ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/JSON-XS/Changes
Revision: 1.120
Committed: Mon Sep 29 03:09:27 2008 UTC (15 years, 7 months ago) by root
Branch: MAIN
CVS Tags: rel-2_23
Changes since 1.119: +3 -0 lines
Log Message:
2.23

File Contents

# Content
1 Revision history for Perl extension JSON::XS
2
3 2.23 Mon Sep 29 05:08:29 CEST 2008
4 - fix a compilation problem when perl is not using char * as, well,
5 char *.
6 - use PL_hexdigit in favour of rolling our own.
7
8 2.2222 Sun Jul 20 18:49:00 CEST 2008
9 - same game again, broken 5.10 finds yet another assertion
10 failure, and the workaround causes additional runtime warnings.
11 Work around the next assertion AND the warning. 5.10 seriously
12 needs to adjust it's attitude against working code.
13
14 2.222 Sat Jul 19 06:15:34 CEST 2008
15 - you work around one -DDEBUGGING assertion bug in perl 5.10
16 just to hit the next one. work around this one, too.
17
18 2.22 Tue Jul 15 13:26:51 CEST 2008
19 - allow higher nesting levels in incremental parser.
20 - error out earlier in some cases in the incremental parser
21 (as suggested by Yuval Kogman).
22 - improve incr-parser test (Yuval Kogman).
23
24 2.21 Tue Jun 3 08:43:23 CEST 2008
25 - (hopefully) work around a perl 5.10 bug with -DDEBUGGING.
26 - remove the experimental status of the incremental parser interface.
27 - move =encoding around again, to avoid bugs with search.cpan.org.
28 when can we finally have utf-8 in pod???
29 - add ->incr_reset method.
30
31 2.2 Wed Apr 16 20:37:25 CEST 2008
32 - lifted the log2 rounding restriction of max_depth and max_size.
33 - make booleans mutable by creating a copy instead of handing out
34 the same scalar (reported by pasha sadri).
35 - added support for incremental json parsing (still EXPERIMENTAL).
36 - implemented and added a json_xs command line utility that can convert
37 from/to a number of serialisation formats - tell me if you need more.
38 - implement allow_unknown/get_allow_unknown methods.
39 - fixed documentation of max_depth w.r.t. higher and equal.
40 - moved down =encoding directive a bit, too much breaks if it's the first
41 pod directive :/.
42 - removed documentation section on other modules, it became somewhat
43 outdated and is nowadays mostly of historical interest.
44
45 2.1 Wed Mar 19 23:23:18 CET 2008
46 - update documentation here and there: add a large section
47 about utf8/latin1/ascii flags, add a security consideration
48 and extend and clarify the JSON and YAML section.
49 - medium speed enhancements when encoding/decoding non-ascii chars.
50 - minor speedup in number encoding case.
51 - extend and clarify the section on incompatibilities
52 between YAML and JSON.
53 - switch to static inline from just inline when using gcc.
54 - add =encoding utf-8 to the manpage, now that perl 5.10 supports it.
55 - fix some issues with UV to JSON conversion of unknown impact.
56 - published the yahoo locals search result used in benchmarks as the
57 original url changes so comparison is impossible.
58
59 2.01 Wed Dec 5 11:40:28 CET 2007
60 - INCOMPATIBLE API CHANGE: to_json and from_json have been
61 renamed to encode_json/decode_json for JSON.pm compatibility.
62 The old functions croak and might be replaced by JSON.pm
63 comaptible versions in some later release.
64
65 2.0 Tue Dec 4 11:30:46 CET 2007
66 - this is supposed to be the first version of JSON::XS
67 compatible with version 2.0+ of the JSON module.
68 Using the JSON module as frontend to JSON::XS should be
69 as fast as using JSON::XS directly, so consider using it
70 instead.
71 - added get_* methods for all "simple" options.
72 - make JSON::XS subclassable.
73
74 1.53 Tue Nov 13 23:58:33 CET 2007
75 - minor doc clarifications.
76 - fixed many doc typos (patch by Thomas L. Shinnick).
77
78 1.52 Mon Oct 15 03:22:06 CEST 2007
79 - remove =encoding pod directive again, it confuses too many pod
80 parsers :/.
81
82 1.51 Sat Oct 13 03:55:56 CEST 2007
83 - encode empty arrays/hashes in a compact way when pretty is enabled.
84 - apparently JSON::XS was used to find some bugs in the
85 JSON_checker testsuite, so add (the corrected) JSON_checker tests to
86 the testsuite.
87 - quite a bit of doc updates/extension.
88 - require 5.8.2, as this seems to be the first unicode-stable version.
89
90 1.5 Tue Aug 28 04:05:38 CEST 2007
91 - add support for tied hashes, based on ideas and testcase by
92 Marcus Holland-Moritz.
93 - implemented relaxed parsing mode where some extensions are being
94 accepted. generation is still JSON-only.
95
96 1.44 Wed Aug 22 01:02:44 CEST 2007
97 - very experimental process-emulation support, slowing everything down.
98 the horribly broken perl threads are still not supported - YMMV.
99
100 1.43 Thu Jul 26 13:26:37 CEST 2007
101 - convert big json numbers exclusively consisting of digits to NV
102 only when there is no loss of precision, otherwise to string.
103
104 1.42 Tue Jul 24 00:51:18 CEST 2007
105 - fix a crash caused by not handling missing array elements
106 (report and testcase by Jay Kuri).
107
108 1.41 Tue Jul 10 18:21:44 CEST 2007
109 - fix compilation with NDEBUG (assert side-effect),
110 affects convert_blessed only.
111 - fix a bug in decode filters calling ENTER; SAVETMPS;
112 one time too often.
113 - catch a typical error in TO_JSON methods.
114 - antique-ised XS.xs again to work with outdated
115 C compilers (windows...).
116
117 1.4 Mon Jul 2 10:06:30 CEST 2007
118 - add convert_blessed setting.
119 - encode did not catch all blessed objects, encoding their
120 contents in most cases. This has been fixed by introducing
121 the allow_blessed setting.
122 - added filter_json_object and filter_json_single_key_object
123 settings that specify a callback to be called when
124 all/specific json objects are encountered.
125 - assume that most object keys are simple ascii words and
126 optimise this case, penalising the general case. This can
127 speed up decoding by 30% in typical cases and gives
128 a smaller and faster perl hash.
129 - implemented simpleminded, optional resource size checking
130 in decode_json.
131 - remove objToJson/jsonToObj aliases, as the next version
132 of JSON will not have them either.
133 - bit the bullet and converted the very simple json object
134 into a more complex one.
135 - work around a bug where perl wrongly claims an integer
136 is not an integer.
137 - unbundle JSON::XS::Boolean into own pm file so Storable
138 and similar modules can resolve the overloading when thawing.
139
140 1.3 Sun Jun 24 01:55:02 CEST 2007
141 - make JSON::XS::true and false special overloaded objects
142 and return those instead of 1 and 0 for those json atoms
143 (JSON::PP compatibility is NOT achieved yet).
144 - add JSON::XS::is_bool predicate to test for those special
145 values.
146 - add a reference to
147 http://jpsykes.com/47/practical-csrf-and-json-security.
148 - removed require 5.8.8 again, it is just not very expert-friendly.
149 Also try to be more compatible with slightly older versions,
150 which are not recommended (because they are buggy).
151
152 1.24 Mon Jun 11 05:40:49 CEST 2007
153 - added informative section on JSON-as-YAML.
154 - get rid of some c99-isms again.
155 - localise dec->cur in decode_str, speeding up
156 string decoding considerably (>15% on my amd64 + gcc).
157 - increased SHORT_STRING_LEN to 16kb: stack space is
158 usually plenty, and this actually saves memory
159 when !shrinking as short strings will fit perfectly.
160
161 1.23 Wed Jun 6 20:13:06 CEST 2007
162 - greatly improved small integer encoding and decoding speed.
163 - implement a number of µ-optimisations.
164 - updated benchmarks.
165
166 1.22 Thu May 24 00:07:25 CEST 2007
167 - require 5.8.8 explicitly as older perls do not seem to offer
168 the required macros.
169 - possibly made it compile on so-called C compilers by microsoft.
170
171 1.21 Wed May 9 18:40:32 CEST 2007
172 - character offset reported for trailing garbage was random.
173
174 1.2 Wed May 9 18:35:01 CEST 2007
175 - decode did not work with magical scalars (doh!).
176 - added latin1 flag to produce JSON texts in the latin1 subset
177 of unicode.
178 - flag trailing garbage as error.
179 - new decode_prefix method that returns the number
180 of characters consumed by a decode.
181 - max octets/char in perls UTF-X is actually 13, not 11,
182 as pointed out by Glenn Linderman.
183 - fixed typoe reported by YAMASHINA Hio.
184
185 1.11 Mon Apr 9 07:05:49 CEST 2007
186 - properly 0-terminate sv's returned by encode to help
187 C libraries that expect that 0 to be there.
188 - partially "port" JSON from C to microsofts fucking broken
189 pseudo-C. They should be burned to the ground for pissing
190 on standards. And I should be stoned for even trying to
191 support this filthy excuse for a c compiler.
192
193 1.1 Wed Apr 4 01:45:00 CEST 2007
194 - clarify documentation (pointed out by Quinn Weaver).
195 - decode_utf8 sometimes did not correctly flag errors,
196 leading to segfaults.
197 - further reduced default nesting depth to 512 due to the test
198 failure by that anonymous "chris" whose e-mail address seems
199 to be impossible to get. Tests on other freebsd systems indicate
200 that this is likely a problem in his/her configuration and not this
201 module.
202 - renamed json => JSON in error messages.
203 - corrected the character offset in some error messages.
204
205 1.01 Sat Mar 31 16:15:40 CEST 2007
206 - do not segfault when from_json/decode gets passed
207 a non-string object (reported by Florian Ragwitz).
208 This has no effect on normal operation.
209
210 1.0 Thu Mar 29 04:43:34 CEST 2007
211 - the long awaited (by me) 1.0 version.
212 - add \0 (JSON::XS::false) and \1 (JSON::XS::true) mappings to JSON
213 true and false.
214 - add some more notes to shrink, as suggested by Alex Efros.
215 - improve testsuite.
216 - halve the default nesting depth limit, to hopefully make it
217 work on Freebsd (unfortunately, the cpan tester did not
218 send me his report, so I cannot ask about the stack limit on fbsd).
219
220 0.8 Mon Mar 26 00:10:48 CEST 2007
221 - fix a memleak when decoding hashes.
222 - export jsonToBj and objToJson as aliases
223 to to_json and from_json, to reduce incompatibilities
224 between JSON/JSON::PC and JSON::XS. (experimental).
225 - implement a maximum nesting depth for both en- and de-coding.
226 - added a security considerations sections.
227
228 0.7 Sun Mar 25 01:46:30 CET 2007
229 - code cleanup.
230 - fix a memory overflow bug when indenting.
231 - pretty-printing now up to 15% faster.
232 - improve decoding speed of strings by
233 up to 50% by specialcasing short strings.
234 - further decoding speedups for strings using
235 lots of \u escapes.
236 - improve utf8 decoding speed for U+80 .. U+7FF.
237
238 0.5 Sat Mar 24 20:41:51 CET 2007
239 - added the UTF-16 encoding example hinted at in previous
240 versions.
241 - minor documentation fixes.
242 - fix a bug in and optimise canonicalising fastpath
243 (reported by Craig Manley).
244 - remove a subtest that breaks with bleadperl (reported
245 by Andreas König).
246
247 0.31 Sat Mar 24 02:14:34 CET 2007
248 - documentation updates.
249 - do some casting to hopefully fix Andreas' problem.
250 - nuke bogus json rpc stuff.
251
252 0.3 Fri Mar 23 19:33:21 CET 2007
253 - remove spurious PApp::Util reference (John McNamara).
254 - adapted lots of tests from other json modules
255 (idea by Chris Carline).
256 - documented mapping from json to perl and vice versa.
257 - improved the documentation by adding more examples.
258 - added short escaping forms, reducing the created
259 json texts a bit.
260 - added shrink flag.
261 - when flag methods are called without enable argument
262 they will by default enable their flag.
263 - considerably improved string encoding speed (at least
264 with gcc 4).
265 - added a test that covers lots of different characters.
266 - clarified some error messages.
267 - error messages now use correct character offset
268 with F_UTF8.
269 - improve the "no bytes" and "no warnings" hacks in
270 case the called functions do... stuff.
271 - croak when encoding to ascii and an out-of-range
272 (non-unicode) codepoint is encountered.
273
274 0.2 Fri Mar 23 00:23:34 CET 2007
275 - the "could not sleep without debugging release".
276 it should basically work now, with many bugs as
277 no production tests have been run yet.
278 - added more testcases.
279 - the expected shitload of bugfixes.
280 - handle utf8 flag correctly in decode.
281 - fix segfault in decoder.
282 - utf8n_to_uvuni sets retlen to -1, but retlen is an
283 unsigned types (argh).
284 - fix decoding of utf-8 strings.
285 - improved error diagnostics.
286 - fix decoding of 'null'.
287 - fix parsing of empty array/hashes
288 - silence warnings when we prepare the croak message.
289
290 0.1 Thu Mar 22 22:13:43 CET 2007
291 - first release, very untested, basically just to claim
292 the namespace.
293
294 0.01 Thu Mar 22 06:08:12 CET 2007
295 - original version; cloned from Convert-Scalar
296