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

Comparing JSON-XS/Changes (file contents):
Revision 1.58 by root, Mon Jun 25 06:57:42 2007 UTC vs.
Revision 1.171 by root, Thu Sep 1 11:38:58 2016 UTC

1Revision history for Perl extension JSON::XS 1Revision history for Perl extension JSON::XS
2 2
3TODO: blessed object tests + maxsize 3TODO: maybe detetc and croak on more invalid inputs (e.g. +-inf/nan)
41.4 4TODO: maybe avoid the reblessing and better support readonly objects.
5TODO: http://stevehanov.ca/blog/index.php?id=104 compression
6TODO: how to cope with tagged values and standard json decoders
7TODO: investigate magic (Eric Brine)
8TODO: rfc7464
9TODO: Victor Efimov
10
11 - json_xs: toformat failure error message fixed.
12
133.02 Fri Feb 26 22:45:20 CET 2016
14 - allow_nonref now affects booleans (\1, $Types::Serialiser::Boolean)
15 as well (reported by Alex Efros).
16 - allow literal tabs in strings in relaxed mode (patch by
17 lubo.rintel@gooddata.com).
18 - support "cbor" format in json_xs tool.
19 - support (and fix) calling encode and decode in list context
20 (reported by Вадим Власов).
21 - work around a bug in older perls crashing when presented
22 with shared hash keys (Reini Urban).
23 - use stability canary.
24
253.01 Tue Oct 29 16:55:15 CET 2013
26 - backport to perls < 5.18 (reported by Paul Howarth).
27
283.0 Tue Oct 29 01:35:37 CET 2013
29 - implemented an object tagging extension (using the
30 Types::Serialiser serialisation protocol).
31 - reworked the documentation regarding object serialisation,
32 add a new OBJECT SERIALISATION section that explains th
33 whole process.
34 - new setting: allow_tags.
35 - switch to Types::Serialiser booleans.
36 - remove to_json/from_json.
37 - other minor improvements to the documentation.
38
392.34 Thu May 23 11:30:34 CEST 2013
40 - work around bugs in perl 5.18 breaking more than 100
41 widely used modules, without a fix in sight because
42 p5pers don't care about CPAN anymore.
43 - when canonicalising, only allocate up to 64 hash key
44 pointers on the stack. for larger hashes, use the heap,
45 to avoid using too much stackspace.
46 - discuss the problem with setlocale (reported by a few victims).
47
482.33 Wed Aug 1 21:03:52 CEST 2012
49 - internal encode/decode XS wrappers did not expect stack
50 moves caused by callbacks (analyzed and testcase by Jesse Luehrs).
51 - add bencode as to/from option in bin/json_xs.
52 - add -e option to json_xs, and none and string in/out formats.
53
542.32 Thu Aug 11 19:06:38 CEST 2011
55 - fix a bug in the initial whitespace accumulation.
56
572.31 Wed Jul 27 17:53:05 CEST 2011
58 - don't accumulate initial whitespace in the incremental buffer
59 (this can be useful to allow whitespace-keepalive on a tcp
60 connection without triggering the max_size limit).
61 - properly croak on some invalid inputs that are not strings
62 (e.g. undef) when trying to decode a json text (reported
63 and analyzed by Goro Fuji).
64
652.3 Wed Aug 18 01:26:47 CEST 2010
66 - make sure decoder doesn't change the decoding in the incremental
67 parser (testcase provided by Hendrik Schumacher).
68 - applied patch by DaTa for Data::Dumper support in json_xs.
69 - added -t dump support to json_xs, using Data::Dump.
70 - added -f eval support to json_xs.
71
722.29 Wed Mar 17 02:39:12 CET 2010
73 - fix a memory leak when callbacks set using filter_json_object
74 or filter_json_single_key_object were called (great testcase
75 by Eric Wilhelm).
76
772.28 Thu Mar 11 20:30:46 CET 2010
78 - implement our own atof function - perl's can be orders of
79 magnitudes slower than even the system one. on the positive
80 side, ours seems to be more exact in general than perl's.
81 (testcase provided by Tim Meadowcroft).
82 - clarify floating point conversion issues a bit.
83 - update jpsykes csrf article url.
84 - updated benchmark section - JSON::PP became much faster!
85
862.27 Thu Jan 7 07:35:08 CET 2010
87 - support relaxed option inside the incremental parser
88 (testcase provided by IKEGAMI via Makamaka).
89
902.26 Sat Oct 10 03:26:19 CEST 2009
91 - big integers could become truncated (based on patch
92 by Strobl Anton).
93 - output format change: indent now adds a final newline, which is
94 more expected and more true to the documentation.
95
962.25 Sat Aug 8 12:04:41 CEST 2009
97 - the perl debugger completely breaks lvalue subs - try to work
98 around the issue.
99 - ignore RMAGICAL hashes w.r.t. CANONICAL.
100 - try to work around a possible char signedness issue on aix.
101 - require common sense.
102
1032.24 Sat May 30 08:25:45 CEST 2009
104 - the incremental parser did not update its parse offset
105 pointer correctly when parsing utf8-strings (nicely
106 debugged by Martin Evans).
107 - appending a non-utf8-string to the incremental parser
108 in utf8 mode failed to upgrade the string.
109 - wording of parse error messages has been improved.
110
1112.232 Sun Feb 22 11:12:25 CET 2009
112 - use an exponential algorithm to extend strings, to
113 help platforms with bad or abysmal==windows memory
114 allocater performance, at the expense of some memory
115 wastage (use shrink to recover this extra memory).
116 (nicely analysed by Dmitry Karasik).
117
1182.2311 Thu Feb 19 02:12:54 CET 2009
119 - add a section "JSON and ECMAscript" to explain some
120 incompatibilities between the two (problem was noted by
121 various people).
122 - add t/20_faihu.t.
123
1242.231 Thu Nov 20 04:59:08 CET 2008
125 - work around 5.10.0 magic bugs where manipulating magic values
126 (such as $1) would permanently damage them as perl would
127 ignore the magicalness, by making a full copy of the string,
128 reported by Dmitry Karasik.
129 - work around spurious warnings under older perl 5.8's.
130
1312.23 Mon Sep 29 05:08:29 CEST 2008
132 - fix a compilation problem when perl is not using char * as, well,
133 char *.
134 - use PL_hexdigit in favour of rolling our own.
135
1362.2222 Sun Jul 20 18:49:00 CEST 2008
137 - same game again, broken 5.10 finds yet another assertion
138 failure, and the workaround causes additional runtime warnings.
139 Work around the next assertion AND the warning. 5.10 seriously
140 needs to adjust it's attitude against working code.
141
1422.222 Sat Jul 19 06:15:34 CEST 2008
143 - you work around one -DDEBUGGING assertion bug in perl 5.10
144 just to hit the next one. work around this one, too.
145
1462.22 Tue Jul 15 13:26:51 CEST 2008
147 - allow higher nesting levels in incremental parser.
148 - error out earlier in some cases in the incremental parser
149 (as suggested by Yuval Kogman).
150 - improve incr-parser test (Yuval Kogman).
151
1522.21 Tue Jun 3 08:43:23 CEST 2008
153 - (hopefully) work around a perl 5.10 bug with -DDEBUGGING.
154 - remove the experimental status of the incremental parser interface.
155 - move =encoding around again, to avoid bugs with search.cpan.org.
156 when can we finally have utf-8 in pod???
157 - add ->incr_reset method.
158
1592.2 Wed Apr 16 20:37:25 CEST 2008
160 - lifted the log2 rounding restriction of max_depth and max_size.
161 - make booleans mutable by creating a copy instead of handing out
162 the same scalar (reported by pasha sadri).
163 - added support for incremental json parsing (still EXPERIMENTAL).
164 - implemented and added a json_xs command line utility that can convert
165 from/to a number of serialisation formats - tell me if you need more.
166 - implement allow_unknown/get_allow_unknown methods.
167 - fixed documentation of max_depth w.r.t. higher and equal.
168 - moved down =encoding directive a bit, too much breaks if it's the first
169 pod directive :/.
170 - removed documentation section on other modules, it became somewhat
171 outdated and is nowadays mostly of historical interest.
172
1732.1 Wed Mar 19 23:23:18 CET 2008
174 - update documentation here and there: add a large section
175 about utf8/latin1/ascii flags, add a security consideration
176 and extend and clarify the JSON and YAML section.
177 - medium speed enhancements when encoding/decoding non-ascii chars.
178 - minor speedup in number encoding case.
179 - extend and clarify the section on incompatibilities
180 between YAML and JSON.
181 - switch to static inline from just inline when using gcc.
182 - add =encoding utf-8 to the manpage, now that perl 5.10 supports it.
183 - fix some issues with UV to JSON conversion of unknown impact.
184 - published the yahoo locals search result used in benchmarks as the
185 original url changes so comparison is impossible.
186
1872.01 Wed Dec 5 11:40:28 CET 2007
188 - INCOMPATIBLE API CHANGE: to_json and from_json have been
189 renamed to encode_json/decode_json for JSON.pm compatibility.
190 The old functions croak and might be replaced by JSON.pm
191 comaptible versions in some later release.
192
1932.0 Tue Dec 4 11:30:46 CET 2007
194 - this is supposed to be the first version of JSON::XS
195 compatible with version 2.0+ of the JSON module.
196 Using the JSON module as frontend to JSON::XS should be
197 as fast as using JSON::XS directly, so consider using it
198 instead.
199 - added get_* methods for all "simple" options.
200 - make JSON::XS subclassable.
201
2021.53 Tue Nov 13 23:58:33 CET 2007
203 - minor doc clarifications.
204 - fixed many doc typos (patch by Thomas L. Shinnick).
205
2061.52 Mon Oct 15 03:22:06 CEST 2007
207 - remove =encoding pod directive again, it confuses too many pod
208 parsers :/.
209
2101.51 Sat Oct 13 03:55:56 CEST 2007
211 - encode empty arrays/hashes in a compact way when pretty is enabled.
212 - apparently JSON::XS was used to find some bugs in the
213 JSON_checker testsuite, so add (the corrected) JSON_checker tests to
214 the testsuite.
215 - quite a bit of doc updates/extension.
216 - require 5.8.2, as this seems to be the first unicode-stable version.
217
2181.5 Tue Aug 28 04:05:38 CEST 2007
219 - add support for tied hashes, based on ideas and testcase by
220 Marcus Holland-Moritz.
221 - implemented relaxed parsing mode where some extensions are being
222 accepted. generation is still JSON-only.
223
2241.44 Wed Aug 22 01:02:44 CEST 2007
225 - very experimental process-emulation support, slowing everything down.
226 the horribly broken perl threads are still not supported - YMMV.
227
2281.43 Thu Jul 26 13:26:37 CEST 2007
229 - convert big json numbers exclusively consisting of digits to NV
230 only when there is no loss of precision, otherwise to string.
231
2321.42 Tue Jul 24 00:51:18 CEST 2007
233 - fix a crash caused by not handling missing array elements
234 (report and testcase by Jay Kuri).
235
2361.41 Tue Jul 10 18:21:44 CEST 2007
237 - fix compilation with NDEBUG (assert side-effect),
238 affects convert_blessed only.
239 - fix a bug in decode filters calling ENTER; SAVETMPS;
240 one time too often.
241 - catch a typical error in TO_JSON methods.
242 - antique-ised XS.xs again to work with outdated
243 C compilers (windows...).
244
2451.4 Mon Jul 2 10:06:30 CEST 2007
5 - add convert_blessed option. 246 - add convert_blessed setting.
6 - encode did not catch all blessed objects, encoding their 247 - encode did not catch all blessed objects, encoding their
7 contents in most cases. This has been fixed by introducing 248 contents in most cases. This has been fixed by introducing
8 the allow_blessed setting. 249 the allow_blessed setting.
250 - added filter_json_object and filter_json_single_key_object
251 settings that specify a callback to be called when
252 all/specific json objects are encountered.
253 - assume that most object keys are simple ascii words and
254 optimise this case, penalising the general case. This can
255 speed up decoding by 30% in typical cases and gives
256 a smaller and faster perl hash.
9 - implemented simpleminded, optional resource size checking 257 - implemented simpleminded, optional resource size checking
10 in decode_json. 258 in decode_json.
259 - remove objToJson/jsonToObj aliases, as the next version
260 of JSON will not have them either.
261 - bit the bullet and converted the very simple json object
262 into a more complex one.
263 - work around a bug where perl wrongly claims an integer
264 is not an integer.
265 - unbundle JSON::XS::Boolean into own pm file so Storable
266 and similar modules can resolve the overloading when thawing.
11 267
121.3 Sun Jun 24 01:55:02 CEST 2007 2681.3 Sun Jun 24 01:55:02 CEST 2007
13 - make JSON::XS::true and false special overloaded objects 269 - make JSON::XS::true and false special overloaded objects
14 and return those instead of 1 and 0 for those json atoms 270 and return those instead of 1 and 0 for those json atoms
15 (JSON::PP compatibility is NOT achieved yet). 271 (JSON::PP compatibility is NOT achieved yet).
16 - add JSON::XS::is_bool predicate to test for those special 272 - add JSON::XS::is_bool predicate to test for those special
17 values. 273 values.
18 - add a reference to 274 - add a reference to
19 http://jpsykes.com/47/practical-csrf-and-json-security. 275 http://jpsykes.com/47/practical-csrf-and-json-security.
20 - removed require 5.8.8 again, it just not very expert-friendly, 276 - removed require 5.8.8 again, it is just not very expert-friendly.
21 and try to be more compatible with slightly older versions, 277 Also try to be more compatible with slightly older versions,
22 which are not recommended (because they are buggy). 278 which are not recommended (because they are buggy).
23 279
241.24 Mon Jun 11 05:40:49 CEST 2007 2801.24 Mon Jun 11 05:40:49 CEST 2007
25 - added informative section on JSON-as-YAML. 281 - added informative section on JSON-as-YAML.
26 - get rid of some c99-isms again. 282 - get rid of some c99-isms again.
142 case the called functions do... stuff. 398 case the called functions do... stuff.
143 - croak when encoding to ascii and an out-of-range 399 - croak when encoding to ascii and an out-of-range
144 (non-unicode) codepoint is encountered. 400 (non-unicode) codepoint is encountered.
145 401
1460.2 Fri Mar 23 00:23:34 CET 2007 4020.2 Fri Mar 23 00:23:34 CET 2007
147 - the "could not sleep without debuggign release". 403 - the "could not sleep without debugging release".
148 it should basically work now, with many bugs as 404 it should basically work now, with many bugs as
149 no production tests have been run yet. 405 no production tests have been run yet.
150 - added more testcases. 406 - added more testcases.
151 - the expected shitload of bugfixes. 407 - the expected shitload of bugfixes.
152 - handle utf8 flag correctly in decode. 408 - handle utf8 flag correctly in decode.

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines