ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/JSON-XS/Changes
Revision: 1.62
Committed: Sun Jul 1 22:20:00 2007 UTC (16 years, 10 months ago) by root
Branch: MAIN
Changes since 1.61: +2 -1 lines
Log Message:
*** empty log message ***

File Contents

# Content
1 Revision history for Perl extension JSON::XS
2
3 TODO: object filtering (decode_filtered?)
4 TODO: make xs-level object bigger, longer
5 1.4
6 - add convert_blessed option.
7 - encode did not catch all blessed objects, encoding their
8 contents in most cases. This has been fixed by introducing
9 the allow_blessed setting.
10 - assume that most object keys are simple ascii words and
11 optimise this case, penalising the general case. This can
12 speed up decoding by 30% in typical cases and gives
13 a smaller and faster perl hash.
14 - implemented simpleminded, optional resource size checking
15 in decode_json.
16 - remove objToJson/jsonToObj aliases, as the next version
17 of JSON will not have them either.
18 - bit the bullet and converted the very simple json object
19 into a more complex one.
20
21 1.3 Sun Jun 24 01:55:02 CEST 2007
22 - make JSON::XS::true and false special overloaded objects
23 and return those instead of 1 and 0 for those json atoms
24 (JSON::PP compatibility is NOT achieved yet).
25 - add JSON::XS::is_bool predicate to test for those special
26 values.
27 - add a reference to
28 http://jpsykes.com/47/practical-csrf-and-json-security.
29 - removed require 5.8.8 again, it is just not very expert-friendly.
30 Also try to be more compatible with slightly older versions,
31 which are not recommended (because they are buggy).
32
33 1.24 Mon Jun 11 05:40:49 CEST 2007
34 - added informative section on JSON-as-YAML.
35 - get rid of some c99-isms again.
36 - localise dec->cur in decode_str, speeding up
37 string decoding considerably (>15% on my amd64 + gcc).
38 - increased SHORT_STRING_LEN to 16kb: stack space is
39 usually plenty, and this actually saves memory
40 when !shrinking as short strings will fit perfectly.
41
42 1.23 Wed Jun 6 20:13:06 CEST 2007
43 - greatly improved small integer encoding and decoding speed.
44 - implement a number of µ-optimisations.
45 - updated benchmarks.
46
47 1.22 Thu May 24 00:07:25 CEST 2007
48 - require 5.8.8 explicitly as older perls do not seem to offer
49 the required macros.
50 - possibly made it compile on so-called C compilers by microsoft.
51
52 1.21 Wed May 9 18:40:32 CEST 2007
53 - character offset reported for trailing garbage was random.
54
55 1.2 Wed May 9 18:35:01 CEST 2007
56 - decode did not work with magical scalars (doh!).
57 - added latin1 flag to produce JSON texts in the latin1 subset
58 of unicode.
59 - flag trailing garbage as error.
60 - new decode_prefix method that returns the number
61 of characters consumed by a decode.
62 - max octets/char in perls UTF-X is actually 13, not 11,
63 as pointed out by Glenn Linderman.
64 - fixed typoe reported by YAMASHINA Hio.
65
66 1.11 Mon Apr 9 07:05:49 CEST 2007
67 - properly 0-terminate sv's returned by encode to help
68 C libraries that expect that 0 to be there.
69 - partially "port" JSON from C to microsofts fucking broken
70 pseudo-C. They should be burned to the ground for pissing
71 on standards. And I should be stoned for even trying to
72 support this filthy excuse for a c compiler.
73
74 1.1 Wed Apr 4 01:45:00 CEST 2007
75 - clarify documentation (pointed out by Quinn Weaver).
76 - decode_utf8 sometimes did not correctly flag errors,
77 leading to segfaults.
78 - further reduced default nesting depth to 512 due to the test
79 failure by that anonymous "chris" whose e-mail address seems
80 to be impossible to get. Tests on other freebsd systems indicate
81 that this is likely a problem in his/her configuration and not this
82 module.
83 - renamed json => JSON in error messages.
84 - corrected the character offset in some error messages.
85
86 1.01 Sat Mar 31 16:15:40 CEST 2007
87 - do not segfault when from_json/decode gets passed
88 a non-string object (reported by Florian Ragwitz).
89 This has no effect on normal operation.
90
91 1.0 Thu Mar 29 04:43:34 CEST 2007
92 - the long awaited (by me) 1.0 version.
93 - add \0 (JSON::XS::false) and \1 (JSON::XS::true) mappings to JSON
94 true and false.
95 - add some more notes to shrink, as suggested by Alex Efros.
96 - improve testsuite.
97 - halve the default nesting depth limit, to hopefully make it
98 work on Freebsd (unfortunately, the cpan tester did not
99 send me his report, so I cannot ask about the stack limit on fbsd).
100
101 0.8 Mon Mar 26 00:10:48 CEST 2007
102 - fix a memleak when decoding hashes.
103 - export jsonToBj and objToJson as aliases
104 to to_json and from_json, to reduce incompatibilities
105 between JSON/JSON::PC and JSON::XS. (experimental).
106 - implement a maximum nesting depth for both en- and de-coding.
107 - added a security considerations sections.
108
109 0.7 Sun Mar 25 01:46:30 CET 2007
110 - code cleanup.
111 - fix a memory overflow bug when indenting.
112 - pretty-printing now up to 15% faster.
113 - improve decoding speed of strings by
114 up to 50% by specialcasing short strings.
115 - further decoding speedups for strings using
116 lots of \u escapes.
117 - improve utf8 decoding speed for U+80 .. U+7FF.
118
119 0.5 Sat Mar 24 20:41:51 CET 2007
120 - added the UTF-16 encoding example hinted at in previous
121 versions.
122 - minor documentation fixes.
123 - fix a bug in and optimise canonicalising fastpath
124 (reported by Craig Manley).
125 - remove a subtest that breaks with bleadperl (reported
126 by Andreas König).
127
128 0.31 Sat Mar 24 02:14:34 CET 2007
129 - documentation updates.
130 - do some casting to hopefully fix Andreas' problem.
131 - nuke bogus json rpc stuff.
132
133 0.3 Fri Mar 23 19:33:21 CET 2007
134 - remove spurious PApp::Util reference (John McNamara).
135 - adapted lots of tests from other json modules
136 (idea by Chris Carline).
137 - documented mapping from json to perl and vice versa.
138 - improved the documentation by adding more examples.
139 - added short escaping forms, reducing the created
140 json texts a bit.
141 - added shrink flag.
142 - when flag methods are called without enable argument
143 they will by default enable their flag.
144 - considerably improved string encoding speed (at least
145 with gcc 4).
146 - added a test that covers lots of different characters.
147 - clarified some error messages.
148 - error messages now use correct character offset
149 with F_UTF8.
150 - improve the "no bytes" and "no warnings" hacks in
151 case the called functions do... stuff.
152 - croak when encoding to ascii and an out-of-range
153 (non-unicode) codepoint is encountered.
154
155 0.2 Fri Mar 23 00:23:34 CET 2007
156 - the "could not sleep without debuggign release".
157 it should basically work now, with many bugs as
158 no production tests have been run yet.
159 - added more testcases.
160 - the expected shitload of bugfixes.
161 - handle utf8 flag correctly in decode.
162 - fix segfault in decoder.
163 - utf8n_to_uvuni sets retlen to -1, but retlen is an
164 unsigned types (argh).
165 - fix decoding of utf-8 strings.
166 - improved error diagnostics.
167 - fix decoding of 'null'.
168 - fix parsing of empty array/hashes
169 - silence warnings when we prepare the croak message.
170
171 0.1 Thu Mar 22 22:13:43 CET 2007
172 - first release, very untested, basically just to claim
173 the namespace.
174
175 0.01 Thu Mar 22 06:08:12 CET 2007
176 - original version; cloned from Convert-Scalar
177