ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/JSON-XS/Changes
Revision: 1.163
Committed: Wed Oct 30 22:11:01 2013 UTC (10 years, 6 months ago) by root
Branch: MAIN
Changes since 1.162: +1 -0 lines
Log Message:
*** empty log message ***

File Contents

# Content
1 Revision history for Perl extension JSON::XS
2
3 TODO: maybe detetc and croak on more invalid inputs (e.g. +-inf/nan)
4 TODO: maybe avoid the reblessing and better support readonly objects.
5 TODO: http://stevehanov.ca/blog/index.php?id=104 compression
6
7 TODO: how to cope with tagged values and standard json decoders
8 3.01 Tue Oct 29 16:55:15 CET 2013
9 - backport to perls < 5.18 (reported by Paul Howarth).
10
11 3.0 Tue Oct 29 01:35:37 CET 2013
12 - implemented an object tagging extension (using the
13 Types::Serialiser serialisation protocol).
14 - reworked the documentation regarding object serialisation,
15 add a new OBJECT SERIALISATION section that explains the
16 whole process.
17 - new setting: allow_tags.
18 - switch to Types::Serialiser booleans.
19 - remove to_json/from_json.
20 - other minor improvements to the documentation.
21
22 2.34 Thu May 23 11:30:34 CEST 2013
23 - work around bugs in perl 5.18 breaking more than 100
24 widely used modules, without a fix in sight because
25 p5pers don't care about CPAN anymore.
26 - when canonicalising, only allocate up to 64 hash key
27 pointers on the stack. for larger hashes, use the heap,
28 to avoid using too much stackspace.
29 - discuss the problem with setlocale (reported by a few victims).
30
31 2.33 Wed Aug 1 21:03:52 CEST 2012
32 - internal encode/decode XS wrappers did not expect stack
33 moves caused by callbacks (analyzed and testcase by Jesse Luehrs).
34 - add bencode as to/from option in bin/json_xs.
35 - add -e option to json_xs, and none and string in/out formats.
36
37 2.32 Thu Aug 11 19:06:38 CEST 2011
38 - fix a bug in the initial whitespace accumulation.
39
40 2.31 Wed Jul 27 17:53:05 CEST 2011
41 - don't accumulate initial whitespace in the incremental buffer
42 (this can be useful to allow whitespace-keepalive on a tcp
43 connection without triggering the max_size limit).
44 - properly croak on some invalid inputs that are not strings
45 (e.g. undef) when trying to decode a json text (reported
46 and analyzed by Goro Fuji).
47
48 2.3 Wed Aug 18 01:26:47 CEST 2010
49 - make sure decoder doesn't change the decoding in the incremental
50 parser (testcase provided by Hendrik Schumacher).
51 - applied patch by DaTa for Data::Dumper support in json_xs.
52 - added -t dump support to json_xs, using Data::Dump.
53 - added -f eval support to json_xs.
54
55 2.29 Wed Mar 17 02:39:12 CET 2010
56 - fix a memory leak when callbacks set using filter_json_object
57 or filter_json_single_key_object were called (great testcase
58 by Eric Wilhelm).
59
60 2.28 Thu Mar 11 20:30:46 CET 2010
61 - implement our own atof function - perl's can be orders of
62 magnitudes slower than even the system one. on the positive
63 side, ours seems to be more exact in general than perl's.
64 (testcase provided by Tim Meadowcroft).
65 - clarify floating point conversion issues a bit.
66 - update jpsykes csrf article url.
67 - updated benchmark section - JSON::PP became much faster!
68
69 2.27 Thu Jan 7 07:35:08 CET 2010
70 - support relaxed option inside the incremental parser
71 (testcase provided by IKEGAMI via Makamaka).
72
73 2.26 Sat Oct 10 03:26:19 CEST 2009
74 - big integers could become truncated (based on patch
75 by Strobl Anton).
76 - output format change: indent now adds a final newline, which is
77 more expected and more true to the documentation.
78
79 2.25 Sat Aug 8 12:04:41 CEST 2009
80 - the perl debugger completely breaks lvalue subs - try to work
81 around the issue.
82 - ignore RMAGICAL hashes w.r.t. CANONICAL.
83 - try to work around a possible char signedness issue on aix.
84 - require common sense.
85
86 2.24 Sat May 30 08:25:45 CEST 2009
87 - the incremental parser did not update its parse offset
88 pointer correctly when parsing utf8-strings (nicely
89 debugged by Martin Evans).
90 - appending a non-utf8-string to the incremental parser
91 in utf8 mode failed to upgrade the string.
92 - wording of parse error messages has been improved.
93
94 2.232 Sun Feb 22 11:12:25 CET 2009
95 - use an exponential algorithm to extend strings, to
96 help platforms with bad or abysmal==windows memory
97 allocater performance, at the expense of some memory
98 wastage (use shrink to recover this extra memory).
99 (nicely analysed by Dmitry Karasik).
100
101 2.2311 Thu Feb 19 02:12:54 CET 2009
102 - add a section "JSON and ECMAscript" to explain some
103 incompatibilities between the two (problem was noted by
104 various people).
105 - add t/20_faihu.t.
106
107 2.231 Thu Nov 20 04:59:08 CET 2008
108 - work around 5.10.0 magic bugs where manipulating magic values
109 (such as $1) would permanently damage them as perl would
110 ignore the magicalness, by making a full copy of the string,
111 reported by Dmitry Karasik.
112 - work around spurious warnings under older perl 5.8's.
113
114 2.23 Mon Sep 29 05:08:29 CEST 2008
115 - fix a compilation problem when perl is not using char * as, well,
116 char *.
117 - use PL_hexdigit in favour of rolling our own.
118
119 2.2222 Sun Jul 20 18:49:00 CEST 2008
120 - same game again, broken 5.10 finds yet another assertion
121 failure, and the workaround causes additional runtime warnings.
122 Work around the next assertion AND the warning. 5.10 seriously
123 needs to adjust it's attitude against working code.
124
125 2.222 Sat Jul 19 06:15:34 CEST 2008
126 - you work around one -DDEBUGGING assertion bug in perl 5.10
127 just to hit the next one. work around this one, too.
128
129 2.22 Tue Jul 15 13:26:51 CEST 2008
130 - allow higher nesting levels in incremental parser.
131 - error out earlier in some cases in the incremental parser
132 (as suggested by Yuval Kogman).
133 - improve incr-parser test (Yuval Kogman).
134
135 2.21 Tue Jun 3 08:43:23 CEST 2008
136 - (hopefully) work around a perl 5.10 bug with -DDEBUGGING.
137 - remove the experimental status of the incremental parser interface.
138 - move =encoding around again, to avoid bugs with search.cpan.org.
139 when can we finally have utf-8 in pod???
140 - add ->incr_reset method.
141
142 2.2 Wed Apr 16 20:37:25 CEST 2008
143 - lifted the log2 rounding restriction of max_depth and max_size.
144 - make booleans mutable by creating a copy instead of handing out
145 the same scalar (reported by pasha sadri).
146 - added support for incremental json parsing (still EXPERIMENTAL).
147 - implemented and added a json_xs command line utility that can convert
148 from/to a number of serialisation formats - tell me if you need more.
149 - implement allow_unknown/get_allow_unknown methods.
150 - fixed documentation of max_depth w.r.t. higher and equal.
151 - moved down =encoding directive a bit, too much breaks if it's the first
152 pod directive :/.
153 - removed documentation section on other modules, it became somewhat
154 outdated and is nowadays mostly of historical interest.
155
156 2.1 Wed Mar 19 23:23:18 CET 2008
157 - update documentation here and there: add a large section
158 about utf8/latin1/ascii flags, add a security consideration
159 and extend and clarify the JSON and YAML section.
160 - medium speed enhancements when encoding/decoding non-ascii chars.
161 - minor speedup in number encoding case.
162 - extend and clarify the section on incompatibilities
163 between YAML and JSON.
164 - switch to static inline from just inline when using gcc.
165 - add =encoding utf-8 to the manpage, now that perl 5.10 supports it.
166 - fix some issues with UV to JSON conversion of unknown impact.
167 - published the yahoo locals search result used in benchmarks as the
168 original url changes so comparison is impossible.
169
170 2.01 Wed Dec 5 11:40:28 CET 2007
171 - INCOMPATIBLE API CHANGE: to_json and from_json have been
172 renamed to encode_json/decode_json for JSON.pm compatibility.
173 The old functions croak and might be replaced by JSON.pm
174 comaptible versions in some later release.
175
176 2.0 Tue Dec 4 11:30:46 CET 2007
177 - this is supposed to be the first version of JSON::XS
178 compatible with version 2.0+ of the JSON module.
179 Using the JSON module as frontend to JSON::XS should be
180 as fast as using JSON::XS directly, so consider using it
181 instead.
182 - added get_* methods for all "simple" options.
183 - make JSON::XS subclassable.
184
185 1.53 Tue Nov 13 23:58:33 CET 2007
186 - minor doc clarifications.
187 - fixed many doc typos (patch by Thomas L. Shinnick).
188
189 1.52 Mon Oct 15 03:22:06 CEST 2007
190 - remove =encoding pod directive again, it confuses too many pod
191 parsers :/.
192
193 1.51 Sat Oct 13 03:55:56 CEST 2007
194 - encode empty arrays/hashes in a compact way when pretty is enabled.
195 - apparently JSON::XS was used to find some bugs in the
196 JSON_checker testsuite, so add (the corrected) JSON_checker tests to
197 the testsuite.
198 - quite a bit of doc updates/extension.
199 - require 5.8.2, as this seems to be the first unicode-stable version.
200
201 1.5 Tue Aug 28 04:05:38 CEST 2007
202 - add support for tied hashes, based on ideas and testcase by
203 Marcus Holland-Moritz.
204 - implemented relaxed parsing mode where some extensions are being
205 accepted. generation is still JSON-only.
206
207 1.44 Wed Aug 22 01:02:44 CEST 2007
208 - very experimental process-emulation support, slowing everything down.
209 the horribly broken perl threads are still not supported - YMMV.
210
211 1.43 Thu Jul 26 13:26:37 CEST 2007
212 - convert big json numbers exclusively consisting of digits to NV
213 only when there is no loss of precision, otherwise to string.
214
215 1.42 Tue Jul 24 00:51:18 CEST 2007
216 - fix a crash caused by not handling missing array elements
217 (report and testcase by Jay Kuri).
218
219 1.41 Tue Jul 10 18:21:44 CEST 2007
220 - fix compilation with NDEBUG (assert side-effect),
221 affects convert_blessed only.
222 - fix a bug in decode filters calling ENTER; SAVETMPS;
223 one time too often.
224 - catch a typical error in TO_JSON methods.
225 - antique-ised XS.xs again to work with outdated
226 C compilers (windows...).
227
228 1.4 Mon Jul 2 10:06:30 CEST 2007
229 - add convert_blessed setting.
230 - encode did not catch all blessed objects, encoding their
231 contents in most cases. This has been fixed by introducing
232 the allow_blessed setting.
233 - added filter_json_object and filter_json_single_key_object
234 settings that specify a callback to be called when
235 all/specific json objects are encountered.
236 - assume that most object keys are simple ascii words and
237 optimise this case, penalising the general case. This can
238 speed up decoding by 30% in typical cases and gives
239 a smaller and faster perl hash.
240 - implemented simpleminded, optional resource size checking
241 in decode_json.
242 - remove objToJson/jsonToObj aliases, as the next version
243 of JSON will not have them either.
244 - bit the bullet and converted the very simple json object
245 into a more complex one.
246 - work around a bug where perl wrongly claims an integer
247 is not an integer.
248 - unbundle JSON::XS::Boolean into own pm file so Storable
249 and similar modules can resolve the overloading when thawing.
250
251 1.3 Sun Jun 24 01:55:02 CEST 2007
252 - make JSON::XS::true and false special overloaded objects
253 and return those instead of 1 and 0 for those json atoms
254 (JSON::PP compatibility is NOT achieved yet).
255 - add JSON::XS::is_bool predicate to test for those special
256 values.
257 - add a reference to
258 http://jpsykes.com/47/practical-csrf-and-json-security.
259 - removed require 5.8.8 again, it is just not very expert-friendly.
260 Also try to be more compatible with slightly older versions,
261 which are not recommended (because they are buggy).
262
263 1.24 Mon Jun 11 05:40:49 CEST 2007
264 - added informative section on JSON-as-YAML.
265 - get rid of some c99-isms again.
266 - localise dec->cur in decode_str, speeding up
267 string decoding considerably (>15% on my amd64 + gcc).
268 - increased SHORT_STRING_LEN to 16kb: stack space is
269 usually plenty, and this actually saves memory
270 when !shrinking as short strings will fit perfectly.
271
272 1.23 Wed Jun 6 20:13:06 CEST 2007
273 - greatly improved small integer encoding and decoding speed.
274 - implement a number of µ-optimisations.
275 - updated benchmarks.
276
277 1.22 Thu May 24 00:07:25 CEST 2007
278 - require 5.8.8 explicitly as older perls do not seem to offer
279 the required macros.
280 - possibly made it compile on so-called C compilers by microsoft.
281
282 1.21 Wed May 9 18:40:32 CEST 2007
283 - character offset reported for trailing garbage was random.
284
285 1.2 Wed May 9 18:35:01 CEST 2007
286 - decode did not work with magical scalars (doh!).
287 - added latin1 flag to produce JSON texts in the latin1 subset
288 of unicode.
289 - flag trailing garbage as error.
290 - new decode_prefix method that returns the number
291 of characters consumed by a decode.
292 - max octets/char in perls UTF-X is actually 13, not 11,
293 as pointed out by Glenn Linderman.
294 - fixed typoe reported by YAMASHINA Hio.
295
296 1.11 Mon Apr 9 07:05:49 CEST 2007
297 - properly 0-terminate sv's returned by encode to help
298 C libraries that expect that 0 to be there.
299 - partially "port" JSON from C to microsofts fucking broken
300 pseudo-C. They should be burned to the ground for pissing
301 on standards. And I should be stoned for even trying to
302 support this filthy excuse for a c compiler.
303
304 1.1 Wed Apr 4 01:45:00 CEST 2007
305 - clarify documentation (pointed out by Quinn Weaver).
306 - decode_utf8 sometimes did not correctly flag errors,
307 leading to segfaults.
308 - further reduced default nesting depth to 512 due to the test
309 failure by that anonymous "chris" whose e-mail address seems
310 to be impossible to get. Tests on other freebsd systems indicate
311 that this is likely a problem in his/her configuration and not this
312 module.
313 - renamed json => JSON in error messages.
314 - corrected the character offset in some error messages.
315
316 1.01 Sat Mar 31 16:15:40 CEST 2007
317 - do not segfault when from_json/decode gets passed
318 a non-string object (reported by Florian Ragwitz).
319 This has no effect on normal operation.
320
321 1.0 Thu Mar 29 04:43:34 CEST 2007
322 - the long awaited (by me) 1.0 version.
323 - add \0 (JSON::XS::false) and \1 (JSON::XS::true) mappings to JSON
324 true and false.
325 - add some more notes to shrink, as suggested by Alex Efros.
326 - improve testsuite.
327 - halve the default nesting depth limit, to hopefully make it
328 work on Freebsd (unfortunately, the cpan tester did not
329 send me his report, so I cannot ask about the stack limit on fbsd).
330
331 0.8 Mon Mar 26 00:10:48 CEST 2007
332 - fix a memleak when decoding hashes.
333 - export jsonToBj and objToJson as aliases
334 to to_json and from_json, to reduce incompatibilities
335 between JSON/JSON::PC and JSON::XS. (experimental).
336 - implement a maximum nesting depth for both en- and de-coding.
337 - added a security considerations sections.
338
339 0.7 Sun Mar 25 01:46:30 CET 2007
340 - code cleanup.
341 - fix a memory overflow bug when indenting.
342 - pretty-printing now up to 15% faster.
343 - improve decoding speed of strings by
344 up to 50% by specialcasing short strings.
345 - further decoding speedups for strings using
346 lots of \u escapes.
347 - improve utf8 decoding speed for U+80 .. U+7FF.
348
349 0.5 Sat Mar 24 20:41:51 CET 2007
350 - added the UTF-16 encoding example hinted at in previous
351 versions.
352 - minor documentation fixes.
353 - fix a bug in and optimise canonicalising fastpath
354 (reported by Craig Manley).
355 - remove a subtest that breaks with bleadperl (reported
356 by Andreas König).
357
358 0.31 Sat Mar 24 02:14:34 CET 2007
359 - documentation updates.
360 - do some casting to hopefully fix Andreas' problem.
361 - nuke bogus json rpc stuff.
362
363 0.3 Fri Mar 23 19:33:21 CET 2007
364 - remove spurious PApp::Util reference (John McNamara).
365 - adapted lots of tests from other json modules
366 (idea by Chris Carline).
367 - documented mapping from json to perl and vice versa.
368 - improved the documentation by adding more examples.
369 - added short escaping forms, reducing the created
370 json texts a bit.
371 - added shrink flag.
372 - when flag methods are called without enable argument
373 they will by default enable their flag.
374 - considerably improved string encoding speed (at least
375 with gcc 4).
376 - added a test that covers lots of different characters.
377 - clarified some error messages.
378 - error messages now use correct character offset
379 with F_UTF8.
380 - improve the "no bytes" and "no warnings" hacks in
381 case the called functions do... stuff.
382 - croak when encoding to ascii and an out-of-range
383 (non-unicode) codepoint is encountered.
384
385 0.2 Fri Mar 23 00:23:34 CET 2007
386 - the "could not sleep without debugging release".
387 it should basically work now, with many bugs as
388 no production tests have been run yet.
389 - added more testcases.
390 - the expected shitload of bugfixes.
391 - handle utf8 flag correctly in decode.
392 - fix segfault in decoder.
393 - utf8n_to_uvuni sets retlen to -1, but retlen is an
394 unsigned types (argh).
395 - fix decoding of utf-8 strings.
396 - improved error diagnostics.
397 - fix decoding of 'null'.
398 - fix parsing of empty array/hashes
399 - silence warnings when we prepare the croak message.
400
401 0.1 Thu Mar 22 22:13:43 CET 2007
402 - first release, very untested, basically just to claim
403 the namespace.
404
405 0.01 Thu Mar 22 06:08:12 CET 2007
406 - original version; cloned from Convert-Scalar
407