Refactor t/globs.t to avoid the side effects of a change to strict.pm
[p5sagit/Devel-Size.git] / CHANGES
CommitLineData
e98cedbf 1Revision history for Perl extension Devel::Size.
2
88d3c90b 30.77_50 2012-02-10 nicholas
4 * t/globs.t was failing on 5.15.6 and later due to side effects of a change
5 to strict.pm [CPAN #73998]
6
76ab8e98 70.77 2011-05-16 nicholas
8 [no changes]
9
3d18ea10 100.76_50 2011-05-12 nicholas
11 * Split out HEK size calculation into hek_size(). Add the shared HE overhead.
924d9c4e 12 * Handle shared hash key scalars correctly.
638a265a 13 * GvNAME() is shared from 5.10 onwards.
b3a37f1a 14 * Count HvNAME(), the HV "aux" struct, the mro_meta struct, and ENAMEs.
3d18ea10 15
4a51050d 160.76 2011-05-11 nicholas
17 * Just fix the version number in the line below.
18
190.75_52 2011-05-09 nicholas
b6558d1d 20 * Use a table for SV body sizes. These incorporate the space saving post 5.8.x
95dc1714 21 * Correctly handle SvOOK scalars. 5.12 and later don't use SvIVX().
9847261d 22 * Magic vtables aren't freed when magic is freed, so don't count them.
23 (They are static structures. Anything that assumes otherwise is buggy.)
d6158a76 24 * Reinstate 5.6.x "support". (i.e. it compiles and passes tests.)
177ebd37 25 * Reinstate 5.005_xx "support". (i.e. it compiles and passes tests.)
b6558d1d 26
ec404c23 270.75_51 2011-05-05 nicholas
28 * Only use a static array of vtables on gcc.
29
d9b022a1 300.75_50 2011-05-04 nicholas
b7130948 31 * The core's magic vtables are global constants, so aren't part of the size.
d1888d0b 32 * Follow mg_obj and mg_ptr.
d9b022a1 33
ce5fa0d6 340.75 2011-05-04 nicholas
35 [no changes]
36
f404ed48 370.74_53 2011-05-01 nicholas
38 * Fix typo error in bit-vector tracking mechanism. On 64 bit platforms which
39 allocate on 8 byte alignment (rather than 16), a low pointer bit could get
40 lost, resulting in new pointers being considered already "seen".
41 "Found" by BinGOs' smoker, fixed on spectre.mongueurs.net. Thanks.
42
574d9fd9 430.74_52 2011-04-23 nicholas
44 * Fix potential SEGVs for OP_AELEMFAST on a lexical (eg $foo[3])
219b7d34 45 * Fix likely SEGVs for PVOPs (missing break)
4a3d023d 46 * Fix potential SEGVs for PVBMs on 5.10 and later
13683e3a 47 * Move hash and array traversal from total_size() to sv_size()
48 - this allows total_size() and size() to be XS ALIASes.
574d9fd9 49
6ec51ae0 500.74_51 2011-04-22 nicholas
a52ceccd 51 * Don't count PL_sv_{undef,no,yes} in the size returned
81f1c018 52 * total_size() was double-counting entries in typeglobs
53 * sv_size() was double-counting the PVGV size if GvEGV() looped back
6ec51ae0 54 * fix classic off-by-one error - the answer is strlen() + 1, not just strlen()
a52ceccd 55
db519f11 560.74_50 2011-04-19 nicholas
57 * Ensure that size() doesn't add the referent's size for non SVt_RV references
58
efc072e6 590.74 2011-04-19 nicholas
1507f6e0 60 * Correct the Makefile.PL - LICENSE was added to ExtUtils::MakeMaker in 6.31
61
eee00145 620.73_51 2011-04-17 nicholas
63 * Refactor the C code to accumulate the size inside the tracking structure
64 - this means that the C *_size() functions now return void
65 - The XS functions size() and total_size() now return UV, not IV
66
5f04c81d 670.73_50 2011-04-17 nicholas
68 * Avoid using assert() on ithreaded 5.8.8 and earlier, as it needs my_perl
5daca940 69 * Fix the test added in 0.73, as it tripped over a combination of bugs
5f04c81d 70
1f9c451d 710.73 2011-04-16 nicholas
72 * Revert a bad assertion introduced in 0.72_50, which was logically wrong.
73 - and a test to demonstrate one case that would trigger it
74
750.72_52 2011-04-15 nicholas
fb53b4e7 76 * Add a test for the non-exceptional warnings.
77
1f9c451d 780.72_51 2011-04-15 nicholas
265a0548 79 * Add PERL_NO_GET_CONTEXT to improve performance under multiplicity
80
795fc84c 810.72_50 2011-04-14 nicholas
1a36ac09 82 * Exception handling is totally MSVC specific, so only use it there
83 - this means that we don't need to use a C++ compiler anywhere
30fe4f47 84 * Rework bit-vector tracking mechanism to use a 256-way tree. This avoids
85 making assumptions about 64-bit platforms' memory layouts, and eliminates
86 the fatal error introduced in 0.72 when the assumption was violated.
568039a0 87 * Convert to XSLoader
795fc84c 88 * Resolve CPAN #49437 (Devel::Size adds magic in Perl 5.10)
66f50dda 89 * Resolve CPAN #58484 and #58485 (related to CVs that are XSUBs)
795fc84c 90
7e4f879b 910.72 2008-10-14 BrowserUk 70 tests
9fc9ab86 92 * Added bit-vector pointer tracking mechanism.
93 - new fatal error (64-bit platforms only)
94 * Added exception handling
95 - 4 new warnings (disabled by default)
96 * Updated POD to reflect above
97 * Added basic.t test 13
98 * replaced Module::Install with hand crafted EU::MM Makefile.pl
99 (With many thanks to Sisyphus)
100 because we couldn't work out how to add C++ options to allow
101 exception handling.
102
b7621729 1030.71 2008-08-24 Tels 69 tests
104 * adapt patch from Reini Urban to fix failing RV under 5.10 and 5.11. AV
105 and HV were pushed directly onto the pending_array, and not the RV,
106 which caused #33530. So he rewrote the logic to deref the RV inside
107 the array traversal. Applied this with one small omission, which
108 caused test faiures.
109 * Fixed 5.11 RV/IV logic. (Thanx Reini Urban!)
110 * Removed one duplicate total_size arrayref test. (Thanx Reini Urban!)
111 * changed //printf to dbg_printf() (Thanx Reini again!)
112
f44772ad 1130.70 2008-08-23 Tels 69 tests
114 * fix SEGFAULTS under v5.10 (Thanx Reini Urban!)
6ea94d90 115 * fix compilation under blead (Thanx Reini Urban!)
116 * require Perl 5.006 (to stop CPANTESTERS sending bogus reports
117 about failed compiling on ancient Perls)
f44772ad 118 * update the bundles Module::Install to v0.77
6ea94d90 119
5a83b7cf 1200.69 2007-08-11 Tels 69 tests
121 * fix compilation under Perl v5.9.5 and v5.10 (Thanx Steve Peters!)
122 * clarify the license by specifying Perl v5.8.8's license
b7621729 123 * small doc fixes, add a README file
5a83b7cf 124
c8db37d3 1250.68 2007-06-12 Tels 69 tests
126 * remove a bit of duplicate code in op_size, the second instance
127 was a no-op anyway
128 * fix the failing tests by rewriting the test logic to be sane
129 * the count of array slots was off by one
130
b1e5ad85 1310.67 2007-03-15 Tels 20 tests
132 * rip out Build.PL since it doesn't actually work
133 * fix bug #1653: All SVPV subtype may contain an RV instead of a PV
134 (Thanx a lot Jan, and sorry for the truly horrible long delay!)
135 * fix bug #24846: Does not correctly recurse into references in PVNV
136 (Thanx STEVIEO)
137 * add t/recurse.t that tests the two bugfixes in combination
138
1390.66 2007-03-02 Tels 15 tests
140 * fix the failing test (the code was right, but the test wrong)
141
0430b7f7 1420.65 2007-02-24 Tels 15 tests
143 * fix the test failings under 64 bit (bugs #21037, #18596, #21404)
144 * fix wrong size for strings (bug #17586)
145 * reverse Changelog (bug #16636)
146 * Size.xs: remove unused variable "count"
147 * Size.xs: fix warnings about missing ()
148 * Size.xs: fix size of array refs when AvALLOC == 0
149 * Modernize package:
150 + maintained by me, signed with my key
151 + bundle Module::Install and use it
152 + add POD tests
153 + rewrite tests to use Test::More; add more tests
e98cedbf 154
0430b7f7 1550.64 Mon Dec 19 18:46:00 2005
156 - Patch to make Devel::Size compile on bleadperl (Courtesy of
9fc9ab86 157 Nick Clark)
158 - Fix up the PVIV string size calcs (Courtesy of Andrew Shirrayev)
fea63ffa 159
0430b7f7 1600.63 Thu Jun 30 14:29:43 2005
161 - Turns out that OP_TRANS is a basic op, not a loop op. This
9fc9ab86 162 is a bug in versions of perl 5.8.x before 5.8.7. Work around
163 it.
b98fcdb9 164
0430b7f7 1650.62 Tue Jun 28 11:59:00 2005
166 - Took out // comments
9fc9ab86 167 - Added in copyright notice
168 - Some small amount of regex parsing
169 - Suppress multiple copies of each warning on each call
966a1570 170
0430b7f7 1710.61 Mon Jun 27 16:19:00 2005
172 - Added more checks for formats
9fc9ab86 173 - Got CVs sizing right
78dfb4e7 174
0430b7f7 1750.59 Sat Nov 27 16:42:42 2004
176 - Applied documentation and sane warning patch from Nigel Sandever
177 - Taught Devel::Size how to size up IO and globs properly
78dfb4e7 178
1790.58 Fri Jul 18 11:42:32 2003
180 - Fix for problems triggered by perl 5.8.0 and up, more tests, and
9fc9ab86 181 removal of an "our" for better backwards compatibility. (Courtesy
182 of Marcus Holland-Moritz <mhx-perl@gmx.net>)
5073b933 183
0430b7f7 1840.57 Thu Mar 20 13:21:14 2003
185 - setting $Devel::Size::warn to 0 disables not complete warnings
a3ec2202 186
0430b7f7 1870.56 Mon Feb 24 12:10:13 2003
188 - Chopped out some C++ comments. D'oh! Version incremented for CPAN
98ecbbc6 189
0430b7f7 1900.55 Sat Feb 22 17:21:00 2003
191 - Fixed a bad size calculation (we were overestimating by one byte)
9fc9ab86 192 - Updated the docs to show some of the places that there might be 'dark'
193 memory that Devel::Size can't see.
194 - Added in tests from Ken Williams
62691e7c 195
0430b7f7 1960.54 Sat Oct 12 14:11:00 2002
9fc9ab86 197 - Applied a patch to get it working on 5.8.0 under Tru64
0430b7f7 198
1990.53 Thu Oct 10 12:30:00 2002
200 - Finally started updating Changes file
9fc9ab86 201 - Applied doc patch from Ann Barcomb
202 - Got globs sizing right
0430b7f7 203
2040.01 Mon Oct 7 01:05:32 2002
9fc9ab86 205 - original version; created by h2xs 1.2 with options
206 -A -n Devel::Size
24d37977 207