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

Comparing JSON-XS/Changes (file contents):
Revision 1.6 by root, Fri Mar 23 15:11:06 2007 UTC vs.
Revision 1.100 by root, Wed Mar 19 15:17:53 2008 UTC

1Revision history for Perl extension JSON::XS 1Revision history for Perl extension JSON::XS
2 2
3TODO: document conversion algorithm 3 - update documentation here and there, add a large section
4 about utf8/latin1/ascii flags, add a security consideration.
5 - medium speed enhancements when encoding/decoding non-ascii chars.
6 - minor speedup in number encoding case.
7 - extend and clarify the section on incompatibilities
8 between YAML and JSON.
9 - switch to static inline from just inline when using gcc.
10 - add =encoding utf-8 to the manpage, now that perl 5.10 supports it.
11 - fix some issues with UV to JSON conversion of unknown impact.
12 - published the yahoo locals search result used in benchmarks as the
13 original url changes so comparison is impossible.
4 14
50.3 152.01 Wed Dec 5 11:40:28 CET 2007
16 - INCOMPATIBLE API CHANGE: to_json and from_json have been
17 renamed to encode_json/decode_json for JSON.pm compatibility.
18 The old functions croak and might be replaced by JSON.pm
19 comaptible versions in some later release.
20
212.0 Tue Dec 4 11:30:46 CET 2007
22 - this is supposed to be the first version of JSON::XS
23 compatible with version 2.0+ of the JSON module.
24 Using the JSON module as frontend to JSON::XS should be
25 as fast as using JSON::XS directly, so consider using it
26 instead.
27 - added get_* methods for all "simple" options.
28 - make JSON::XS subclassable.
29
301.53 Tue Nov 13 23:58:33 CET 2007
31 - minor doc clarifications.
32 - fixed many doc typos (patch by Thomas L. Shinnick).
33
341.52 Mon Oct 15 03:22:06 CEST 2007
35 - remove =encoding pod directive again, it confuses too many pod
36 parsers :/.
37
381.51 Sat Oct 13 03:55:56 CEST 2007
39 - encode empty arrays/hashes in a compact way when pretty is enabled.
40 - apparently JSON::XS was used to find some bugs in the
41 JSON_checker testsuite, so add (the corrected) JSON_checker tests to
42 the testsuite.
43 - quite a bit of doc updates/extension.
44 - require 5.8.2, as this seems to be the first unicode-stable version.
45
461.5 Tue Aug 28 04:05:38 CEST 2007
47 - add support for tied hashes, based on ideas and testcase by
48 Marcus Holland-Moritz.
49 - implemented relaxed parsing mode where some extensions are being
50 accepted. generation is still JSON-only.
51
521.44 Wed Aug 22 01:02:44 CEST 2007
53 - very experimental process-emulation support, slowing everything down.
54 the horribly broken perl threads are still not supported - YMMV.
55
561.43 Thu Jul 26 13:26:37 CEST 2007
57 - convert big json numbers exclusively consisting of digits to NV
58 only when there is no loss of precision, otherwise to string.
59
601.42 Tue Jul 24 00:51:18 CEST 2007
61 - fix a crash caused by not handling missing array elements
62 (report and testcase by Jay Kuri).
63
641.41 Tue Jul 10 18:21:44 CEST 2007
65 - fix compilation with NDEBUG (assert side-effect),
66 affects convert_blessed only.
67 - fix a bug in decode filters calling ENTER; SAVETMPS;
68 one time too often.
69 - catch a typical error in TO_JSON methods.
70 - antique-ised XS.xs again to work with outdated
71 C compilers (windows...).
72
731.4 Mon Jul 2 10:06:30 CEST 2007
74 - add convert_blessed setting.
75 - encode did not catch all blessed objects, encoding their
76 contents in most cases. This has been fixed by introducing
77 the allow_blessed setting.
78 - added filter_json_object and filter_json_single_key_object
79 settings that specify a callback to be called when
80 all/specific json objects are encountered.
81 - assume that most object keys are simple ascii words and
82 optimise this case, penalising the general case. This can
83 speed up decoding by 30% in typical cases and gives
84 a smaller and faster perl hash.
85 - implemented simpleminded, optional resource size checking
86 in decode_json.
87 - remove objToJson/jsonToObj aliases, as the next version
88 of JSON will not have them either.
89 - bit the bullet and converted the very simple json object
90 into a more complex one.
91 - work around a bug where perl wrongly claims an integer
92 is not an integer.
93 - unbundle JSON::XS::Boolean into own pm file so Storable
94 and similar modules can resolve the overloading when thawing.
95
961.3 Sun Jun 24 01:55:02 CEST 2007
97 - make JSON::XS::true and false special overloaded objects
98 and return those instead of 1 and 0 for those json atoms
99 (JSON::PP compatibility is NOT achieved yet).
100 - add JSON::XS::is_bool predicate to test for those special
101 values.
102 - add a reference to
103 http://jpsykes.com/47/practical-csrf-and-json-security.
104 - removed require 5.8.8 again, it is just not very expert-friendly.
105 Also try to be more compatible with slightly older versions,
106 which are not recommended (because they are buggy).
107
1081.24 Mon Jun 11 05:40:49 CEST 2007
109 - added informative section on JSON-as-YAML.
110 - get rid of some c99-isms again.
111 - localise dec->cur in decode_str, speeding up
112 string decoding considerably (>15% on my amd64 + gcc).
113 - increased SHORT_STRING_LEN to 16kb: stack space is
114 usually plenty, and this actually saves memory
115 when !shrinking as short strings will fit perfectly.
116
1171.23 Wed Jun 6 20:13:06 CEST 2007
118 - greatly improved small integer encoding and decoding speed.
119 - implement a number of µ-optimisations.
120 - updated benchmarks.
121
1221.22 Thu May 24 00:07:25 CEST 2007
123 - require 5.8.8 explicitly as older perls do not seem to offer
124 the required macros.
125 - possibly made it compile on so-called C compilers by microsoft.
126
1271.21 Wed May 9 18:40:32 CEST 2007
128 - character offset reported for trailing garbage was random.
129
1301.2 Wed May 9 18:35:01 CEST 2007
131 - decode did not work with magical scalars (doh!).
132 - added latin1 flag to produce JSON texts in the latin1 subset
133 of unicode.
134 - flag trailing garbage as error.
135 - new decode_prefix method that returns the number
136 of characters consumed by a decode.
137 - max octets/char in perls UTF-X is actually 13, not 11,
138 as pointed out by Glenn Linderman.
139 - fixed typoe reported by YAMASHINA Hio.
140
1411.11 Mon Apr 9 07:05:49 CEST 2007
142 - properly 0-terminate sv's returned by encode to help
143 C libraries that expect that 0 to be there.
144 - partially "port" JSON from C to microsofts fucking broken
145 pseudo-C. They should be burned to the ground for pissing
146 on standards. And I should be stoned for even trying to
147 support this filthy excuse for a c compiler.
148
1491.1 Wed Apr 4 01:45:00 CEST 2007
150 - clarify documentation (pointed out by Quinn Weaver).
151 - decode_utf8 sometimes did not correctly flag errors,
152 leading to segfaults.
153 - further reduced default nesting depth to 512 due to the test
154 failure by that anonymous "chris" whose e-mail address seems
155 to be impossible to get. Tests on other freebsd systems indicate
156 that this is likely a problem in his/her configuration and not this
157 module.
158 - renamed json => JSON in error messages.
159 - corrected the character offset in some error messages.
160
1611.01 Sat Mar 31 16:15:40 CEST 2007
162 - do not segfault when from_json/decode gets passed
163 a non-string object (reported by Florian Ragwitz).
164 This has no effect on normal operation.
165
1661.0 Thu Mar 29 04:43:34 CEST 2007
167 - the long awaited (by me) 1.0 version.
168 - add \0 (JSON::XS::false) and \1 (JSON::XS::true) mappings to JSON
169 true and false.
170 - add some more notes to shrink, as suggested by Alex Efros.
171 - improve testsuite.
172 - halve the default nesting depth limit, to hopefully make it
173 work on Freebsd (unfortunately, the cpan tester did not
174 send me his report, so I cannot ask about the stack limit on fbsd).
175
1760.8 Mon Mar 26 00:10:48 CEST 2007
177 - fix a memleak when decoding hashes.
178 - export jsonToBj and objToJson as aliases
179 to to_json and from_json, to reduce incompatibilities
180 between JSON/JSON::PC and JSON::XS. (experimental).
181 - implement a maximum nesting depth for both en- and de-coding.
182 - added a security considerations sections.
183
1840.7 Sun Mar 25 01:46:30 CET 2007
185 - code cleanup.
186 - fix a memory overflow bug when indenting.
187 - pretty-printing now up to 15% faster.
188 - improve decoding speed of strings by
189 up to 50% by specialcasing short strings.
190 - further decoding speedups for strings using
191 lots of \u escapes.
192 - improve utf8 decoding speed for U+80 .. U+7FF.
193
1940.5 Sat Mar 24 20:41:51 CET 2007
195 - added the UTF-16 encoding example hinted at in previous
196 versions.
197 - minor documentation fixes.
198 - fix a bug in and optimise canonicalising fastpath
199 (reported by Craig Manley).
200 - remove a subtest that breaks with bleadperl (reported
201 by Andreas König).
202
2030.31 Sat Mar 24 02:14:34 CET 2007
204 - documentation updates.
205 - do some casting to hopefully fix Andreas' problem.
206 - nuke bogus json rpc stuff.
207
2080.3 Fri Mar 23 19:33:21 CET 2007
6 - remove spurious PApp::Util reference (John McNamara). 209 - remove spurious PApp::Util reference (John McNamara).
7 - adapted lots of tests from other json modules 210 - adapted lots of tests from other json modules
8 (idea by Chris Carline). 211 (idea by Chris Carline).
212 - documented mapping from json to perl and vice versa.
213 - improved the documentation by adding more examples.
9 - added short escaping forms, reducing the created 214 - added short escaping forms, reducing the created
10 json texts a bit. 215 json texts a bit.
11 - added shrink flag. 216 - added shrink flag.
12 - when flag methods are called without enable argument 217 - when flag methods are called without enable argument
13 they will by default enable their flag. 218 they will by default enable their flag.
14 - considerably improved string encoding speed (at least 219 - considerably improved string encoding speed (at least
15 with gcc 4). 220 with gcc 4).
221 - added a test that covers lots of different characters.
222 - clarified some error messages.
223 - error messages now use correct character offset
224 with F_UTF8.
225 - improve the "no bytes" and "no warnings" hacks in
226 case the called functions do... stuff.
227 - croak when encoding to ascii and an out-of-range
228 (non-unicode) codepoint is encountered.
16 229
170.2 Fri Mar 23 00:23:34 CET 2007 2300.2 Fri Mar 23 00:23:34 CET 2007
18 - the "could not sleep without debuggign release". 231 - the "could not sleep without debuggign release".
19 it should basically work now, with many bugs as 232 it should basically work now, with many bugs as
20 no production tests have been run yet. 233 no production tests have been run yet.

Diff Legend

– Removed lines
+ Added lines
< Changed lines
> Changed lines