| 1 |
Revision history for Perl extension JSON::XS |
| 2 |
|
| 3 |
- added support for incremental json parsing (still EXPERIMENTAL). |
| 4 |
- implemented and added a json_xs command line utility that can convert |
| 5 |
from/to a number of serialisation formats - tell me if you need more. |
| 6 |
- moved down =encoding directive a bit, too much breaks if its the first |
| 7 |
pod directive :/. |
| 8 |
|
| 9 |
2.1 Wed Mar 19 23:23:18 CET 2008 |
| 10 |
- update documentation here and there: add a large section |
| 11 |
about utf8/latin1/ascii flags, add a security consideration |
| 12 |
and extend and clarify the JSON and YAML section. |
| 13 |
- medium speed enhancements when encoding/decoding non-ascii chars. |
| 14 |
- minor speedup in number encoding case. |
| 15 |
- extend and clarify the section on incompatibilities |
| 16 |
between YAML and JSON. |
| 17 |
- switch to static inline from just inline when using gcc. |
| 18 |
- add =encoding utf-8 to the manpage, now that perl 5.10 supports it. |
| 19 |
- fix some issues with UV to JSON conversion of unknown impact. |
| 20 |
- published the yahoo locals search result used in benchmarks as the |
| 21 |
original url changes so comparison is impossible. |
| 22 |
|
| 23 |
2.01 Wed Dec 5 11:40:28 CET 2007 |
| 24 |
- INCOMPATIBLE API CHANGE: to_json and from_json have been |
| 25 |
renamed to encode_json/decode_json for JSON.pm compatibility. |
| 26 |
The old functions croak and might be replaced by JSON.pm |
| 27 |
comaptible versions in some later release. |
| 28 |
|
| 29 |
2.0 Tue Dec 4 11:30:46 CET 2007 |
| 30 |
- this is supposed to be the first version of JSON::XS |
| 31 |
compatible with version 2.0+ of the JSON module. |
| 32 |
Using the JSON module as frontend to JSON::XS should be |
| 33 |
as fast as using JSON::XS directly, so consider using it |
| 34 |
instead. |
| 35 |
- added get_* methods for all "simple" options. |
| 36 |
- make JSON::XS subclassable. |
| 37 |
|
| 38 |
1.53 Tue Nov 13 23:58:33 CET 2007 |
| 39 |
- minor doc clarifications. |
| 40 |
- fixed many doc typos (patch by Thomas L. Shinnick). |
| 41 |
|
| 42 |
1.52 Mon Oct 15 03:22:06 CEST 2007 |
| 43 |
- remove =encoding pod directive again, it confuses too many pod |
| 44 |
parsers :/. |
| 45 |
|
| 46 |
1.51 Sat Oct 13 03:55:56 CEST 2007 |
| 47 |
- encode empty arrays/hashes in a compact way when pretty is enabled. |
| 48 |
- apparently JSON::XS was used to find some bugs in the |
| 49 |
JSON_checker testsuite, so add (the corrected) JSON_checker tests to |
| 50 |
the testsuite. |
| 51 |
- quite a bit of doc updates/extension. |
| 52 |
- require 5.8.2, as this seems to be the first unicode-stable version. |
| 53 |
|
| 54 |
1.5 Tue Aug 28 04:05:38 CEST 2007 |
| 55 |
- add support for tied hashes, based on ideas and testcase by |
| 56 |
Marcus Holland-Moritz. |
| 57 |
- implemented relaxed parsing mode where some extensions are being |
| 58 |
accepted. generation is still JSON-only. |
| 59 |
|
| 60 |
1.44 Wed Aug 22 01:02:44 CEST 2007 |
| 61 |
- very experimental process-emulation support, slowing everything down. |
| 62 |
the horribly broken perl threads are still not supported - YMMV. |
| 63 |
|
| 64 |
1.43 Thu Jul 26 13:26:37 CEST 2007 |
| 65 |
- convert big json numbers exclusively consisting of digits to NV |
| 66 |
only when there is no loss of precision, otherwise to string. |
| 67 |
|
| 68 |
1.42 Tue Jul 24 00:51:18 CEST 2007 |
| 69 |
- fix a crash caused by not handling missing array elements |
| 70 |
(report and testcase by Jay Kuri). |
| 71 |
|
| 72 |
1.41 Tue Jul 10 18:21:44 CEST 2007 |
| 73 |
- fix compilation with NDEBUG (assert side-effect), |
| 74 |
affects convert_blessed only. |
| 75 |
- fix a bug in decode filters calling ENTER; SAVETMPS; |
| 76 |
one time too often. |
| 77 |
- catch a typical error in TO_JSON methods. |
| 78 |
- antique-ised XS.xs again to work with outdated |
| 79 |
C compilers (windows...). |
| 80 |
|
| 81 |
1.4 Mon Jul 2 10:06:30 CEST 2007 |
| 82 |
- add convert_blessed setting. |
| 83 |
- encode did not catch all blessed objects, encoding their |
| 84 |
contents in most cases. This has been fixed by introducing |
| 85 |
the allow_blessed setting. |
| 86 |
- added filter_json_object and filter_json_single_key_object |
| 87 |
settings that specify a callback to be called when |
| 88 |
all/specific json objects are encountered. |
| 89 |
- assume that most object keys are simple ascii words and |
| 90 |
optimise this case, penalising the general case. This can |
| 91 |
speed up decoding by 30% in typical cases and gives |
| 92 |
a smaller and faster perl hash. |
| 93 |
- implemented simpleminded, optional resource size checking |
| 94 |
in decode_json. |
| 95 |
- remove objToJson/jsonToObj aliases, as the next version |
| 96 |
of JSON will not have them either. |
| 97 |
- bit the bullet and converted the very simple json object |
| 98 |
into a more complex one. |
| 99 |
- work around a bug where perl wrongly claims an integer |
| 100 |
is not an integer. |
| 101 |
- unbundle JSON::XS::Boolean into own pm file so Storable |
| 102 |
and similar modules can resolve the overloading when thawing. |
| 103 |
|
| 104 |
1.3 Sun Jun 24 01:55:02 CEST 2007 |
| 105 |
- make JSON::XS::true and false special overloaded objects |
| 106 |
and return those instead of 1 and 0 for those json atoms |
| 107 |
(JSON::PP compatibility is NOT achieved yet). |
| 108 |
- add JSON::XS::is_bool predicate to test for those special |
| 109 |
values. |
| 110 |
- add a reference to |
| 111 |
http://jpsykes.com/47/practical-csrf-and-json-security. |
| 112 |
- removed require 5.8.8 again, it is just not very expert-friendly. |
| 113 |
Also try to be more compatible with slightly older versions, |
| 114 |
which are not recommended (because they are buggy). |
| 115 |
|
| 116 |
1.24 Mon Jun 11 05:40:49 CEST 2007 |
| 117 |
- added informative section on JSON-as-YAML. |
| 118 |
- get rid of some c99-isms again. |
| 119 |
- localise dec->cur in decode_str, speeding up |
| 120 |
string decoding considerably (>15% on my amd64 + gcc). |
| 121 |
- increased SHORT_STRING_LEN to 16kb: stack space is |
| 122 |
usually plenty, and this actually saves memory |
| 123 |
when !shrinking as short strings will fit perfectly. |
| 124 |
|
| 125 |
1.23 Wed Jun 6 20:13:06 CEST 2007 |
| 126 |
- greatly improved small integer encoding and decoding speed. |
| 127 |
- implement a number of µ-optimisations. |
| 128 |
- updated benchmarks. |
| 129 |
|
| 130 |
1.22 Thu May 24 00:07:25 CEST 2007 |
| 131 |
- require 5.8.8 explicitly as older perls do not seem to offer |
| 132 |
the required macros. |
| 133 |
- possibly made it compile on so-called C compilers by microsoft. |
| 134 |
|
| 135 |
1.21 Wed May 9 18:40:32 CEST 2007 |
| 136 |
- character offset reported for trailing garbage was random. |
| 137 |
|
| 138 |
1.2 Wed May 9 18:35:01 CEST 2007 |
| 139 |
- decode did not work with magical scalars (doh!). |
| 140 |
- added latin1 flag to produce JSON texts in the latin1 subset |
| 141 |
of unicode. |
| 142 |
- flag trailing garbage as error. |
| 143 |
- new decode_prefix method that returns the number |
| 144 |
of characters consumed by a decode. |
| 145 |
- max octets/char in perls UTF-X is actually 13, not 11, |
| 146 |
as pointed out by Glenn Linderman. |
| 147 |
- fixed typoe reported by YAMASHINA Hio. |
| 148 |
|
| 149 |
1.11 Mon Apr 9 07:05:49 CEST 2007 |
| 150 |
- properly 0-terminate sv's returned by encode to help |
| 151 |
C libraries that expect that 0 to be there. |
| 152 |
- partially "port" JSON from C to microsofts fucking broken |
| 153 |
pseudo-C. They should be burned to the ground for pissing |
| 154 |
on standards. And I should be stoned for even trying to |
| 155 |
support this filthy excuse for a c compiler. |
| 156 |
|
| 157 |
1.1 Wed Apr 4 01:45:00 CEST 2007 |
| 158 |
- clarify documentation (pointed out by Quinn Weaver). |
| 159 |
- decode_utf8 sometimes did not correctly flag errors, |
| 160 |
leading to segfaults. |
| 161 |
- further reduced default nesting depth to 512 due to the test |
| 162 |
failure by that anonymous "chris" whose e-mail address seems |
| 163 |
to be impossible to get. Tests on other freebsd systems indicate |
| 164 |
that this is likely a problem in his/her configuration and not this |
| 165 |
module. |
| 166 |
- renamed json => JSON in error messages. |
| 167 |
- corrected the character offset in some error messages. |
| 168 |
|
| 169 |
1.01 Sat Mar 31 16:15:40 CEST 2007 |
| 170 |
- do not segfault when from_json/decode gets passed |
| 171 |
a non-string object (reported by Florian Ragwitz). |
| 172 |
This has no effect on normal operation. |
| 173 |
|
| 174 |
1.0 Thu Mar 29 04:43:34 CEST 2007 |
| 175 |
- the long awaited (by me) 1.0 version. |
| 176 |
- add \0 (JSON::XS::false) and \1 (JSON::XS::true) mappings to JSON |
| 177 |
true and false. |
| 178 |
- add some more notes to shrink, as suggested by Alex Efros. |
| 179 |
- improve testsuite. |
| 180 |
- halve the default nesting depth limit, to hopefully make it |
| 181 |
work on Freebsd (unfortunately, the cpan tester did not |
| 182 |
send me his report, so I cannot ask about the stack limit on fbsd). |
| 183 |
|
| 184 |
0.8 Mon Mar 26 00:10:48 CEST 2007 |
| 185 |
- fix a memleak when decoding hashes. |
| 186 |
- export jsonToBj and objToJson as aliases |
| 187 |
to to_json and from_json, to reduce incompatibilities |
| 188 |
between JSON/JSON::PC and JSON::XS. (experimental). |
| 189 |
- implement a maximum nesting depth for both en- and de-coding. |
| 190 |
- added a security considerations sections. |
| 191 |
|
| 192 |
0.7 Sun Mar 25 01:46:30 CET 2007 |
| 193 |
- code cleanup. |
| 194 |
- fix a memory overflow bug when indenting. |
| 195 |
- pretty-printing now up to 15% faster. |
| 196 |
- improve decoding speed of strings by |
| 197 |
up to 50% by specialcasing short strings. |
| 198 |
- further decoding speedups for strings using |
| 199 |
lots of \u escapes. |
| 200 |
- improve utf8 decoding speed for U+80 .. U+7FF. |
| 201 |
|
| 202 |
0.5 Sat Mar 24 20:41:51 CET 2007 |
| 203 |
- added the UTF-16 encoding example hinted at in previous |
| 204 |
versions. |
| 205 |
- minor documentation fixes. |
| 206 |
- fix a bug in and optimise canonicalising fastpath |
| 207 |
(reported by Craig Manley). |
| 208 |
- remove a subtest that breaks with bleadperl (reported |
| 209 |
by Andreas König). |
| 210 |
|
| 211 |
0.31 Sat Mar 24 02:14:34 CET 2007 |
| 212 |
- documentation updates. |
| 213 |
- do some casting to hopefully fix Andreas' problem. |
| 214 |
- nuke bogus json rpc stuff. |
| 215 |
|
| 216 |
0.3 Fri Mar 23 19:33:21 CET 2007 |
| 217 |
- remove spurious PApp::Util reference (John McNamara). |
| 218 |
- adapted lots of tests from other json modules |
| 219 |
(idea by Chris Carline). |
| 220 |
- documented mapping from json to perl and vice versa. |
| 221 |
- improved the documentation by adding more examples. |
| 222 |
- added short escaping forms, reducing the created |
| 223 |
json texts a bit. |
| 224 |
- added shrink flag. |
| 225 |
- when flag methods are called without enable argument |
| 226 |
they will by default enable their flag. |
| 227 |
- considerably improved string encoding speed (at least |
| 228 |
with gcc 4). |
| 229 |
- added a test that covers lots of different characters. |
| 230 |
- clarified some error messages. |
| 231 |
- error messages now use correct character offset |
| 232 |
with F_UTF8. |
| 233 |
- improve the "no bytes" and "no warnings" hacks in |
| 234 |
case the called functions do... stuff. |
| 235 |
- croak when encoding to ascii and an out-of-range |
| 236 |
(non-unicode) codepoint is encountered. |
| 237 |
|
| 238 |
0.2 Fri Mar 23 00:23:34 CET 2007 |
| 239 |
- the "could not sleep without debuggign release". |
| 240 |
it should basically work now, with many bugs as |
| 241 |
no production tests have been run yet. |
| 242 |
- added more testcases. |
| 243 |
- the expected shitload of bugfixes. |
| 244 |
- handle utf8 flag correctly in decode. |
| 245 |
- fix segfault in decoder. |
| 246 |
- utf8n_to_uvuni sets retlen to -1, but retlen is an |
| 247 |
unsigned types (argh). |
| 248 |
- fix decoding of utf-8 strings. |
| 249 |
- improved error diagnostics. |
| 250 |
- fix decoding of 'null'. |
| 251 |
- fix parsing of empty array/hashes |
| 252 |
- silence warnings when we prepare the croak message. |
| 253 |
|
| 254 |
0.1 Thu Mar 22 22:13:43 CET 2007 |
| 255 |
- first release, very untested, basically just to claim |
| 256 |
the namespace. |
| 257 |
|
| 258 |
0.01 Thu Mar 22 06:08:12 CET 2007 |
| 259 |
- original version; cloned from Convert-Scalar |
| 260 |
|
