1 Revision history for Perl extension Devel::Size.
3 0.79_53 2015-03-25 nicholas
4 * Avoid t/recurse.t failing on earlier versions on 32 bit platforms
5 (subroutines are smaller than it assumes)
6 * Count the size of padlist names in CVs (for v5.18 and later - earlier TODO)
7 * Handle PADNAMELIST/PADNAME introduced in v5.21.7
9 0.79_52 2015-03-20 nicholas
10 two patches from Zefram:
11 * handle the new METHOP. [CPAN #101071]
12 * handle the multideref op, and to a limited extent the UNOP_AUX op class in
13 general. [CPAN #102911]
15 * As of eacbb37937698a03 (v5.21.6), can't call CvPADLIST on an XSUB.
16 More usefully, the slot is now used for XS versioning protection, which
17 means that we can be confident that nothing out there is using it for data.
18 Reported as [CPAN #102909].
19 * Avoid t/recurse.t failing because Test::More::is() has been refactored and
20 is now larger than it was. [CPAN #102910].
22 0.79_51 2015-02-28 nicholas
23 * as of 5.20.0, s/// is no longer a reliable test for OOK [CPAN #95493]
25 0.79_50 2015-02-28 nicholas
27 * wrong test for magicalness [CPAN #88180]
28 * reading tied scalar shrinks it [CPAN #88182]
30 0.79 2013-05-12 nicholas
33 0.78_52 2013-05-10 nicholas
34 * Fix the bug in the previous fix. Oops :-)
36 0.78_51 2013-05-10 nicholas
37 * v5.17.1 and later handle CopSTASH differently under ithreads
39 0.78_50 2013-05-10 nicholas
40 * Alternative test for CvOUTSIDE, and one for CvPADLIST
41 * v5.17 changes the type of PADLIST [CPAN #83904]
42 * stubbed subs no longer have CvOUTSIDE set [CPAN #77913]
43 * Fix for v5.17.2's OP slab allocator [CPAN #83903]
44 * Spelling fix to docs from gregor herrmann [CPAN #78766]
46 0.78 2012-07-26 nicholas
49 0.77_51 2012-07-17 nicholas
50 * Fix for MSVC builds from bulk 88 [CPAN #77589]
52 0.77_50 2012-02-10 nicholas
53 * t/globs.t was failing on 5.15.6 and later due to side effects of a change
54 to strict.pm [CPAN #73998]
55 * skip tests in t/magic.t that use formline on 5.8.1 and 5.8.2, as those
56 versions have a buggy formline that can trigger an assertion failure.
58 0.77 2011-05-16 nicholas
61 0.76_50 2011-05-12 nicholas
62 * Split out HEK size calculation into hek_size(). Add the shared HE overhead.
63 * Handle shared hash key scalars correctly.
64 * GvNAME() is shared from 5.10 onwards.
65 * Count HvNAME(), the HV "aux" struct, the mro_meta struct, and ENAMEs.
67 0.76 2011-05-11 nicholas
68 * Just fix the version number in the line below.
70 0.75_52 2011-05-09 nicholas
71 * Use a table for SV body sizes. These incorporate the space saving post 5.8.x
72 * Correctly handle SvOOK scalars. 5.12 and later don't use SvIVX().
73 * Magic vtables aren't freed when magic is freed, so don't count them.
74 (They are static structures. Anything that assumes otherwise is buggy.)
75 * Reinstate 5.6.x "support". (i.e. it compiles and passes tests.)
76 * Reinstate 5.005_xx "support". (i.e. it compiles and passes tests.)
78 0.75_51 2011-05-05 nicholas
79 * Only use a static array of vtables on gcc.
81 0.75_50 2011-05-04 nicholas
82 * The core's magic vtables are global constants, so aren't part of the size.
83 * Follow mg_obj and mg_ptr.
85 0.75 2011-05-04 nicholas
88 0.74_53 2011-05-01 nicholas
89 * Fix typo error in bit-vector tracking mechanism. On 64 bit platforms which
90 allocate on 8 byte alignment (rather than 16), a low pointer bit could get
91 lost, resulting in new pointers being considered already "seen".
92 "Found" by BinGOs' smoker, fixed on spectre.mongueurs.net. Thanks.
94 0.74_52 2011-04-23 nicholas
95 * Fix potential SEGVs for OP_AELEMFAST on a lexical (eg $foo[3])
96 * Fix likely SEGVs for PVOPs (missing break)
97 * Fix potential SEGVs for PVBMs on 5.10 and later
98 * Move hash and array traversal from total_size() to sv_size()
99 - this allows total_size() and size() to be XS ALIASes.
101 0.74_51 2011-04-22 nicholas
102 * Don't count PL_sv_{undef,no,yes} in the size returned
103 * total_size() was double-counting entries in typeglobs
104 * sv_size() was double-counting the PVGV size if GvEGV() looped back
105 * fix classic off-by-one error - the answer is strlen() + 1, not just strlen()
107 0.74_50 2011-04-19 nicholas
108 * Ensure that size() doesn't add the referent's size for non SVt_RV references
110 0.74 2011-04-19 nicholas
111 * Correct the Makefile.PL - LICENSE was added to ExtUtils::MakeMaker in 6.31
113 0.73_51 2011-04-17 nicholas
114 * Refactor the C code to accumulate the size inside the tracking structure
115 - this means that the C *_size() functions now return void
116 - The XS functions size() and total_size() now return UV, not IV
118 0.73_50 2011-04-17 nicholas
119 * Avoid using assert() on ithreaded 5.8.8 and earlier, as it needs my_perl
120 * Fix the test added in 0.73, as it tripped over a combination of bugs
122 0.73 2011-04-16 nicholas
123 * Revert a bad assertion introduced in 0.72_50, which was logically wrong.
124 - and a test to demonstrate one case that would trigger it
126 0.72_52 2011-04-15 nicholas
127 * Add a test for the non-exceptional warnings.
129 0.72_51 2011-04-15 nicholas
130 * Add PERL_NO_GET_CONTEXT to improve performance under multiplicity
132 0.72_50 2011-04-14 nicholas
133 * Exception handling is totally MSVC specific, so only use it there
134 - this means that we don't need to use a C++ compiler anywhere
135 * Rework bit-vector tracking mechanism to use a 256-way tree. This avoids
136 making assumptions about 64-bit platforms' memory layouts, and eliminates
137 the fatal error introduced in 0.72 when the assumption was violated.
138 * Convert to XSLoader
139 * Resolve CPAN #49437 (Devel::Size adds magic in Perl 5.10)
140 * Resolve CPAN #58484 and #58485 (related to CVs that are XSUBs)
142 0.72 2008-10-14 BrowserUk 70 tests
143 * Added bit-vector pointer tracking mechanism.
144 - new fatal error (64-bit platforms only)
145 * Added exception handling
146 - 4 new warnings (disabled by default)
147 * Updated POD to reflect above
148 * Added basic.t test 13
149 * replaced Module::Install with hand crafted EU::MM Makefile.pl
150 (With many thanks to Sisyphus)
151 because we couldn't work out how to add C++ options to allow
154 0.71 2008-08-24 Tels 69 tests
155 * adapt patch from Reini Urban to fix failing RV under 5.10 and 5.11. AV
156 and HV were pushed directly onto the pending_array, and not the RV,
157 which caused #33530. So he rewrote the logic to deref the RV inside
158 the array traversal. Applied this with one small omission, which
160 * Fixed 5.11 RV/IV logic. (Thanx Reini Urban!)
161 * Removed one duplicate total_size arrayref test. (Thanx Reini Urban!)
162 * changed //printf to dbg_printf() (Thanx Reini again!)
164 0.70 2008-08-23 Tels 69 tests
165 * fix SEGFAULTS under v5.10 (Thanx Reini Urban!)
166 * fix compilation under blead (Thanx Reini Urban!)
167 * require Perl 5.006 (to stop CPANTESTERS sending bogus reports
168 about failed compiling on ancient Perls)
169 * update the bundles Module::Install to v0.77
171 0.69 2007-08-11 Tels 69 tests
172 * fix compilation under Perl v5.9.5 and v5.10 (Thanx Steve Peters!)
173 * clarify the license by specifying Perl v5.8.8's license
174 * small doc fixes, add a README file
176 0.68 2007-06-12 Tels 69 tests
177 * remove a bit of duplicate code in op_size, the second instance
179 * fix the failing tests by rewriting the test logic to be sane
180 * the count of array slots was off by one
182 0.67 2007-03-15 Tels 20 tests
183 * rip out Build.PL since it doesn't actually work
184 * fix bug #1653: All SVPV subtype may contain an RV instead of a PV
185 (Thanx a lot Jan, and sorry for the truly horrible long delay!)
186 * fix bug #24846: Does not correctly recurse into references in PVNV
188 * add t/recurse.t that tests the two bugfixes in combination
190 0.66 2007-03-02 Tels 15 tests
191 * fix the failing test (the code was right, but the test wrong)
193 0.65 2007-02-24 Tels 15 tests
194 * fix the test failings under 64 bit (bugs #21037, #18596, #21404)
195 * fix wrong size for strings (bug #17586)
196 * reverse Changelog (bug #16636)
197 * Size.xs: remove unused variable "count"
198 * Size.xs: fix warnings about missing ()
199 * Size.xs: fix size of array refs when AvALLOC == 0
201 + maintained by me, signed with my key
202 + bundle Module::Install and use it
204 + rewrite tests to use Test::More; add more tests
206 0.64 Mon Dec 19 18:46:00 2005
207 - Patch to make Devel::Size compile on bleadperl (Courtesy of
209 - Fix up the PVIV string size calcs (Courtesy of Andrew Shirrayev)
211 0.63 Thu Jun 30 14:29:43 2005
212 - Turns out that OP_TRANS is a basic op, not a loop op. This
213 is a bug in versions of perl 5.8.x before 5.8.7. Work around
216 0.62 Tue Jun 28 11:59:00 2005
217 - Took out // comments
218 - Added in copyright notice
219 - Some small amount of regex parsing
220 - Suppress multiple copies of each warning on each call
222 0.61 Mon Jun 27 16:19:00 2005
223 - Added more checks for formats
224 - Got CVs sizing right
226 0.59 Sat Nov 27 16:42:42 2004
227 - Applied documentation and sane warning patch from Nigel Sandever
228 - Taught Devel::Size how to size up IO and globs properly
230 0.58 Fri Jul 18 11:42:32 2003
231 - Fix for problems triggered by perl 5.8.0 and up, more tests, and
232 removal of an "our" for better backwards compatibility. (Courtesy
233 of Marcus Holland-Moritz <mhx-perl@gmx.net>)
235 0.57 Thu Mar 20 13:21:14 2003
236 - setting $Devel::Size::warn to 0 disables not complete warnings
238 0.56 Mon Feb 24 12:10:13 2003
239 - Chopped out some C++ comments. D'oh! Version incremented for CPAN
241 0.55 Sat Feb 22 17:21:00 2003
242 - Fixed a bad size calculation (we were overestimating by one byte)
243 - Updated the docs to show some of the places that there might be 'dark'
244 memory that Devel::Size can't see.
245 - Added in tests from Ken Williams
247 0.54 Sat Oct 12 14:11:00 2002
248 - Applied a patch to get it working on 5.8.0 under Tru64
250 0.53 Thu Oct 10 12:30:00 2002
251 - Finally started updating Changes file
252 - Applied doc patch from Ann Barcomb
253 - Got globs sizing right
255 0.01 Mon Oct 7 01:05:32 2002
256 - original version; created by h2xs 1.2 with options