1 Revision history for indirect
3 0.29 2013-03-05 01:30 UTC
4 + Fix : [RT #83659] : false positives
5 Proper method calls in string-like environments (like
6 "@{[ $x->new ]}" will no longer be reported as indirect.
7 This was a regression in 0.28.
8 Thanks Andrew Main for reporting.
9 + Fix : Broken linkage on Windows with gcc 3.4, which appears in
10 particular when using ActivePerl's default compiler suite.
11 For those setups, the indirect shared library will now be
12 linked against the perl dll directly (instead of the import
15 0.28 2013-02-26 17:05 UTC
16 + Fix : [RT #83450] : newlines confuse indirect
17 Perl sometimes resets the line buffer between the object and
18 the method name (e.g. for "sort Class\n->method" outside of
19 eval), and this could cause direct method calls to be reported
21 Thanks Gianni Ceccarelli for reporting.
22 + Fix : Check functions are now replaced and restored in a thread-safe
23 manner, either by using the wrap_op_checker() function from perl
24 when it is available (starting from perl 5.16) or by taking the
25 OP_REFCNT mutex on older perls.
27 0.27 2013-01-30 19:00 UTC
28 + Fix : [RT #82562] : indirect/Devel::CallParser interaction
29 indirect has been taught to play nicely with Devel::CallParser.
30 Thanks Andrew Main for the patch.
31 + Tst : Author tests overhaul.
33 0.26 2011-10-23 14:25 UTC
34 + Add : "no indirect 'global'" enables the pragma for the whole program,
35 except for lexical scopes that "use indirect" explicitely.
36 + Chg : Passing both the 'fatal' and 'hook' options to unimport() is
37 now considered an error, and will result in an exception.
38 unimport() used to consider only the first passed option of
39 those two, and silently ignored the other.
40 + Tst : Test failures of t/41-threads-teardown.t and t/50-external.t on
41 Cygwin should have been addressed.
42 + Tst : Threads tests will not fail anymore if resources constraints
43 prevent the system from creating all the required threads.
45 0.25 2011-08-24 15:40 UTC
46 + Fix : RT #69291 is now also fixed for perl 5.8. The pragma will no
47 longer vivify the "indirect" entry in the hints hash %^H on
49 + Tst : Attempt to make t/50-external.t pass on Cygwin.
51 0.24 2011-07-17 23:15 UTC
52 + Fix : [RT #64521] : "no indirect" leaking into eval.
53 This is currently only fixed for perl 5.10 (perl 5.12 and
54 higher were never affected). It was caused by a very stupid
55 mistake of mine that was introduced in indirect version 0.23.
56 Thanks Michael G Schwern for reporting.
57 + Fix : [RT #69291] : indirect.pm breaks %^H.
58 This was caused by the same mistake as for the previous bug,
59 and as such it is also only fixed for perl 5.10 (and never
61 Thanks Andrew Main for reporting.
62 + Doc : C++ compilers are officially NOT supported.
64 0.23 2010-10-03 00:15 UTC
65 + Fix : Some indirect constructs could be incorrectly reported when
66 several modules were used in the same scope. This caused
67 t/30-scope.t to fail randomly.
68 + Tst : Threads tests are now only run on perl 5.13.4 and higher.
69 They could segfault randomly because of what seems to be an
70 internal bug of Perl, which has been addressed in 5.13.4.
71 There is also an environment variable that allows you to
72 forcefully run those tests, but it should be set only for
73 author testing and not for end users.
75 0.22 2010-08-16 16:00 UTC
76 + Add : Indirect constructs are now reported for code interpolated
77 in quote-like environments, like "${\( ... )}", "@{[ ... ]}",
78 s/pattern/ ... /e, qr/(?{ ... })/ or qr/(??{ ... })/.
79 + Add : You can now make the pragma lethal by passing anything matching
80 /^:?fatal$/i to import(), including "FATAL" and ":Fatal".
81 + Fix : [RT #60378] : segmentation fault on indirect_ck_method.
82 This caused constructs like "@{[ $obj->$meth ]}" to segfault
83 when $meth was a lexical.
84 Thanks Tokuhiro Matsuno for reporting.
86 0.21 2010-05-31 23:10 UTC
87 + Chg : perl 5.8.1 is now required (instead of 5.8.0).
88 + Fix : [RT #57699] : indirect fail with 64-bit int on 5.13.1.
89 It was actually a problem with thread destructors segfaulting
90 because they weren't called at the right time anymore.
91 Thanks Andrew Main for reporting.
92 + Tst : A few more regression tests about the scope leak bug.
94 0.20 2010-04-18 21:25 UTC
95 + Fix : [RT #50570] : "indirect" leaking into LWP.
96 Thanks Andrew Main for reporting.
97 More generally, the require propagation workaround on 5.8-5.10.0
98 has been overhauled, and other scope leaks should be fixed.
99 + Fix : Test failures with 5.12 on Windows where Strawberry Perl crashes
100 because the SystemRoot environment variable is missing.
101 + Fix : Work around Kwalitee test misfailures.
103 0.19 2009-08-28 18:40 UTC
104 + Add : The new constant I_FORKSAFE can be tested to know whether the
105 module will behave nicely when fork()ing. It's currently always
106 true except on Windows where you need perl 5.10.1 for it to be
108 + Fix : I_THREADSAFE and I_FORKSAFE ought to be true when
109 PERL_INDIRECT_PM_DISABLE is set.
110 + Fix : The pragma could still leak if you passed to the "hook" option
111 a reference to the same (named) subroutine from two different
112 require scopes. The fix also provides a better solution for
114 + Fix : Segfaults when indirect is loaded for the first time from
116 + Fix : Leaks of memory associated with the root interpreter.
117 + Opt : Less memory will be used for non-threaded perls version 5.10.0
118 and below, and for threaded perls from version 5.10.1.
120 0.18 2009-08-23 16:15 UTC
121 + Add : When set, the PERL_INDIRECT_PM_DISABLE environment variable
122 disables the pragma globally.
124 0.17 2009-07-16 12:10 UTC
125 + Fix : [RT #47902] : "no indirect" leaking again.
126 This actually turned out to be a bug in how the hook coderefs
127 were stored in the hints hash.
128 Thanks Andrew Main for reporting once again.
129 + Fix : t/80-regressions.t failing on Windows.
130 + Tst : Yet more cleanups.
132 0.16 2009-07-14 16:50 UTC
133 + Add : Indirect calls on blocks are now reported. For those, '{' is
134 passed to the hook as the object description.
135 + Add : The new indirect::msg() function publicizes the default
136 warning/exception message.
137 + Fix : [RT #47866] : Segfault with UTF-8 regexps.
138 Thanks Andrew Main for reporting.
141 0.15 2009-07-08 22:55 UTC
142 + Fix : Invalid constructs with the same method and package name were
144 + Fix : The error line number used to point to the end of the expression
145 instead of its beginning.
147 0.14 2009-06-04 21:55 UTC
148 + Fix : Prevent bogus invalid syntaxes caused by reallocated memory
149 chunks. Thanks Andrew Main for reporting with a reproducible
152 0.13 2009-05-24 18:50 UTC
153 + Add : The hook now receives the file name and the line where the
154 error happened in respectively $_[2] and $_[3].
155 + Fix : Pass mortalized copies of the method name and object to the
156 hook. This seems to fix some rare crashes.
157 + Fix : Work around a bug in perl 5.10.0 and lower. Thanks Andrew Main
158 for teaching me about this issue.
159 + Fix : Report the correct file in error messages (a regression from the
162 0.12 2009-05-03 14:30 UTC
163 + Add : You can specify the action to execute for each indirect
164 construct encountered with the new "hook => $coderef" unimport()
166 + Chg : A ptable is now used internally for the op => position mapping.
167 + Fix : The pragma is now built with thread safety features enabled
168 whenever possible (a notable exception is perl 5.8.x on Win32,
169 as something seems wrong with its context handling).
170 The new indirect::I_THREADSAFE() constant reflects this.
171 + Fix : A negation precedence nit in indirect_ck_entersub().
172 + Tst : "use/no indirect" while parsing an indirect construct.
173 + Tst : Thread safety.
175 0.11 2009-02-08 18:35 UTC
176 + Fix : Potential collisions by hashing pointers with a wrong format.
177 + Upd : Resources in META.yml.
179 0.10 2009-01-17 12:40 UTC
180 Re-release 0.09_01 as stable.
182 0.09_01 2008-12-08 17:55 UTC
183 + Fix : Invalid memory read with "${\(new Foo)}" constructs. The
184 corresponding test is turned back on.
185 + Tst : Refinements in t/30-scope.t
187 0.09 2008-12-05 20:35 UTC
188 + Add : Support for perl 5.8.
189 + Tst : Skip a test in t/10-good.t that randomly segfaults for (I guess)
190 systems stricter than linux in the way they manage their memory.
192 0.08 2008-10-22 14:45 UTC
193 + Fix : A rare edge case for package whose names are prefix of 'main'.
194 + Tst : Test $$ as variable and state variables.
196 0.07_03 2008-10-17 20:10 UTC
197 + Add : Support and tests for variables with spaces after the sigil.
198 + Upd : META.yml spec updated to 1.4.
200 0.07_02 2008-10-15 21:10 UTC
201 + Add : Support and tests for package variables.
202 + Tst : Coverage improved by removing dead code.
204 0.07_01 2008-10-15 16:00 UTC
205 + Fix : [RT #40055] : Not handling RV2SV => GV(SV) correctly, which
206 could cause 'no indirect; print' segfaults. Thanks Goro Fuji for
209 0.06 2008-10-11 16:45 UTC
211 + Tst : Test "no indirect 'anything'", "foo Class->bar", and indirect
212 uses of exec() and system().
214 0.05 2008-10-02 14:40 UTC
215 + Chg : Now the old check function is always called before storing an op
217 + Fix : Misc code and docs refinements.
219 0.04 2008-08-30 19:00 UTC
220 + Fix : Clean up the op->src hash when we're done with an entersub.
221 + Tst : No longer fork for testing. IPC::Cmd isn't required anymore.
223 0.03 2008-08-12 15:25 UTC
224 This release is kindly supported by Copenhagen Hotel Centrum WiFi.
225 + Fix : Tests used not to pass PERL5OPTS to their kids. This lead to
226 failures under CPAN. I think.
229 0.02 2008-08-11 15:55 UTC
230 + Fix : Some building failures with old gcc versions that didn't seem to
231 like the ((hint == 2) ? croak : warn)(msg) construct. I think.
232 + Rem : Unused cruft from a previous implementation.
233 + Tst : Fail more gracefully when we can't capture buffers or when the
234 child returned an error.
236 0.01 2008-08-10 20:40 UTC
237 First version, released on an unsuspecting world.