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

Comparing JSON-XS/Changes (file contents):
Revision 1.21 by root, Sun Mar 25 21:19:13 2007 UTC vs.
Revision 1.158 by root, Mon Oct 28 23:19:54 2013 UTC

1Revision history for Perl extension JSON::XS 1Revision history for Perl extension JSON::XS
2 2
3TODO: maybe simply use \0 and \1 as true/false when encoding (and overloaded/blessed versions thereof on 3TODO: maybe detetc and croak on more invalid inputs (e.g. +-inf/nan)
4 decoding?) 4TODO: maybe avoid the reblessing and better support readonly objects.
5TODO: http://stevehanov.ca/blog/index.php?id=104 compression
5 6
7 - switch to Types::Serialiser booleans.
8 - remove to_json/from_json.
9 - minor improvements to the documentation.
10
112.34 Thu May 23 11:30:34 CEST 2013
12 - work around bugs in perl 5.18 breaking more than 100
13 widely used modules, without a fix in sight because
14 p5pers don't care about CPAN anymore.
15 - when canonicalising, only allocate up to 64 hash key
16 pointers on the stack. for larger hashes, use the heap,
17 to avoid using too much stackspace.
18 - discuss the problem with setlocale (reported by a few victims).
19
202.33 Wed Aug 1 21:03:52 CEST 2012
21 - internal encode/decode XS wrappers did not expect stack
22 moves caused by callbacks (analyzed and testcase by Jesse Luehrs).
23 - add bencode as to/from option in bin/json_xs.
24 - add -e option to json_xs, and none and string in/out formats.
25
262.32 Thu Aug 11 19:06:38 CEST 2011
27 - fix a bug in the initial whitespace accumulation.
28
292.31 Wed Jul 27 17:53:05 CEST 2011
30 - don't accumulate initial whitespace in the incremental buffer
31 (this can be useful to allow whitespace-keepalive on a tcp
32 connection without triggering the max_size limit).
33 - properly croak on some invalid inputs that are not strings
34 (e.g. undef) when trying to decode a json text (reported
35 and analyzed by Goro Fuji).
36
372.3 Wed Aug 18 01:26:47 CEST 2010
38 - make sure decoder doesn't change the decoding in the incremental
39 parser (testcase provided by Hendrik Schumacher).
40 - applied patch by DaTa for Data::Dumper support in json_xs.
41 - added -t dump support to json_xs, using Data::Dump.
42 - added -f eval support to json_xs.
43
442.29 Wed Mar 17 02:39:12 CET 2010
45 - fix a memory leak when callbacks set using filter_json_object
46 or filter_json_single_key_object were called (great testcase
47 by Eric Wilhelm).
48
492.28 Thu Mar 11 20:30:46 CET 2010
50 - implement our own atof function - perl's can be orders of
51 magnitudes slower than even the system one. on the positive
52 side, ours seems to be more exact in general than perl's.
53 (testcase provided by Tim Meadowcroft).
54 - clarify floating point conversion issues a bit.
55 - update jpsykes csrf article url.
56 - updated benchmark section - JSON::PP became much faster!
57
582.27 Thu Jan 7 07:35:08 CET 2010
59 - support relaxed option inside the incremental parser
60 (testcase provided by IKEGAMI via Makamaka).
61
622.26 Sat Oct 10 03:26:19 CEST 2009
63 - big integers could become truncated (based on patch
64 by Strobl Anton).
65 - output format change: indent now adds a final newline, which is
66 more expected and more true to the documentation.
67
682.25 Sat Aug 8 12:04:41 CEST 2009
69 - the perl debugger completely breaks lvalue subs - try to work
70 around the issue.
71 - ignore RMAGICAL hashes w.r.t. CANONICAL.
72 - try to work around a possible char signedness issue on aix.
73 - require common sense.
74
752.24 Sat May 30 08:25:45 CEST 2009
76 - the incremental parser did not update its parse offset
77 pointer correctly when parsing utf8-strings (nicely
78 debugged by Martin Evans).
79 - appending a non-utf8-string to the incremental parser
80 in utf8 mode failed to upgrade the string.
81 - wording of parse error messages has been improved.
82
832.232 Sun Feb 22 11:12:25 CET 2009
84 - use an exponential algorithm to extend strings, to
85 help platforms with bad or abysmal==windows memory
86 allocater performance, at the expense of some memory
87 wastage (use shrink to recover this extra memory).
88 (nicely analysed by Dmitry Karasik).
89
902.2311 Thu Feb 19 02:12:54 CET 2009
91 - add a section "JSON and ECMAscript" to explain some
92 incompatibilities between the two (problem was noted by
93 various people).
94 - add t/20_faihu.t.
95
962.231 Thu Nov 20 04:59:08 CET 2008
97 - work around 5.10.0 magic bugs where manipulating magic values
98 (such as $1) would permanently damage them as perl would
99 ignore the magicalness, by making a full copy of the string,
100 reported by Dmitry Karasik.
101 - work around spurious warnings under older perl 5.8's.
102
1032.23 Mon Sep 29 05:08:29 CEST 2008
104 - fix a compilation problem when perl is not using char * as, well,
105 char *.
106 - use PL_hexdigit in favour of rolling our own.
107
1082.2222 Sun Jul 20 18:49:00 CEST 2008
109 - same game again, broken 5.10 finds yet another assertion
110 failure, and the workaround causes additional runtime warnings.
111 Work around the next assertion AND the warning. 5.10 seriously
112 needs to adjust it's attitude against working code.
113
1142.222 Sat Jul 19 06:15:34 CEST 2008
115 - you work around one -DDEBUGGING assertion bug in perl 5.10
116 just to hit the next one. work around this one, too.
117
1182.22 Tue Jul 15 13:26:51 CEST 2008
119 - allow higher nesting levels in incremental parser.
120 - error out earlier in some cases in the incremental parser
121 (as suggested by Yuval Kogman).
122 - improve incr-parser test (Yuval Kogman).
123
1242.21 Tue Jun 3 08:43:23 CEST 2008
125 - (hopefully) work around a perl 5.10 bug with -DDEBUGGING.
126 - remove the experimental status of the incremental parser interface.
127 - move =encoding around again, to avoid bugs with search.cpan.org.
128 when can we finally have utf-8 in pod???
129 - add ->incr_reset method.
130
1312.2 Wed Apr 16 20:37:25 CEST 2008
132 - lifted the log2 rounding restriction of max_depth and max_size.
133 - make booleans mutable by creating a copy instead of handing out
134 the same scalar (reported by pasha sadri).
135 - added support for incremental json parsing (still EXPERIMENTAL).
136 - implemented and added a json_xs command line utility that can convert
137 from/to a number of serialisation formats - tell me if you need more.
138 - implement allow_unknown/get_allow_unknown methods.
139 - fixed documentation of max_depth w.r.t. higher and equal.
140 - moved down =encoding directive a bit, too much breaks if it's the first
141 pod directive :/.
142 - removed documentation section on other modules, it became somewhat
143 outdated and is nowadays mostly of historical interest.
144
1452.1 Wed Mar 19 23:23:18 CET 2008
146 - update documentation here and there: add a large section
147 about utf8/latin1/ascii flags, add a security consideration
148 and extend and clarify the JSON and YAML section.
149 - medium speed enhancements when encoding/decoding non-ascii chars.
150 - minor speedup in number encoding case.
151 - extend and clarify the section on incompatibilities
152 between YAML and JSON.
153 - switch to static inline from just inline when using gcc.
154 - add =encoding utf-8 to the manpage, now that perl 5.10 supports it.
155 - fix some issues with UV to JSON conversion of unknown impact.
156 - published the yahoo locals search result used in benchmarks as the
157 original url changes so comparison is impossible.
158
1592.01 Wed Dec 5 11:40:28 CET 2007
160 - INCOMPATIBLE API CHANGE: to_json and from_json have been
161 renamed to encode_json/decode_json for JSON.pm compatibility.
162 The old functions croak and might be replaced by JSON.pm
163 comaptible versions in some later release.
164
1652.0 Tue Dec 4 11:30:46 CET 2007
166 - this is supposed to be the first version of JSON::XS
167 compatible with version 2.0+ of the JSON module.
168 Using the JSON module as frontend to JSON::XS should be
169 as fast as using JSON::XS directly, so consider using it
170 instead.
171 - added get_* methods for all "simple" options.
172 - make JSON::XS subclassable.
173
1741.53 Tue Nov 13 23:58:33 CET 2007
175 - minor doc clarifications.
176 - fixed many doc typos (patch by Thomas L. Shinnick).
177
1781.52 Mon Oct 15 03:22:06 CEST 2007
179 - remove =encoding pod directive again, it confuses too many pod
180 parsers :/.
181
1821.51 Sat Oct 13 03:55:56 CEST 2007
183 - encode empty arrays/hashes in a compact way when pretty is enabled.
184 - apparently JSON::XS was used to find some bugs in the
185 JSON_checker testsuite, so add (the corrected) JSON_checker tests to
186 the testsuite.
187 - quite a bit of doc updates/extension.
188 - require 5.8.2, as this seems to be the first unicode-stable version.
189
1901.5 Tue Aug 28 04:05:38 CEST 2007
191 - add support for tied hashes, based on ideas and testcase by
192 Marcus Holland-Moritz.
193 - implemented relaxed parsing mode where some extensions are being
194 accepted. generation is still JSON-only.
195
1961.44 Wed Aug 22 01:02:44 CEST 2007
197 - very experimental process-emulation support, slowing everything down.
198 the horribly broken perl threads are still not supported - YMMV.
199
2001.43 Thu Jul 26 13:26:37 CEST 2007
201 - convert big json numbers exclusively consisting of digits to NV
202 only when there is no loss of precision, otherwise to string.
203
2041.42 Tue Jul 24 00:51:18 CEST 2007
205 - fix a crash caused by not handling missing array elements
206 (report and testcase by Jay Kuri).
207
2081.41 Tue Jul 10 18:21:44 CEST 2007
209 - fix compilation with NDEBUG (assert side-effect),
210 affects convert_blessed only.
211 - fix a bug in decode filters calling ENTER; SAVETMPS;
212 one time too often.
213 - catch a typical error in TO_JSON methods.
214 - antique-ised XS.xs again to work with outdated
215 C compilers (windows...).
216
2171.4 Mon Jul 2 10:06:30 CEST 2007
218 - add convert_blessed setting.
219 - encode did not catch all blessed objects, encoding their
220 contents in most cases. This has been fixed by introducing
221 the allow_blessed setting.
222 - added filter_json_object and filter_json_single_key_object
223 settings that specify a callback to be called when
224 all/specific json objects are encountered.
225 - assume that most object keys are simple ascii words and
226 optimise this case, penalising the general case. This can
227 speed up decoding by 30% in typical cases and gives
228 a smaller and faster perl hash.
229 - implemented simpleminded, optional resource size checking
230 in decode_json.
231 - remove objToJson/jsonToObj aliases, as the next version
232 of JSON will not have them either.
233 - bit the bullet and converted the very simple json object
234 into a more complex one.
235 - work around a bug where perl wrongly claims an integer
236 is not an integer.
237 - unbundle JSON::XS::Boolean into own pm file so Storable
238 and similar modules can resolve the overloading when thawing.
239
2401.3 Sun Jun 24 01:55:02 CEST 2007
241 - make JSON::XS::true and false special overloaded objects
242 and return those instead of 1 and 0 for those json atoms
243 (JSON::PP compatibility is NOT achieved yet).
244 - add JSON::XS::is_bool predicate to test for those special
245 values.
246 - add a reference to
247 http://jpsykes.com/47/practical-csrf-and-json-security.
248 - removed require 5.8.8 again, it is just not very expert-friendly.
249 Also try to be more compatible with slightly older versions,
250 which are not recommended (because they are buggy).
251
2521.24 Mon Jun 11 05:40:49 CEST 2007
253 - added informative section on JSON-as-YAML.
254 - get rid of some c99-isms again.
255 - localise dec->cur in decode_str, speeding up
256 string decoding considerably (>15% on my amd64 + gcc).
257 - increased SHORT_STRING_LEN to 16kb: stack space is
258 usually plenty, and this actually saves memory
259 when !shrinking as short strings will fit perfectly.
260
2611.23 Wed Jun 6 20:13:06 CEST 2007
262 - greatly improved small integer encoding and decoding speed.
263 - implement a number of ยต-optimisations.
264 - updated benchmarks.
265
2661.22 Thu May 24 00:07:25 CEST 2007
267 - require 5.8.8 explicitly as older perls do not seem to offer
268 the required macros.
269 - possibly made it compile on so-called C compilers by microsoft.
270
2711.21 Wed May 9 18:40:32 CEST 2007
272 - character offset reported for trailing garbage was random.
273
2741.2 Wed May 9 18:35:01 CEST 2007
275 - decode did not work with magical scalars (doh!).
276 - added latin1 flag to produce JSON texts in the latin1 subset
277 of unicode.
278 - flag trailing garbage as error.
279 - new decode_prefix method that returns the number
280 of characters consumed by a decode.
281 - max octets/char in perls UTF-X is actually 13, not 11,
282 as pointed out by Glenn Linderman.
283 - fixed typoe reported by YAMASHINA Hio.
284
2851.11 Mon Apr 9 07:05:49 CEST 2007
286 - properly 0-terminate sv's returned by encode to help
287 C libraries that expect that 0 to be there.
288 - partially "port" JSON from C to microsofts fucking broken
289 pseudo-C. They should be burned to the ground for pissing
290 on standards. And I should be stoned for even trying to
291 support this filthy excuse for a c compiler.
292
2931.1 Wed Apr 4 01:45:00 CEST 2007
294 - clarify documentation (pointed out by Quinn Weaver).
295 - decode_utf8 sometimes did not correctly flag errors,
296 leading to segfaults.
297 - further reduced default nesting depth to 512 due to the test
298 failure by that anonymous "chris" whose e-mail address seems
299 to be impossible to get. Tests on other freebsd systems indicate
300 that this is likely a problem in his/her configuration and not this
301 module.
302 - renamed json => JSON in error messages.
303 - corrected the character offset in some error messages.
304
3051.01 Sat Mar 31 16:15:40 CEST 2007
306 - do not segfault when from_json/decode gets passed
307 a non-string object (reported by Florian Ragwitz).
308 This has no effect on normal operation.
309
3101.0 Thu Mar 29 04:43:34 CEST 2007
311 - the long awaited (by me) 1.0 version.
312 - add \0 (JSON::XS::false) and \1 (JSON::XS::true) mappings to JSON
313 true and false.
314 - add some more notes to shrink, as suggested by Alex Efros.
315 - improve testsuite.
316 - halve the default nesting depth limit, to hopefully make it
317 work on Freebsd (unfortunately, the cpan tester did not
318 send me his report, so I cannot ask about the stack limit on fbsd).
319
3200.8 Mon Mar 26 00:10:48 CEST 2007
6 - fix a memleak when decoding hashes. 321 - fix a memleak when decoding hashes.
7 - export jsonToBj and objToJson as aliases 322 - export jsonToBj and objToJson as aliases
8 to to_json and from_json, to reduce incompatibilities 323 to to_json and from_json, to reduce incompatibilities
9 between JSON/JSON::PC and JSON::XS. (experimental). 324 between JSON/JSON::PC and JSON::XS. (experimental).
10 - implement a maximum nesting depth for both en- and de-coding. 325 - implement a maximum nesting depth for both en- and de-coding.
… …
55 case the called functions do... stuff. 370 case the called functions do... stuff.
56 - croak when encoding to ascii and an out-of-range 371 - croak when encoding to ascii and an out-of-range
57 (non-unicode) codepoint is encountered. 372 (non-unicode) codepoint is encountered.
58 373
590.2 Fri Mar 23 00:23:34 CET 2007 3740.2 Fri Mar 23 00:23:34 CET 2007
60 - the "could not sleep without debuggign release". 375 - the "could not sleep without debugging release".
61 it should basically work now, with many bugs as 376 it should basically work now, with many bugs as
62 no production tests have been run yet. 377 no production tests have been run yet.
63 - added more testcases. 378 - added more testcases.
64 - the expected shitload of bugfixes. 379 - the expected shitload of bugfixes.
65 - handle utf8 flag correctly in decode. 380 - handle utf8 flag correctly in decode.

Diff Legend

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