1 Revision history for Variable-Magic
3 0.35 2009-05-15 20:50 UTC
4 This is a maintenance release. Users of 0.34 can skip this update.
5 + Chg : uvar magic used to be enabled for 5.9.5. It now requires 5.10 or
7 + Fix : Silence compilation warnings for cxinc().
8 + Rem : The workaround introduced in 0.31 for the "Unknown errors" issue
9 was removed, as regression tests show that 0.34 provides a
11 + Tst : Skip t/40-threads.t on 5.8.x with old versions of threads and
14 0.34 2009-04-19 16:55 UTC
15 + Fix : Destruction of wizards in require.
16 + Fix : "panic: restartop" in 5.10 when dieing at compile time triggers
19 0.33 2009-03-26 00:00 UTC
20 + Chg : cast() and getsig() now croak too when an invalid signature is
22 + Fix : It was possible to generate with gensig() the same signature
23 twice before actually defining the wizards.
24 + Tst : More stash tests.
25 + Tst : Test when the magic takes place in some cases.
26 + Tst : Improved coverage for the op object feature.
28 0.32 2009-03-01 13:45 UTC
29 + Chg : dispell() and getdata() now croak when an invalid signature is
31 + Doc : More examples and a brand new synopsis.
32 + Fix : The signature is no longer stored in the mg_private member of
33 the MAGIC struct. This fixes possible clashes with magics from
35 + Fix : op info objects weren't blessed into the right class.
36 + Fix : Races when initializing the op names cache under threads.
38 0.31 2009-02-19 23:50 UTC
39 + Doc : A somewhat better introduction. The FUNCTION section was moved
40 before CONSTANTS and HISTORY since it's more important. And
41 'assignation' is really 'assignment', says Ricardo. :)
42 + Fix : The op_info features are now thread safe.
43 + Fix : Segfaults that occured when trying to get the op name during
45 + Fix : Segfaults and wrong "Unknown error" exceptions that happened
46 when dieing in require caused a free callback to fire.
47 Thanks Florian Ragwitz for reporting.
49 0.30 2009-02-12 17:05 UTC
50 + Add : You can now get in all the callbacks either the name or a B::OP
51 object representing the current op if you pass the new 'op_info'
52 option to wizard() with respective values VMG_OP_INFO_NAME and
54 + Doc : Document magic on array/hashes values.
55 + Fix : Completely skip free callbacks during global destruction.
56 + Tst : Tests now pass again on 5.8.3 and lower (they were broken since
58 + Tst : Test magic on symbol table and on array/hash values.
60 0.29 2009-02-08 11:10 UTC
61 + Add : Set $_[2] to the default length for len magic on scalars.
62 + Chg : Perl 5.8 is required.
63 + Chg : The CLONE method will no longer be defined for non-threaded
66 + Fix : The logic for turning off the GMAGICAL flag on hashes when no
67 uvar magic is involved could do nasty things on tied hashes.
68 + Upd : More resources in META.yml.
70 0.28 2009-01-24 17:05 UTC
71 + Add : The 'copy_key' option to wizard(), that enables the safe edition
72 of $_[2] in uvar callbacks to redirect actions to another key.
73 + Fix : Exception objects thrown from callbacks at compile time were not
75 + Fix : free callbacks triggered during global destruction could lead to
76 segfaults because the wizard was already freed.
77 + Fix : Refcount of scalars with free callback.
78 + Fix : Segfaults when using get or uvar magic simultaneously with clear
80 + Doc : The documentation about what the callbacks take and should
81 return is now somewhat clearer.
83 0.27 2009-01-18 23:40 UTC
84 + Fix : Offset of 1 with len magic on scalars.
85 + Fix : Segfaults that seem to happen when one croak in a callback at
86 compile time (thanks Ash Berlin for the testcase).
87 + Upd : Resources in META.yml (Florian Ragwitz)
88 + Tst : Large cleanup resulting in a wider set of tested behaviours.
90 0.26 2008-11-29 22:05 UTC
91 + Fix : Leaks of cloned coderefs that access lexicals.
92 + Fix : Building with maint-5.8 and blead (thanks Andreas Koenig).
94 0.25 2008-11-12 19:40 UTC
95 + Fix : Compatibility with 5.8.9.
96 + Fix : Old Pod::Coverage don't ignore CLONE.
97 + Upd : META.yml spec updated to 1.4.
99 0.24 2008-10-12 14:55 UTC
100 + Fix : Really fix it.
102 0.23 2008-10-11 17:25 UTC
103 + Fix : Building on perls with multiplicity but not ithreads.
105 0.22 2008-09-29 19:40 UTC
106 + Doc : Explain how magic differs from overloading.
107 + Fix : Disable thread safety for perl 5.8 on Windows. If you want it
108 enabled, reliable feedback is welcome.
110 0.21_02 2008-09-28 23:10 UTC
111 + Fix : Hide previously deleted magic tokens in vmg_svt_free (Yuval
114 0.21_01 2008-09-22 13:25 UTC
115 + Chg : Shrink the context data.
116 + Fix : Enable thread safety features for perls that just lack
118 + Tst : Output threads and threads::shared versions in threads tests.
120 0.20 2008-09-20 23:35 UTC
121 + Fix : The module is now thread safe. You can test if it had been
122 built with thread safety features enabled through the
123 VMG_THREADSAFE constant.
124 + Fix : Correct backtrace when wizard() croaks.
125 + Fix : 'cast $a, undef' should throw an 'invalid wizard' error instead
126 of an 'invalid signature'.
128 0.19 2008-07-05 15:05 UTC
129 + Add : The patchlevel with which the module has been built is now
130 exported with the VMG_PERL_PATCHLEVEL constant.
131 + Chg : Version checking macros overhaul.
132 + Fix : 'uvar' magic should only be enabled in maint-5.10 and higher
133 since it's never going to be backported to 5.8.
134 + Fix : Likewise, the copy callback API change should only be enabled in
136 + Fix : Instead of those ackward fake refcount increments in the wizard
137 destructor, flag the wizard with SVf_BREAK and force temps
138 cleanup. This should fix the wizard destruction for all perls,
139 and solves the leak introduced by the previous solution.
141 0.18 2008-07-04 12:20 UTC
142 + Fix : Enable uvar magic when patchlevel >= 28419.
143 + Fix : Set VMG_COMPAT_ARRAY_PUSH_NOLEN for Windows ActiveState Perl
146 0.17 2008-05-11 09:05 UTC
147 + Fix : 5.8.4 and lower don't have mPUSHi. Gah. Yeah, I know, I should
150 0.16 2008-05-10 22:05 UTC
151 + Add : The samples/copy.pl script.
152 + Chg : The sv_magical() + vmg_mg_magical() combo was simplified into
154 + Tst : t/33-code.t was testing scalars, not code.
155 + Tst : is() and like() are better than ok().
156 + Tst : 100% coverage reached.
158 0.15 2008-04-11 18:25 UTC
159 + Chg : Factor vmg_cb_call{1,2,3}() into one function.
160 + Fix : len magic is no longer called when taking the length() of a
161 magical scalar since p32969. The VMG_COMPAT_SCALAR_LENGTH_NOLEN
162 constant was added to cover this.
163 + Tst : More tests for t/22-len.t.
165 0.14 2008-03-24 12:35 UTC
166 + Fix : t/16-huf.t failures on Solaris and FreeBSD caused by not
167 updating mg->mg_ptr after Renew-ing it on dispell.
168 + Fix : Undefining MGf_DUP caused the wizard's magic to be wrongly
169 initialized. MGf_DUP now appears as true from the user point of
170 view, but the dup callback isn't actually set.
171 + Fix : Warnings with blead due to copy callbacks now taking an I32 (API
173 + Fix : vmg_svt_val() and vmg_uvar_del() aren't required if no uvar
175 + Tst : Tests now output the version of optional modules loaded.
177 0.13 2008-03-19 14:35 UTC
178 + Doc : Link to coverage report.
179 + Fix : Correct dependencies listing in META.yml.
180 + Tst : Improved test coverage.
181 + Tst : Print the patchlevel as a comment.
182 + Tst : Use is() where it's relevant.
183 + Tst : t/16-huf.t now really tests interaction with H::U::FH.
185 0.12 2008-02-07 18:15 UTC
186 + Fix : POD error. Thanks to Chris Williams (BinGOs) for the quick
189 0.11 2008-02-07 17:55 UTC
190 + Add : Copy callbacks now receive the current key/index in $_[2].
191 The current value/element is now in $_[3].
192 + Chg : The trigger of copy magic on list assignment with perls greater
193 than 5.10.0 was caused by uvar magic. Hence,
194 VMG_COMPAT_HASH_LISTASSIGN_COPY was removed. Use VMG_UVAR
196 + Fix : Build failures on patched perls.
197 + Tst : Added missing exported symbols to 01-import.t.
199 0.10 2008-02-04 11:30 UTC
200 + Add : New script : samples/vm_vs_tie.pl, that benchmarks our uvar
201 magic versus tied hashes.
202 + Add : The VMG_COMPAT_* constants can be used from userspace to check
203 perl magic abilities.
204 + Fix : Callbacks that returned undef made us croak, breaking the
205 variable behaviour (regression test in 14-callbacks.t).
206 + Fix : uvar callbacks weren't tested for non-NULL-ity before being
207 called (regression test in 28-uvar.t).
208 + Tst : Fix typo in 25-copy.t that prevented Tie::Hash tests to be ran.
210 0.09 2008-02-02 11:30 UTC
211 + Doc : Explicitely say that uvar callbacks are safely ignored for
213 + Doc : Document caveats and fix the usual set of typos.
214 + Fix : vmg_dispell() didn't check if the ext magic were ours when
215 counting wizards that have uvar callbacks, resulting in a
216 possible memory misread.
217 + Fix : getdata() now returns directly the data object, and no longer a
218 copy. This caused a leak.
219 + Tst : Prefix author tests by 9*-.
220 + Tst : New optional author test : 95-portability-files.t, that uses
221 Test::Portability::Files when it's present.
222 + Tst : New test : 14-self.t, that tests application of magic on the
224 + Tst : Move Hash::Util::FieldHash tests out of 11-multiple.t to
227 0.08 2008-02-01 16:55 UTC
228 + Add : copy magic for tied arrays/hashes.
230 + Add : uvar magics : fetch, store, exists, delete for hashes.
233 Internal development versions.
235 0.06 2007-11-20 10:10 UTC
236 + Chg : 5.7.3 is now officially required.
237 + Fix : "data" test failures on 5.8.{0,2}.
238 + Fix : Drand01() vs rand().
240 0.05 2007-11-19 09:10 UTC
241 + Fix : 5.10.0_RC1 compatibility fix.
243 0.04 2007-08-28 12:25 UTC
244 + Chg : s/require (XSLoader)/use $1/.
245 + Fix : Tests are now strict.
246 + Fix : Complete dependencies.
248 0.03 2007-08-01 17:20 UTC
249 + Add : Passing the signature of an already defined magic to wizard()
250 now returns the corresponding magic object.
251 + Add : You can pass the numeric signature as the wizard argument of
252 cast(), dispell() and getdata().
253 + Add : Any argument specified after the wizard (or the signature) in a
254 call to cast() is now passed to the private data constructor in
256 + Chg : $_[0] is now always a reference to the magic variable in all
257 callbacks. The reason for changing from the previous behaviour
258 is that one may want to apply the same magic to a plain scalar
259 and to a scalar reference, and needs a way to distinguish
260 between them in the callback (say, ref()).
261 + Fix : Wizard object destruction used not to free the signature.
263 0.02 2007-07-27 13:50 UTC
264 + Fix : In response to test report 548152 :
265 Newx() and SvMAGIC_set() not present on older perls.
266 + Fix : In response to test report 548275 :
267 Since perl 5.9.5, 'clear' magic is invoked when an array is
268 undefined (bug #43357). Moreover, 'len' magic is no longer
269 called by pushing an element since perl 5.9.3.
270 + Fix : Missing glob test in MANIFEST.
272 0.01 2007-07-25 16:15 UTC
273 First version, released on an unsuspecting world.