1 Revision history for Perl extension Devel::Size.
3 0.76_50 2011-05-12 nicholas
4 * Split out HEK size calculation into hek_size(). Add the shared HE overhead.
5 * Handle shared hash key scalars correctly.
6 * GvNAME() is shared from 5.10 onwards.
7 * Count HvNAME(), the HV "aux" struct, and the mro_meta struct.
9 0.76 2011-05-11 nicholas
10 * Just fix the version number in the line below.
12 0.75_52 2011-05-09 nicholas
13 * Use a table for SV body sizes. These incorporate the space saving post 5.8.x
14 * Correctly handle SvOOK scalars. 5.12 and later don't use SvIVX().
15 * Magic vtables aren't freed when magic is freed, so don't count them.
16 (They are static structures. Anything that assumes otherwise is buggy.)
17 * Reinstate 5.6.x "support". (i.e. it compiles and passes tests.)
18 * Reinstate 5.005_xx "support". (i.e. it compiles and passes tests.)
20 0.75_51 2011-05-05 nicholas
21 * Only use a static array of vtables on gcc.
23 0.75_50 2011-05-04 nicholas
24 * The core's magic vtables are global constants, so aren't part of the size.
25 * Follow mg_obj and mg_ptr.
27 0.75 2011-05-04 nicholas
30 0.74_53 2011-05-01 nicholas
31 * Fix typo error in bit-vector tracking mechanism. On 64 bit platforms which
32 allocate on 8 byte alignment (rather than 16), a low pointer bit could get
33 lost, resulting in new pointers being considered already "seen".
34 "Found" by BinGOs' smoker, fixed on spectre.mongueurs.net. Thanks.
36 0.74_52 2011-04-23 nicholas
37 * Fix potential SEGVs for OP_AELEMFAST on a lexical (eg $foo[3])
38 * Fix likely SEGVs for PVOPs (missing break)
39 * Fix potential SEGVs for PVBMs on 5.10 and later
40 * Move hash and array traversal from total_size() to sv_size()
41 - this allows total_size() and size() to be XS ALIASes.
43 0.74_51 2011-04-22 nicholas
44 * Don't count PL_sv_{undef,no,yes} in the size returned
45 * total_size() was double-counting entries in typeglobs
46 * sv_size() was double-counting the PVGV size if GvEGV() looped back
47 * fix classic off-by-one error - the answer is strlen() + 1, not just strlen()
49 0.74_50 2011-04-19 nicholas
50 * Ensure that size() doesn't add the referent's size for non SVt_RV references
52 0.74 2011-04-19 nicholas
53 * Correct the Makefile.PL - LICENSE was added to ExtUtils::MakeMaker in 6.31
55 0.73_51 2011-04-17 nicholas
56 * Refactor the C code to accumulate the size inside the tracking structure
57 - this means that the C *_size() functions now return void
58 - The XS functions size() and total_size() now return UV, not IV
60 0.73_50 2011-04-17 nicholas
61 * Avoid using assert() on ithreaded 5.8.8 and earlier, as it needs my_perl
62 * Fix the test added in 0.73, as it tripped over a combination of bugs
64 0.73 2011-04-16 nicholas
65 * Revert a bad assertion introduced in 0.72_50, which was logically wrong.
66 - and a test to demonstrate one case that would trigger it
68 0.72_52 2011-04-15 nicholas
69 * Add a test for the non-exceptional warnings.
71 0.72_51 2011-04-15 nicholas
72 * Add PERL_NO_GET_CONTEXT to improve performance under multiplicity
74 0.72_50 2011-04-14 nicholas
75 * Exception handling is totally MSVC specific, so only use it there
76 - this means that we don't need to use a C++ compiler anywhere
77 * Rework bit-vector tracking mechanism to use a 256-way tree. This avoids
78 making assumptions about 64-bit platforms' memory layouts, and eliminates
79 the fatal error introduced in 0.72 when the assumption was violated.
81 * Resolve CPAN #49437 (Devel::Size adds magic in Perl 5.10)
82 * Resolve CPAN #58484 and #58485 (related to CVs that are XSUBs)
84 0.72 2008-10-14 BrowserUk 70 tests
85 * Added bit-vector pointer tracking mechanism.
86 - new fatal error (64-bit platforms only)
87 * Added exception handling
88 - 4 new warnings (disabled by default)
89 * Updated POD to reflect above
90 * Added basic.t test 13
91 * replaced Module::Install with hand crafted EU::MM Makefile.pl
92 (With many thanks to Sisyphus)
93 because we couldn't work out how to add C++ options to allow
96 0.71 2008-08-24 Tels 69 tests
97 * adapt patch from Reini Urban to fix failing RV under 5.10 and 5.11. AV
98 and HV were pushed directly onto the pending_array, and not the RV,
99 which caused #33530. So he rewrote the logic to deref the RV inside
100 the array traversal. Applied this with one small omission, which
102 * Fixed 5.11 RV/IV logic. (Thanx Reini Urban!)
103 * Removed one duplicate total_size arrayref test. (Thanx Reini Urban!)
104 * changed //printf to dbg_printf() (Thanx Reini again!)
106 0.70 2008-08-23 Tels 69 tests
107 * fix SEGFAULTS under v5.10 (Thanx Reini Urban!)
108 * fix compilation under blead (Thanx Reini Urban!)
109 * require Perl 5.006 (to stop CPANTESTERS sending bogus reports
110 about failed compiling on ancient Perls)
111 * update the bundles Module::Install to v0.77
113 0.69 2007-08-11 Tels 69 tests
114 * fix compilation under Perl v5.9.5 and v5.10 (Thanx Steve Peters!)
115 * clarify the license by specifying Perl v5.8.8's license
116 * small doc fixes, add a README file
118 0.68 2007-06-12 Tels 69 tests
119 * remove a bit of duplicate code in op_size, the second instance
121 * fix the failing tests by rewriting the test logic to be sane
122 * the count of array slots was off by one
124 0.67 2007-03-15 Tels 20 tests
125 * rip out Build.PL since it doesn't actually work
126 * fix bug #1653: All SVPV subtype may contain an RV instead of a PV
127 (Thanx a lot Jan, and sorry for the truly horrible long delay!)
128 * fix bug #24846: Does not correctly recurse into references in PVNV
130 * add t/recurse.t that tests the two bugfixes in combination
132 0.66 2007-03-02 Tels 15 tests
133 * fix the failing test (the code was right, but the test wrong)
135 0.65 2007-02-24 Tels 15 tests
136 * fix the test failings under 64 bit (bugs #21037, #18596, #21404)
137 * fix wrong size for strings (bug #17586)
138 * reverse Changelog (bug #16636)
139 * Size.xs: remove unused variable "count"
140 * Size.xs: fix warnings about missing ()
141 * Size.xs: fix size of array refs when AvALLOC == 0
143 + maintained by me, signed with my key
144 + bundle Module::Install and use it
146 + rewrite tests to use Test::More; add more tests
148 0.64 Mon Dec 19 18:46:00 2005
149 - Patch to make Devel::Size compile on bleadperl (Courtesy of
151 - Fix up the PVIV string size calcs (Courtesy of Andrew Shirrayev)
153 0.63 Thu Jun 30 14:29:43 2005
154 - Turns out that OP_TRANS is a basic op, not a loop op. This
155 is a bug in versions of perl 5.8.x before 5.8.7. Work around
158 0.62 Tue Jun 28 11:59:00 2005
159 - Took out // comments
160 - Added in copyright notice
161 - Some small amount of regex parsing
162 - Suppress multiple copies of each warning on each call
164 0.61 Mon Jun 27 16:19:00 2005
165 - Added more checks for formats
166 - Got CVs sizing right
168 0.59 Sat Nov 27 16:42:42 2004
169 - Applied documentation and sane warning patch from Nigel Sandever
170 - Taught Devel::Size how to size up IO and globs properly
172 0.58 Fri Jul 18 11:42:32 2003
173 - Fix for problems triggered by perl 5.8.0 and up, more tests, and
174 removal of an "our" for better backwards compatibility. (Courtesy
175 of Marcus Holland-Moritz <mhx-perl@gmx.net>)
177 0.57 Thu Mar 20 13:21:14 2003
178 - setting $Devel::Size::warn to 0 disables not complete warnings
180 0.56 Mon Feb 24 12:10:13 2003
181 - Chopped out some C++ comments. D'oh! Version incremented for CPAN
183 0.55 Sat Feb 22 17:21:00 2003
184 - Fixed a bad size calculation (we were overestimating by one byte)
185 - Updated the docs to show some of the places that there might be 'dark'
186 memory that Devel::Size can't see.
187 - Added in tests from Ken Williams
189 0.54 Sat Oct 12 14:11:00 2002
190 - Applied a patch to get it working on 5.8.0 under Tru64
192 0.53 Thu Oct 10 12:30:00 2002
193 - Finally started updating Changes file
194 - Applied doc patch from Ann Barcomb
195 - Got globs sizing right
197 0.01 Mon Oct 7 01:05:32 2002
198 - original version; created by h2xs 1.2 with options