1 Revision history for indirect
3 0.17 2009-07-16 12:10 UTC
4 + Fix : [RT #47902] : "no indirect" leaking again.
5 This actually turned out to be a bug in how the hook coderefs
6 were stored in the hints hash.
7 Thanks Andrew Main for reporting once again.
8 + Fix : t/80-regressions.t failing on Windows.
9 + Tst : Yet more cleanups.
11 0.16 2009-07-14 16:50 UTC
12 + Add : Indirect calls on blocks are now reported. For those, '{' is
13 passed to the hook as the object description.
14 + Add : The new indirect::msg() function publicizes the default
15 warning/exception message.
16 + Fix : [RT #47866] : Segfault with UTF-8 regexps.
17 Thanks Andrew Main for reporting.
20 0.15 2009-07-08 22:55 UTC
21 + Fix : Invalid constructs with the same method and package name were
23 + Fix : The error line number used to point to the end of the expression
24 instead of its beginning.
26 0.14 2009-06-04 21:55 UTC
27 + Fix : Prevent bogus invalid syntaxes caused by reallocated memory
28 chunks. Thanks Andrew Main for reporting with a reproducible
31 0.13 2009-05-24 18:50 UTC
32 + Add : The hook now receives the file name and the line where the
33 error happened in respectively $_[2] and $_[3].
34 + Fix : Pass mortalized copies of the method name and object to the
35 hook. This seems to fix some rare crashes.
36 + Fix : Work around a bug in perl 5.10.0 and lower. Thanks Andrew Main
37 for teaching me about this issue.
38 + Fix : Report the correct file in error messages (a regression from the
41 0.12 2009-05-03 14:30 UTC
42 + Add : You can specify the action to execute for each indirect
43 construct encountered with the new "hook => $coderef" unimport()
45 + Chg : A ptable is now used internally for the op => position mapping.
46 + Fix : The pragma is now built with thread safety features enabled
47 whenever possible (a notable exception is perl 5.8.x on Win32,
48 as something seems wrong with its context handling).
49 The new indirect::I_THREADSAFE() constant reflects this.
50 + Fix : A negation precedence nit in indirect_ck_entersub().
51 + Tst : "use/no indirect" while parsing an indirect construct.
52 + Tst : Thread safety.
54 0.11 2009-02-08 18:35 UTC
55 + Fix : Potential collisions by hashing pointers with a wrong format.
56 + Upd : Resources in META.yml.
58 0.10 2009-01-17 12:40 UTC
59 Re-release 0.09_01 as stable.
61 0.09_01 2008-12-08 17:55 UTC
62 + Fix : Invalid memory read with "${\(new Foo)}" constructs. The
63 corresponding test is turned back on.
64 + Tst : Refinements in t/30-scope.t
66 0.09 2008-12-05 20:35 UTC
67 + Add : Support for perl 5.8.
68 + Tst : Skip a test in t/10-good.t that randomly segfaults for (I guess)
69 systems stricter than linux in the way they manage their memory.
71 0.08 2008-10-22 14:45 UTC
72 + Fix : A rare edge case for package whose names are prefix of 'main'.
73 + Tst : Test $$ as variable and state variables.
75 0.07_03 2008-10-17 20:10 UTC
76 + Add : Support and tests for variables with spaces after the sigil.
77 + Upd : META.yml spec updated to 1.4.
79 0.07_02 2008-10-15 21:10 UTC
80 + Add : Support and tests for package variables.
81 + Tst : Coverage improved by removing dead code.
83 0.07_01 2008-10-15 16:00 UTC
84 + Fix : [RT #40055] : Not handling RV2SV => GV(SV) correctly, which
85 could cause 'no indirect; print' segfaults. Thanks Goro Fuji for
88 0.06 2008-10-11 16:45 UTC
90 + Tst : Test "no indirect 'anything'", "foo Class->bar", and indirect
91 uses of exec() and system().
93 0.05 2008-10-02 14:40 UTC
94 + Chg : Now the old check function is always called before storing an op
96 + Fix : Misc code and docs refinements.
98 0.04 2008-08-30 19:00 UTC
99 + Fix : Clean up the op->src hash when we're done with an entersub.
100 + Tst : No longer fork for testing. IPC::Cmd isn't required anymore.
102 0.03 2008-08-12 15:25 UTC
103 This release is kindly supported by Copenhagen Hotel Centrum WiFi.
104 + Fix : Tests used not to pass PERL5OPTS to their kids. This lead to
105 failures under CPAN. I think.
108 0.02 2008-08-11 15:55 UTC
109 + Fix : Some building failures with old gcc versions that didn't seem to
110 like the ((hint == 2) ? croak : warn)(msg) construct. I think.
111 + Rem : Unused cruft from a previous implementation.
112 + Tst : Fail more gracefully when we can't capture buffers or when the
113 child returned an error.
115 0.01 2008-08-10 20:40 UTC
116 First version, released on an unsuspecting world.