1 Revision history for Variable-Magic
3 0.29 2009-02-08 11:10 UTC
4 + Add : Set $_[2] to the default length for len magic on scalars.
5 + Chg : Perl 5.8 is required.
6 + Chg : The CLONE method will no longer be defined for non-threaded
9 + Fix : The logic for turning off the GMAGICAL flag on hashes when no
10 uvar magic is involved could do nasty things on tied hashes.
11 + Upd : More resources in META.yml.
13 0.28 2009-01-24 17:05 UTC
14 + Add : The 'copy_key' option to wizard(), that enables the safe edition
15 of $_[2] in uvar callbacks to redirect actions to another key.
16 + Fix : Exception objects thrown from callbacks at compile time were not
18 + Fix : free callbacks triggered during global destruction could lead to
19 segfaults because the wizard was already freed.
20 + Fix : Refcount of scalars with free callback.
21 + Fix : Segfaults when using get or uvar magic simultaneously with clear
23 + Doc : The documentation about what the callbacks take and should
24 return is now somewhat clearer.
26 0.27 2009-01-18 23:40 UTC
27 + Fix : Offset of 1 with len magic on scalars.
28 + Fix : Segfaults that seem to happen when one croak in a callback at
29 compile time (thanks Ash Berlin for the testcase).
30 + Upd : Resources in META.yml (Florian Ragwitz)
31 + Tst : Large cleanup resulting in a wider set of tested behaviours.
33 0.26 2008-11-29 22:05 UTC
34 + Fix : Leaks of cloned coderefs that access lexicals.
35 + Fix : Building with maint-5.8 and blead (thanks Andreas Koenig).
37 0.25 2008-11-12 19:40 UTC
38 + Fix : Compatibility with 5.8.9.
39 + Fix : Old Pod::Coverage don't ignore CLONE.
40 + Upd : META.yml spec updated to 1.4.
42 0.24 2008-10-12 14:55 UTC
43 + Fix : Really fix it.
45 0.23 2008-10-11 17:25 UTC
46 + Fix : Building on perls with multiplicity but not ithreads.
48 0.22 2008-09-29 19:40 UTC
49 + Doc : Explain how magic differs from overloading.
50 + Fix : Disable thread safety for perl 5.8 on Windows. If you want it
51 enabled, reliable feedback is welcome.
53 0.21_02 2008-09-28 23:10 UTC
54 + Fix : Hide previously deleted magic tokens in vmg_svt_free (Yuval
57 0.21_01 2008-09-22 13:25 UTC
58 + Chg : Shrink the context data.
59 + Fix : Enable thread safety features for perls that just lack
61 + Tst : Output threads and threads::shared versions in threads tests.
63 0.20 2008-09-20 23:35 UTC
64 + Fix : The module is now thread safe. You can test if it had been
65 built with thread safety features enabled through the
66 VMG_THREADSAFE constant.
67 + Fix : Correct backtrace when wizard() croaks.
68 + Fix : 'cast $a, undef' should throw an 'invalid wizard' error instead
69 of an 'invalid signature'.
71 0.19 2008-07-05 15:05 UTC
72 + Add : The patchlevel with which the module has been built is now
73 exported with the VMG_PERL_PATCHLEVEL constant.
74 + Chg : Version checking macros overhaul.
75 + Fix : 'uvar' magic should only be enabled in maint-5.10 and higher
76 since it's never going to be backported to 5.8.
77 + Fix : Likewise, the copy callback API change should only be enabled in
79 + Fix : Instead of those ackward fake refcount increments in the wizard
80 destructor, flag the wizard with SVf_BREAK and force temps
81 cleanup. This should fix the wizard destruction for all perls,
82 and solves the leak introduced by the previous solution.
84 0.18 2008-07-04 12:20 UTC
85 + Fix : Enable uvar magic when patchlevel >= 28419.
86 + Fix : Set VMG_COMPAT_ARRAY_PUSH_NOLEN for Windows ActiveState Perl
89 0.17 2008-05-11 09:05 UTC
90 + Fix : 5.8.4 and lower don't have mPUSHi. Gah. Yeah, I know, I should
93 0.16 2008-05-10 22:05 UTC
94 + Add : The samples/copy.pl script.
95 + Chg : The sv_magical() + vmg_mg_magical() combo was simplified into
97 + Tst : t/33-code.t was testing scalars, not code.
98 + Tst : is() and like() are better than ok().
99 + Tst : 100% coverage reached.
101 0.15 2008-04-11 18:25 UTC
102 + Chg : Factor vmg_cb_call{1,2,3}() into one function.
103 + Fix : len magic is no longer called when taking the length() of a
104 magical scalar since p32969. The VMG_COMPAT_SCALAR_LENGTH_NOLEN
105 constant was added to cover this.
106 + Tst : More tests for t/22-len.t.
108 0.14 2008-03-24 12:35 UTC
109 + Fix : t/16-huf.t failures on Solaris and FreeBSD caused by not
110 updating mg->mg_ptr after Renew-ing it on dispell.
111 + Fix : Undefining MGf_DUP caused the wizard's magic to be wrongly
112 initialized. MGf_DUP now appears as true from the user point of
113 view, but the dup callback isn't actually set.
114 + Fix : Warnings with blead due to copy callbacks now taking an I32 (API
116 + Fix : vmg_svt_val() and vmg_uvar_del() aren't required if no uvar
118 + Tst : Tests now output the version of optional modules loaded.
120 0.13 2008-03-19 14:35 UTC
121 + Doc : Link to coverage report.
122 + Fix : Correct dependencies listing in META.yml.
123 + Tst : Improved test coverage.
124 + Tst : Print the patchlevel as a comment.
125 + Tst : Use is() where it's relevant.
126 + Tst : t/16-huf.t now really tests interaction with H::U::FH.
128 0.12 2008-02-07 18:15 UTC
129 + Fix : POD error. Thanks to Chris Williams (BinGOs) for the quick
132 0.11 2008-02-07 17:55 UTC
133 + Add : Copy callbacks now receive the current key/index in $_[2].
134 The current value/element is now in $_[3].
135 + Chg : The trigger of copy magic on list assignment with perls greater
136 than 5.10.0 was caused by uvar magic. Hence,
137 VMG_COMPAT_HASH_LISTASSIGN_COPY was removed. Use VMG_UVAR
139 + Fix : Build failures on patched perls.
140 + Tst : Added missing exported symbols to 01-import.t.
142 0.10 2008-02-04 11:30 UTC
143 + Add : New script : samples/vm_vs_tie.pl, that benchmarks our uvar
144 magic versus tied hashes.
145 + Add : The VMG_COMPAT_* constants can be used from userspace to check
146 perl magic abilities.
147 + Fix : Callbacks that returned undef made us croak, breaking the
148 variable behaviour (regression test in 14-callbacks.t).
149 + Fix : uvar callbacks weren't tested for non-NULL-ity before being
150 called (regression test in 28-uvar.t).
151 + Tst : Fix typo in 25-copy.t that prevented Tie::Hash tests to be ran.
153 0.09 2008-02-02 11:30 UTC
154 + Doc : Explicitely say that uvar callbacks are safely ignored for
156 + Doc : Document caveats and fix the usual set of typos.
157 + Fix : vmg_dispell() didn't check if the ext magic were ours when
158 counting wizards that have uvar callbacks, resulting in a
159 possible memory misread.
160 + Fix : getdata() now returns directly the data object, and no longer a
161 copy. This caused a leak.
162 + Tst : Prefix author tests by 9*-.
163 + Tst : New optional author test : 95-portability-files.t, that uses
164 Test::Portability::Files when it's present.
165 + Tst : New test : 14-self.t, that tests application of magic on the
167 + Tst : Move Hash::Util::FieldHash tests out of 11-multiple.t to
170 0.08 2008-02-01 16:55 UTC
171 + Add : copy magic for tied arrays/hashes.
173 + Add : uvar magics : fetch, store, exists, delete for hashes.
176 Internal development versions.
178 0.06 2007-11-20 10:10 UTC
179 + Chg : 5.7.3 is now officially required.
180 + Fix : "data" test failures on 5.8.{0,2}.
181 + Fix : Drand01() vs rand().
183 0.05 2007-11-19 09:10 UTC
184 + Fix : 5.10.0_RC1 compatibility fix.
186 0.04 2007-08-28 12:25 UTC
187 + Chg : s/require (XSLoader)/use $1/.
188 + Fix : Tests are now strict.
189 + Fix : Complete dependencies.
191 0.03 2007-08-01 17:20 UTC
192 + Add : Passing the signature of an already defined magic to wizard()
193 now returns the corresponding magic object.
194 + Add : You can pass the numeric signature as the wizard argument of
195 cast(), dispell() and getdata().
196 + Add : Any argument specified after the wizard (or the signature) in a
197 call to cast() is now passed to the private data constructor in
199 + Chg : $_[0] is now always a reference to the magic variable in all
200 callbacks. The reason for changing from the previous behaviour
201 is that one may want to apply the same magic to a plain scalar
202 and to a scalar reference, and needs a way to distinguish
203 between them in the callback (say, ref()).
204 + Fix : Wizard object destruction used not to free the signature.
206 0.02 2007-07-27 13:50 UTC
207 + Fix : In response to test report 548152 :
208 Newx() and SvMAGIC_set() not present on older perls.
209 + Fix : In response to test report 548275 :
210 Since perl 5.9.5, 'clear' magic is invoked when an array is
211 undefined (bug #43357). Moreover, 'len' magic is no longer
212 called by pushing an element since perl 5.9.3.
213 + Fix : Missing glob test in MANIFEST.
215 0.01 2007-07-25 16:15 UTC
216 First version, released on an unsuspecting world.