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

Comparing JSON-XS/Changes (file contents):
Revision 1.49 by root, Mon Jun 11 02:58:10 2007 UTC vs.
Revision 1.169 by root, Fri Feb 26 21:46:45 2016 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 3TODO: maybe detetc and croak on more invalid inputs (e.g. +-inf/nan)
4 not 0 or 1 in perl. or use attached magic? 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)
5 8
6investigate the possibility of a yaml-compatible output format. 93.02 Fri Feb 26 22:45:20 CET 2016
10 - allow_nonref now affects booleans (\1, $Types::Serialiser::Boolean)
11 as well (reported by Alex Efros).
12 - allow literal tabs in strings in relaxed mode (patch by
13 lubo.rintel@gooddata.com).
14 - support "cbor" format in json_xs tool.
15 - support (and fix) calling encode and decode in list context
16 (reported by Вадим Власов).
17 - work around a bug in older perls crashing when presented
18 with shared hash keys (Reini Urban).
19 - use stability canary.
7 20
213.01 Tue Oct 29 16:55:15 CET 2013
22 - backport to perls < 5.18 (reported by Paul Howarth).
23
243.0 Tue Oct 29 01:35:37 CET 2013
25 - implemented an object tagging extension (using the
26 Types::Serialiser serialisation protocol).
27 - reworked the documentation regarding object serialisation,
28 add a new OBJECT SERIALISATION section that explains th
29 whole process.
30 - new setting: allow_tags.
31 - switch to Types::Serialiser booleans.
32 - remove to_json/from_json.
33 - other minor improvements to the documentation.
34
352.34 Thu May 23 11:30:34 CEST 2013
36 - work around bugs in perl 5.18 breaking more than 100
37 widely used modules, without a fix in sight because
38 p5pers don't care about CPAN anymore.
39 - when canonicalising, only allocate up to 64 hash key
40 pointers on the stack. for larger hashes, use the heap,
41 to avoid using too much stackspace.
42 - discuss the problem with setlocale (reported by a few victims).
43
442.33 Wed Aug 1 21:03:52 CEST 2012
45 - internal encode/decode XS wrappers did not expect stack
46 moves caused by callbacks (analyzed and testcase by Jesse Luehrs).
47 - add bencode as to/from option in bin/json_xs.
48 - add -e option to json_xs, and none and string in/out formats.
49
502.32 Thu Aug 11 19:06:38 CEST 2011
51 - fix a bug in the initial whitespace accumulation.
52
532.31 Wed Jul 27 17:53:05 CEST 2011
54 - don't accumulate initial whitespace in the incremental buffer
55 (this can be useful to allow whitespace-keepalive on a tcp
56 connection without triggering the max_size limit).
57 - properly croak on some invalid inputs that are not strings
58 (e.g. undef) when trying to decode a json text (reported
59 and analyzed by Goro Fuji).
60
612.3 Wed Aug 18 01:26:47 CEST 2010
62 - make sure decoder doesn't change the decoding in the incremental
63 parser (testcase provided by Hendrik Schumacher).
64 - applied patch by DaTa for Data::Dumper support in json_xs.
65 - added -t dump support to json_xs, using Data::Dump.
66 - added -f eval support to json_xs.
67
682.29 Wed Mar 17 02:39:12 CET 2010
69 - fix a memory leak when callbacks set using filter_json_object
70 or filter_json_single_key_object were called (great testcase
71 by Eric Wilhelm).
72
732.28 Thu Mar 11 20:30:46 CET 2010
74 - implement our own atof function - perl's can be orders of
75 magnitudes slower than even the system one. on the positive
76 side, ours seems to be more exact in general than perl's.
77 (testcase provided by Tim Meadowcroft).
78 - clarify floating point conversion issues a bit.
79 - update jpsykes csrf article url.
80 - updated benchmark section - JSON::PP became much faster!
81
822.27 Thu Jan 7 07:35:08 CET 2010
83 - support relaxed option inside the incremental parser
84 (testcase provided by IKEGAMI via Makamaka).
85
862.26 Sat Oct 10 03:26:19 CEST 2009
87 - big integers could become truncated (based on patch
88 by Strobl Anton).
89 - output format change: indent now adds a final newline, which is
90 more expected and more true to the documentation.
91
922.25 Sat Aug 8 12:04:41 CEST 2009
93 - the perl debugger completely breaks lvalue subs - try to work
94 around the issue.
95 - ignore RMAGICAL hashes w.r.t. CANONICAL.
96 - try to work around a possible char signedness issue on aix.
97 - require common sense.
98
992.24 Sat May 30 08:25:45 CEST 2009
100 - the incremental parser did not update its parse offset
101 pointer correctly when parsing utf8-strings (nicely
102 debugged by Martin Evans).
103 - appending a non-utf8-string to the incremental parser
104 in utf8 mode failed to upgrade the string.
105 - wording of parse error messages has been improved.
106
1072.232 Sun Feb 22 11:12:25 CET 2009
108 - use an exponential algorithm to extend strings, to
109 help platforms with bad or abysmal==windows memory
110 allocater performance, at the expense of some memory
111 wastage (use shrink to recover this extra memory).
112 (nicely analysed by Dmitry Karasik).
113
1142.2311 Thu Feb 19 02:12:54 CET 2009
115 - add a section "JSON and ECMAscript" to explain some
116 incompatibilities between the two (problem was noted by
117 various people).
118 - add t/20_faihu.t.
119
1202.231 Thu Nov 20 04:59:08 CET 2008
121 - work around 5.10.0 magic bugs where manipulating magic values
122 (such as $1) would permanently damage them as perl would
123 ignore the magicalness, by making a full copy of the string,
124 reported by Dmitry Karasik.
125 - work around spurious warnings under older perl 5.8's.
126
1272.23 Mon Sep 29 05:08:29 CEST 2008
128 - fix a compilation problem when perl is not using char * as, well,
129 char *.
130 - use PL_hexdigit in favour of rolling our own.
131
1322.2222 Sun Jul 20 18:49:00 CEST 2008
133 - same game again, broken 5.10 finds yet another assertion
134 failure, and the workaround causes additional runtime warnings.
135 Work around the next assertion AND the warning. 5.10 seriously
136 needs to adjust it's attitude against working code.
137
1382.222 Sat Jul 19 06:15:34 CEST 2008
139 - you work around one -DDEBUGGING assertion bug in perl 5.10
140 just to hit the next one. work around this one, too.
141
1422.22 Tue Jul 15 13:26:51 CEST 2008
143 - allow higher nesting levels in incremental parser.
144 - error out earlier in some cases in the incremental parser
145 (as suggested by Yuval Kogman).
146 - improve incr-parser test (Yuval Kogman).
147
1482.21 Tue Jun 3 08:43:23 CEST 2008
149 - (hopefully) work around a perl 5.10 bug with -DDEBUGGING.
150 - remove the experimental status of the incremental parser interface.
151 - move =encoding around again, to avoid bugs with search.cpan.org.
152 when can we finally have utf-8 in pod???
153 - add ->incr_reset method.
154
1552.2 Wed Apr 16 20:37:25 CEST 2008
156 - lifted the log2 rounding restriction of max_depth and max_size.
157 - make booleans mutable by creating a copy instead of handing out
158 the same scalar (reported by pasha sadri).
159 - added support for incremental json parsing (still EXPERIMENTAL).
160 - implemented and added a json_xs command line utility that can convert
161 from/to a number of serialisation formats - tell me if you need more.
162 - implement allow_unknown/get_allow_unknown methods.
163 - fixed documentation of max_depth w.r.t. higher and equal.
164 - moved down =encoding directive a bit, too much breaks if it's the first
165 pod directive :/.
166 - removed documentation section on other modules, it became somewhat
167 outdated and is nowadays mostly of historical interest.
168
1692.1 Wed Mar 19 23:23:18 CET 2008
170 - update documentation here and there: add a large section
171 about utf8/latin1/ascii flags, add a security consideration
172 and extend and clarify the JSON and YAML section.
173 - medium speed enhancements when encoding/decoding non-ascii chars.
174 - minor speedup in number encoding case.
175 - extend and clarify the section on incompatibilities
176 between YAML and JSON.
177 - switch to static inline from just inline when using gcc.
178 - add =encoding utf-8 to the manpage, now that perl 5.10 supports it.
179 - fix some issues with UV to JSON conversion of unknown impact.
180 - published the yahoo locals search result used in benchmarks as the
181 original url changes so comparison is impossible.
182
1832.01 Wed Dec 5 11:40:28 CET 2007
184 - INCOMPATIBLE API CHANGE: to_json and from_json have been
185 renamed to encode_json/decode_json for JSON.pm compatibility.
186 The old functions croak and might be replaced by JSON.pm
187 comaptible versions in some later release.
188
1892.0 Tue Dec 4 11:30:46 CET 2007
190 - this is supposed to be the first version of JSON::XS
191 compatible with version 2.0+ of the JSON module.
192 Using the JSON module as frontend to JSON::XS should be
193 as fast as using JSON::XS directly, so consider using it
194 instead.
195 - added get_* methods for all "simple" options.
196 - make JSON::XS subclassable.
197
1981.53 Tue Nov 13 23:58:33 CET 2007
199 - minor doc clarifications.
200 - fixed many doc typos (patch by Thomas L. Shinnick).
201
2021.52 Mon Oct 15 03:22:06 CEST 2007
203 - remove =encoding pod directive again, it confuses too many pod
204 parsers :/.
205
2061.51 Sat Oct 13 03:55:56 CEST 2007
207 - encode empty arrays/hashes in a compact way when pretty is enabled.
208 - apparently JSON::XS was used to find some bugs in the
209 JSON_checker testsuite, so add (the corrected) JSON_checker tests to
210 the testsuite.
211 - quite a bit of doc updates/extension.
212 - require 5.8.2, as this seems to be the first unicode-stable version.
213
2141.5 Tue Aug 28 04:05:38 CEST 2007
215 - add support for tied hashes, based on ideas and testcase by
216 Marcus Holland-Moritz.
217 - implemented relaxed parsing mode where some extensions are being
218 accepted. generation is still JSON-only.
219
2201.44 Wed Aug 22 01:02:44 CEST 2007
221 - very experimental process-emulation support, slowing everything down.
222 the horribly broken perl threads are still not supported - YMMV.
223
2241.43 Thu Jul 26 13:26:37 CEST 2007
225 - convert big json numbers exclusively consisting of digits to NV
226 only when there is no loss of precision, otherwise to string.
227
2281.42 Tue Jul 24 00:51:18 CEST 2007
229 - fix a crash caused by not handling missing array elements
230 (report and testcase by Jay Kuri).
231
2321.41 Tue Jul 10 18:21:44 CEST 2007
233 - fix compilation with NDEBUG (assert side-effect),
234 affects convert_blessed only.
235 - fix a bug in decode filters calling ENTER; SAVETMPS;
236 one time too often.
237 - catch a typical error in TO_JSON methods.
238 - antique-ised XS.xs again to work with outdated
239 C compilers (windows...).
240
2411.4 Mon Jul 2 10:06:30 CEST 2007
242 - add convert_blessed setting.
243 - encode did not catch all blessed objects, encoding their
244 contents in most cases. This has been fixed by introducing
245 the allow_blessed setting.
246 - added filter_json_object and filter_json_single_key_object
247 settings that specify a callback to be called when
248 all/specific json objects are encountered.
249 - assume that most object keys are simple ascii words and
250 optimise this case, penalising the general case. This can
251 speed up decoding by 30% in typical cases and gives
252 a smaller and faster perl hash.
253 - implemented simpleminded, optional resource size checking
254 in decode_json.
255 - remove objToJson/jsonToObj aliases, as the next version
256 of JSON will not have them either.
257 - bit the bullet and converted the very simple json object
258 into a more complex one.
259 - work around a bug where perl wrongly claims an integer
260 is not an integer.
261 - unbundle JSON::XS::Boolean into own pm file so Storable
262 and similar modules can resolve the overloading when thawing.
263
2641.3 Sun Jun 24 01:55:02 CEST 2007
265 - make JSON::XS::true and false special overloaded objects
266 and return those instead of 1 and 0 for those json atoms
267 (JSON::PP compatibility is NOT achieved yet).
268 - add JSON::XS::is_bool predicate to test for those special
269 values.
270 - add a reference to
271 http://jpsykes.com/47/practical-csrf-and-json-security.
272 - removed require 5.8.8 again, it is just not very expert-friendly.
273 Also try to be more compatible with slightly older versions,
274 which are not recommended (because they are buggy).
275
2761.24 Mon Jun 11 05:40:49 CEST 2007
8 - added informative section on JSON-as-YAML. 277 - added informative section on JSON-as-YAML.
278 - get rid of some c99-isms again.
279 - localise dec->cur in decode_str, speeding up
280 string decoding considerably (>15% on my amd64 + gcc).
281 - increased SHORT_STRING_LEN to 16kb: stack space is
282 usually plenty, and this actually saves memory
283 when !shrinking as short strings will fit perfectly.
9 284
101.23 Wed Jun 6 20:13:06 CEST 2007 2851.23 Wed Jun 6 20:13:06 CEST 2007
11 - greatly improved small integer encoding and decoding speed. 286 - greatly improved small integer encoding and decoding speed.
12 - implement a number of µ-optimisations. 287 - implement a number of µ-optimisations.
13 - updated benchmarks. 288 - updated benchmarks.
119 case the called functions do... stuff. 394 case the called functions do... stuff.
120 - croak when encoding to ascii and an out-of-range 395 - croak when encoding to ascii and an out-of-range
121 (non-unicode) codepoint is encountered. 396 (non-unicode) codepoint is encountered.
122 397
1230.2 Fri Mar 23 00:23:34 CET 2007 3980.2 Fri Mar 23 00:23:34 CET 2007
124 - the "could not sleep without debuggign release". 399 - the "could not sleep without debugging release".
125 it should basically work now, with many bugs as 400 it should basically work now, with many bugs as
126 no production tests have been run yet. 401 no production tests have been run yet.
127 - added more testcases. 402 - added more testcases.
128 - the expected shitload of bugfixes. 403 - the expected shitload of bugfixes.
129 - handle utf8 flag correctly in decode. 404 - handle utf8 flag correctly in decode.

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines