1 Revision history for indirect
3 0.26 2011-10-23 14:25 UTC
4 + Add : "no indirect 'global'" enables the pragma for the whole program,
5 except for lexical scopes that "use indirect" explicitely.
6 + Chg : Passing both the 'fatal' and 'hook' options to unimport() is
7 now considered an error, and will result in an exception.
8 unimport() used to consider only the first passed option of
9 those two, and silently ignored the other.
10 + Tst : Test failures of t/41-threads-teardown.t and t/50-external.t on
11 Cygwin should have been addressed.
12 + Tst : Threads tests will not fail anymore if resources constraints
13 prevent the system from creating all the required threads.
15 0.25 2011-08-24 15:40 UTC
16 + Fix : RT #69291 is now also fixed for perl 5.8. The pragma will no
17 longer vivify the "indirect" entry in the hints hash %^H on
19 + Tst : Attempt to make t/50-external.t pass on Cygwin.
21 0.24 2011-07-17 23:15 UTC
22 + Fix : [RT #64521] : "no indirect" leaking into eval.
23 This is currently only fixed for perl 5.10 (perl 5.12 and
24 higher were never affected). It was caused by a very stupid
25 mistake of mine that was introduced in indirect version 0.23.
26 Thanks Michael G Schwern for reporting.
27 + Fix : [RT #69291] : indirect.pm breaks %^H.
28 This was caused by the same mistake as for the previous bug,
29 and as such it is also only fixed for perl 5.10 (and never
31 Thanks Andrew Main for reporting.
32 + Doc : C++ compilers are officially NOT supported.
34 0.23 2010-10-03 00:15 UTC
35 + Fix : Some indirect constructs could be incorrectly reported when
36 several modules were used in the same scope. This caused
37 t/30-scope.t to fail randomly.
38 + Tst : Threads tests are now only run on perl 5.13.4 and higher.
39 They could segfault randomly because of what seems to be an
40 internal bug of Perl, which has been addressed in 5.13.4.
41 There is also an environment variable that allows you to
42 forcefully run those tests, but it should be set only for
43 author testing and not for end users.
45 0.22 2010-08-16 16:00 UTC
46 + Add : Indirect constructs are now reported for code interpolated
47 in quote-like environments, like "${\( ... )}", "@{[ ... ]}",
48 s/pattern/ ... /e, qr/(?{ ... })/ or qr/(??{ ... })/.
49 + Add : You can now make the pragma lethal by passing anything matching
50 /^:?fatal$/i to import(), including "FATAL" and ":Fatal".
51 + Fix : [RT #60378] : segmentation fault on indirect_ck_method.
52 This caused constructs like "@{[ $obj->$meth ]}" to segfault
53 when $meth was a lexical.
54 Thanks Tokuhiro Matsuno for reporting.
56 0.21 2010-05-31 23:10 UTC
57 + Chg : perl 5.8.1 is now required (instead of 5.8.0).
58 + Fix : [RT #57699] : indirect fail with 64-bit int on 5.13.1.
59 It was actually a problem with thread destructors segfaulting
60 because they weren't called at the right time anymore.
61 Thanks Andrew Main for reporting.
62 + Tst : A few more regression tests about the scope leak bug.
64 0.20 2010-04-18 21:25 UTC
65 + Fix : [RT #50570] : "indirect" leaking into LWP.
66 Thanks Andrew Main for reporting.
67 More generally, the require propagation workaround on 5.8-5.10.0
68 has been overhauled, and other scope leaks should be fixed.
69 + Fix : Test failures with 5.12 on Windows where Strawberry Perl crashes
70 because the SystemRoot environment variable is missing.
71 + Fix : Work around Kwalitee test misfailures.
73 0.19 2009-08-28 18:40 UTC
74 + Add : The new constant I_FORKSAFE can be tested to know whether the
75 module will behave nicely when fork()ing. It's currently always
76 true except on Windows where you need perl 5.10.1 for it to be
78 + Fix : I_THREADSAFE and I_FORKSAFE ought to be true when
79 PERL_INDIRECT_PM_DISABLE is set.
80 + Fix : The pragma could still leak if you passed to the "hook" option
81 a reference to the same (named) subroutine from two different
82 require scopes. The fix also provides a better solution for
84 + Fix : Segfaults when indirect is loaded for the first time from
86 + Fix : Leaks of memory associated with the root interpreter.
87 + Opt : Less memory will be used for non-threaded perls version 5.10.0
88 and below, and for threaded perls from version 5.10.1.
90 0.18 2009-08-23 16:15 UTC
91 + Add : When set, the PERL_INDIRECT_PM_DISABLE environment variable
92 disables the pragma globally.
94 0.17 2009-07-16 12:10 UTC
95 + Fix : [RT #47902] : "no indirect" leaking again.
96 This actually turned out to be a bug in how the hook coderefs
97 were stored in the hints hash.
98 Thanks Andrew Main for reporting once again.
99 + Fix : t/80-regressions.t failing on Windows.
100 + Tst : Yet more cleanups.
102 0.16 2009-07-14 16:50 UTC
103 + Add : Indirect calls on blocks are now reported. For those, '{' is
104 passed to the hook as the object description.
105 + Add : The new indirect::msg() function publicizes the default
106 warning/exception message.
107 + Fix : [RT #47866] : Segfault with UTF-8 regexps.
108 Thanks Andrew Main for reporting.
111 0.15 2009-07-08 22:55 UTC
112 + Fix : Invalid constructs with the same method and package name were
114 + Fix : The error line number used to point to the end of the expression
115 instead of its beginning.
117 0.14 2009-06-04 21:55 UTC
118 + Fix : Prevent bogus invalid syntaxes caused by reallocated memory
119 chunks. Thanks Andrew Main for reporting with a reproducible
122 0.13 2009-05-24 18:50 UTC
123 + Add : The hook now receives the file name and the line where the
124 error happened in respectively $_[2] and $_[3].
125 + Fix : Pass mortalized copies of the method name and object to the
126 hook. This seems to fix some rare crashes.
127 + Fix : Work around a bug in perl 5.10.0 and lower. Thanks Andrew Main
128 for teaching me about this issue.
129 + Fix : Report the correct file in error messages (a regression from the
132 0.12 2009-05-03 14:30 UTC
133 + Add : You can specify the action to execute for each indirect
134 construct encountered with the new "hook => $coderef" unimport()
136 + Chg : A ptable is now used internally for the op => position mapping.
137 + Fix : The pragma is now built with thread safety features enabled
138 whenever possible (a notable exception is perl 5.8.x on Win32,
139 as something seems wrong with its context handling).
140 The new indirect::I_THREADSAFE() constant reflects this.
141 + Fix : A negation precedence nit in indirect_ck_entersub().
142 + Tst : "use/no indirect" while parsing an indirect construct.
143 + Tst : Thread safety.
145 0.11 2009-02-08 18:35 UTC
146 + Fix : Potential collisions by hashing pointers with a wrong format.
147 + Upd : Resources in META.yml.
149 0.10 2009-01-17 12:40 UTC
150 Re-release 0.09_01 as stable.
152 0.09_01 2008-12-08 17:55 UTC
153 + Fix : Invalid memory read with "${\(new Foo)}" constructs. The
154 corresponding test is turned back on.
155 + Tst : Refinements in t/30-scope.t
157 0.09 2008-12-05 20:35 UTC
158 + Add : Support for perl 5.8.
159 + Tst : Skip a test in t/10-good.t that randomly segfaults for (I guess)
160 systems stricter than linux in the way they manage their memory.
162 0.08 2008-10-22 14:45 UTC
163 + Fix : A rare edge case for package whose names are prefix of 'main'.
164 + Tst : Test $$ as variable and state variables.
166 0.07_03 2008-10-17 20:10 UTC
167 + Add : Support and tests for variables with spaces after the sigil.
168 + Upd : META.yml spec updated to 1.4.
170 0.07_02 2008-10-15 21:10 UTC
171 + Add : Support and tests for package variables.
172 + Tst : Coverage improved by removing dead code.
174 0.07_01 2008-10-15 16:00 UTC
175 + Fix : [RT #40055] : Not handling RV2SV => GV(SV) correctly, which
176 could cause 'no indirect; print' segfaults. Thanks Goro Fuji for
179 0.06 2008-10-11 16:45 UTC
181 + Tst : Test "no indirect 'anything'", "foo Class->bar", and indirect
182 uses of exec() and system().
184 0.05 2008-10-02 14:40 UTC
185 + Chg : Now the old check function is always called before storing an op
187 + Fix : Misc code and docs refinements.
189 0.04 2008-08-30 19:00 UTC
190 + Fix : Clean up the op->src hash when we're done with an entersub.
191 + Tst : No longer fork for testing. IPC::Cmd isn't required anymore.
193 0.03 2008-08-12 15:25 UTC
194 This release is kindly supported by Copenhagen Hotel Centrum WiFi.
195 + Fix : Tests used not to pass PERL5OPTS to their kids. This lead to
196 failures under CPAN. I think.
199 0.02 2008-08-11 15:55 UTC
200 + Fix : Some building failures with old gcc versions that didn't seem to
201 like the ((hint == 2) ? croak : warn)(msg) construct. I think.
202 + Rem : Unused cruft from a previous implementation.
203 + Tst : Fail more gracefully when we can't capture buffers or when the
204 child returned an error.
206 0.01 2008-08-10 20:40 UTC
207 First version, released on an unsuspecting world.