilities.
265 + Fix : Callbacks that returned undef made us croak, breaking the
266 variable behaviour (regression test in 14-callbacks.t).
267 + Fix : uvar callbacks weren't tested for non-NULL-ity before being
268 called (regression test in 28-uvar.t).
269 + Tst : Fix typo in 25-copy.t that prevented Tie::Hash tests to be ran.
271 0.09 2008-02-02 11:30 UTC
272 + Doc : Explicitely say that uvar callbacks are safely ignored for
274 + Doc : Document caveats and fix the usual set of typos.
275 + Fix : vmg_dispell() didn't check if the ext magic were ours when
276 counting wizards that have uvar callbacks, resulting in a
277 possible memory misread.
278 + Fix : getdata() now returns directly the data object, and no longer a
279 copy. This caused a leak.
280 + Tst : Prefix author tests by 9*-.
281 + Tst : New optional author test : 95-portability-files.t, that uses
282 Test::Portability::Files when it's present.
283 + Tst : New test : 14-self.t, that tests application of magic on the
285 + Tst : Move Hash::Util::FieldHash tests out of 11-multiple.t to
288 0.08 2008-02-01 16:55 UTC
289 + Add : copy magic for tied arrays/hashes.
291 + Add : uvar magics : fetch, store, exists, delete for hashes.
294 Internal development versions.
296 0.06 2007-11-20 10:10 UTC
297 + Chg : 5.7.3 is now officially required.
298 + Fix : "data" test failures on 5.8.{0,2}.
299 + Fix : Drand01() vs rand().
301 0.05 2007-11-19 09:10 UTC
302 + Fix : 5.10.0_RC1 compatibility fix.
304 0.04 2007-08-28 12:25 UTC
305 + Chg : s/require (XSLoader)/use $1/.
306 + Fix : Tests are now strict.
307 + Fix : Complete dependencies.
309 0.03 2007-08-01 17:20 UTC
310 + Add : Passing the signature of an already defined magic to wizard()
311 now returns the corresponding magic object.
312 + Add : You can pass the numeric signature as the wizard argument of
313 cast(), dispell() and getdata().
314 + Add : Any argument specified after the wizard (or the signature) in a
315 call to cast() is now passed to the private data constructor in
317 + Chg : $_[0] is now always a reference to the magic variable in all
318 callbacks. The reason for changing from the previous behaviour
319 is that one may want to apply the same magic to a plain scalar
320 and to a scalar reference, and needs a way to distinguish
321 between them in the callback (say, ref()).
322 + Fix : Wizard object destruction used not to free the signature.
324 0.02 2007-07-27 13:50 UTC
325 + Fix : In response to test report 548152 :
326 Newx() and SvMAGIC_set() not present on older perls.
327 + Fix : In response to test report 548275 :
328 Since perl 5.9.5, 'clear' magic is invoked when an array is
329 undefined (bug #43357). Moreover, 'len' magic is no longer
330 called by pushing an element since perl 5.9.3.
331 + Fix : Missing glob test in MANIFEST.
333 0.01 2007-07-25 16:15 UTC
334 First version, released on an unsuspecting world.