1 Revision history for indirect
3 0.34 2015-04-02 19:50 UTC
4 + Chg : The new environment variable to enable thread tests on older
5 perls is PERL_FORCE_TEST_THREADS. Note that this variable
6 should only be turned on by authors.
7 + Fix : [RT #100068] : add link to historical tchrist post
8 The link has been added to the documentation. Thanks Olivier
10 + Fix : Segfaults when the module is loaded by several threads (or
11 Windows emulated processes) ran in parallel.
12 + Fix : Update the Windows ActivePerl + gcc 3.4 workaround for
13 ExtUtils::MakeMaker 7.04. Thanks Christian Walde for reporting
14 and feedback on this issue.
15 + Fix : Be really compatible with the optional OP_PARENT feature.
16 + Tst : $ENV{$Config{ldlibpthname}} is now preserved on all platforms,
17 which will address failures of t/41-threads-teardown.t and
18 t/50-external.t with unusual compilers (like icc) that link all
19 their compiled objects to their own libraries.
21 0.33 2014-09-29 20:20 UTC
22 + Fix : [RT #99083] : Breaks eval in an END block in Win32 pseudo-forks.
23 Thanks Graham Knop for reporting.
24 + Fix : Segfaults during global destruction of a thread or a
27 0.32 2014-09-21 20:15 UTC
28 + Add : Support for the PERL_OP_PARENT optional feature introduced in
30 + Fix : [RT #92806] : Tests that use run_perl() fail on Android
31 Thanks Brian Fraser for the patch.
32 + Fix : indirect constructs will no longer segfault while inside
33 the empty package on perl 5.8.x. This fix may also prevent
34 some segfaults during global destruction.
36 0.31 2013-09-05 16:45 UTC
37 + Fix : [RT #88428] : no indirect in eval can trigger for direct calls
39 Thanks Graham Knop for reporting.
40 + Tst : Author tests are no longer bundled with this distribution.
41 They are only made available to authors in the git repository.
43 0.30 2013-05-16 15:55 UTC
44 + Fix : [RT #83806] : false positives with Devel::Declare
45 [RT #83839] : false positive using ? : syntax
46 Thanks Andrew Main for the patch.
47 However, please note that the reason this patch seems to fix
48 thinks has not been explained.
49 + Fix : [RT #84649] : incorrect RT link in metadata
50 Thanks Karen Etheridge for reporting.
52 0.29 2013-03-05 01:30 UTC
53 + Fix : [RT #83659] : false positives
54 Proper method calls in string-like environments (like
55 "@{[ $x->new ]}" will no longer be reported as indirect.
56 This was a regression in 0.28.
57 Thanks Andrew Main for reporting.
58 + Fix : Broken linkage on Windows with gcc 3.4, which appears in
59 particular when using ActivePerl's default compiler suite.
60 For those setups, the indirect shared library will now be
61 linked against the perl dll directly (instead of the import
64 0.28 2013-02-26 17:05 UTC
65 + Fix : [RT #83450] : newlines confuse indirect
66 Perl sometimes resets the line buffer between the object and
67 the method name (e.g. for "sort Class\n->method" outside of
68 eval), and this could cause direct method calls to be reported
70 Thanks Gianni Ceccarelli for reporting.
71 + Fix : Check functions are now replaced and restored in a thread-safe
72 manner, either by using the wrap_op_checker() function from perl
73 when it is available (starting from perl 5.16) or by taking the
74 OP_REFCNT mutex on older perls.
76 0.27 2013-01-30 19:00 UTC
77 + Fix : [RT #82562] : indirect/Devel::CallParser interaction
78 indirect has been taught to play nicely with Devel::CallParser.
79 Thanks Andrew Main for the patch.
80 + Tst : Author tests overhaul.
82 0.26 2011-10-23 14:25 UTC
83 + Add : "no indirect 'global'" enables the pragma for the whole program,
84 except for lexical scopes that "use indirect" explicitely.
85 + Chg : Passing both the 'fatal' and 'hook' options to unimport() is
86 now considered an error, and will result in an exception.
87 unimport() used to consider only the first passed option of
88 those two, and silently ignored the other.
89 + Tst : Test failures of t/41-threads-teardown.t and t/50-external.t on
90 Cygwin should have been addressed.
91 + Tst : Threads tests will not fail anymore if resources constraints
92 prevent the system from creating all the required threads.
94 0.25 2011-08-24 15:40 UTC
95 + Fix : RT #69291 is now also fixed for perl 5.8. The pragma will no
96 longer vivify the "indirect" entry in the hints hash %^H on
98 + Tst : Attempt to make t/50-external.t pass on Cygwin.
100 0.24 2011-07-17 23:15 UTC
101 + Fix : [RT #64521] : "no indirect" leaking into eval.
102 This is currently only fixed for perl 5.10 (perl 5.12 and
103 higher were never affected). It was caused by a very stupid
104 mistake of mine that was introduced in indirect version 0.23.
105 Thanks Michael G Schwern for reporting.
106 + Fix : [RT #69291] : indirect.pm breaks %^H.
107 This was caused by the same mistake as for the previous bug,
108 and as such it is also only fixed for perl 5.10 (and never
110 Thanks Andrew Main for reporting.
111 + Doc : C++ compilers are officially NOT supported.
113 0.23 2010-10-03 00:15 UTC
114 + Fix : Some indirect constructs could be incorrectly reported when
115 several modules were used in the same scope. This caused
116 t/30-scope.t to fail randomly.
117 + Tst : Threads tests are now only run on perl 5.13.4 and higher.
118 They could segfault randomly because of what seems to be an
119 internal bug of Perl, which has been addressed in 5.13.4.
120 There is also an environment variable that allows you to
121 forcefully run those tests, but it should be set only for
122 author testing and not for end users.
124 0.22 2010-08-16 16:00 UTC
125 + Add : Indirect constructs are now reported for code interpolated
126 in quote-like environments, like "${\( ... )}", "@{[ ... ]}",
127 s/pattern/ ... /e, qr/(?{ ... })/ or qr/(??{ ... })/.
128 + Add : You can now make the pragma lethal by passing anything matching
129 /^:?fatal$/i to import(), including "FATAL" and ":Fatal".
130 + Fix : [RT #60378] : segmentation fault on indirect_ck_method.
131 This caused constructs like "@{[ $obj->$meth ]}" to segfault
132 when $meth was a lexical.
133 Thanks Tokuhiro Matsuno for reporting.
135 0.21 2010-05-31 23:10 UTC
136 + Chg : perl 5.8.1 is now required (instead of 5.8.0).
137 + Fix : [RT #57699] : indirect fail with 64-bit int on 5.13.1.
138 It was actually a problem with thread destructors segfaulting
139 because they weren't called at the right time anymore.
140 Thanks Andrew Main for reporting.
141 + Tst : A few more regression tests about the scope leak bug.
143 0.20 2010-04-18 21:25 UTC
144 + Fix : [RT #50570] : "indirect" leaking into LWP.
145 Thanks Andrew Main for reporting.
146 More generally, the require propagation workaround on 5.8-5.10.0
147 has been overhauled, and other scope leaks should be fixed.
148 + Fix : Test failures with 5.12 on Windows where Strawberry Perl crashes
149 because the SystemRoot environment variable is missing.
150 + Fix : Work around Kwalitee test misfailures.
152 0.19 2009-08-28 18:40 UTC
153 + Add : The new constant I_FORKSAFE can be tested to know whether the
154 module will behave nicely when fork()ing. It's currently always
155 true except on Windows where you need perl 5.10.1 for it to be
157 + Fix : I_THREADSAFE and I_FORKSAFE ought to be true when
158 PERL_INDIRECT_PM_DISABLE is set.
159 + Fix : The pragma could still leak if you passed to the "hook" option
160 a reference to the same (named) subroutine from two different
161 require scopes. The fix also provides a better solution for
163 + Fix : Segfaults when indirect is loaded for the first time from
165 + Fix : Leaks of memory associated with the root interpreter.
166 + Opt : Less memory will be used for non-threaded perls version 5.10.0
167 and below, and for threaded perls from version 5.10.1.
169 0.18 2009-08-23 16:15 UTC
170 + Add : When set, the PERL_INDIRECT_PM_DISABLE environment variable
171 disables the pragma globally.
173 0.17 2009-07-16 12:10 UTC
174 + Fix : [RT #47902] : "no indirect" leaking again.
175 This actually turned out to be a bug in how the hook coderefs
176 were stored in the hints hash.
177 Thanks Andrew Main for reporting once again.
178 + Fix : t/80-regressions.t failing on Windows.
179 + Tst : Yet more cleanups.
181 0.16 2009-07-14 16:50 UTC
182 + Add : Indirect calls on blocks are now reported. For those, '{' is
183 passed to the hook as the object description.
184 + Add : The new indirect::msg() function publicizes the default
185 warning/exception message.
186 + Fix : [RT #47866] : Segfault with UTF-8 regexps.
187 Thanks Andrew Main for reporting.
190 0.15 2009-07-08 22:55 UTC
191 + Fix : Invalid constructs with the same method and package name were
193 + Fix : The error line number used to point to the end of the expression
194 instead of its beginning.
196 0.14 2009-06-04 21:55 UTC
197 + Fix : Prevent bogus invalid syntaxes caused by reallocated memory
198 chunks. Thanks Andrew Main for reporting with a reproducible
201 0.13 2009-05-24 18:50 UTC
202 + Add : The hook now receives the file name and the line where the
203 error happened in respectively $_[2] and $_[3].
204 + Fix : Pass mortalized copies of the method name and object to the
205 hook. This seems to fix some rare crashes.
206 + Fix : Work around a bug in perl 5.10.0 and lower. Thanks Andrew Main
207 for teaching me about this issue.
208 + Fix : Report the correct file in error messages (a regression from the
211 0.12 2009-05-03 14:30 UTC
212 + Add : You can specify the action to execute for each indirect
213 construct encountered with the new "hook => $coderef" unimport()
215 + Chg : A ptable is now used internally for the op => position mapping.
216 + Fix : The pragma is now built with thread safety features enabled
217 whenever possible (a notable exception is perl 5.8.x on Win32,
218 as something seems wrong with its context handling).
219 The new indirect::I_THREADSAFE() constant reflects this.
220 + Fix : A negation precedence nit in indirect_ck_entersub().
221 + Tst : "use/no indirect" while parsing an indirect construct.
222 + Tst : Thread safety.
224 0.11 2009-02-08 18:35 UTC
225 + Fix : Potential collisions by hashing pointers with a wrong format.
226 + Upd : Resources in META.yml.
228 0.10 2009-01-17 12:40 UTC
229 Re-release 0.09_01 as stable.
231 0.09_01 2008-12-08 17:55 UTC
232 + Fix : Invalid memory read with "${\(new Foo)}" constructs. The
233 corresponding test is turned back on.
234 + Tst : Refinements in t/30-scope.t
236 0.09 2008-12-05 20:35 UTC
237 + Add : Support for perl 5.8.
238 + Tst : Skip a test in t/10-good.t that randomly segfaults for (I guess)
239 systems stricter than linux in the way they manage their memory.
241 0.08 2008-10-22 14:45 UTC
242 + Fix : A rare edge case for package whose names are prefix of 'main'.
243 + Tst : Test $$ as variable and state variables.
245 0.07_03 2008-10-17 20:10 UTC
246 + Add : Support and tests for variables with spaces after the sigil.
247 + Upd : META.yml spec updated to 1.4.
249 0.07_02 2008-10-15 21:10 UTC
250 + Add : Support and tests for package variables.
251 + Tst : Coverage improved by removing dead code.
253 0.07_01 2008-10-15 16:00 UTC
254 + Fix : [RT #40055] : Not handling RV2SV => GV(SV) correctly, which
255 could cause 'no indirect; print' segfaults. Thanks Goro Fuji for
258 0.06 2008-10-11 16:45 UTC
260 + Tst : Test "no indirect 'anything'", "foo Class->bar", and indirect
261 uses of exec() and system().
263 0.05 2008-10-02 14:40 UTC
264 + Chg : Now the old check function is always called before storing an op
266 + Fix : Misc code and docs refinements.
268 0.04 2008-08-30 19:00 UTC
269 + Fix : Clean up the op->src hash when we're done with an entersub.
270 + Tst : No longer fork for testing. IPC::Cmd isn't required anymore.
272 0.03 2008-08-12 15:25 UTC
273 This release is kindly supported by Copenhagen Hotel Centrum WiFi.
274 + Fix : Tests used not to pass PERL5OPTS to their kids. This lead to
275 failures under CPAN. I think.
278 0.02 2008-08-11 15:55 UTC
279 + Fix : Some building failures with old gcc versions that didn't seem to
280 like the ((hint == 2) ? croak : warn)(msg) construct. I think.
281 + Rem : Unused cruft from a previous implementation.
282 + Tst : Fail more gracefully when we can't capture buffers or when the
283 child returned an error.
285 0.01 2008-08-10 20:40 UTC
286 First version, released on an unsuspecting world.