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

Comparing JSON-XS/Changes (file contents):
Revision 1.66 by root, Mon Jul 2 01:12:27 2007 UTC vs.
Revision 1.170 by root, Thu Sep 1 11:38:39 2016 UTC

1Revision history for Perl extension JSON::XS 1Revision history for Perl extension JSON::XS
2 2
31.4 3TODO: maybe detetc and croak on more invalid inputs (e.g. +-inf/nan)
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
113.02 Fri Feb 26 22:45:20 CET 2016
12 - allow_nonref now affects booleans (\1, $Types::Serialiser::Boolean)
13 as well (reported by Alex Efros).
14 - allow literal tabs in strings in relaxed mode (patch by
15 lubo.rintel@gooddata.com).
16 - support "cbor" format in json_xs tool.
17 - support (and fix) calling encode and decode in list context
18 (reported by Вадим Власов).
19 - work around a bug in older perls crashing when presented
20 with shared hash keys (Reini Urban).
21 - use stability canary.
22
233.01 Tue Oct 29 16:55:15 CET 2013
24 - backport to perls < 5.18 (reported by Paul Howarth).
25
263.0 Tue Oct 29 01:35:37 CET 2013
27 - implemented an object tagging extension (using the
28 Types::Serialiser serialisation protocol).
29 - reworked the documentation regarding object serialisation,
30 add a new OBJECT SERIALISATION section that explains th
31 whole process.
32 - new setting: allow_tags.
33 - switch to Types::Serialiser booleans.
34 - remove to_json/from_json.
35 - other minor improvements to the documentation.
36
372.34 Thu May 23 11:30:34 CEST 2013
38 - work around bugs in perl 5.18 breaking more than 100
39 widely used modules, without a fix in sight because
40 p5pers don't care about CPAN anymore.
41 - when canonicalising, only allocate up to 64 hash key
42 pointers on the stack. for larger hashes, use the heap,
43 to avoid using too much stackspace.
44 - discuss the problem with setlocale (reported by a few victims).
45
462.33 Wed Aug 1 21:03:52 CEST 2012
47 - internal encode/decode XS wrappers did not expect stack
48 moves caused by callbacks (analyzed and testcase by Jesse Luehrs).
49 - add bencode as to/from option in bin/json_xs.
50 - add -e option to json_xs, and none and string in/out formats.
51
522.32 Thu Aug 11 19:06:38 CEST 2011
53 - fix a bug in the initial whitespace accumulation.
54
552.31 Wed Jul 27 17:53:05 CEST 2011
56 - don't accumulate initial whitespace in the incremental buffer
57 (this can be useful to allow whitespace-keepalive on a tcp
58 connection without triggering the max_size limit).
59 - properly croak on some invalid inputs that are not strings
60 (e.g. undef) when trying to decode a json text (reported
61 and analyzed by Goro Fuji).
62
632.3 Wed Aug 18 01:26:47 CEST 2010
64 - make sure decoder doesn't change the decoding in the incremental
65 parser (testcase provided by Hendrik Schumacher).
66 - applied patch by DaTa for Data::Dumper support in json_xs.
67 - added -t dump support to json_xs, using Data::Dump.
68 - added -f eval support to json_xs.
69
702.29 Wed Mar 17 02:39:12 CET 2010
71 - fix a memory leak when callbacks set using filter_json_object
72 or filter_json_single_key_object were called (great testcase
73 by Eric Wilhelm).
74
752.28 Thu Mar 11 20:30:46 CET 2010
76 - implement our own atof function - perl's can be orders of
77 magnitudes slower than even the system one. on the positive
78 side, ours seems to be more exact in general than perl's.
79 (testcase provided by Tim Meadowcroft).
80 - clarify floating point conversion issues a bit.
81 - update jpsykes csrf article url.
82 - updated benchmark section - JSON::PP became much faster!
83
842.27 Thu Jan 7 07:35:08 CET 2010
85 - support relaxed option inside the incremental parser
86 (testcase provided by IKEGAMI via Makamaka).
87
882.26 Sat Oct 10 03:26:19 CEST 2009
89 - big integers could become truncated (based on patch
90 by Strobl Anton).
91 - output format change: indent now adds a final newline, which is
92 more expected and more true to the documentation.
93
942.25 Sat Aug 8 12:04:41 CEST 2009
95 - the perl debugger completely breaks lvalue subs - try to work
96 around the issue.
97 - ignore RMAGICAL hashes w.r.t. CANONICAL.
98 - try to work around a possible char signedness issue on aix.
99 - require common sense.
100
1012.24 Sat May 30 08:25:45 CEST 2009
102 - the incremental parser did not update its parse offset
103 pointer correctly when parsing utf8-strings (nicely
104 debugged by Martin Evans).
105 - appending a non-utf8-string to the incremental parser
106 in utf8 mode failed to upgrade the string.
107 - wording of parse error messages has been improved.
108
1092.232 Sun Feb 22 11:12:25 CET 2009
110 - use an exponential algorithm to extend strings, to
111 help platforms with bad or abysmal==windows memory
112 allocater performance, at the expense of some memory
113 wastage (use shrink to recover this extra memory).
114 (nicely analysed by Dmitry Karasik).
115
1162.2311 Thu Feb 19 02:12:54 CET 2009
117 - add a section "JSON and ECMAscript" to explain some
118 incompatibilities between the two (problem was noted by
119 various people).
120 - add t/20_faihu.t.
121
1222.231 Thu Nov 20 04:59:08 CET 2008
123 - work around 5.10.0 magic bugs where manipulating magic values
124 (such as $1) would permanently damage them as perl would
125 ignore the magicalness, by making a full copy of the string,
126 reported by Dmitry Karasik.
127 - work around spurious warnings under older perl 5.8's.
128
1292.23 Mon Sep 29 05:08:29 CEST 2008
130 - fix a compilation problem when perl is not using char * as, well,
131 char *.
132 - use PL_hexdigit in favour of rolling our own.
133
1342.2222 Sun Jul 20 18:49:00 CEST 2008
135 - same game again, broken 5.10 finds yet another assertion
136 failure, and the workaround causes additional runtime warnings.
137 Work around the next assertion AND the warning. 5.10 seriously
138 needs to adjust it's attitude against working code.
139
1402.222 Sat Jul 19 06:15:34 CEST 2008
141 - you work around one -DDEBUGGING assertion bug in perl 5.10
142 just to hit the next one. work around this one, too.
143
1442.22 Tue Jul 15 13:26:51 CEST 2008
145 - allow higher nesting levels in incremental parser.
146 - error out earlier in some cases in the incremental parser
147 (as suggested by Yuval Kogman).
148 - improve incr-parser test (Yuval Kogman).
149
1502.21 Tue Jun 3 08:43:23 CEST 2008
151 - (hopefully) work around a perl 5.10 bug with -DDEBUGGING.
152 - remove the experimental status of the incremental parser interface.
153 - move =encoding around again, to avoid bugs with search.cpan.org.
154 when can we finally have utf-8 in pod???
155 - add ->incr_reset method.
156
1572.2 Wed Apr 16 20:37:25 CEST 2008
158 - lifted the log2 rounding restriction of max_depth and max_size.
159 - make booleans mutable by creating a copy instead of handing out
160 the same scalar (reported by pasha sadri).
161 - added support for incremental json parsing (still EXPERIMENTAL).
162 - implemented and added a json_xs command line utility that can convert
163 from/to a number of serialisation formats - tell me if you need more.
164 - implement allow_unknown/get_allow_unknown methods.
165 - fixed documentation of max_depth w.r.t. higher and equal.
166 - moved down =encoding directive a bit, too much breaks if it's the first
167 pod directive :/.
168 - removed documentation section on other modules, it became somewhat
169 outdated and is nowadays mostly of historical interest.
170
1712.1 Wed Mar 19 23:23:18 CET 2008
172 - update documentation here and there: add a large section
173 about utf8/latin1/ascii flags, add a security consideration
174 and extend and clarify the JSON and YAML section.
175 - medium speed enhancements when encoding/decoding non-ascii chars.
176 - minor speedup in number encoding case.
177 - extend and clarify the section on incompatibilities
178 between YAML and JSON.
179 - switch to static inline from just inline when using gcc.
180 - add =encoding utf-8 to the manpage, now that perl 5.10 supports it.
181 - fix some issues with UV to JSON conversion of unknown impact.
182 - published the yahoo locals search result used in benchmarks as the
183 original url changes so comparison is impossible.
184
1852.01 Wed Dec 5 11:40:28 CET 2007
186 - INCOMPATIBLE API CHANGE: to_json and from_json have been
187 renamed to encode_json/decode_json for JSON.pm compatibility.
188 The old functions croak and might be replaced by JSON.pm
189 comaptible versions in some later release.
190
1912.0 Tue Dec 4 11:30:46 CET 2007
192 - this is supposed to be the first version of JSON::XS
193 compatible with version 2.0+ of the JSON module.
194 Using the JSON module as frontend to JSON::XS should be
195 as fast as using JSON::XS directly, so consider using it
196 instead.
197 - added get_* methods for all "simple" options.
198 - make JSON::XS subclassable.
199
2001.53 Tue Nov 13 23:58:33 CET 2007
201 - minor doc clarifications.
202 - fixed many doc typos (patch by Thomas L. Shinnick).
203
2041.52 Mon Oct 15 03:22:06 CEST 2007
205 - remove =encoding pod directive again, it confuses too many pod
206 parsers :/.
207
2081.51 Sat Oct 13 03:55:56 CEST 2007
209 - encode empty arrays/hashes in a compact way when pretty is enabled.
210 - apparently JSON::XS was used to find some bugs in the
211 JSON_checker testsuite, so add (the corrected) JSON_checker tests to
212 the testsuite.
213 - quite a bit of doc updates/extension.
214 - require 5.8.2, as this seems to be the first unicode-stable version.
215
2161.5 Tue Aug 28 04:05:38 CEST 2007
217 - add support for tied hashes, based on ideas and testcase by
218 Marcus Holland-Moritz.
219 - implemented relaxed parsing mode where some extensions are being
220 accepted. generation is still JSON-only.
221
2221.44 Wed Aug 22 01:02:44 CEST 2007
223 - very experimental process-emulation support, slowing everything down.
224 the horribly broken perl threads are still not supported - YMMV.
225
2261.43 Thu Jul 26 13:26:37 CEST 2007
227 - convert big json numbers exclusively consisting of digits to NV
228 only when there is no loss of precision, otherwise to string.
229
2301.42 Tue Jul 24 00:51:18 CEST 2007
231 - fix a crash caused by not handling missing array elements
232 (report and testcase by Jay Kuri).
233
2341.41 Tue Jul 10 18:21:44 CEST 2007
235 - fix compilation with NDEBUG (assert side-effect),
236 affects convert_blessed only.
237 - fix a bug in decode filters calling ENTER; SAVETMPS;
238 one time too often.
239 - catch a typical error in TO_JSON methods.
240 - antique-ised XS.xs again to work with outdated
241 C compilers (windows...).
242
2431.4 Mon Jul 2 10:06:30 CEST 2007
4 - add convert_blessed setting. 244 - add convert_blessed setting.
5 - encode did not catch all blessed objects, encoding their 245 - encode did not catch all blessed objects, encoding their
6 contents in most cases. This has been fixed by introducing 246 contents in most cases. This has been fixed by introducing
7 the allow_blessed setting. 247 the allow_blessed setting.
8 - added filter_json_object and filter_json_single_key_object 248 - added filter_json_object and filter_json_single_key_object
9 settings that specify a callback to be called when 249 settings that specify a callback to be called when
10 json objects are encountered. 250 all/specific json objects are encountered.
11 - assume that most object keys are simple ascii words and 251 - assume that most object keys are simple ascii words and
12 optimise this case, penalising the general case. This can 252 optimise this case, penalising the general case. This can
13 speed up decoding by 30% in typical cases and gives 253 speed up decoding by 30% in typical cases and gives
14 a smaller and faster perl hash. 254 a smaller and faster perl hash.
15 - implemented simpleminded, optional resource size checking 255 - implemented simpleminded, optional resource size checking
156 case the called functions do... stuff. 396 case the called functions do... stuff.
157 - croak when encoding to ascii and an out-of-range 397 - croak when encoding to ascii and an out-of-range
158 (non-unicode) codepoint is encountered. 398 (non-unicode) codepoint is encountered.
159 399
1600.2 Fri Mar 23 00:23:34 CET 2007 4000.2 Fri Mar 23 00:23:34 CET 2007
161 - the "could not sleep without debuggign release". 401 - the "could not sleep without debugging release".
162 it should basically work now, with many bugs as 402 it should basically work now, with many bugs as
163 no production tests have been run yet. 403 no production tests have been run yet.
164 - added more testcases. 404 - added more testcases.
165 - the expected shitload of bugfixes. 405 - the expected shitload of bugfixes.
166 - handle utf8 flag correctly in decode. 406 - handle utf8 flag correctly in decode.

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines