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

Comparing JSON-XS/Changes (file contents):
Revision 1.52 by root, Thu Jun 14 23:58:57 2007 UTC vs.
Revision 1.193 by root, Tue Oct 27 18:05:58 2020 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 detect 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: how to cope with tagged values and standard json decoders
6TODO: investigate magic (Eric Brine)
7TODO: [PATCH] Types::Serialiser: Inline true(), false() and error() functions
8TODO: replace bool_stash by BOOL_STASH seems to work with mod_perl, make a compile time option?
5 9
104.03 Tue Oct 27 19:05:01 CET 2020
11 - when parsing comments in relaxed mode, JSON::XS would detect garbage
12 after the JSON text if the comment is after the end and does not end in
13 a newline (reported by Felipe Gasper).
14
154.02 Wed Mar 6 08:31:24 CET 2019
16 - undo the fix from 4.01, it breaks more things than it fixes
17 (another testcase by Wesley Schwengle).
18 - try a proper fix this time.
19
204.01 Sun Feb 24 05:03:30 CET 2019
21 - fix some stack corruption caused mostly when calling methods
22 in list context (testcase by Wesley Schwengle).
23
244.0 Fri Nov 16 00:06:54 CET 2018
25 - SECURITY IMPLICATION: this release enables allow_nonref by default
26 for compatibility with RFC 7159 and newer. See "old" vs. "new"
27 JSON under SECURITY CONSIDERATIONS.
28 - reworked the "old" vs. "new" JSON section.
29 - add ->boolean_values to provide the values to which booleans
30 decode (requested by Aristotle Pagaltzis).
31 - decode would wrongly accept ASCII NUL characters instead of
32 reporting them as trailing garbage.
33 - work around what smells like a perl bug w.r.t. exceptions
34 thrown in callbacks.
35 - incremental parser now more or less respects allow_nonref.
36 - json_xs json-pretty now enables canonical mode.
37 - add documentation section about I-JSON.
38 - minor documentation fixes/updates.
39
403.04 Thu Aug 17 04:30:47 CEST 2017
41 - change exponential realloc algorithm on encoding and string decoding to be
42 really exponential (this helps slow/debugging allocators such as libumem)
43 (reported by Matthew Horsfall).
44 - string encoding would needlessly overallocate output space
45 (testcase by Matthew Horsfall).
46 - be very paranoid about extending buffer lengths and croak if buffers get too large,
47 which might (or might not) improve security.
48 - add cbor-packed type to json_xs.
49 - switch from YAML to YAML::XS in json_xs, as YAML is way too buggy and outdated.
50
513.03 Wed Nov 16 20:20:59 CET 2016
52 - fix a bug introduced by a perl bug workaround that would cause
53 incremental parsing to fail with a sv_chop panic.
54 - json_xs: toformat failure error message fixed.
55 - json_xs: allow cyclic data structures in CBOR.
56
573.02 Fri Feb 26 22:45:20 CET 2016
58 - allow_nonref now affects booleans (\1, $Types::Serialiser::Boolean)
59 as well (reported by Alex Efros).
60 - allow literal tabs in strings in relaxed mode (patch by
61 lubo.rintel@gooddata.com).
62 - support "cbor" format in json_xs tool.
63 - support (and fix) calling encode and decode in list context
64 (reported by Вадим Власов).
65 - work around a bug in older perls crashing when presented
66 with shared hash keys (Reini Urban).
67 - use stability canary.
68
693.01 Tue Oct 29 16:55:15 CET 2013
70 - backport to perls < 5.18 (reported by Paul Howarth).
71
723.0 Tue Oct 29 01:35:37 CET 2013
73 - implemented an object tagging extension (using the
74 Types::Serialiser serialisation protocol).
75 - reworked the documentation regarding object serialisation,
76 add a new OBJECT SERIALISATION section that explains th
77 whole process.
78 - new setting: allow_tags.
79 - switch to Types::Serialiser booleans.
80 - remove to_json/from_json.
81 - other minor improvements to the documentation.
82
832.34 Thu May 23 11:30:34 CEST 2013
84 - work around bugs in perl 5.18 breaking more than 100
85 widely used modules, without a fix in sight because
86 p5pers don't care about CPAN anymore.
87 - when canonicalising, only allocate up to 64 hash key
88 pointers on the stack. for larger hashes, use the heap,
89 to avoid using too much stackspace.
90 - discuss the problem with setlocale (reported by a few victims).
91
922.33 Wed Aug 1 21:03:52 CEST 2012
93 - internal encode/decode XS wrappers did not expect stack
94 moves caused by callbacks (analyzed and testcase by Jesse Luehrs).
95 - add bencode as to/from option in bin/json_xs.
96 - add -e option to json_xs, and none and string in/out formats.
97
982.32 Thu Aug 11 19:06:38 CEST 2011
99 - fix a bug in the initial whitespace accumulation.
100
1012.31 Wed Jul 27 17:53:05 CEST 2011
102 - don't accumulate initial whitespace in the incremental buffer
103 (this can be useful to allow whitespace-keepalive on a tcp
104 connection without triggering the max_size limit).
105 - properly croak on some invalid inputs that are not strings
106 (e.g. undef) when trying to decode a json text (reported
107 and analyzed by Goro Fuji).
108
1092.3 Wed Aug 18 01:26:47 CEST 2010
110 - make sure decoder doesn't change the decoding in the incremental
111 parser (testcase provided by Hendrik Schumacher).
112 - applied patch by DaTa for Data::Dumper support in json_xs.
113 - added -t dump support to json_xs, using Data::Dump.
114 - added -f eval support to json_xs.
115
1162.29 Wed Mar 17 02:39:12 CET 2010
117 - fix a memory leak when callbacks set using filter_json_object
118 or filter_json_single_key_object were called (great testcase
119 by Eric Wilhelm).
120
1212.28 Thu Mar 11 20:30:46 CET 2010
122 - implement our own atof function - perl's can be orders of
123 magnitudes slower than even the system one. on the positive
124 side, ours seems to be more exact in general than perl's.
125 (testcase provided by Tim Meadowcroft).
126 - clarify floating point conversion issues a bit.
127 - update jpsykes csrf article url.
128 - updated benchmark section - JSON::PP became much faster!
129
1302.27 Thu Jan 7 07:35:08 CET 2010
131 - support relaxed option inside the incremental parser
132 (testcase provided by IKEGAMI via Makamaka).
133
1342.26 Sat Oct 10 03:26:19 CEST 2009
135 - big integers could become truncated (based on patch
136 by Strobl Anton).
137 - output format change: indent now adds a final newline, which is
138 more expected and more true to the documentation.
139
1402.25 Sat Aug 8 12:04:41 CEST 2009
141 - the perl debugger completely breaks lvalue subs - try to work
142 around the issue.
143 - ignore RMAGICAL hashes w.r.t. CANONICAL.
144 - try to work around a possible char signedness issue on aix.
145 - require common sense.
146
1472.24 Sat May 30 08:25:45 CEST 2009
148 - the incremental parser did not update its parse offset
149 pointer correctly when parsing utf8-strings (nicely
150 debugged by Martin Evans).
151 - appending a non-utf8-string to the incremental parser
152 in utf8 mode failed to upgrade the string.
153 - wording of parse error messages has been improved.
154
1552.232 Sun Feb 22 11:12:25 CET 2009
156 - use an exponential algorithm to extend strings, to
157 help platforms with bad or abysmal==windows memory
158 allocater performance, at the expense of some memory
159 wastage (use shrink to recover this extra memory).
160 (nicely analysed by Dmitry Karasik).
161
1622.2311 Thu Feb 19 02:12:54 CET 2009
163 - add a section "JSON and ECMAscript" to explain some
164 incompatibilities between the two (problem was noted by
165 various people).
166 - add t/20_faihu.t.
167
1682.231 Thu Nov 20 04:59:08 CET 2008
169 - work around 5.10.0 magic bugs where manipulating magic values
170 (such as $1) would permanently damage them as perl would
171 ignore the magicalness, by making a full copy of the string,
172 reported by Dmitry Karasik.
173 - work around spurious warnings under older perl 5.8's.
174
1752.23 Mon Sep 29 05:08:29 CEST 2008
176 - fix a compilation problem when perl is not using char * as, well,
177 char *.
178 - use PL_hexdigit in favour of rolling our own.
179
1802.2222 Sun Jul 20 18:49:00 CEST 2008
181 - same game again, broken 5.10 finds yet another assertion
182 failure, and the workaround causes additional runtime warnings.
183 Work around the next assertion AND the warning. 5.10 seriously
184 needs to adjust it's attitude against working code.
185
1862.222 Sat Jul 19 06:15:34 CEST 2008
187 - you work around one -DDEBUGGING assertion bug in perl 5.10
188 just to hit the next one. work around this one, too.
189
1902.22 Tue Jul 15 13:26:51 CEST 2008
191 - allow higher nesting levels in incremental parser.
192 - error out earlier in some cases in the incremental parser
193 (as suggested by Yuval Kogman).
194 - improve incr-parser test (Yuval Kogman).
195
1962.21 Tue Jun 3 08:43:23 CEST 2008
197 - (hopefully) work around a perl 5.10 bug with -DDEBUGGING.
198 - remove the experimental status of the incremental parser interface.
199 - move =encoding around again, to avoid bugs with search.cpan.org.
200 when can we finally have utf-8 in pod???
201 - add ->incr_reset method.
202
2032.2 Wed Apr 16 20:37:25 CEST 2008
204 - lifted the log2 rounding restriction of max_depth and max_size.
205 - make booleans mutable by creating a copy instead of handing out
206 the same scalar (reported by pasha sadri).
207 - added support for incremental json parsing (still EXPERIMENTAL).
208 - implemented and added a json_xs command line utility that can convert
209 from/to a number of serialisation formats - tell me if you need more.
210 - implement allow_unknown/get_allow_unknown methods.
211 - fixed documentation of max_depth w.r.t. higher and equal.
212 - moved down =encoding directive a bit, too much breaks if it's the first
213 pod directive :/.
214 - removed documentation section on other modules, it became somewhat
215 outdated and is nowadays mostly of historical interest.
216
2172.1 Wed Mar 19 23:23:18 CET 2008
218 - update documentation here and there: add a large section
219 about utf8/latin1/ascii flags, add a security consideration
220 and extend and clarify the JSON and YAML section.
221 - medium speed enhancements when encoding/decoding non-ascii chars.
222 - minor speedup in number encoding case.
223 - extend and clarify the section on incompatibilities
224 between YAML and JSON.
225 - switch to static inline from just inline when using gcc.
226 - add =encoding utf-8 to the manpage, now that perl 5.10 supports it.
227 - fix some issues with UV to JSON conversion of unknown impact.
228 - published the yahoo locals search result used in benchmarks as the
229 original url changes so comparison is impossible.
230
2312.01 Wed Dec 5 11:40:28 CET 2007
232 - INCOMPATIBLE API CHANGE: to_json and from_json have been
233 renamed to encode_json/decode_json for JSON.pm compatibility.
234 The old functions croak and might be replaced by JSON.pm
235 comaptible versions in some later release.
236
2372.0 Tue Dec 4 11:30:46 CET 2007
238 - this is supposed to be the first version of JSON::XS
239 compatible with version 2.0+ of the JSON module.
240 Using the JSON module as frontend to JSON::XS should be
241 as fast as using JSON::XS directly, so consider using it
242 instead.
243 - added get_* methods for all "simple" options.
244 - make JSON::XS subclassable.
245
2461.53 Tue Nov 13 23:58:33 CET 2007
247 - minor doc clarifications.
248 - fixed many doc typos (patch by Thomas L. Shinnick).
249
2501.52 Mon Oct 15 03:22:06 CEST 2007
251 - remove =encoding pod directive again, it confuses too many pod
252 parsers :/.
253
2541.51 Sat Oct 13 03:55:56 CEST 2007
255 - encode empty arrays/hashes in a compact way when pretty is enabled.
256 - apparently JSON::XS was used to find some bugs in the
257 JSON_checker testsuite, so add (the corrected) JSON_checker tests to
258 the testsuite.
259 - quite a bit of doc updates/extension.
260 - require 5.8.2, as this seems to be the first unicode-stable version.
261
2621.5 Tue Aug 28 04:05:38 CEST 2007
263 - add support for tied hashes, based on ideas and testcase by
264 Marcus Holland-Moritz.
265 - implemented relaxed parsing mode where some extensions are being
266 accepted. generation is still JSON-only.
267
2681.44 Wed Aug 22 01:02:44 CEST 2007
269 - very experimental process-emulation support, slowing everything down.
270 the horribly broken perl threads are still not supported - YMMV.
271
2721.43 Thu Jul 26 13:26:37 CEST 2007
273 - convert big json numbers exclusively consisting of digits to NV
274 only when there is no loss of precision, otherwise to string.
275
2761.42 Tue Jul 24 00:51:18 CEST 2007
277 - fix a crash caused by not handling missing array elements
278 (report and testcase by Jay Kuri).
279
2801.41 Tue Jul 10 18:21:44 CEST 2007
281 - fix compilation with NDEBUG (assert side-effect),
282 affects convert_blessed only.
283 - fix a bug in decode filters calling ENTER; SAVETMPS;
284 one time too often.
285 - catch a typical error in TO_JSON methods.
286 - antique-ised XS.xs again to work with outdated
287 C compilers (windows...).
288
2891.4 Mon Jul 2 10:06:30 CEST 2007
290 - add convert_blessed setting.
291 - encode did not catch all blessed objects, encoding their
292 contents in most cases. This has been fixed by introducing
293 the allow_blessed setting.
294 - added filter_json_object and filter_json_single_key_object
295 settings that specify a callback to be called when
296 all/specific json objects are encountered.
297 - assume that most object keys are simple ascii words and
298 optimise this case, penalising the general case. This can
299 speed up decoding by 30% in typical cases and gives
300 a smaller and faster perl hash.
301 - implemented simpleminded, optional resource size checking
302 in decode_json.
303 - remove objToJson/jsonToObj aliases, as the next version
304 of JSON will not have them either.
305 - bit the bullet and converted the very simple json object
306 into a more complex one.
307 - work around a bug where perl wrongly claims an integer
308 is not an integer.
309 - unbundle JSON::XS::Boolean into own pm file so Storable
310 and similar modules can resolve the overloading when thawing.
311
3121.3 Sun Jun 24 01:55:02 CEST 2007
313 - make JSON::XS::true and false special overloaded objects
314 and return those instead of 1 and 0 for those json atoms
315 (JSON::PP compatibility is NOT achieved yet).
316 - add JSON::XS::is_bool predicate to test for those special
317 values.
6 - add a reference to 318 - add a reference to
7 http://jpsykes.com/47/practical-csrf-and-json-security. 319 http://jpsykes.com/47/practical-csrf-and-json-security.
320 - removed require 5.8.8 again, it is just not very expert-friendly.
321 Also try to be more compatible with slightly older versions,
322 which are not recommended (because they are buggy).
8 323
91.24 Mon Jun 11 05:40:49 CEST 2007 3241.24 Mon Jun 11 05:40:49 CEST 2007
10 - added informative section on JSON-as-YAML. 325 - added informative section on JSON-as-YAML.
11 - get rid of some c99-isms again. 326 - get rid of some c99-isms again.
12 - localise dec->cur in decode_str, speeding up 327 - localise dec->cur in decode_str, speeding up
127 case the called functions do... stuff. 442 case the called functions do... stuff.
128 - croak when encoding to ascii and an out-of-range 443 - croak when encoding to ascii and an out-of-range
129 (non-unicode) codepoint is encountered. 444 (non-unicode) codepoint is encountered.
130 445
1310.2 Fri Mar 23 00:23:34 CET 2007 4460.2 Fri Mar 23 00:23:34 CET 2007
132 - the "could not sleep without debuggign release". 447 - the "could not sleep without debugging release".
133 it should basically work now, with many bugs as 448 it should basically work now, with many bugs as
134 no production tests have been run yet. 449 no production tests have been run yet.
135 - added more testcases. 450 - added more testcases.
136 - the expected shitload of bugfixes. 451 - the expected shitload of bugfixes.
137 - handle utf8 flag correctly in decode. 452 - handle utf8 flag correctly in decode.

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines