ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/JSON-XS/Changes
Revision: 1.79
Committed: Thu Oct 11 23:07:43 2007 UTC (16 years, 7 months ago) by root
Branch: MAIN
Changes since 1.78: +2 -0 lines
Log Message:
*** empty log message ***

File Contents

# Content
1 Revision history for Perl extension JSON::XS
2
3 - add JSON_checker tests to testsuite.
4 - some doc updates.
5 - require 5.8.2, as this seems to be the first unicode-stable version.
6
7 1.5 Tue Aug 28 04:05:38 CEST 2007
8 - add support for tied hashes, based on ideas and testcase by
9 Marcus Holland-Moritz.
10 - implemented relaxed parsing mode where some extensions are being
11 accepted. generation is still JSON-only.
12
13 1.44 Wed Aug 22 01:02:44 CEST 2007
14 - very experimental process-emulation support, slowing everything down.
15 the horribly broken perl threads are still not supported - YMMV.
16
17 1.43 Thu Jul 26 13:26:37 CEST 2007
18 - convert big json numbers exclusively consisting of digits to NV
19 only when there is no loss of precision, otherwise to string.
20
21 1.42 Tue Jul 24 00:51:18 CEST 2007
22 - fix a crash caused by not handling missing array elements
23 (report and testcase by Jay Kuri).
24
25 1.41 Tue Jul 10 18:21:44 CEST 2007
26 - fix compilation with NDEBUG (assert side-effect),
27 affects convert_blessed only.
28 - fix a bug in decode filters calling ENTER; SAVETMPS;
29 one time too often.
30 - catch a typical error in TO_JSON methods.
31 - antique-ised XS.xs again to work with outdated
32 C compilers (windows...).
33
34 1.4 Mon Jul 2 10:06:30 CEST 2007
35 - add convert_blessed setting.
36 - encode did not catch all blessed objects, encoding their
37 contents in most cases. This has been fixed by introducing
38 the allow_blessed setting.
39 - added filter_json_object and filter_json_single_key_object
40 settings that specify a callback to be called when
41 all/specific json objects are encountered.
42 - assume that most object keys are simple ascii words and
43 optimise this case, penalising the general case. This can
44 speed up decoding by 30% in typical cases and gives
45 a smaller and faster perl hash.
46 - implemented simpleminded, optional resource size checking
47 in decode_json.
48 - remove objToJson/jsonToObj aliases, as the next version
49 of JSON will not have them either.
50 - bit the bullet and converted the very simple json object
51 into a more complex one.
52 - work around a bug where perl wrongly claims an integer
53 is not an integer.
54 - unbundle JSON::XS::Boolean into own pm file so Storable
55 and similar modules can resolve the overloading when thawing.
56
57 1.3 Sun Jun 24 01:55:02 CEST 2007
58 - make JSON::XS::true and false special overloaded objects
59 and return those instead of 1 and 0 for those json atoms
60 (JSON::PP compatibility is NOT achieved yet).
61 - add JSON::XS::is_bool predicate to test for those special
62 values.
63 - add a reference to
64 http://jpsykes.com/47/practical-csrf-and-json-security.
65 - removed require 5.8.8 again, it is just not very expert-friendly.
66 Also try to be more compatible with slightly older versions,
67 which are not recommended (because they are buggy).
68
69 1.24 Mon Jun 11 05:40:49 CEST 2007
70 - added informative section on JSON-as-YAML.
71 - get rid of some c99-isms again.
72 - localise dec->cur in decode_str, speeding up
73 string decoding considerably (>15% on my amd64 + gcc).
74 - increased SHORT_STRING_LEN to 16kb: stack space is
75 usually plenty, and this actually saves memory
76 when !shrinking as short strings will fit perfectly.
77
78 1.23 Wed Jun 6 20:13:06 CEST 2007
79 - greatly improved small integer encoding and decoding speed.
80 - implement a number of µ-optimisations.
81 - updated benchmarks.
82
83 1.22 Thu May 24 00:07:25 CEST 2007
84 - require 5.8.8 explicitly as older perls do not seem to offer
85 the required macros.
86 - possibly made it compile on so-called C compilers by microsoft.
87
88 1.21 Wed May 9 18:40:32 CEST 2007
89 - character offset reported for trailing garbage was random.
90
91 1.2 Wed May 9 18:35:01 CEST 2007
92 - decode did not work with magical scalars (doh!).
93 - added latin1 flag to produce JSON texts in the latin1 subset
94 of unicode.
95 - flag trailing garbage as error.
96 - new decode_prefix method that returns the number
97 of characters consumed by a decode.
98 - max octets/char in perls UTF-X is actually 13, not 11,
99 as pointed out by Glenn Linderman.
100 - fixed typoe reported by YAMASHINA Hio.
101
102 1.11 Mon Apr 9 07:05:49 CEST 2007
103 - properly 0-terminate sv's returned by encode to help
104 C libraries that expect that 0 to be there.
105 - partially "port" JSON from C to microsofts fucking broken
106 pseudo-C. They should be burned to the ground for pissing
107 on standards. And I should be stoned for even trying to
108 support this filthy excuse for a c compiler.
109
110 1.1 Wed Apr 4 01:45:00 CEST 2007
111 - clarify documentation (pointed out by Quinn Weaver).
112 - decode_utf8 sometimes did not correctly flag errors,
113 leading to segfaults.
114 - further reduced default nesting depth to 512 due to the test
115 failure by that anonymous "chris" whose e-mail address seems
116 to be impossible to get. Tests on other freebsd systems indicate
117 that this is likely a problem in his/her configuration and not this
118 module.
119 - renamed json => JSON in error messages.
120 - corrected the character offset in some error messages.
121
122 1.01 Sat Mar 31 16:15:40 CEST 2007
123 - do not segfault when from_json/decode gets passed
124 a non-string object (reported by Florian Ragwitz).
125 This has no effect on normal operation.
126
127 1.0 Thu Mar 29 04:43:34 CEST 2007
128 - the long awaited (by me) 1.0 version.
129 - add \0 (JSON::XS::false) and \1 (JSON::XS::true) mappings to JSON
130 true and false.
131 - add some more notes to shrink, as suggested by Alex Efros.
132 - improve testsuite.
133 - halve the default nesting depth limit, to hopefully make it
134 work on Freebsd (unfortunately, the cpan tester did not
135 send me his report, so I cannot ask about the stack limit on fbsd).
136
137 0.8 Mon Mar 26 00:10:48 CEST 2007
138 - fix a memleak when decoding hashes.
139 - export jsonToBj and objToJson as aliases
140 to to_json and from_json, to reduce incompatibilities
141 between JSON/JSON::PC and JSON::XS. (experimental).
142 - implement a maximum nesting depth for both en- and de-coding.
143 - added a security considerations sections.
144
145 0.7 Sun Mar 25 01:46:30 CET 2007
146 - code cleanup.
147 - fix a memory overflow bug when indenting.
148 - pretty-printing now up to 15% faster.
149 - improve decoding speed of strings by
150 up to 50% by specialcasing short strings.
151 - further decoding speedups for strings using
152 lots of \u escapes.
153 - improve utf8 decoding speed for U+80 .. U+7FF.
154
155 0.5 Sat Mar 24 20:41:51 CET 2007
156 - added the UTF-16 encoding example hinted at in previous
157 versions.
158 - minor documentation fixes.
159 - fix a bug in and optimise canonicalising fastpath
160 (reported by Craig Manley).
161 - remove a subtest that breaks with bleadperl (reported
162 by Andreas König).
163
164 0.31 Sat Mar 24 02:14:34 CET 2007
165 - documentation updates.
166 - do some casting to hopefully fix Andreas' problem.
167 - nuke bogus json rpc stuff.
168
169 0.3 Fri Mar 23 19:33:21 CET 2007
170 - remove spurious PApp::Util reference (John McNamara).
171 - adapted lots of tests from other json modules
172 (idea by Chris Carline).
173 - documented mapping from json to perl and vice versa.
174 - improved the documentation by adding more examples.
175 - added short escaping forms, reducing the created
176 json texts a bit.
177 - added shrink flag.
178 - when flag methods are called without enable argument
179 they will by default enable their flag.
180 - considerably improved string encoding speed (at least
181 with gcc 4).
182 - added a test that covers lots of different characters.
183 - clarified some error messages.
184 - error messages now use correct character offset
185 with F_UTF8.
186 - improve the "no bytes" and "no warnings" hacks in
187 case the called functions do... stuff.
188 - croak when encoding to ascii and an out-of-range
189 (non-unicode) codepoint is encountered.
190
191 0.2 Fri Mar 23 00:23:34 CET 2007
192 - the "could not sleep without debuggign release".
193 it should basically work now, with many bugs as
194 no production tests have been run yet.
195 - added more testcases.
196 - the expected shitload of bugfixes.
197 - handle utf8 flag correctly in decode.
198 - fix segfault in decoder.
199 - utf8n_to_uvuni sets retlen to -1, but retlen is an
200 unsigned types (argh).
201 - fix decoding of utf-8 strings.
202 - improved error diagnostics.
203 - fix decoding of 'null'.
204 - fix parsing of empty array/hashes
205 - silence warnings when we prepare the croak message.
206
207 0.1 Thu Mar 22 22:13:43 CET 2007
208 - first release, very untested, basically just to claim
209 the namespace.
210
211 0.01 Thu Mar 22 06:08:12 CET 2007
212 - original version; cloned from Convert-Scalar
213