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

Comparing JSON-XS/Changes (file contents):
Revision 1.63 by root, Sun Jul 1 23:40:07 2007 UTC vs.
Revision 1.151 by root, Thu Sep 22 19:37:24 2011 UTC

1Revision history for Perl extension JSON::XS 1Revision history for Perl extension JSON::XS
2 2
3TODO: object filtering (decode_filtered?) 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.
5
6 - add bencode as to/from option in bin/json_xs.
7
82.32 Thu Aug 11 19:06:38 CEST 2011
9 - fix a bug in the initial whitespace accumulation.
10
112.31 Wed Jul 27 17:53:05 CEST 2011
12 - don't accumulate initial whitespace in the incremental buffer
13 (this can be useful to allow whitespace-keepalive on a tcp
14 connection without triggering the max_size limit).
15 - properly croak on some invalid inputs that are not strings
16 (e.g. undef) when trying to decode a json text (reported
17 and analyzed by Goro Fuji).
18
192.3 Wed Aug 18 01:26:47 CEST 2010
20 - make sure decoder doesn't change the decoding in the incremental
21 parser (testcase provided by Hendrik Schumacher).
22 - applied patch by DaTa for Data::Dumper support in json_xs.
23 - added -t dump support to json_xs, using Data::Dump.
24 - added -f eval support to json_xs.
25
262.29 Wed Mar 17 02:39:12 CET 2010
27 - fix a memory leak when callbacks set using filter_json_object
28 or filter_json_single_key_object were called (great testcase
29 by Eric Wilhelm).
30
312.28 Thu Mar 11 20:30:46 CET 2010
32 - implement our own atof function - perl's can be orders of
33 magnitudes slower than even the system one. on the positive
34 side, ours seems to be more exact in general than perl's.
35 (testcase provided by Tim Meadowcroft).
36 - clarify floating point conversion issues a bit.
37 - update jpsykes csrf article url.
38 - updated benchmark section - JSON::PP became much faster!
39
402.27 Thu Jan 7 07:35:08 CET 2010
41 - support relaxed option inside the incremental parser
42 (testcase provided by IKEGAMI via Makamaka).
43
442.26 Sat Oct 10 03:26:19 CEST 2009
45 - big integers could become truncated (based on patch
46 by Strobl Anton).
47 - output format change: indent now adds a final newline, which is
48 more expected and more true to the documentation.
49
502.25 Sat Aug 8 12:04:41 CEST 2009
51 - the perl debugger completely breaks lvalue subs - try to work
52 around the issue.
53 - ignore RMAGICAL hashes w.r.t. CANONICAL.
54 - try to work around a possible char signedness issue on aix.
55 - require common sense.
56
572.24 Sat May 30 08:25:45 CEST 2009
58 - the incremental parser did not update its parse offset
59 pointer correctly when parsing utf8-strings (nicely
60 debugged by Martin Evans).
61 - appending a non-utf8-string to the incremental parser
62 in utf8 mode failed to upgrade the string.
63 - wording of parse error messages has been improved.
64
652.232 Sun Feb 22 11:12:25 CET 2009
66 - use an exponential algorithm to extend strings, to
67 help platforms with bad or abysmal==windows memory
68 allocater performance, at the expense of some memory
69 wastage (use shrink to recover this extra memory).
70 (nicely analysed by Dmitry Karasik).
71
722.2311 Thu Feb 19 02:12:54 CET 2009
73 - add a section "JSON and ECMAscript" to explain some
74 incompatibilities between the two (problem was noted by
75 various people).
76 - add t/20_faihu.t.
77
782.231 Thu Nov 20 04:59:08 CET 2008
79 - work around 5.10.0 magic bugs where manipulating magic values
80 (such as $1) would permanently damage them as perl would
81 ignore the magicalness, by making a full copy of the string,
82 reported by Dmitry Karasik.
83 - work around spurious warnings under older perl 5.8's.
84
852.23 Mon Sep 29 05:08:29 CEST 2008
86 - fix a compilation problem when perl is not using char * as, well,
87 char *.
88 - use PL_hexdigit in favour of rolling our own.
89
902.2222 Sun Jul 20 18:49:00 CEST 2008
91 - same game again, broken 5.10 finds yet another assertion
92 failure, and the workaround causes additional runtime warnings.
93 Work around the next assertion AND the warning. 5.10 seriously
94 needs to adjust it's attitude against working code.
95
962.222 Sat Jul 19 06:15:34 CEST 2008
97 - you work around one -DDEBUGGING assertion bug in perl 5.10
98 just to hit the next one. work around this one, too.
99
1002.22 Tue Jul 15 13:26:51 CEST 2008
101 - allow higher nesting levels in incremental parser.
102 - error out earlier in some cases in the incremental parser
103 (as suggested by Yuval Kogman).
104 - improve incr-parser test (Yuval Kogman).
105
1062.21 Tue Jun 3 08:43:23 CEST 2008
107 - (hopefully) work around a perl 5.10 bug with -DDEBUGGING.
108 - remove the experimental status of the incremental parser interface.
109 - move =encoding around again, to avoid bugs with search.cpan.org.
110 when can we finally have utf-8 in pod???
111 - add ->incr_reset method.
112
1132.2 Wed Apr 16 20:37:25 CEST 2008
114 - lifted the log2 rounding restriction of max_depth and max_size.
115 - make booleans mutable by creating a copy instead of handing out
116 the same scalar (reported by pasha sadri).
117 - added support for incremental json parsing (still EXPERIMENTAL).
118 - implemented and added a json_xs command line utility that can convert
119 from/to a number of serialisation formats - tell me if you need more.
120 - implement allow_unknown/get_allow_unknown methods.
121 - fixed documentation of max_depth w.r.t. higher and equal.
122 - moved down =encoding directive a bit, too much breaks if it's the first
123 pod directive :/.
124 - removed documentation section on other modules, it became somewhat
125 outdated and is nowadays mostly of historical interest.
126
1272.1 Wed Mar 19 23:23:18 CET 2008
128 - update documentation here and there: add a large section
129 about utf8/latin1/ascii flags, add a security consideration
130 and extend and clarify the JSON and YAML section.
131 - medium speed enhancements when encoding/decoding non-ascii chars.
132 - minor speedup in number encoding case.
133 - extend and clarify the section on incompatibilities
134 between YAML and JSON.
135 - switch to static inline from just inline when using gcc.
136 - add =encoding utf-8 to the manpage, now that perl 5.10 supports it.
137 - fix some issues with UV to JSON conversion of unknown impact.
138 - published the yahoo locals search result used in benchmarks as the
139 original url changes so comparison is impossible.
140
1412.01 Wed Dec 5 11:40:28 CET 2007
142 - INCOMPATIBLE API CHANGE: to_json and from_json have been
143 renamed to encode_json/decode_json for JSON.pm compatibility.
144 The old functions croak and might be replaced by JSON.pm
145 comaptible versions in some later release.
146
1472.0 Tue Dec 4 11:30:46 CET 2007
148 - this is supposed to be the first version of JSON::XS
149 compatible with version 2.0+ of the JSON module.
150 Using the JSON module as frontend to JSON::XS should be
151 as fast as using JSON::XS directly, so consider using it
152 instead.
153 - added get_* methods for all "simple" options.
154 - make JSON::XS subclassable.
155
1561.53 Tue Nov 13 23:58:33 CET 2007
157 - minor doc clarifications.
158 - fixed many doc typos (patch by Thomas L. Shinnick).
159
1601.52 Mon Oct 15 03:22:06 CEST 2007
161 - remove =encoding pod directive again, it confuses too many pod
162 parsers :/.
163
1641.51 Sat Oct 13 03:55:56 CEST 2007
165 - encode empty arrays/hashes in a compact way when pretty is enabled.
166 - apparently JSON::XS was used to find some bugs in the
167 JSON_checker testsuite, so add (the corrected) JSON_checker tests to
168 the testsuite.
169 - quite a bit of doc updates/extension.
170 - require 5.8.2, as this seems to be the first unicode-stable version.
171
1721.5 Tue Aug 28 04:05:38 CEST 2007
173 - add support for tied hashes, based on ideas and testcase by
174 Marcus Holland-Moritz.
175 - implemented relaxed parsing mode where some extensions are being
176 accepted. generation is still JSON-only.
177
1781.44 Wed Aug 22 01:02:44 CEST 2007
179 - very experimental process-emulation support, slowing everything down.
180 the horribly broken perl threads are still not supported - YMMV.
181
1821.43 Thu Jul 26 13:26:37 CEST 2007
183 - convert big json numbers exclusively consisting of digits to NV
184 only when there is no loss of precision, otherwise to string.
185
1861.42 Tue Jul 24 00:51:18 CEST 2007
187 - fix a crash caused by not handling missing array elements
188 (report and testcase by Jay Kuri).
189
1901.41 Tue Jul 10 18:21:44 CEST 2007
191 - fix compilation with NDEBUG (assert side-effect),
192 affects convert_blessed only.
193 - fix a bug in decode filters calling ENTER; SAVETMPS;
194 one time too often.
195 - catch a typical error in TO_JSON methods.
196 - antique-ised XS.xs again to work with outdated
197 C compilers (windows...).
198
1991.4 Mon Jul 2 10:06:30 CEST 2007
5 - add convert_blessed setting. 200 - add convert_blessed setting.
6 - encode did not catch all blessed objects, encoding their 201 - encode did not catch all blessed objects, encoding their
7 contents in most cases. This has been fixed by introducing 202 contents in most cases. This has been fixed by introducing
8 the allow_blessed setting. 203 the allow_blessed setting.
204 - added filter_json_object and filter_json_single_key_object
205 settings that specify a callback to be called when
206 all/specific json objects are encountered.
9 - assume that most object keys are simple ascii words and 207 - assume that most object keys are simple ascii words and
10 optimise this case, penalising the general case. This can 208 optimise this case, penalising the general case. This can
11 speed up decoding by 30% in typical cases and gives 209 speed up decoding by 30% in typical cases and gives
12 a smaller and faster perl hash. 210 a smaller and faster perl hash.
13 - implemented simpleminded, optional resource size checking 211 - implemented simpleminded, optional resource size checking
14 in decode_json. 212 in decode_json.
15 - remove objToJson/jsonToObj aliases, as the next version 213 - remove objToJson/jsonToObj aliases, as the next version
16 of JSON will not have them either. 214 of JSON will not have them either.
17 - bit the bullet and converted the very simple json object 215 - bit the bullet and converted the very simple json object
18 into a more complex one. 216 into a more complex one.
217 - work around a bug where perl wrongly claims an integer
218 is not an integer.
219 - unbundle JSON::XS::Boolean into own pm file so Storable
220 and similar modules can resolve the overloading when thawing.
19 221
201.3 Sun Jun 24 01:55:02 CEST 2007 2221.3 Sun Jun 24 01:55:02 CEST 2007
21 - make JSON::XS::true and false special overloaded objects 223 - make JSON::XS::true and false special overloaded objects
22 and return those instead of 1 and 0 for those json atoms 224 and return those instead of 1 and 0 for those json atoms
23 (JSON::PP compatibility is NOT achieved yet). 225 (JSON::PP compatibility is NOT achieved yet).
… …
150 case the called functions do... stuff. 352 case the called functions do... stuff.
151 - croak when encoding to ascii and an out-of-range 353 - croak when encoding to ascii and an out-of-range
152 (non-unicode) codepoint is encountered. 354 (non-unicode) codepoint is encountered.
153 355
1540.2 Fri Mar 23 00:23:34 CET 2007 3560.2 Fri Mar 23 00:23:34 CET 2007
155 - the "could not sleep without debuggign release". 357 - the "could not sleep without debugging release".
156 it should basically work now, with many bugs as 358 it should basically work now, with many bugs as
157 no production tests have been run yet. 359 no production tests have been run yet.
158 - added more testcases. 360 - added more testcases.
159 - the expected shitload of bugfixes. 361 - the expected shitload of bugfixes.
160 - handle utf8 flag correctly in decode. 362 - handle utf8 flag correctly in decode.

Diff Legend

– Removed lines
+ Added lines
< Changed lines
> Changed lines