ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/JSON-XS/Changes
(Generate patch)

Comparing JSON-XS/Changes (file contents):
Revision 1.123 by root, Tue Feb 17 23:29:38 2009 UTC vs.
Revision 1.148 by root, Tue Mar 8 10:18:50 2011 UTC

1Revision history for Perl extension JSON::XS 1Revision history for Perl extension JSON::XS
2 2
3TODO: maybe detetc and croak on more invalid inputs (e.g. +-inf/nan)
4
5 - properly croak on some invalid inputs that are not strings
6 (e.g. undef) when trying to decode a json text (reported
7 and analyzed by Goro Fuji).
8
92.3 Wed Aug 18 01:26:47 CEST 2010
10 - make sure decoder doesn't change the decoding in the incremental
11 parser (testcase provided by Hendrik Schumacher).
12 - applied patch by DaTa for Data::Dumper support in json_xs.
13 - added -t dump support to json_xs, using Data::Dump.
14 - added -f eval support to json_xs.
15
162.29 Wed Mar 17 02:39:12 CET 2010
17 - fix a memory leak when callbacks set using filter_json_object
18 or filter_json_single_key_object were called (great testcase
19 by Eric Wilhelm).
20
212.28 Thu Mar 11 20:30:46 CET 2010
22 - implement our own atof function - perl's can be orders of
23 magnitudes slower than even the system one. on the positive
24 side, ours seems to be more exact in general than perl's.
25 (testcase provided by Tim Meadowcroft).
26 - clarify floating point conversion issues a bit.
27 - update jpsykes csrf article url.
28 - updated benchmark section - JSON::PP became much faster!
29
302.27 Thu Jan 7 07:35:08 CET 2010
31 - support relaxed option inside the incremental parser
32 (testcase provided by IKEGAMI via Makamaka).
33
342.26 Sat Oct 10 03:26:19 CEST 2009
35 - big integers could become truncated (based on patch
36 by Strobl Anton).
37 - output format change: indent now adds a final newline, which is
38 more expected and more true to the documentation.
39
402.25 Sat Aug 8 12:04:41 CEST 2009
41 - the perl debugger completely breaks lvalue subs - try to work
42 around the issue.
43 - ignore RMAGICAL hashes w.r.t. CANONICAL.
44 - try to work around a possible char signedness issue on aix.
45 - require common sense.
46
472.24 Sat May 30 08:25:45 CEST 2009
48 - the incremental parser did not update its parse offset
49 pointer correctly when parsing utf8-strings (nicely
50 debugged by Martin Evans).
51 - appending a non-utf8-string to the incremental parser
52 in utf8 mode failed to upgrade the string.
53 - wording of parse error messages has been improved.
54
552.232 Sun Feb 22 11:12:25 CET 2009
56 - use an exponential algorithm to extend strings, to
57 help platforms with bad or abysmal==windows memory
58 allocater performance, at the expense of some memory
59 wastage (use shrink to recover this extra memory).
60 (nicely analysed by Dmitry Karasik).
61
622.2311 Thu Feb 19 02:12:54 CET 2009
3 - add a section "JSON and ECMAscript" to explain some 63 - add a section "JSON and ECMAscript" to explain some
4 incompatibilities between the two (problem was noted by 64 incompatibilities between the two (problem was noted by
5 various people). 65 various people).
6 - add t/20_faihu.t. 66 - add t/20_faihu.t.
7 67
82.231 Thu Nov 20 04:59:08 CET 2008 682.231 Thu Nov 20 04:59:08 CET 2008
9 - work around 5.10.0 magic bugs where manipulating magic values 69 - work around 5.10.0 magic bugs where manipulating magic values
10 (such as $1) would permanently damage them as perl would 70 (such as $1) would permanently damage them as perl would
11 ignore the magicalness, by making a full copy of the string, 71 ignore the magicalness, by making a full copy of the string,
12 reported by Dmitry Karasik. 72 reported by Dmitry Karasik.
13 - work around spurious wanrings under older perl 5.8's. 73 - work around spurious warnings under older perl 5.8's.
14 74
152.23 Mon Sep 29 05:08:29 CEST 2008 752.23 Mon Sep 29 05:08:29 CEST 2008
16 - fix a compilation problem when perl is not using char * as, well, 76 - fix a compilation problem when perl is not using char * as, well,
17 char *. 77 char *.
18 - use PL_hexdigit in favour of rolling our own. 78 - use PL_hexdigit in favour of rolling our own.

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines