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

Comparing JSON-XS/Changes (file contents):
Revision 1.28 by root, Tue Apr 3 01:25:40 2007 UTC vs.
Revision 1.127 by root, Sat May 30 06:26:05 2009 UTC

1Revision history for Perl extension JSON::XS 1Revision history for Perl extension JSON::XS
2 2
3TODO: maybe use some overload magic to decode true and false into something 32.24 Sat May 30 08:25:45 CEST 2009
4 not 0 or 1 in perl. or use attached magic? 4 - the incremental parser did not update its parse offset
5 pointer correctly when parsing utf8-strings (nicely
6 debugged by Martin Evans).
7 - appending a non-utf8-string to the incremental parser
8 in utf8 mode failed to upgrade the string.
9 - wording of parse error messages has been improved.
5 10
112.232 Sun Feb 22 11:12:25 CET 2009
12 - use an exponential algorithm to extend strings, to
13 help platforms with bad or abysmal==windows memory
14 allocater performance, at the expense of some memory
15 wastage (use shrink to recover this extra memory).
16 (nicely analysed by Dmitry Karasik).
17
182.2311 Thu Feb 19 02:12:54 CET 2009
19 - add a section "JSON and ECMAscript" to explain some
20 incompatibilities between the two (problem was noted by
21 various people).
22 - add t/20_faihu.t.
23
242.231 Thu Nov 20 04:59:08 CET 2008
25 - work around 5.10.0 magic bugs where manipulating magic values
26 (such as $1) would permanently damage them as perl would
27 ignore the magicalness, by making a full copy of the string,
28 reported by Dmitry Karasik.
29 - work around spurious warnings under older perl 5.8's.
30
312.23 Mon Sep 29 05:08:29 CEST 2008
32 - fix a compilation problem when perl is not using char * as, well,
33 char *.
34 - use PL_hexdigit in favour of rolling our own.
35
362.2222 Sun Jul 20 18:49:00 CEST 2008
37 - same game again, broken 5.10 finds yet another assertion
38 failure, and the workaround causes additional runtime warnings.
39 Work around the next assertion AND the warning. 5.10 seriously
40 needs to adjust it's attitude against working code.
41
422.222 Sat Jul 19 06:15:34 CEST 2008
43 - you work around one -DDEBUGGING assertion bug in perl 5.10
44 just to hit the next one. work around this one, too.
45
462.22 Tue Jul 15 13:26:51 CEST 2008
47 - allow higher nesting levels in incremental parser.
48 - error out earlier in some cases in the incremental parser
49 (as suggested by Yuval Kogman).
50 - improve incr-parser test (Yuval Kogman).
51
522.21 Tue Jun 3 08:43:23 CEST 2008
53 - (hopefully) work around a perl 5.10 bug with -DDEBUGGING.
54 - remove the experimental status of the incremental parser interface.
55 - move =encoding around again, to avoid bugs with search.cpan.org.
56 when can we finally have utf-8 in pod???
57 - add ->incr_reset method.
58
592.2 Wed Apr 16 20:37:25 CEST 2008
60 - lifted the log2 rounding restriction of max_depth and max_size.
61 - make booleans mutable by creating a copy instead of handing out
62 the same scalar (reported by pasha sadri).
63 - added support for incremental json parsing (still EXPERIMENTAL).
64 - implemented and added a json_xs command line utility that can convert
65 from/to a number of serialisation formats - tell me if you need more.
66 - implement allow_unknown/get_allow_unknown methods.
67 - fixed documentation of max_depth w.r.t. higher and equal.
68 - moved down =encoding directive a bit, too much breaks if it's the first
69 pod directive :/.
70 - removed documentation section on other modules, it became somewhat
71 outdated and is nowadays mostly of historical interest.
72
732.1 Wed Mar 19 23:23:18 CET 2008
74 - update documentation here and there: add a large section
75 about utf8/latin1/ascii flags, add a security consideration
76 and extend and clarify the JSON and YAML section.
77 - medium speed enhancements when encoding/decoding non-ascii chars.
78 - minor speedup in number encoding case.
79 - extend and clarify the section on incompatibilities
80 between YAML and JSON.
81 - switch to static inline from just inline when using gcc.
82 - add =encoding utf-8 to the manpage, now that perl 5.10 supports it.
83 - fix some issues with UV to JSON conversion of unknown impact.
84 - published the yahoo locals search result used in benchmarks as the
85 original url changes so comparison is impossible.
86
872.01 Wed Dec 5 11:40:28 CET 2007
88 - INCOMPATIBLE API CHANGE: to_json and from_json have been
89 renamed to encode_json/decode_json for JSON.pm compatibility.
90 The old functions croak and might be replaced by JSON.pm
91 comaptible versions in some later release.
92
932.0 Tue Dec 4 11:30:46 CET 2007
94 - this is supposed to be the first version of JSON::XS
95 compatible with version 2.0+ of the JSON module.
96 Using the JSON module as frontend to JSON::XS should be
97 as fast as using JSON::XS directly, so consider using it
98 instead.
99 - added get_* methods for all "simple" options.
100 - make JSON::XS subclassable.
101
1021.53 Tue Nov 13 23:58:33 CET 2007
103 - minor doc clarifications.
104 - fixed many doc typos (patch by Thomas L. Shinnick).
105
1061.52 Mon Oct 15 03:22:06 CEST 2007
107 - remove =encoding pod directive again, it confuses too many pod
108 parsers :/.
109
1101.51 Sat Oct 13 03:55:56 CEST 2007
111 - encode empty arrays/hashes in a compact way when pretty is enabled.
112 - apparently JSON::XS was used to find some bugs in the
113 JSON_checker testsuite, so add (the corrected) JSON_checker tests to
114 the testsuite.
115 - quite a bit of doc updates/extension.
116 - require 5.8.2, as this seems to be the first unicode-stable version.
117
1181.5 Tue Aug 28 04:05:38 CEST 2007
119 - add support for tied hashes, based on ideas and testcase by
120 Marcus Holland-Moritz.
121 - implemented relaxed parsing mode where some extensions are being
122 accepted. generation is still JSON-only.
123
1241.44 Wed Aug 22 01:02:44 CEST 2007
125 - very experimental process-emulation support, slowing everything down.
126 the horribly broken perl threads are still not supported - YMMV.
127
1281.43 Thu Jul 26 13:26:37 CEST 2007
129 - convert big json numbers exclusively consisting of digits to NV
130 only when there is no loss of precision, otherwise to string.
131
1321.42 Tue Jul 24 00:51:18 CEST 2007
133 - fix a crash caused by not handling missing array elements
134 (report and testcase by Jay Kuri).
135
1361.41 Tue Jul 10 18:21:44 CEST 2007
137 - fix compilation with NDEBUG (assert side-effect),
138 affects convert_blessed only.
139 - fix a bug in decode filters calling ENTER; SAVETMPS;
140 one time too often.
141 - catch a typical error in TO_JSON methods.
142 - antique-ised XS.xs again to work with outdated
143 C compilers (windows...).
144
1451.4 Mon Jul 2 10:06:30 CEST 2007
146 - add convert_blessed setting.
147 - encode did not catch all blessed objects, encoding their
148 contents in most cases. This has been fixed by introducing
149 the allow_blessed setting.
150 - added filter_json_object and filter_json_single_key_object
151 settings that specify a callback to be called when
152 all/specific json objects are encountered.
153 - assume that most object keys are simple ascii words and
154 optimise this case, penalising the general case. This can
155 speed up decoding by 30% in typical cases and gives
156 a smaller and faster perl hash.
157 - implemented simpleminded, optional resource size checking
158 in decode_json.
159 - remove objToJson/jsonToObj aliases, as the next version
160 of JSON will not have them either.
161 - bit the bullet and converted the very simple json object
162 into a more complex one.
163 - work around a bug where perl wrongly claims an integer
164 is not an integer.
165 - unbundle JSON::XS::Boolean into own pm file so Storable
166 and similar modules can resolve the overloading when thawing.
167
1681.3 Sun Jun 24 01:55:02 CEST 2007
169 - make JSON::XS::true and false special overloaded objects
170 and return those instead of 1 and 0 for those json atoms
171 (JSON::PP compatibility is NOT achieved yet).
172 - add JSON::XS::is_bool predicate to test for those special
173 values.
174 - add a reference to
175 http://jpsykes.com/47/practical-csrf-and-json-security.
176 - removed require 5.8.8 again, it is just not very expert-friendly.
177 Also try to be more compatible with slightly older versions,
178 which are not recommended (because they are buggy).
179
1801.24 Mon Jun 11 05:40:49 CEST 2007
181 - added informative section on JSON-as-YAML.
182 - get rid of some c99-isms again.
183 - localise dec->cur in decode_str, speeding up
184 string decoding considerably (>15% on my amd64 + gcc).
185 - increased SHORT_STRING_LEN to 16kb: stack space is
186 usually plenty, and this actually saves memory
187 when !shrinking as short strings will fit perfectly.
188
1891.23 Wed Jun 6 20:13:06 CEST 2007
190 - greatly improved small integer encoding and decoding speed.
191 - implement a number of ยต-optimisations.
192 - updated benchmarks.
193
1941.22 Thu May 24 00:07:25 CEST 2007
195 - require 5.8.8 explicitly as older perls do not seem to offer
196 the required macros.
197 - possibly made it compile on so-called C compilers by microsoft.
198
1991.21 Wed May 9 18:40:32 CEST 2007
200 - character offset reported for trailing garbage was random.
201
2021.2 Wed May 9 18:35:01 CEST 2007
203 - decode did not work with magical scalars (doh!).
204 - added latin1 flag to produce JSON texts in the latin1 subset
205 of unicode.
206 - flag trailing garbage as error.
207 - new decode_prefix method that returns the number
208 of characters consumed by a decode.
209 - max octets/char in perls UTF-X is actually 13, not 11,
210 as pointed out by Glenn Linderman.
211 - fixed typoe reported by YAMASHINA Hio.
212
2131.11 Mon Apr 9 07:05:49 CEST 2007
214 - properly 0-terminate sv's returned by encode to help
215 C libraries that expect that 0 to be there.
216 - partially "port" JSON from C to microsofts fucking broken
217 pseudo-C. They should be burned to the ground for pissing
218 on standards. And I should be stoned for even trying to
219 support this filthy excuse for a c compiler.
220
2211.1 Wed Apr 4 01:45:00 CEST 2007
6 - clarify documentation (pointed out by Quinn Weaver). 222 - clarify documentation (pointed out by Quinn Weaver).
223 - decode_utf8 sometimes did not correctly flag errors,
224 leading to segfaults.
225 - further reduced default nesting depth to 512 due to the test
226 failure by that anonymous "chris" whose e-mail address seems
227 to be impossible to get. Tests on other freebsd systems indicate
228 that this is likely a problem in his/her configuration and not this
229 module.
230 - renamed json => JSON in error messages.
231 - corrected the character offset in some error messages.
7 232
81.01 Sat Mar 31 16:15:40 CEST 2007 2331.01 Sat Mar 31 16:15:40 CEST 2007
9 - do not segfault when from_json/decode gets passed 234 - do not segfault when from_json/decode gets passed
10 a non-string object (reported by Florian Ragwitz). 235 a non-string object (reported by Florian Ragwitz).
11 This has no effect on normal operation. 236 This has no effect on normal operation.
… …
73 case the called functions do... stuff. 298 case the called functions do... stuff.
74 - croak when encoding to ascii and an out-of-range 299 - croak when encoding to ascii and an out-of-range
75 (non-unicode) codepoint is encountered. 300 (non-unicode) codepoint is encountered.
76 301
770.2 Fri Mar 23 00:23:34 CET 2007 3020.2 Fri Mar 23 00:23:34 CET 2007
78 - the "could not sleep without debuggign release". 303 - the "could not sleep without debugging release".
79 it should basically work now, with many bugs as 304 it should basically work now, with many bugs as
80 no production tests have been run yet. 305 no production tests have been run yet.
81 - added more testcases. 306 - added more testcases.
82 - the expected shitload of bugfixes. 307 - the expected shitload of bugfixes.
83 - handle utf8 flag correctly in decode. 308 - handle utf8 flag correctly in decode.

Diff Legend

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