1 | =head1 NAME |
1 | =head1 NAME |
2 | |
2 | |
3 | JSON::XS - JSON serialising/deserialising, done correctly and fast |
3 | JSON::XS - JSON serialising/deserialising, done correctly and fast |
4 | |
4 | |
|
|
5 | JSON::XS - 正しくて高速な JSON シリアライザ/デシリアライザ |
|
|
6 | (http://fleur.hio.jp/perldoc/mix/lib/JSON/XS.html) |
|
|
7 | |
5 | =head1 SYNOPSIS |
8 | =head1 SYNOPSIS |
6 | |
9 | |
7 | use JSON::XS; |
10 | use JSON::XS; |
8 | |
11 | |
9 | # exported functions, croak on error |
12 | # exported functions, they croak on error |
|
|
13 | # and expect/generate UTF-8 |
10 | |
14 | |
11 | $utf8_encoded_json_text = to_json $perl_hash_or_arrayref; |
15 | $utf8_encoded_json_text = to_json $perl_hash_or_arrayref; |
12 | $perl_hash_or_arrayref = from_json $utf8_encoded_json_text; |
16 | $perl_hash_or_arrayref = from_json $utf8_encoded_json_text; |
13 | |
17 | |
14 | # oo-interface |
18 | # OO-interface |
15 | |
19 | |
16 | $coder = JSON::XS->new->ascii->pretty->allow_nonref; |
20 | $coder = JSON::XS->new->ascii->pretty->allow_nonref; |
17 | $pretty_printed_unencoded = $coder->encode ($perl_scalar); |
21 | $pretty_printed_unencoded = $coder->encode ($perl_scalar); |
18 | $perl_scalar = $coder->decode ($unicode_json_text); |
22 | $perl_scalar = $coder->decode ($unicode_json_text); |
19 | |
23 | |
… | |
… | |
36 | |
40 | |
37 | =head2 FEATURES |
41 | =head2 FEATURES |
38 | |
42 | |
39 | =over 4 |
43 | =over 4 |
40 | |
44 | |
41 | =item * correct handling of unicode issues |
45 | =item * correct Unicode handling |
42 | |
46 | |
43 | This module knows how to handle Unicode, and even documents how and when |
47 | This module knows how to handle Unicode, and even documents how and when |
44 | it does so. |
48 | it does so. |
45 | |
49 | |
46 | =item * round-trip integrity |
50 | =item * round-trip integrity |
47 | |
51 | |
48 | When you serialise a perl data structure using only datatypes supported |
52 | When you serialise a perl data structure using only datatypes supported |
49 | by JSON, the deserialised data structure is identical on the Perl level. |
53 | by JSON, the deserialised data structure is identical on the Perl level. |
50 | (e.g. the string "2.0" doesn't suddenly become "2"). |
54 | (e.g. the string "2.0" doesn't suddenly become "2" just because it looks |
|
|
55 | like a number). |
51 | |
56 | |
52 | =item * strict checking of JSON correctness |
57 | =item * strict checking of JSON correctness |
53 | |
58 | |
54 | There is no guessing, no generating of illegal JSON texts by default, |
59 | There is no guessing, no generating of illegal JSON texts by default, |
55 | and only JSON is accepted as input by default (the latter is a security |
60 | and only JSON is accepted as input by default (the latter is a security |
… | |
… | |
65 | This module has both a simple functional interface as well as an OO |
70 | This module has both a simple functional interface as well as an OO |
66 | interface. |
71 | interface. |
67 | |
72 | |
68 | =item * reasonably versatile output formats |
73 | =item * reasonably versatile output formats |
69 | |
74 | |
70 | You can choose between the most compact guarenteed single-line format |
75 | You can choose between the most compact guaranteed single-line format |
71 | possible (nice for simple line-based protocols), a pure-ascii format (for |
76 | possible (nice for simple line-based protocols), a pure-ascii format |
72 | when your transport is not 8-bit clean), or a pretty-printed format (for |
77 | (for when your transport is not 8-bit clean, still supports the whole |
73 | when you want to read that stuff). Or you can combine those features in |
78 | Unicode range), or a pretty-printed format (for when you want to read that |
74 | whatever way you like. |
79 | stuff). Or you can combine those features in whatever way you like. |
75 | |
80 | |
76 | =back |
81 | =back |
77 | |
82 | |
78 | =cut |
83 | =cut |
79 | |
84 | |
80 | package JSON::XS; |
85 | package JSON::XS; |
81 | |
86 | |
82 | BEGIN { |
87 | use strict; |
83 | $VERSION = '0.31'; |
88 | |
|
|
89 | our $VERSION = '1.52'; |
84 | @ISA = qw(Exporter); |
90 | our @ISA = qw(Exporter); |
85 | |
91 | |
86 | @EXPORT = qw(to_json from_json); |
92 | our @EXPORT = qw(to_json from_json); |
87 | require Exporter; |
|
|
88 | |
93 | |
89 | require XSLoader; |
94 | use Exporter; |
90 | XSLoader::load JSON::XS::, $VERSION; |
95 | use XSLoader; |
91 | } |
|
|
92 | |
96 | |
93 | =head1 FUNCTIONAL INTERFACE |
97 | =head1 FUNCTIONAL INTERFACE |
94 | |
98 | |
95 | The following convinience methods are provided by this module. They are |
99 | The following convenience methods are provided by this module. They are |
96 | exported by default: |
100 | exported by default: |
97 | |
101 | |
98 | =over 4 |
102 | =over 4 |
99 | |
103 | |
100 | =item $json_text = to_json $perl_scalar |
104 | =item $json_text = to_json $perl_scalar |
101 | |
105 | |
102 | Converts the given Perl data structure (a simple scalar or a reference to |
106 | Converts the given Perl data structure to a UTF-8 encoded, binary string |
103 | a hash or array) to a UTF-8 encoded, binary string (that is, the string contains |
107 | (that is, the string contains octets only). Croaks on error. |
104 | octets only). Croaks on error. |
|
|
105 | |
108 | |
106 | This function call is functionally identical to: |
109 | This function call is functionally identical to: |
107 | |
110 | |
108 | $json_text = JSON::XS->new->utf8->encode ($perl_scalar) |
111 | $json_text = JSON::XS->new->utf8->encode ($perl_scalar) |
109 | |
112 | |
110 | except being faster. |
113 | except being faster. |
111 | |
114 | |
112 | =item $perl_scalar = from_json $json_text |
115 | =item $perl_scalar = from_json $json_text |
113 | |
116 | |
114 | The opposite of C<to_json>: expects an UTF-8 (binary) string and tries to |
117 | The opposite of C<to_json>: expects an UTF-8 (binary) string and tries |
115 | parse that as an UTF-8 encoded JSON text, returning the resulting simple |
118 | to parse that as an UTF-8 encoded JSON text, returning the resulting |
116 | scalar or reference. Croaks on error. |
119 | reference. Croaks on error. |
117 | |
120 | |
118 | This function call is functionally identical to: |
121 | This function call is functionally identical to: |
119 | |
122 | |
120 | $perl_scalar = JSON::XS->new->utf8->decode ($json_text) |
123 | $perl_scalar = JSON::XS->new->utf8->decode ($json_text) |
121 | |
124 | |
122 | except being faster. |
125 | except being faster. |
123 | |
126 | |
|
|
127 | =item $is_boolean = JSON::XS::is_bool $scalar |
|
|
128 | |
|
|
129 | Returns true if the passed scalar represents either JSON::XS::true or |
|
|
130 | JSON::XS::false, two constants that act like C<1> and C<0>, respectively |
|
|
131 | and are used to represent JSON C<true> and C<false> values in Perl. |
|
|
132 | |
|
|
133 | See MAPPING, below, for more information on how JSON values are mapped to |
|
|
134 | Perl. |
|
|
135 | |
124 | =back |
136 | =back |
|
|
137 | |
|
|
138 | |
|
|
139 | =head1 A FEW NOTES ON UNICODE AND PERL |
|
|
140 | |
|
|
141 | Since this often leads to confusion, here are a few very clear words on |
|
|
142 | how Unicode works in Perl, modulo bugs. |
|
|
143 | |
|
|
144 | =over 4 |
|
|
145 | |
|
|
146 | =item 1. Perl strings can store characters with ordinal values > 255. |
|
|
147 | |
|
|
148 | This enables you to store Unicode characters as single characters in a |
|
|
149 | Perl string - very natural. |
|
|
150 | |
|
|
151 | =item 2. Perl does I<not> associate an encoding with your strings. |
|
|
152 | |
|
|
153 | Unless you force it to, e.g. when matching it against a regex, or printing |
|
|
154 | the scalar to a file, in which case Perl either interprets your string as |
|
|
155 | locale-encoded text, octets/binary, or as Unicode, depending on various |
|
|
156 | settings. In no case is an encoding stored together with your data, it is |
|
|
157 | I<use> that decides encoding, not any magical metadata. |
|
|
158 | |
|
|
159 | =item 3. The internal utf-8 flag has no meaning with regards to the |
|
|
160 | encoding of your string. |
|
|
161 | |
|
|
162 | Just ignore that flag unless you debug a Perl bug, a module written in |
|
|
163 | XS or want to dive into the internals of perl. Otherwise it will only |
|
|
164 | confuse you, as, despite the name, it says nothing about how your string |
|
|
165 | is encoded. You can have Unicode strings with that flag set, with that |
|
|
166 | flag clear, and you can have binary data with that flag set and that flag |
|
|
167 | clear. Other possibilities exist, too. |
|
|
168 | |
|
|
169 | If you didn't know about that flag, just the better, pretend it doesn't |
|
|
170 | exist. |
|
|
171 | |
|
|
172 | =item 4. A "Unicode String" is simply a string where each character can be |
|
|
173 | validly interpreted as a Unicode codepoint. |
|
|
174 | |
|
|
175 | If you have UTF-8 encoded data, it is no longer a Unicode string, but a |
|
|
176 | Unicode string encoded in UTF-8, giving you a binary string. |
|
|
177 | |
|
|
178 | =item 5. A string containing "high" (> 255) character values is I<not> a UTF-8 string. |
|
|
179 | |
|
|
180 | It's a fact. Learn to live with it. |
|
|
181 | |
|
|
182 | =back |
|
|
183 | |
|
|
184 | I hope this helps :) |
|
|
185 | |
125 | |
186 | |
126 | =head1 OBJECT-ORIENTED INTERFACE |
187 | =head1 OBJECT-ORIENTED INTERFACE |
127 | |
188 | |
128 | The object oriented interface lets you configure your own encoding or |
189 | The object oriented interface lets you configure your own encoding or |
129 | decoding style, within the limits of supported formats. |
190 | decoding style, within the limits of supported formats. |
… | |
… | |
143 | |
204 | |
144 | =item $json = $json->ascii ([$enable]) |
205 | =item $json = $json->ascii ([$enable]) |
145 | |
206 | |
146 | If C<$enable> is true (or missing), then the C<encode> method will not |
207 | If C<$enable> is true (or missing), then the C<encode> method will not |
147 | generate characters outside the code range C<0..127> (which is ASCII). Any |
208 | generate characters outside the code range C<0..127> (which is ASCII). Any |
148 | unicode characters outside that range will be escaped using either a |
209 | Unicode characters outside that range will be escaped using either a |
149 | single \uXXXX (BMP characters) or a double \uHHHH\uLLLLL escape sequence, |
210 | single \uXXXX (BMP characters) or a double \uHHHH\uLLLLL escape sequence, |
150 | as per RFC4627. |
211 | as per RFC4627. The resulting encoded JSON text can be treated as a native |
|
|
212 | Unicode string, an ascii-encoded, latin1-encoded or UTF-8 encoded string, |
|
|
213 | or any other superset of ASCII. |
151 | |
214 | |
152 | If C<$enable> is false, then the C<encode> method will not escape Unicode |
215 | If C<$enable> is false, then the C<encode> method will not escape Unicode |
153 | characters unless required by the JSON syntax. This results in a faster |
216 | characters unless required by the JSON syntax or other flags. This results |
154 | and more compact format. |
217 | in a faster and more compact format. |
|
|
218 | |
|
|
219 | The main use for this flag is to produce JSON texts that can be |
|
|
220 | transmitted over a 7-bit channel, as the encoded JSON texts will not |
|
|
221 | contain any 8 bit characters. |
155 | |
222 | |
156 | JSON::XS->new->ascii (1)->encode ([chr 0x10401]) |
223 | JSON::XS->new->ascii (1)->encode ([chr 0x10401]) |
157 | => ["\ud801\udc01"] |
224 | => ["\ud801\udc01"] |
|
|
225 | |
|
|
226 | =item $json = $json->latin1 ([$enable]) |
|
|
227 | |
|
|
228 | If C<$enable> is true (or missing), then the C<encode> method will encode |
|
|
229 | the resulting JSON text as latin1 (or iso-8859-1), escaping any characters |
|
|
230 | outside the code range C<0..255>. The resulting string can be treated as a |
|
|
231 | latin1-encoded JSON text or a native Unicode string. The C<decode> method |
|
|
232 | will not be affected in any way by this flag, as C<decode> by default |
|
|
233 | expects Unicode, which is a strict superset of latin1. |
|
|
234 | |
|
|
235 | If C<$enable> is false, then the C<encode> method will not escape Unicode |
|
|
236 | characters unless required by the JSON syntax or other flags. |
|
|
237 | |
|
|
238 | The main use for this flag is efficiently encoding binary data as JSON |
|
|
239 | text, as most octets will not be escaped, resulting in a smaller encoded |
|
|
240 | size. The disadvantage is that the resulting JSON text is encoded |
|
|
241 | in latin1 (and must correctly be treated as such when storing and |
|
|
242 | transferring), a rare encoding for JSON. It is therefore most useful when |
|
|
243 | you want to store data structures known to contain binary data efficiently |
|
|
244 | in files or databases, not when talking to other JSON encoders/decoders. |
|
|
245 | |
|
|
246 | JSON::XS->new->latin1->encode (["\x{89}\x{abc}"] |
|
|
247 | => ["\x{89}\\u0abc"] # (perl syntax, U+abc escaped, U+89 not) |
158 | |
248 | |
159 | =item $json = $json->utf8 ([$enable]) |
249 | =item $json = $json->utf8 ([$enable]) |
160 | |
250 | |
161 | If C<$enable> is true (or missing), then the C<encode> method will encode |
251 | If C<$enable> is true (or missing), then the C<encode> method will encode |
162 | the JSON result into UTF-8, as required by many protocols, while the |
252 | the JSON result into UTF-8, as required by many protocols, while the |
… | |
… | |
165 | range C<0..255>, they are thus useful for bytewise/binary I/O. In future |
255 | range C<0..255>, they are thus useful for bytewise/binary I/O. In future |
166 | versions, enabling this option might enable autodetection of the UTF-16 |
256 | versions, enabling this option might enable autodetection of the UTF-16 |
167 | and UTF-32 encoding families, as described in RFC4627. |
257 | and UTF-32 encoding families, as described in RFC4627. |
168 | |
258 | |
169 | If C<$enable> is false, then the C<encode> method will return the JSON |
259 | If C<$enable> is false, then the C<encode> method will return the JSON |
170 | string as a (non-encoded) unicode string, while C<decode> expects thus a |
260 | string as a (non-encoded) Unicode string, while C<decode> expects thus a |
171 | unicode string. Any decoding or encoding (e.g. to UTF-8 or UTF-16) needs |
261 | Unicode string. Any decoding or encoding (e.g. to UTF-8 or UTF-16) needs |
172 | to be done yourself, e.g. using the Encode module. |
262 | to be done yourself, e.g. using the Encode module. |
173 | |
263 | |
174 | Example, output UTF-16BE-encoded JSON: |
264 | Example, output UTF-16BE-encoded JSON: |
175 | |
265 | |
176 | use Encode; |
266 | use Encode; |
… | |
… | |
200 | |
290 | |
201 | =item $json = $json->indent ([$enable]) |
291 | =item $json = $json->indent ([$enable]) |
202 | |
292 | |
203 | If C<$enable> is true (or missing), then the C<encode> method will use a multiline |
293 | If C<$enable> is true (or missing), then the C<encode> method will use a multiline |
204 | format as output, putting every array member or object/hash key-value pair |
294 | format as output, putting every array member or object/hash key-value pair |
205 | into its own line, identing them properly. |
295 | into its own line, indenting them properly. |
206 | |
296 | |
207 | If C<$enable> is false, no newlines or indenting will be produced, and the |
297 | If C<$enable> is false, no newlines or indenting will be produced, and the |
208 | resulting JSON text is guarenteed not to contain any C<newlines>. |
298 | resulting JSON text is guaranteed not to contain any C<newlines>. |
209 | |
299 | |
210 | This setting has no effect when decoding JSON texts. |
300 | This setting has no effect when decoding JSON texts. |
211 | |
301 | |
212 | =item $json = $json->space_before ([$enable]) |
302 | =item $json = $json->space_before ([$enable]) |
213 | |
303 | |
… | |
… | |
238 | |
328 | |
239 | Example, space_before and indent disabled, space_after enabled: |
329 | Example, space_before and indent disabled, space_after enabled: |
240 | |
330 | |
241 | {"key": "value"} |
331 | {"key": "value"} |
242 | |
332 | |
|
|
333 | =item $json = $json->relaxed ([$enable]) |
|
|
334 | |
|
|
335 | If C<$enable> is true (or missing), then C<decode> will accept some |
|
|
336 | extensions to normal JSON syntax (see below). C<encode> will not be |
|
|
337 | affected in anyway. I<Be aware that this option makes you accept invalid |
|
|
338 | JSON texts as if they were valid!>. I suggest only to use this option to |
|
|
339 | parse application-specific files written by humans (configuration files, |
|
|
340 | resource files etc.) |
|
|
341 | |
|
|
342 | If C<$enable> is false (the default), then C<decode> will only accept |
|
|
343 | valid JSON texts. |
|
|
344 | |
|
|
345 | Currently accepted extensions are: |
|
|
346 | |
|
|
347 | =over 4 |
|
|
348 | |
|
|
349 | =item * list items can have an end-comma |
|
|
350 | |
|
|
351 | JSON I<separates> array elements and key-value pairs with commas. This |
|
|
352 | can be annoying if you write JSON texts manually and want to be able to |
|
|
353 | quickly append elements, so this extension accepts comma at the end of |
|
|
354 | such items not just between them: |
|
|
355 | |
|
|
356 | [ |
|
|
357 | 1, |
|
|
358 | 2, <- this comma not normally allowed |
|
|
359 | ] |
|
|
360 | { |
|
|
361 | "k1": "v1", |
|
|
362 | "k2": "v2", <- this comma not normally allowed |
|
|
363 | } |
|
|
364 | |
|
|
365 | =item * shell-style '#'-comments |
|
|
366 | |
|
|
367 | Whenever JSON allows whitespace, shell-style comments are additionally |
|
|
368 | allowed. They are terminated by the first carriage-return or line-feed |
|
|
369 | character, after which more white-space and comments are allowed. |
|
|
370 | |
|
|
371 | [ |
|
|
372 | 1, # this comment not allowed in JSON |
|
|
373 | # neither this one... |
|
|
374 | ] |
|
|
375 | |
|
|
376 | =back |
|
|
377 | |
243 | =item $json = $json->canonical ([$enable]) |
378 | =item $json = $json->canonical ([$enable]) |
244 | |
379 | |
245 | If C<$enable> is true (or missing), then the C<encode> method will output JSON objects |
380 | If C<$enable> is true (or missing), then the C<encode> method will output JSON objects |
246 | by sorting their keys. This is adding a comparatively high overhead. |
381 | by sorting their keys. This is adding a comparatively high overhead. |
247 | |
382 | |
… | |
… | |
249 | pairs in the order Perl stores them (which will likely change between runs |
384 | pairs in the order Perl stores them (which will likely change between runs |
250 | of the same script). |
385 | of the same script). |
251 | |
386 | |
252 | This option is useful if you want the same data structure to be encoded as |
387 | This option is useful if you want the same data structure to be encoded as |
253 | the same JSON text (given the same overall settings). If it is disabled, |
388 | the same JSON text (given the same overall settings). If it is disabled, |
254 | the same hash migh be encoded differently even if contains the same data, |
389 | the same hash might be encoded differently even if contains the same data, |
255 | as key-value pairs have no inherent ordering in Perl. |
390 | as key-value pairs have no inherent ordering in Perl. |
256 | |
391 | |
257 | This setting has no effect when decoding JSON texts. |
392 | This setting has no effect when decoding JSON texts. |
258 | |
393 | |
259 | =item $json = $json->allow_nonref ([$enable]) |
394 | =item $json = $json->allow_nonref ([$enable]) |
… | |
… | |
272 | resulting in an invalid JSON text: |
407 | resulting in an invalid JSON text: |
273 | |
408 | |
274 | JSON::XS->new->allow_nonref->encode ("Hello, World!") |
409 | JSON::XS->new->allow_nonref->encode ("Hello, World!") |
275 | => "Hello, World!" |
410 | => "Hello, World!" |
276 | |
411 | |
|
|
412 | =item $json = $json->allow_blessed ([$enable]) |
|
|
413 | |
|
|
414 | If C<$enable> is true (or missing), then the C<encode> method will not |
|
|
415 | barf when it encounters a blessed reference. Instead, the value of the |
|
|
416 | B<convert_blessed> option will decide whether C<null> (C<convert_blessed> |
|
|
417 | disabled or no C<to_json> method found) or a representation of the |
|
|
418 | object (C<convert_blessed> enabled and C<to_json> method found) is being |
|
|
419 | encoded. Has no effect on C<decode>. |
|
|
420 | |
|
|
421 | If C<$enable> is false (the default), then C<encode> will throw an |
|
|
422 | exception when it encounters a blessed object. |
|
|
423 | |
|
|
424 | =item $json = $json->convert_blessed ([$enable]) |
|
|
425 | |
|
|
426 | If C<$enable> is true (or missing), then C<encode>, upon encountering a |
|
|
427 | blessed object, will check for the availability of the C<TO_JSON> method |
|
|
428 | on the object's class. If found, it will be called in scalar context |
|
|
429 | and the resulting scalar will be encoded instead of the object. If no |
|
|
430 | C<TO_JSON> method is found, the value of C<allow_blessed> will decide what |
|
|
431 | to do. |
|
|
432 | |
|
|
433 | The C<TO_JSON> method may safely call die if it wants. If C<TO_JSON> |
|
|
434 | returns other blessed objects, those will be handled in the same |
|
|
435 | way. C<TO_JSON> must take care of not causing an endless recursion cycle |
|
|
436 | (== crash) in this case. The name of C<TO_JSON> was chosen because other |
|
|
437 | methods called by the Perl core (== not by the user of the object) are |
|
|
438 | usually in upper case letters and to avoid collisions with the C<to_json> |
|
|
439 | function. |
|
|
440 | |
|
|
441 | This setting does not yet influence C<decode> in any way, but in the |
|
|
442 | future, global hooks might get installed that influence C<decode> and are |
|
|
443 | enabled by this setting. |
|
|
444 | |
|
|
445 | If C<$enable> is false, then the C<allow_blessed> setting will decide what |
|
|
446 | to do when a blessed object is found. |
|
|
447 | |
|
|
448 | =item $json = $json->filter_json_object ([$coderef->($hashref)]) |
|
|
449 | |
|
|
450 | When C<$coderef> is specified, it will be called from C<decode> each |
|
|
451 | time it decodes a JSON object. The only argument is a reference to the |
|
|
452 | newly-created hash. If the code references returns a single scalar (which |
|
|
453 | need not be a reference), this value (i.e. a copy of that scalar to avoid |
|
|
454 | aliasing) is inserted into the deserialised data structure. If it returns |
|
|
455 | an empty list (NOTE: I<not> C<undef>, which is a valid scalar), the |
|
|
456 | original deserialised hash will be inserted. This setting can slow down |
|
|
457 | decoding considerably. |
|
|
458 | |
|
|
459 | When C<$coderef> is omitted or undefined, any existing callback will |
|
|
460 | be removed and C<decode> will not change the deserialised hash in any |
|
|
461 | way. |
|
|
462 | |
|
|
463 | Example, convert all JSON objects into the integer 5: |
|
|
464 | |
|
|
465 | my $js = JSON::XS->new->filter_json_object (sub { 5 }); |
|
|
466 | # returns [5] |
|
|
467 | $js->decode ('[{}]') |
|
|
468 | # throw an exception because allow_nonref is not enabled |
|
|
469 | # so a lone 5 is not allowed. |
|
|
470 | $js->decode ('{"a":1, "b":2}'); |
|
|
471 | |
|
|
472 | =item $json = $json->filter_json_single_key_object ($key [=> $coderef->($value)]) |
|
|
473 | |
|
|
474 | Works remotely similar to C<filter_json_object>, but is only called for |
|
|
475 | JSON objects having a single key named C<$key>. |
|
|
476 | |
|
|
477 | This C<$coderef> is called before the one specified via |
|
|
478 | C<filter_json_object>, if any. It gets passed the single value in the JSON |
|
|
479 | object. If it returns a single value, it will be inserted into the data |
|
|
480 | structure. If it returns nothing (not even C<undef> but the empty list), |
|
|
481 | the callback from C<filter_json_object> will be called next, as if no |
|
|
482 | single-key callback were specified. |
|
|
483 | |
|
|
484 | If C<$coderef> is omitted or undefined, the corresponding callback will be |
|
|
485 | disabled. There can only ever be one callback for a given key. |
|
|
486 | |
|
|
487 | As this callback gets called less often then the C<filter_json_object> |
|
|
488 | one, decoding speed will not usually suffer as much. Therefore, single-key |
|
|
489 | objects make excellent targets to serialise Perl objects into, especially |
|
|
490 | as single-key JSON objects are as close to the type-tagged value concept |
|
|
491 | as JSON gets (it's basically an ID/VALUE tuple). Of course, JSON does not |
|
|
492 | support this in any way, so you need to make sure your data never looks |
|
|
493 | like a serialised Perl hash. |
|
|
494 | |
|
|
495 | Typical names for the single object key are C<__class_whatever__>, or |
|
|
496 | C<$__dollars_are_rarely_used__$> or C<}ugly_brace_placement>, or even |
|
|
497 | things like C<__class_md5sum(classname)__>, to reduce the risk of clashing |
|
|
498 | with real hashes. |
|
|
499 | |
|
|
500 | Example, decode JSON objects of the form C<< { "__widget__" => <id> } >> |
|
|
501 | into the corresponding C<< $WIDGET{<id>} >> object: |
|
|
502 | |
|
|
503 | # return whatever is in $WIDGET{5}: |
|
|
504 | JSON::XS |
|
|
505 | ->new |
|
|
506 | ->filter_json_single_key_object (__widget__ => sub { |
|
|
507 | $WIDGET{ $_[0] } |
|
|
508 | }) |
|
|
509 | ->decode ('{"__widget__": 5') |
|
|
510 | |
|
|
511 | # this can be used with a TO_JSON method in some "widget" class |
|
|
512 | # for serialisation to json: |
|
|
513 | sub WidgetBase::TO_JSON { |
|
|
514 | my ($self) = @_; |
|
|
515 | |
|
|
516 | unless ($self->{id}) { |
|
|
517 | $self->{id} = ..get..some..id..; |
|
|
518 | $WIDGET{$self->{id}} = $self; |
|
|
519 | } |
|
|
520 | |
|
|
521 | { __widget__ => $self->{id} } |
|
|
522 | } |
|
|
523 | |
277 | =item $json = $json->shrink ([$enable]) |
524 | =item $json = $json->shrink ([$enable]) |
278 | |
525 | |
279 | Perl usually over-allocates memory a bit when allocating space for |
526 | Perl usually over-allocates memory a bit when allocating space for |
280 | strings. This flag optionally resizes strings generated by either |
527 | strings. This flag optionally resizes strings generated by either |
281 | C<encode> or C<decode> to their minimum size possible. This can save |
528 | C<encode> or C<decode> to their minimum size possible. This can save |
282 | memory when your JSON texts are either very very long or you have many |
529 | memory when your JSON texts are either very very long or you have many |
283 | short strings. It will also try to downgrade any strings to octet-form |
530 | short strings. It will also try to downgrade any strings to octet-form |
284 | if possible: perl stores strings internally either in an encoding called |
531 | if possible: perl stores strings internally either in an encoding called |
285 | UTF-X or in octet-form. The latter cannot store everything but uses less |
532 | UTF-X or in octet-form. The latter cannot store everything but uses less |
286 | space in general. |
533 | space in general (and some buggy Perl or C code might even rely on that |
|
|
534 | internal representation being used). |
287 | |
535 | |
|
|
536 | The actual definition of what shrink does might change in future versions, |
|
|
537 | but it will always try to save space at the expense of time. |
|
|
538 | |
288 | If C<$enable> is true (or missing), the string returned by C<encode> will be shrunk-to-fit, |
539 | If C<$enable> is true (or missing), the string returned by C<encode> will |
289 | while all strings generated by C<decode> will also be shrunk-to-fit. |
540 | be shrunk-to-fit, while all strings generated by C<decode> will also be |
|
|
541 | shrunk-to-fit. |
290 | |
542 | |
291 | If C<$enable> is false, then the normal perl allocation algorithms are used. |
543 | If C<$enable> is false, then the normal perl allocation algorithms are used. |
292 | If you work with your data, then this is likely to be faster. |
544 | If you work with your data, then this is likely to be faster. |
293 | |
545 | |
294 | In the future, this setting might control other things, such as converting |
546 | In the future, this setting might control other things, such as converting |
295 | strings that look like integers or floats into integers or floats |
547 | strings that look like integers or floats into integers or floats |
296 | internally (there is no difference on the Perl level), saving space. |
548 | internally (there is no difference on the Perl level), saving space. |
|
|
549 | |
|
|
550 | =item $json = $json->max_depth ([$maximum_nesting_depth]) |
|
|
551 | |
|
|
552 | Sets the maximum nesting level (default C<512>) accepted while encoding |
|
|
553 | or decoding. If the JSON text or Perl data structure has an equal or |
|
|
554 | higher nesting level then this limit, then the encoder and decoder will |
|
|
555 | stop and croak at that point. |
|
|
556 | |
|
|
557 | Nesting level is defined by number of hash- or arrayrefs that the encoder |
|
|
558 | needs to traverse to reach a given point or the number of C<{> or C<[> |
|
|
559 | characters without their matching closing parenthesis crossed to reach a |
|
|
560 | given character in a string. |
|
|
561 | |
|
|
562 | Setting the maximum depth to one disallows any nesting, so that ensures |
|
|
563 | that the object is only a single hash/object or array. |
|
|
564 | |
|
|
565 | The argument to C<max_depth> will be rounded up to the next highest power |
|
|
566 | of two. If no argument is given, the highest possible setting will be |
|
|
567 | used, which is rarely useful. |
|
|
568 | |
|
|
569 | See SECURITY CONSIDERATIONS, below, for more info on why this is useful. |
|
|
570 | |
|
|
571 | =item $json = $json->max_size ([$maximum_string_size]) |
|
|
572 | |
|
|
573 | Set the maximum length a JSON text may have (in bytes) where decoding is |
|
|
574 | being attempted. The default is C<0>, meaning no limit. When C<decode> |
|
|
575 | is called on a string longer then this number of characters it will not |
|
|
576 | attempt to decode the string but throw an exception. This setting has no |
|
|
577 | effect on C<encode> (yet). |
|
|
578 | |
|
|
579 | The argument to C<max_size> will be rounded up to the next B<highest> |
|
|
580 | power of two (so may be more than requested). If no argument is given, the |
|
|
581 | limit check will be deactivated (same as when C<0> is specified). |
|
|
582 | |
|
|
583 | See SECURITY CONSIDERATIONS, below, for more info on why this is useful. |
297 | |
584 | |
298 | =item $json_text = $json->encode ($perl_scalar) |
585 | =item $json_text = $json->encode ($perl_scalar) |
299 | |
586 | |
300 | Converts the given Perl data structure (a simple scalar or a reference |
587 | Converts the given Perl data structure (a simple scalar or a reference |
301 | to a hash or array) to its JSON representation. Simple scalars will be |
588 | to a hash or array) to its JSON representation. Simple scalars will be |
… | |
… | |
311 | |
598 | |
312 | JSON numbers and strings become simple Perl scalars. JSON arrays become |
599 | JSON numbers and strings become simple Perl scalars. JSON arrays become |
313 | Perl arrayrefs and JSON objects become Perl hashrefs. C<true> becomes |
600 | Perl arrayrefs and JSON objects become Perl hashrefs. C<true> becomes |
314 | C<1>, C<false> becomes C<0> and C<null> becomes C<undef>. |
601 | C<1>, C<false> becomes C<0> and C<null> becomes C<undef>. |
315 | |
602 | |
|
|
603 | =item ($perl_scalar, $characters) = $json->decode_prefix ($json_text) |
|
|
604 | |
|
|
605 | This works like the C<decode> method, but instead of raising an exception |
|
|
606 | when there is trailing garbage after the first JSON object, it will |
|
|
607 | silently stop parsing there and return the number of characters consumed |
|
|
608 | so far. |
|
|
609 | |
|
|
610 | This is useful if your JSON texts are not delimited by an outer protocol |
|
|
611 | (which is not the brightest thing to do in the first place) and you need |
|
|
612 | to know where the JSON text ends. |
|
|
613 | |
|
|
614 | JSON::XS->new->decode_prefix ("[1] the tail") |
|
|
615 | => ([], 3) |
|
|
616 | |
316 | =back |
617 | =back |
|
|
618 | |
317 | |
619 | |
318 | =head1 MAPPING |
620 | =head1 MAPPING |
319 | |
621 | |
320 | This section describes how JSON::XS maps Perl values to JSON values and |
622 | This section describes how JSON::XS maps Perl values to JSON values and |
321 | vice versa. These mappings are designed to "do the right thing" in most |
623 | vice versa. These mappings are designed to "do the right thing" in most |
322 | circumstances automatically, preserving round-tripping characteristics |
624 | circumstances automatically, preserving round-tripping characteristics |
323 | (what you put in comes out as something equivalent). |
625 | (what you put in comes out as something equivalent). |
324 | |
626 | |
325 | For the more enlightened: note that in the following descriptions, |
627 | For the more enlightened: note that in the following descriptions, |
326 | lowercase I<perl> refers to the Perl interpreter, while uppcercase I<Perl> |
628 | lowercase I<perl> refers to the Perl interpreter, while uppercase I<Perl> |
327 | refers to the abstract Perl language itself. |
629 | refers to the abstract Perl language itself. |
328 | |
630 | |
|
|
631 | |
329 | =head2 JSON -> PERL |
632 | =head2 JSON -> PERL |
330 | |
633 | |
331 | =over 4 |
634 | =over 4 |
332 | |
635 | |
333 | =item object |
636 | =item object |
334 | |
637 | |
335 | A JSON object becomes a reference to a hash in Perl. No ordering of object |
638 | A JSON object becomes a reference to a hash in Perl. No ordering of object |
336 | keys is preserved (JSON does not preserver object key ordering itself). |
639 | keys is preserved (JSON does not preserve object key ordering itself). |
337 | |
640 | |
338 | =item array |
641 | =item array |
339 | |
642 | |
340 | A JSON array becomes a reference to an array in Perl. |
643 | A JSON array becomes a reference to an array in Perl. |
341 | |
644 | |
… | |
… | |
345 | are represented by the same codepoints in the Perl string, so no manual |
648 | are represented by the same codepoints in the Perl string, so no manual |
346 | decoding is necessary. |
649 | decoding is necessary. |
347 | |
650 | |
348 | =item number |
651 | =item number |
349 | |
652 | |
350 | A JSON number becomes either an integer or numeric (floating point) |
653 | A JSON number becomes either an integer, numeric (floating point) or |
351 | scalar in perl, depending on its range and any fractional parts. On the |
654 | string scalar in perl, depending on its range and any fractional parts. On |
352 | Perl level, there is no difference between those as Perl handles all the |
655 | the Perl level, there is no difference between those as Perl handles all |
353 | conversion details, but an integer may take slightly less memory and might |
656 | the conversion details, but an integer may take slightly less memory and |
354 | represent more values exactly than (floating point) numbers. |
657 | might represent more values exactly than (floating point) numbers. |
|
|
658 | |
|
|
659 | If the number consists of digits only, JSON::XS will try to represent |
|
|
660 | it as an integer value. If that fails, it will try to represent it as |
|
|
661 | a numeric (floating point) value if that is possible without loss of |
|
|
662 | precision. Otherwise it will preserve the number as a string value. |
|
|
663 | |
|
|
664 | Numbers containing a fractional or exponential part will always be |
|
|
665 | represented as numeric (floating point) values, possibly at a loss of |
|
|
666 | precision. |
|
|
667 | |
|
|
668 | This might create round-tripping problems as numbers might become strings, |
|
|
669 | but as Perl is typeless there is no other way to do it. |
355 | |
670 | |
356 | =item true, false |
671 | =item true, false |
357 | |
672 | |
358 | These JSON atoms become C<0>, C<1>, respectively. Information is lost in |
673 | These JSON atoms become C<JSON::XS::true> and C<JSON::XS::false>, |
359 | this process. Future versions might represent those values differently, |
674 | respectively. They are overloaded to act almost exactly like the numbers |
360 | but they will be guarenteed to act like these integers would normally in |
675 | C<1> and C<0>. You can check whether a scalar is a JSON boolean by using |
361 | Perl. |
676 | the C<JSON::XS::is_bool> function. |
362 | |
677 | |
363 | =item null |
678 | =item null |
364 | |
679 | |
365 | A JSON null atom becomes C<undef> in Perl. |
680 | A JSON null atom becomes C<undef> in Perl. |
366 | |
681 | |
367 | =back |
682 | =back |
|
|
683 | |
368 | |
684 | |
369 | =head2 PERL -> JSON |
685 | =head2 PERL -> JSON |
370 | |
686 | |
371 | The mapping from Perl to JSON is slightly more difficult, as Perl is a |
687 | The mapping from Perl to JSON is slightly more difficult, as Perl is a |
372 | truly typeless language, so we can only guess which JSON type is meant by |
688 | truly typeless language, so we can only guess which JSON type is meant by |
… | |
… | |
375 | =over 4 |
691 | =over 4 |
376 | |
692 | |
377 | =item hash references |
693 | =item hash references |
378 | |
694 | |
379 | Perl hash references become JSON objects. As there is no inherent ordering |
695 | Perl hash references become JSON objects. As there is no inherent ordering |
380 | in hash keys, they will usually be encoded in a pseudo-random order that |
696 | in hash keys (or JSON objects), they will usually be encoded in a |
381 | can change between runs of the same program but stays generally the same |
697 | pseudo-random order that can change between runs of the same program but |
382 | within a single run of a program. JSON::XS can optionally sort the hash |
698 | stays generally the same within a single run of a program. JSON::XS can |
383 | keys (determined by the I<canonical> flag), so the same datastructure |
699 | optionally sort the hash keys (determined by the I<canonical> flag), so |
384 | will serialise to the same JSON text (given same settings and version of |
700 | the same datastructure will serialise to the same JSON text (given same |
385 | JSON::XS), but this incurs a runtime overhead. |
701 | settings and version of JSON::XS), but this incurs a runtime overhead |
|
|
702 | and is only rarely useful, e.g. when you want to compare some JSON text |
|
|
703 | against another for equality. |
386 | |
704 | |
387 | =item array references |
705 | =item array references |
388 | |
706 | |
389 | Perl array references become JSON arrays. |
707 | Perl array references become JSON arrays. |
|
|
708 | |
|
|
709 | =item other references |
|
|
710 | |
|
|
711 | Other unblessed references are generally not allowed and will cause an |
|
|
712 | exception to be thrown, except for references to the integers C<0> and |
|
|
713 | C<1>, which get turned into C<false> and C<true> atoms in JSON. You can |
|
|
714 | also use C<JSON::XS::false> and C<JSON::XS::true> to improve readability. |
|
|
715 | |
|
|
716 | to_json [\0,JSON::XS::true] # yields [false,true] |
|
|
717 | |
|
|
718 | =item JSON::XS::true, JSON::XS::false |
|
|
719 | |
|
|
720 | These special values become JSON true and JSON false values, |
|
|
721 | respectively. You can also use C<\1> and C<\0> directly if you want. |
390 | |
722 | |
391 | =item blessed objects |
723 | =item blessed objects |
392 | |
724 | |
393 | Blessed objects are not allowed. JSON::XS currently tries to encode their |
725 | Blessed objects are not allowed. JSON::XS currently tries to encode their |
394 | underlying representation (hash- or arrayref), but this behaviour might |
726 | underlying representation (hash- or arrayref), but this behaviour might |
… | |
… | |
411 | to_json [$value] # yields ["5"] |
743 | to_json [$value] # yields ["5"] |
412 | |
744 | |
413 | # undef becomes null |
745 | # undef becomes null |
414 | to_json [undef] # yields [null] |
746 | to_json [undef] # yields [null] |
415 | |
747 | |
416 | You can force the type to be a string by stringifying it: |
748 | You can force the type to be a JSON string by stringifying it: |
417 | |
749 | |
418 | my $x = 3.1; # some variable containing a number |
750 | my $x = 3.1; # some variable containing a number |
419 | "$x"; # stringified |
751 | "$x"; # stringified |
420 | $x .= ""; # another, more awkward way to stringify |
752 | $x .= ""; # another, more awkward way to stringify |
421 | print $x; # perl does it for you, too, quite often |
753 | print $x; # perl does it for you, too, quite often |
422 | |
754 | |
423 | You can force the type to be a number by numifying it: |
755 | You can force the type to be a JSON number by numifying it: |
424 | |
756 | |
425 | my $x = "3"; # some variable containing a string |
757 | my $x = "3"; # some variable containing a string |
426 | $x += 0; # numify it, ensuring it will be dumped as a number |
758 | $x += 0; # numify it, ensuring it will be dumped as a number |
427 | $x *= 1; # same thing, the choise is yours. |
759 | $x *= 1; # same thing, the choice is yours. |
428 | |
760 | |
429 | You can not currently output JSON booleans or force the type in other, |
761 | You can not currently force the type in other, less obscure, ways. Tell me |
430 | less obscure, ways. Tell me if you need this capability. |
762 | if you need this capability. |
431 | |
|
|
432 | =item circular data structures |
|
|
433 | |
|
|
434 | Those will be encoded until memory or stackspace runs out. |
|
|
435 | |
763 | |
436 | =back |
764 | =back |
|
|
765 | |
437 | |
766 | |
438 | =head1 COMPARISON |
767 | =head1 COMPARISON |
439 | |
768 | |
440 | As already mentioned, this module was created because none of the existing |
769 | As already mentioned, this module was created because none of the existing |
441 | JSON modules could be made to work correctly. First I will describe the |
770 | JSON modules could be made to work correctly. First I will describe the |
… | |
… | |
447 | |
776 | |
448 | =item JSON 1.07 |
777 | =item JSON 1.07 |
449 | |
778 | |
450 | Slow (but very portable, as it is written in pure Perl). |
779 | Slow (but very portable, as it is written in pure Perl). |
451 | |
780 | |
452 | Undocumented/buggy Unicode handling (how JSON handles unicode values is |
781 | Undocumented/buggy Unicode handling (how JSON handles Unicode values is |
453 | undocumented. One can get far by feeding it unicode strings and doing |
782 | undocumented. One can get far by feeding it Unicode strings and doing |
454 | en-/decoding oneself, but unicode escapes are not working properly). |
783 | en-/decoding oneself, but Unicode escapes are not working properly). |
455 | |
784 | |
456 | No roundtripping (strings get clobbered if they look like numbers, e.g. |
785 | No roundtripping (strings get clobbered if they look like numbers, e.g. |
457 | the string C<2.0> will encode to C<2.0> instead of C<"2.0">, and that will |
786 | the string C<2.0> will encode to C<2.0> instead of C<"2.0">, and that will |
458 | decode into the number 2. |
787 | decode into the number 2. |
459 | |
788 | |
… | |
… | |
481 | Very inflexible (no human-readable format supported, format pretty much |
810 | Very inflexible (no human-readable format supported, format pretty much |
482 | undocumented. I need at least a format for easy reading by humans and a |
811 | undocumented. I need at least a format for easy reading by humans and a |
483 | single-line compact format for use in a protocol, and preferably a way to |
812 | single-line compact format for use in a protocol, and preferably a way to |
484 | generate ASCII-only JSON texts). |
813 | generate ASCII-only JSON texts). |
485 | |
814 | |
486 | Completely broken (and confusingly documented) Unicode handling (unicode |
815 | Completely broken (and confusingly documented) Unicode handling (Unicode |
487 | escapes are not working properly, you need to set ImplicitUnicode to |
816 | escapes are not working properly, you need to set ImplicitUnicode to |
488 | I<different> values on en- and decoding to get symmetric behaviour). |
817 | I<different> values on en- and decoding to get symmetric behaviour). |
489 | |
818 | |
490 | No roundtripping (simple cases work, but this depends on wether the scalar |
819 | No roundtripping (simple cases work, but this depends on whether the scalar |
491 | value was used in a numeric context or not). |
820 | value was used in a numeric context or not). |
492 | |
821 | |
493 | Dumping hashes may skip hash values depending on iterator state. |
822 | Dumping hashes may skip hash values depending on iterator state. |
494 | |
823 | |
495 | Unmaintained (maintainer unresponsive for many months, bugs are not |
824 | Unmaintained (maintainer unresponsive for many months, bugs are not |
496 | getting fixed). |
825 | getting fixed). |
497 | |
826 | |
498 | Does not check input for validity (i.e. will accept non-JSON input and |
827 | Does not check input for validity (i.e. will accept non-JSON input and |
499 | return "something" instead of raising an exception. This is a security |
828 | return "something" instead of raising an exception. This is a security |
500 | issue: imagine two banks transfering money between each other using |
829 | issue: imagine two banks transferring money between each other using |
501 | JSON. One bank might parse a given non-JSON request and deduct money, |
830 | JSON. One bank might parse a given non-JSON request and deduct money, |
502 | while the other might reject the transaction with a syntax error. While a |
831 | while the other might reject the transaction with a syntax error. While a |
503 | good protocol will at least recover, that is extra unnecessary work and |
832 | good protocol will at least recover, that is extra unnecessary work and |
504 | the transaction will still not succeed). |
833 | the transaction will still not succeed). |
505 | |
834 | |
506 | =item JSON::DWIW 0.04 |
835 | =item JSON::DWIW 0.04 |
507 | |
836 | |
508 | Very fast. Very natural. Very nice. |
837 | Very fast. Very natural. Very nice. |
509 | |
838 | |
510 | Undocumented unicode handling (but the best of the pack. Unicode escapes |
839 | Undocumented Unicode handling (but the best of the pack. Unicode escapes |
511 | still don't get parsed properly). |
840 | still don't get parsed properly). |
512 | |
841 | |
513 | Very inflexible. |
842 | Very inflexible. |
514 | |
843 | |
515 | No roundtripping. |
844 | No roundtripping. |
… | |
… | |
518 | result in nothing being output) |
847 | result in nothing being output) |
519 | |
848 | |
520 | Does not check input for validity. |
849 | Does not check input for validity. |
521 | |
850 | |
522 | =back |
851 | =back |
|
|
852 | |
|
|
853 | |
|
|
854 | =head2 JSON and YAML |
|
|
855 | |
|
|
856 | You often hear that JSON is a subset (or a close subset) of YAML. This is, |
|
|
857 | however, a mass hysteria and very far from the truth. In general, there is |
|
|
858 | no way to configure JSON::XS to output a data structure as valid YAML. |
|
|
859 | |
|
|
860 | If you really must use JSON::XS to generate YAML, you should use this |
|
|
861 | algorithm (subject to change in future versions): |
|
|
862 | |
|
|
863 | my $to_yaml = JSON::XS->new->utf8->space_after (1); |
|
|
864 | my $yaml = $to_yaml->encode ($ref) . "\n"; |
|
|
865 | |
|
|
866 | This will usually generate JSON texts that also parse as valid |
|
|
867 | YAML. Please note that YAML has hardcoded limits on (simple) object key |
|
|
868 | lengths that JSON doesn't have, so you should make sure that your hash |
|
|
869 | keys are noticeably shorter than the 1024 characters YAML allows. |
|
|
870 | |
|
|
871 | There might be other incompatibilities that I am not aware of. In general |
|
|
872 | you should not try to generate YAML with a JSON generator or vice versa, |
|
|
873 | or try to parse JSON with a YAML parser or vice versa: chances are high |
|
|
874 | that you will run into severe interoperability problems. |
|
|
875 | |
523 | |
876 | |
524 | =head2 SPEED |
877 | =head2 SPEED |
525 | |
878 | |
526 | It seems that JSON::XS is surprisingly fast, as shown in the following |
879 | It seems that JSON::XS is surprisingly fast, as shown in the following |
527 | tables. They have been generated with the help of the C<eg/bench> program |
880 | tables. They have been generated with the help of the C<eg/bench> program |
528 | in the JSON::XS distribution, to make it easy to compare on your own |
881 | in the JSON::XS distribution, to make it easy to compare on your own |
529 | system. |
882 | system. |
530 | |
883 | |
531 | First comes a comparison between various modules using a very short JSON |
884 | First comes a comparison between various modules using a very short |
532 | string (83 bytes), showing the number of encodes/decodes per second |
885 | single-line JSON string: |
|
|
886 | |
|
|
887 | {"method": "handleMessage", "params": ["user1", "we were just talking"], \ |
|
|
888 | "id": null, "array":[1,11,234,-5,1e5,1e7, true, false]} |
|
|
889 | |
|
|
890 | It shows the number of encodes/decodes per second (JSON::XS uses |
533 | (JSON::XS is the functional interface, while JSON::XS/2 is the OO |
891 | the functional interface, while JSON::XS/2 uses the OO interface |
534 | interface with pretty-printing and hashkey sorting enabled). Higher is |
892 | with pretty-printing and hashkey sorting enabled, JSON::XS/3 enables |
535 | better: |
893 | shrink). Higher is better: |
536 | |
894 | |
|
|
895 | Storable | 15779.925 | 14169.946 | |
|
|
896 | -----------+------------+------------+ |
537 | module | encode | decode | |
897 | module | encode | decode | |
538 | -----------|------------|------------| |
898 | -----------|------------|------------| |
539 | JSON | 14006 | 6820 | |
899 | JSON | 4990.842 | 4088.813 | |
540 | JSON::DWIW | 200937 | 120386 | |
900 | JSON::DWIW | 51653.990 | 71575.154 | |
541 | JSON::PC | 85065 | 129366 | |
901 | JSON::PC | 65948.176 | 74631.744 | |
542 | JSON::Syck | 59898 | 44232 | |
902 | JSON::PP | 8931.652 | 3817.168 | |
543 | JSON::XS | 1171478 | 342435 | |
903 | JSON::Syck | 24877.248 | 27776.848 | |
544 | JSON::XS/2 | 730760 | 328714 | |
904 | JSON::XS | 388361.481 | 227951.304 | |
|
|
905 | JSON::XS/2 | 227951.304 | 218453.333 | |
|
|
906 | JSON::XS/3 | 338250.323 | 218453.333 | |
|
|
907 | Storable | 16500.016 | 135300.129 | |
545 | -----------+------------+------------+ |
908 | -----------+------------+------------+ |
546 | |
909 | |
547 | That is, JSON::XS is 6 times faster than than JSON::DWIW and about 80 |
910 | That is, JSON::XS is about five times faster than JSON::DWIW on encoding, |
|
|
911 | about three times faster on decoding, and over forty times faster |
548 | times faster than JSON, even with pretty-printing and key sorting. |
912 | than JSON, even with pretty-printing and key sorting. It also compares |
|
|
913 | favourably to Storable for small amounts of data. |
549 | |
914 | |
550 | Using a longer test string (roughly 18KB, generated from Yahoo! Locals |
915 | Using a longer test string (roughly 18KB, generated from Yahoo! Locals |
551 | search API (http://nanoref.com/yahooapis/mgPdGg): |
916 | search API (http://nanoref.com/yahooapis/mgPdGg): |
552 | |
917 | |
553 | module | encode | decode | |
918 | module | encode | decode | |
554 | -----------|------------|------------| |
919 | -----------|------------|------------| |
555 | JSON | 673 | 38 | |
920 | JSON | 55.260 | 34.971 | |
556 | JSON::DWIW | 5271 | 770 | |
921 | JSON::DWIW | 825.228 | 1082.513 | |
557 | JSON::PC | 9901 | 2491 | |
922 | JSON::PC | 3571.444 | 2394.829 | |
558 | JSON::Syck | 2360 | 786 | |
923 | JSON::PP | 210.987 | 32.574 | |
559 | JSON::XS | 37398 | 3202 | |
924 | JSON::Syck | 552.551 | 787.544 | |
560 | JSON::XS/2 | 13765 | 3153 | |
925 | JSON::XS | 5780.463 | 4854.519 | |
|
|
926 | JSON::XS/2 | 3869.998 | 4798.975 | |
|
|
927 | JSON::XS/3 | 5862.880 | 4798.975 | |
|
|
928 | Storable | 4445.002 | 5235.027 | |
561 | -----------+------------+------------+ |
929 | -----------+------------+------------+ |
562 | |
930 | |
563 | Again, JSON::XS leads by far in the encoding case, while still beating |
931 | Again, JSON::XS leads by far (except for Storable which non-surprisingly |
564 | every other module in the decoding case. |
932 | decodes faster). |
565 | |
933 | |
566 | On large strings containing lots of unicode characters, some modules |
934 | On large strings containing lots of high Unicode characters, some modules |
567 | (such as JSON::PC) decode faster than JSON::XS, but the result will be |
935 | (such as JSON::PC) seem to decode faster than JSON::XS, but the result |
568 | broken due to missing unicode handling. Others refuse to decode or encode |
936 | will be broken due to missing (or wrong) Unicode handling. Others refuse |
569 | properly, so it was impossible to prepare a fair comparison table for that |
937 | to decode or encode properly, so it was impossible to prepare a fair |
570 | case. |
938 | comparison table for that case. |
571 | |
939 | |
572 | =head1 RESOURCE LIMITS |
|
|
573 | |
940 | |
574 | JSON::XS does not impose any limits on the size of JSON texts or Perl |
941 | =head1 SECURITY CONSIDERATIONS |
575 | values they represent - if your machine can handle it, JSON::XS will |
942 | |
576 | encode or decode it. Future versions might optionally impose structure |
943 | When you are using JSON in a protocol, talking to untrusted potentially |
577 | depth and memory use resource limits. |
944 | hostile creatures requires relatively few measures. |
|
|
945 | |
|
|
946 | First of all, your JSON decoder should be secure, that is, should not have |
|
|
947 | any buffer overflows. Obviously, this module should ensure that and I am |
|
|
948 | trying hard on making that true, but you never know. |
|
|
949 | |
|
|
950 | Second, you need to avoid resource-starving attacks. That means you should |
|
|
951 | limit the size of JSON texts you accept, or make sure then when your |
|
|
952 | resources run out, that's just fine (e.g. by using a separate process that |
|
|
953 | can crash safely). The size of a JSON text in octets or characters is |
|
|
954 | usually a good indication of the size of the resources required to decode |
|
|
955 | it into a Perl structure. While JSON::XS can check the size of the JSON |
|
|
956 | text, it might be too late when you already have it in memory, so you |
|
|
957 | might want to check the size before you accept the string. |
|
|
958 | |
|
|
959 | Third, JSON::XS recurses using the C stack when decoding objects and |
|
|
960 | arrays. The C stack is a limited resource: for instance, on my amd64 |
|
|
961 | machine with 8MB of stack size I can decode around 180k nested arrays but |
|
|
962 | only 14k nested JSON objects (due to perl itself recursing deeply on croak |
|
|
963 | to free the temporary). If that is exceeded, the program crashes. to be |
|
|
964 | conservative, the default nesting limit is set to 512. If your process |
|
|
965 | has a smaller stack, you should adjust this setting accordingly with the |
|
|
966 | C<max_depth> method. |
|
|
967 | |
|
|
968 | And last but least, something else could bomb you that I forgot to think |
|
|
969 | of. In that case, you get to keep the pieces. I am always open for hints, |
|
|
970 | though... |
|
|
971 | |
|
|
972 | If you are using JSON::XS to return packets to consumption |
|
|
973 | by JavaScript scripts in a browser you should have a look at |
|
|
974 | L<http://jpsykes.com/47/practical-csrf-and-json-security> to see whether |
|
|
975 | you are vulnerable to some common attack vectors (which really are browser |
|
|
976 | design bugs, but it is still you who will have to deal with it, as major |
|
|
977 | browser developers care only for features, not about doing security |
|
|
978 | right). |
|
|
979 | |
|
|
980 | |
|
|
981 | =head1 THREADS |
|
|
982 | |
|
|
983 | This module is I<not> guaranteed to be thread safe and there are no |
|
|
984 | plans to change this until Perl gets thread support (as opposed to the |
|
|
985 | horribly slow so-called "threads" which are simply slow and bloated |
|
|
986 | process simulations - use fork, its I<much> faster, cheaper, better). |
|
|
987 | |
|
|
988 | (It might actually work, but you have been warned). |
|
|
989 | |
578 | |
990 | |
579 | =head1 BUGS |
991 | =head1 BUGS |
580 | |
992 | |
581 | While the goal of this module is to be correct, that unfortunately does |
993 | While the goal of this module is to be correct, that unfortunately does |
582 | not mean its bug-free, only that I think its design is bug-free. It is |
994 | not mean its bug-free, only that I think its design is bug-free. It is |
583 | still very young and not well-tested. If you keep reporting bugs they will |
995 | still relatively early in its development. If you keep reporting bugs they |
584 | be fixed swiftly, though. |
996 | will be fixed swiftly, though. |
|
|
997 | |
|
|
998 | Please refrain from using rt.cpan.org or any other bug reporting |
|
|
999 | service. I put the contact address into my modules for a reason. |
585 | |
1000 | |
586 | =cut |
1001 | =cut |
|
|
1002 | |
|
|
1003 | our $true = do { bless \(my $dummy = 1), "JSON::XS::Boolean" }; |
|
|
1004 | our $false = do { bless \(my $dummy = 0), "JSON::XS::Boolean" }; |
|
|
1005 | |
|
|
1006 | sub true() { $true } |
|
|
1007 | sub false() { $false } |
|
|
1008 | |
|
|
1009 | sub is_bool($) { |
|
|
1010 | UNIVERSAL::isa $_[0], "JSON::XS::Boolean" |
|
|
1011 | # or UNIVERSAL::isa $_[0], "JSON::Literal" |
|
|
1012 | } |
|
|
1013 | |
|
|
1014 | XSLoader::load "JSON::XS", $VERSION; |
|
|
1015 | |
|
|
1016 | package JSON::XS::Boolean; |
|
|
1017 | |
|
|
1018 | use overload |
|
|
1019 | "0+" => sub { ${$_[0]} }, |
|
|
1020 | "++" => sub { $_[0] = ${$_[0]} + 1 }, |
|
|
1021 | "--" => sub { $_[0] = ${$_[0]} - 1 }, |
|
|
1022 | fallback => 1; |
587 | |
1023 | |
588 | 1; |
1024 | 1; |
589 | |
1025 | |
590 | =head1 AUTHOR |
1026 | =head1 AUTHOR |
591 | |
1027 | |