1 Revision history for Perl extension Devel::Size.
3 0.79_53 2015-03-21 nicholas
5 0.79_52 2015-03-20 nicholas
6 two patches from Zefram:
7 * handle the new METHOP. [CPAN #101071]
8 * handle the multideref op, and to a limited extent the UNOP_AUX op class in
9 general. [CPAN #102911]
11 * As of eacbb37937698a03 (v5.21.6), can't call CvPADLIST on an XSUB.
12 More usefully, the slot is now used for XS versioning protection, which
13 means that we can be confident that nothing out there is using it for data.
14 Reported as [CPAN #102909].
15 * Avoid t/recurse.t failing because Test::More::is() has been refactored and
16 is now larger than it was. [CPAN #102910].
18 0.79_51 2015-02-28 nicholas
19 * as of 5.20.0, s/// is no longer a reliable test for OOK [CPAN #95493]
21 0.79_50 2015-02-28 nicholas
23 * wrong test for magicalness [CPAN #88180]
24 * reading tied scalar shrinks it [CPAN #88182]
26 0.79 2013-05-12 nicholas
29 0.78_52 2013-05-10 nicholas
30 * Fix the bug in the previous fix. Oops :-)
32 0.78_51 2013-05-10 nicholas
33 * v5.17.1 and later handle CopSTASH differently under ithreads
35 0.78_50 2013-05-10 nicholas
36 * Alternative test for CvOUTSIDE, and one for CvPADLIST
37 * v5.17 changes the type of PADLIST [CPAN #83904]
38 * stubbed subs no longer have CvOUTSIDE set [CPAN #77913]
39 * Fix for v5.17.2's OP slab allocator [CPAN #83903]
40 * Spelling fix to docs from gregor herrmann [CPAN #78766]
42 0.78 2012-07-26 nicholas
45 0.77_51 2012-07-17 nicholas
46 * Fix for MSVC builds from bulk 88 [CPAN #77589]
48 0.77_50 2012-02-10 nicholas
49 * t/globs.t was failing on 5.15.6 and later due to side effects of a change
50 to strict.pm [CPAN #73998]
51 * skip tests in t/magic.t that use formline on 5.8.1 and 5.8.2, as those
52 versions have a buggy formline that can trigger an assertion failure.
54 0.77 2011-05-16 nicholas
57 0.76_50 2011-05-12 nicholas
58 * Split out HEK size calculation into hek_size(). Add the shared HE overhead.
59 * Handle shared hash key scalars correctly.
60 * GvNAME() is shared from 5.10 onwards.
61 * Count HvNAME(), the HV "aux" struct, the mro_meta struct, and ENAMEs.
63 0.76 2011-05-11 nicholas
64 * Just fix the version number in the line below.
66 0.75_52 2011-05-09 nicholas
67 * Use a table for SV body sizes. These incorporate the space saving post 5.8.x
68 * Correctly handle SvOOK scalars. 5.12 and later don't use SvIVX().
69 * Magic vtables aren't freed when magic is freed, so don't count them.
70 (They are static structures. Anything that assumes otherwise is buggy.)
71 * Reinstate 5.6.x "support". (i.e. it compiles and passes tests.)
72 * Reinstate 5.005_xx "support". (i.e. it compiles and passes tests.)
74 0.75_51 2011-05-05 nicholas
75 * Only use a static array of vtables on gcc.
77 0.75_50 2011-05-04 nicholas
78 * The core's magic vtables are global constants, so aren't part of the size.
79 * Follow mg_obj and mg_ptr.
81 0.75 2011-05-04 nicholas
84 0.74_53 2011-05-01 nicholas
85 * Fix typo error in bit-vector tracking mechanism. On 64 bit platforms which
86 allocate on 8 byte alignment (rather than 16), a low pointer bit could get
87 lost, resulting in new pointers being considered already "seen".
88 "Found" by BinGOs' smoker, fixed on spectre.mongueurs.net. Thanks.
90 0.74_52 2011-04-23 nicholas
91 * Fix potential SEGVs for OP_AELEMFAST on a lexical (eg $foo[3])
92 * Fix likely SEGVs for PVOPs (missing break)
93 * Fix potential SEGVs for PVBMs on 5.10 and later
94 * Move hash and array traversal from total_size() to sv_size()
95 - this allows total_size() and size() to be XS ALIASes.
97 0.74_51 2011-04-22 nicholas
98 * Don't count PL_sv_{undef,no,yes} in the size returned
99 * total_size() was double-counting entries in typeglobs
100 * sv_size() was double-counting the PVGV size if GvEGV() looped back
101 * fix classic off-by-one error - the answer is strlen() + 1, not just strlen()
103 0.74_50 2011-04-19 nicholas
104 * Ensure that size() doesn't add the referent's size for non SVt_RV references
106 0.74 2011-04-19 nicholas
107 * Correct the Makefile.PL - LICENSE was added to ExtUtils::MakeMaker in 6.31
109 0.73_51 2011-04-17 nicholas
110 * Refactor the C code to accumulate the size inside the tracking structure
111 - this means that the C *_size() functions now return void
112 - The XS functions size() and total_size() now return UV, not IV
114 0.73_50 2011-04-17 nicholas
115 * Avoid using assert() on ithreaded 5.8.8 and earlier, as it needs my_perl
116 * Fix the test added in 0.73, as it tripped over a combination of bugs
118 0.73 2011-04-16 nicholas
119 * Revert a bad assertion introduced in 0.72_50, which was logically wrong.
120 - and a test to demonstrate one case that would trigger it
122 0.72_52 2011-04-15 nicholas
123 * Add a test for the non-exceptional warnings.
125 0.72_51 2011-04-15 nicholas
126 * Add PERL_NO_GET_CONTEXT to improve performance under multiplicity
128 0.72_50 2011-04-14 nicholas
129 * Exception handling is totally MSVC specific, so only use it there
130 - this means that we don't need to use a C++ compiler anywhere
131 * Rework bit-vector tracking mechanism to use a 256-way tree. This avoids
132 making assumptions about 64-bit platforms' memory layouts, and eliminates
133 the fatal error introduced in 0.72 when the assumption was violated.
134 * Convert to XSLoader
135 * Resolve CPAN #49437 (Devel::Size adds magic in Perl 5.10)
136 * Resolve CPAN #58484 and #58485 (related to CVs that are XSUBs)
138 0.72 2008-10-14 BrowserUk 70 tests
139 * Added bit-vector pointer tracking mechanism.
140 - new fatal error (64-bit platforms only)
141 * Added exception handling
142 - 4 new warnings (disabled by default)
143 * Updated POD to reflect above
144 * Added basic.t test 13
145 * replaced Module::Install with hand crafted EU::MM Makefile.pl
146 (With many thanks to Sisyphus)
147 because we couldn't work out how to add C++ options to allow
150 0.71 2008-08-24 Tels 69 tests
151 * adapt patch from Reini Urban to fix failing RV under 5.10 and 5.11. AV
152 and HV were pushed directly onto the pending_array, and not the RV,
153 which caused #33530. So he rewrote the logic to deref the RV inside
154 the array traversal. Applied this with one small omission, which
156 * Fixed 5.11 RV/IV logic. (Thanx Reini Urban!)
157 * Removed one duplicate total_size arrayref test. (Thanx Reini Urban!)
158 * changed //printf to dbg_printf() (Thanx Reini again!)
160 0.70 2008-08-23 Tels 69 tests
161 * fix SEGFAULTS under v5.10 (Thanx Reini Urban!)
162 * fix compilation under blead (Thanx Reini Urban!)
163 * require Perl 5.006 (to stop CPANTESTERS sending bogus reports
164 about failed compiling on ancient Perls)
165 * update the bundles Module::Install to v0.77
167 0.69 2007-08-11 Tels 69 tests
168 * fix compilation under Perl v5.9.5 and v5.10 (Thanx Steve Peters!)
169 * clarify the license by specifying Perl v5.8.8's license
170 * small doc fixes, add a README file
172 0.68 2007-06-12 Tels 69 tests
173 * remove a bit of duplicate code in op_size, the second instance
175 * fix the failing tests by rewriting the test logic to be sane
176 * the count of array slots was off by one
178 0.67 2007-03-15 Tels 20 tests
179 * rip out Build.PL since it doesn't actually work
180 * fix bug #1653: All SVPV subtype may contain an RV instead of a PV
181 (Thanx a lot Jan, and sorry for the truly horrible long delay!)
182 * fix bug #24846: Does not correctly recurse into references in PVNV
184 * add t/recurse.t that tests the two bugfixes in combination
186 0.66 2007-03-02 Tels 15 tests
187 * fix the failing test (the code was right, but the test wrong)
189 0.65 2007-02-24 Tels 15 tests
190 * fix the test failings under 64 bit (bugs #21037, #18596, #21404)
191 * fix wrong size for strings (bug #17586)
192 * reverse Changelog (bug #16636)
193 * Size.xs: remove unused variable "count"
194 * Size.xs: fix warnings about missing ()
195 * Size.xs: fix size of array refs when AvALLOC == 0
197 + maintained by me, signed with my key
198 + bundle Module::Install and use it
200 + rewrite tests to use Test::More; add more tests
202 0.64 Mon Dec 19 18:46:00 2005
203 - Patch to make Devel::Size compile on bleadperl (Courtesy of
205 - Fix up the PVIV string size calcs (Courtesy of Andrew Shirrayev)
207 0.63 Thu Jun 30 14:29:43 2005
208 - Turns out that OP_TRANS is a basic op, not a loop op. This
209 is a bug in versions of perl 5.8.x before 5.8.7. Work around
212 0.62 Tue Jun 28 11:59:00 2005
213 - Took out // comments
214 - Added in copyright notice
215 - Some small amount of regex parsing
216 - Suppress multiple copies of each warning on each call
218 0.61 Mon Jun 27 16:19:00 2005
219 - Added more checks for formats
220 - Got CVs sizing right
222 0.59 Sat Nov 27 16:42:42 2004
223 - Applied documentation and sane warning patch from Nigel Sandever
224 - Taught Devel::Size how to size up IO and globs properly
226 0.58 Fri Jul 18 11:42:32 2003
227 - Fix for problems triggered by perl 5.8.0 and up, more tests, and
228 removal of an "our" for better backwards compatibility. (Courtesy
229 of Marcus Holland-Moritz <mhx-perl@gmx.net>)
231 0.57 Thu Mar 20 13:21:14 2003
232 - setting $Devel::Size::warn to 0 disables not complete warnings
234 0.56 Mon Feb 24 12:10:13 2003
235 - Chopped out some C++ comments. D'oh! Version incremented for CPAN
237 0.55 Sat Feb 22 17:21:00 2003
238 - Fixed a bad size calculation (we were overestimating by one byte)
239 - Updated the docs to show some of the places that there might be 'dark'
240 memory that Devel::Size can't see.
241 - Added in tests from Ken Williams
243 0.54 Sat Oct 12 14:11:00 2002
244 - Applied a patch to get it working on 5.8.0 under Tru64
246 0.53 Thu Oct 10 12:30:00 2002
247 - Finally started updating Changes file
248 - Applied doc patch from Ann Barcomb
249 - Got globs sizing right
251 0.01 Mon Oct 7 01:05:32 2002
252 - original version; created by h2xs 1.2 with options