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

Comparing JSON-XS/README (file contents):
Revision 1.5 by root, Sat Mar 24 01:15:22 2007 UTC vs.
Revision 1.26 by root, Tue Jun 3 06:43:45 2008 UTC

1NAME 1NAME
2 JSON::XS - JSON serialising/deserialising, done correctly and fast 2 JSON::XS - JSON serialising/deserialising, done correctly and fast
3 3
4 JSON::XS - 正しくて高速な JSON シリアライザ/デシリアライザ
5 (http://fleur.hio.jp/perldoc/mix/lib/JSON/XS.html)
6
4SYNOPSIS 7SYNOPSIS
5 use JSON::XS; 8 use JSON::XS;
6 9
7 # exported functions, croak on error 10 # exported functions, they croak on error
11 # and expect/generate UTF-8
8 12
9 $utf8_encoded_json_text = to_json $perl_hash_or_arrayref; 13 $utf8_encoded_json_text = encode_json $perl_hash_or_arrayref;
10 $perl_hash_or_arrayref = from_json $utf8_encoded_json_text; 14 $perl_hash_or_arrayref = decode_json $utf8_encoded_json_text;
11 15
12 # oo-interface 16 # OO-interface
13 17
14 $coder = JSON::XS->new->ascii->pretty->allow_nonref; 18 $coder = JSON::XS->new->ascii->pretty->allow_nonref;
15 $pretty_printed_unencoded = $coder->encode ($perl_scalar); 19 $pretty_printed_unencoded = $coder->encode ($perl_scalar);
16 $perl_scalar = $coder->decode ($unicode_json_text); 20 $perl_scalar = $coder->decode ($unicode_json_text);
21
22 # Note that JSON version 2.0 and above will automatically use JSON::XS
23 # if available, at virtually no speed overhead either, so you should
24 # be able to just:
25
26 use JSON;
27
28 # and do the same things, except that you have a pure-perl fallback now.
17 29
18DESCRIPTION 30DESCRIPTION
19 This module converts Perl data structures to JSON and vice versa. Its 31 This module converts Perl data structures to JSON and vice versa. Its
20 primary goal is to be *correct* and its secondary goal is to be *fast*. 32 primary goal is to be *correct* and its secondary goal is to be *fast*.
21 To reach the latter goal it was written in C. 33 To reach the latter goal it was written in C.
34
35 Beginning with version 2.0 of the JSON module, when both JSON and
36 JSON::XS are installed, then JSON will fall back on JSON::XS (this can
37 be overridden) with no overhead due to emulation (by inheriting
38 constructor and methods). If JSON::XS is not available, it will fall
39 back to the compatible JSON::PP module as backend, so using JSON instead
40 of JSON::XS gives you a portable JSON API that can be fast when you need
41 and doesn't require a C compiler when that is a problem.
22 42
23 As this is the n-th-something JSON module on CPAN, what was the reason 43 As this is the n-th-something JSON module on CPAN, what was the reason
24 to write yet another JSON module? While it seems there are many JSON 44 to write yet another JSON module? While it seems there are many JSON
25 modules, none of them correctly handle all corner cases, and in most 45 modules, none of them correctly handle all corner cases, and in most
26 cases their maintainers are unresponsive, gone missing, or not listening 46 cases their maintainers are unresponsive, gone missing, or not listening
… …
30 50
31 See MAPPING, below, on how JSON::XS maps perl values to JSON values and 51 See MAPPING, below, on how JSON::XS maps perl values to JSON values and
32 vice versa. 52 vice versa.
33 53
34 FEATURES 54 FEATURES
35 * correct handling of unicode issues 55 * correct Unicode handling
56
36 This module knows how to handle Unicode, and even documents how and 57 This module knows how to handle Unicode, documents how and when it
37 when it does so. 58 does so, and even documents what "correct" means.
38 59
39 * round-trip integrity 60 * round-trip integrity
61
40 When you serialise a perl data structure using only datatypes 62 When you serialise a perl data structure using only data types
41 supported by JSON, the deserialised data structure is identical on 63 supported by JSON, the deserialised data structure is identical on
42 the Perl level. (e.g. the string "2.0" doesn't suddenly become "2"). 64 the Perl level. (e.g. the string "2.0" doesn't suddenly become "2"
65 just because it looks like a number). There minor *are* exceptions
66 to this, read the MAPPING section below to learn about those.
43 67
44 * strict checking of JSON correctness 68 * strict checking of JSON correctness
69
45 There is no guessing, no generating of illegal JSON strings by 70 There is no guessing, no generating of illegal JSON texts by
46 default, and only JSON is accepted as input by default (the latter 71 default, and only JSON is accepted as input by default (the latter
47 is a security feature). 72 is a security feature).
48 73
49 * fast 74 * fast
50 Compared to other JSON modules, this module compares favourably in
51 terms of speed, too.
52 75
76 Compared to other JSON modules and other serialisers such as
77 Storable, this module usually compares favourably in terms of speed,
78 too.
79
53 * simple to use 80 * simple to use
81
54 This module has both a simple functional interface as well as an OO 82 This module has both a simple functional interface as well as an
55 interface. 83 object oriented interface interface.
56 84
57 * reasonably versatile output formats 85 * reasonably versatile output formats
86
58 You can choose between the most compact guarenteed single-line 87 You can choose between the most compact guaranteed-single-line
59 format possible (nice for simple line-based protocols), a pure-ascii 88 format possible (nice for simple line-based protocols), a pure-ASCII
60 format (for when your transport is not 8-bit clean), or a 89 format (for when your transport is not 8-bit clean, still supports
61 pretty-printed format (for when you want to read that stuff). Or you 90 the whole Unicode range), or a pretty-printed format (for when you
62 can combine those features in whatever way you like. 91 want to read that stuff). Or you can combine those features in
92 whatever way you like.
63 93
64FUNCTIONAL INTERFACE 94FUNCTIONAL INTERFACE
65 The following convinience methods are provided by this module. They are 95 The following convenience methods are provided by this module. They are
66 exported by default: 96 exported by default:
67 97
68 $json_string = to_json $perl_scalar 98 $json_text = encode_json $perl_scalar
69 Converts the given Perl data structure (a simple scalar or a 99 Converts the given Perl data structure to a UTF-8 encoded, binary
70 reference to a hash or array) to a UTF-8 encoded, binary string
71 (that is, the string contains octets only). Croaks on error. 100 string (that is, the string contains octets only). Croaks on error.
72 101
73 This function call is functionally identical to 102 This function call is functionally identical to:
103
74 "JSON::XS->new->utf8->encode ($perl_scalar)". 104 $json_text = JSON::XS->new->utf8->encode ($perl_scalar)
75 105
106 Except being faster.
107
76 $perl_scalar = from_json $json_string 108 $perl_scalar = decode_json $json_text
77 The opposite of "to_json": expects an UTF-8 (binary) string and 109 The opposite of "encode_json": expects an UTF-8 (binary) string and
78 tries to parse that as an UTF-8 encoded JSON string, returning the 110 tries to parse that as an UTF-8 encoded JSON text, returning the
79 resulting simple scalar or reference. Croaks on error. 111 resulting reference. Croaks on error.
80 112
81 This function call is functionally identical to 113 This function call is functionally identical to:
114
82 "JSON::XS->new->utf8->decode ($json_string)". 115 $perl_scalar = JSON::XS->new->utf8->decode ($json_text)
116
117 Except being faster.
118
119 $is_boolean = JSON::XS::is_bool $scalar
120 Returns true if the passed scalar represents either JSON::XS::true
121 or JSON::XS::false, two constants that act like 1 and 0,
122 respectively and are used to represent JSON "true" and "false"
123 values in Perl.
124
125 See MAPPING, below, for more information on how JSON values are
126 mapped to Perl.
127
128A FEW NOTES ON UNICODE AND PERL
129 Since this often leads to confusion, here are a few very clear words on
130 how Unicode works in Perl, modulo bugs.
131
132 1. Perl strings can store characters with ordinal values > 255.
133 This enables you to store Unicode characters as single characters in
134 a Perl string - very natural.
135
136 2. Perl does *not* associate an encoding with your strings.
137 ... until you force it to, e.g. when matching it against a regex, or
138 printing the scalar to a file, in which case Perl either interprets
139 your string as locale-encoded text, octets/binary, or as Unicode,
140 depending on various settings. In no case is an encoding stored
141 together with your data, it is *use* that decides encoding, not any
142 magical meta data.
143
144 3. The internal utf-8 flag has no meaning with regards to the encoding
145 of your string.
146 Just ignore that flag unless you debug a Perl bug, a module written
147 in XS or want to dive into the internals of perl. Otherwise it will
148 only confuse you, as, despite the name, it says nothing about how
149 your string is encoded. You can have Unicode strings with that flag
150 set, with that flag clear, and you can have binary data with that
151 flag set and that flag clear. Other possibilities exist, too.
152
153 If you didn't know about that flag, just the better, pretend it
154 doesn't exist.
155
156 4. A "Unicode String" is simply a string where each character can be
157 validly interpreted as a Unicode code point.
158 If you have UTF-8 encoded data, it is no longer a Unicode string,
159 but a Unicode string encoded in UTF-8, giving you a binary string.
160
161 5. A string containing "high" (> 255) character values is *not* a UTF-8
162 string.
163 It's a fact. Learn to live with it.
164
165 I hope this helps :)
83 166
84OBJECT-ORIENTED INTERFACE 167OBJECT-ORIENTED INTERFACE
85 The object oriented interface lets you configure your own encoding or 168 The object oriented interface lets you configure your own encoding or
86 decoding style, within the limits of supported formats. 169 decoding style, within the limits of supported formats.
87 170
… …
91 *disabled*. 174 *disabled*.
92 175
93 The mutators for flags all return the JSON object again and thus 176 The mutators for flags all return the JSON object again and thus
94 calls can be chained: 177 calls can be chained:
95 178
96 my $json = JSON::XS->new->utf8(1)->space_after(1)->encode ({a => [1,2]}) 179 my $json = JSON::XS->new->utf8->space_after->encode ({a => [1,2]})
97 => {"a": [1, 2]} 180 => {"a": [1, 2]}
98 181
99 $json = $json->ascii ([$enable]) 182 $json = $json->ascii ([$enable])
183 $enabled = $json->get_ascii
100 If $enable is true (or missing), then the "encode" method will not 184 If $enable is true (or missing), then the "encode" method will not
101 generate characters outside the code range 0..127. Any unicode 185 generate characters outside the code range 0..127 (which is ASCII).
102 characters outside that range will be escaped using either a single 186 Any Unicode characters outside that range will be escaped using
103 \uXXXX (BMP characters) or a double \uHHHH\uLLLLL escape sequence, 187 either a single \uXXXX (BMP characters) or a double \uHHHH\uLLLLL
104 as per RFC4627. 188 escape sequence, as per RFC4627. The resulting encoded JSON text can
189 be treated as a native Unicode string, an ascii-encoded,
190 latin1-encoded or UTF-8 encoded string, or any other superset of
191 ASCII.
105 192
106 If $enable is false, then the "encode" method will not escape 193 If $enable is false, then the "encode" method will not escape
107 Unicode characters unless necessary. 194 Unicode characters unless required by the JSON syntax or other
195 flags. This results in a faster and more compact format.
108 196
197 See also the section *ENCODING/CODESET FLAG NOTES* later in this
198 document.
199
200 The main use for this flag is to produce JSON texts that can be
201 transmitted over a 7-bit channel, as the encoded JSON texts will not
202 contain any 8 bit characters.
203
109 JSON::XS->new->ascii (1)->encode (chr 0x10401) 204 JSON::XS->new->ascii (1)->encode ([chr 0x10401])
110 => \ud801\udc01 205 => ["\ud801\udc01"]
206
207 $json = $json->latin1 ([$enable])
208 $enabled = $json->get_latin1
209 If $enable is true (or missing), then the "encode" method will
210 encode the resulting JSON text as latin1 (or iso-8859-1), escaping
211 any characters outside the code range 0..255. The resulting string
212 can be treated as a latin1-encoded JSON text or a native Unicode
213 string. The "decode" method will not be affected in any way by this
214 flag, as "decode" by default expects Unicode, which is a strict
215 superset of latin1.
216
217 If $enable is false, then the "encode" method will not escape
218 Unicode characters unless required by the JSON syntax or other
219 flags.
220
221 See also the section *ENCODING/CODESET FLAG NOTES* later in this
222 document.
223
224 The main use for this flag is efficiently encoding binary data as
225 JSON text, as most octets will not be escaped, resulting in a
226 smaller encoded size. The disadvantage is that the resulting JSON
227 text is encoded in latin1 (and must correctly be treated as such
228 when storing and transferring), a rare encoding for JSON. It is
229 therefore most useful when you want to store data structures known
230 to contain binary data efficiently in files or databases, not when
231 talking to other JSON encoders/decoders.
232
233 JSON::XS->new->latin1->encode (["\x{89}\x{abc}"]
234 => ["\x{89}\\u0abc"] # (perl syntax, U+abc escaped, U+89 not)
111 235
112 $json = $json->utf8 ([$enable]) 236 $json = $json->utf8 ([$enable])
237 $enabled = $json->get_utf8
113 If $enable is true (or missing), then the "encode" method will 238 If $enable is true (or missing), then the "encode" method will
114 encode the JSON string into UTF-8, as required by many protocols, 239 encode the JSON result into UTF-8, as required by many protocols,
115 while the "decode" method expects to be handled an UTF-8-encoded 240 while the "decode" method expects to be handled an UTF-8-encoded
116 string. Please note that UTF-8-encoded strings do not contain any 241 string. Please note that UTF-8-encoded strings do not contain any
117 characters outside the range 0..255, they are thus useful for 242 characters outside the range 0..255, they are thus useful for
118 bytewise/binary I/O. 243 bytewise/binary I/O. In future versions, enabling this option might
244 enable autodetection of the UTF-16 and UTF-32 encoding families, as
245 described in RFC4627.
119 246
120 If $enable is false, then the "encode" method will return the JSON 247 If $enable is false, then the "encode" method will return the JSON
121 string as a (non-encoded) unicode string, while "decode" expects 248 string as a (non-encoded) Unicode string, while "decode" expects
122 thus a unicode string. Any decoding or encoding (e.g. to UTF-8 or 249 thus a Unicode string. Any decoding or encoding (e.g. to UTF-8 or
123 UTF-16) needs to be done yourself, e.g. using the Encode module. 250 UTF-16) needs to be done yourself, e.g. using the Encode module.
124 251
252 See also the section *ENCODING/CODESET FLAG NOTES* later in this
253 document.
254
125 Example, output UTF-16-encoded JSON: 255 Example, output UTF-16BE-encoded JSON:
256
257 use Encode;
258 $jsontext = encode "UTF-16BE", JSON::XS->new->encode ($object);
259
260 Example, decode UTF-32LE-encoded JSON:
261
262 use Encode;
263 $object = JSON::XS->new->decode (decode "UTF-32LE", $jsontext);
126 264
127 $json = $json->pretty ([$enable]) 265 $json = $json->pretty ([$enable])
128 This enables (or disables) all of the "indent", "space_before" and 266 This enables (or disables) all of the "indent", "space_before" and
129 "space_after" (and in the future possibly more) flags in one call to 267 "space_after" (and in the future possibly more) flags in one call to
130 generate the most readable (or most compact) form possible. 268 generate the most readable (or most compact) form possible.
… …
139 2 277 2
140 ] 278 ]
141 } 279 }
142 280
143 $json = $json->indent ([$enable]) 281 $json = $json->indent ([$enable])
282 $enabled = $json->get_indent
144 If $enable is true (or missing), then the "encode" method will use a 283 If $enable is true (or missing), then the "encode" method will use a
145 multiline format as output, putting every array member or 284 multiline format as output, putting every array member or
146 object/hash key-value pair into its own line, identing them 285 object/hash key-value pair into its own line, indenting them
147 properly. 286 properly.
148 287
149 If $enable is false, no newlines or indenting will be produced, and 288 If $enable is false, no newlines or indenting will be produced, and
150 the resulting JSON strings is guarenteed not to contain any 289 the resulting JSON text is guaranteed not to contain any "newlines".
151 "newlines".
152 290
153 This setting has no effect when decoding JSON strings. 291 This setting has no effect when decoding JSON texts.
154 292
155 $json = $json->space_before ([$enable]) 293 $json = $json->space_before ([$enable])
294 $enabled = $json->get_space_before
156 If $enable is true (or missing), then the "encode" method will add 295 If $enable is true (or missing), then the "encode" method will add
157 an extra optional space before the ":" separating keys from values 296 an extra optional space before the ":" separating keys from values
158 in JSON objects. 297 in JSON objects.
159 298
160 If $enable is false, then the "encode" method will not add any extra 299 If $enable is false, then the "encode" method will not add any extra
161 space at those places. 300 space at those places.
162 301
163 This setting has no effect when decoding JSON strings. You will also 302 This setting has no effect when decoding JSON texts. You will also
164 most likely combine this setting with "space_after". 303 most likely combine this setting with "space_after".
165 304
166 Example, space_before enabled, space_after and indent disabled: 305 Example, space_before enabled, space_after and indent disabled:
167 306
168 {"key" :"value"} 307 {"key" :"value"}
169 308
170 $json = $json->space_after ([$enable]) 309 $json = $json->space_after ([$enable])
310 $enabled = $json->get_space_after
171 If $enable is true (or missing), then the "encode" method will add 311 If $enable is true (or missing), then the "encode" method will add
172 an extra optional space after the ":" separating keys from values in 312 an extra optional space after the ":" separating keys from values in
173 JSON objects and extra whitespace after the "," separating key-value 313 JSON objects and extra whitespace after the "," separating key-value
174 pairs and array members. 314 pairs and array members.
175 315
176 If $enable is false, then the "encode" method will not add any extra 316 If $enable is false, then the "encode" method will not add any extra
177 space at those places. 317 space at those places.
178 318
179 This setting has no effect when decoding JSON strings. 319 This setting has no effect when decoding JSON texts.
180 320
181 Example, space_before and indent disabled, space_after enabled: 321 Example, space_before and indent disabled, space_after enabled:
182 322
183 {"key": "value"} 323 {"key": "value"}
184 324
325 $json = $json->relaxed ([$enable])
326 $enabled = $json->get_relaxed
327 If $enable is true (or missing), then "decode" will accept some
328 extensions to normal JSON syntax (see below). "encode" will not be
329 affected in anyway. *Be aware that this option makes you accept
330 invalid JSON texts as if they were valid!*. I suggest only to use
331 this option to parse application-specific files written by humans
332 (configuration files, resource files etc.)
333
334 If $enable is false (the default), then "decode" will only accept
335 valid JSON texts.
336
337 Currently accepted extensions are:
338
339 * list items can have an end-comma
340
341 JSON *separates* array elements and key-value pairs with commas.
342 This can be annoying if you write JSON texts manually and want
343 to be able to quickly append elements, so this extension accepts
344 comma at the end of such items not just between them:
345
346 [
347 1,
348 2, <- this comma not normally allowed
349 ]
350 {
351 "k1": "v1",
352 "k2": "v2", <- this comma not normally allowed
353 }
354
355 * shell-style '#'-comments
356
357 Whenever JSON allows whitespace, shell-style comments are
358 additionally allowed. They are terminated by the first
359 carriage-return or line-feed character, after which more
360 white-space and comments are allowed.
361
362 [
363 1, # this comment not allowed in JSON
364 # neither this one...
365 ]
366
185 $json = $json->canonical ([$enable]) 367 $json = $json->canonical ([$enable])
368 $enabled = $json->get_canonical
186 If $enable is true (or missing), then the "encode" method will 369 If $enable is true (or missing), then the "encode" method will
187 output JSON objects by sorting their keys. This is adding a 370 output JSON objects by sorting their keys. This is adding a
188 comparatively high overhead. 371 comparatively high overhead.
189 372
190 If $enable is false, then the "encode" method will output key-value 373 If $enable is false, then the "encode" method will output key-value
191 pairs in the order Perl stores them (which will likely change 374 pairs in the order Perl stores them (which will likely change
192 between runs of the same script). 375 between runs of the same script).
193 376
194 This option is useful if you want the same data structure to be 377 This option is useful if you want the same data structure to be
195 encoded as the same JSON string (given the same overall settings). 378 encoded as the same JSON text (given the same overall settings). If
196 If it is disabled, the same hash migh be encoded differently even if 379 it is disabled, the same hash might be encoded differently even if
197 contains the same data, as key-value pairs have no inherent ordering 380 contains the same data, as key-value pairs have no inherent ordering
198 in Perl. 381 in Perl.
199 382
200 This setting has no effect when decoding JSON strings. 383 This setting has no effect when decoding JSON texts.
201 384
202 $json = $json->allow_nonref ([$enable]) 385 $json = $json->allow_nonref ([$enable])
386 $enabled = $json->get_allow_nonref
203 If $enable is true (or missing), then the "encode" method can 387 If $enable is true (or missing), then the "encode" method can
204 convert a non-reference into its corresponding string, number or 388 convert a non-reference into its corresponding string, number or
205 null JSON value, which is an extension to RFC4627. Likewise, 389 null JSON value, which is an extension to RFC4627. Likewise,
206 "decode" will accept those JSON values instead of croaking. 390 "decode" will accept those JSON values instead of croaking.
207 391
208 If $enable is false, then the "encode" method will croak if it isn't 392 If $enable is false, then the "encode" method will croak if it isn't
209 passed an arrayref or hashref, as JSON strings must either be an 393 passed an arrayref or hashref, as JSON texts must either be an
210 object or array. Likewise, "decode" will croak if given something 394 object or array. Likewise, "decode" will croak if given something
211 that is not a JSON object or array. 395 that is not a JSON object or array.
212 396
213 Example, encode a Perl scalar as JSON value with enabled 397 Example, encode a Perl scalar as JSON value with enabled
214 "allow_nonref", resulting in an invalid JSON text: 398 "allow_nonref", resulting in an invalid JSON text:
215 399
216 JSON::XS->new->allow_nonref->encode ("Hello, World!") 400 JSON::XS->new->allow_nonref->encode ("Hello, World!")
217 => "Hello, World!" 401 => "Hello, World!"
218 402
403 $json = $json->allow_unknown ([$enable])
404 $enabled = $json->get_allow_unknown
405 If $enable is true (or missing), then "encode" will *not* throw an
406 exception when it encounters values it cannot represent in JSON (for
407 example, filehandles) but instead will encode a JSON "null" value.
408 Note that blessed objects are not included here and are handled
409 separately by c<allow_nonref>.
410
411 If $enable is false (the default), then "encode" will throw an
412 exception when it encounters anything it cannot encode as JSON.
413
414 This option does not affect "decode" in any way, and it is
415 recommended to leave it off unless you know your communications
416 partner.
417
418 $json = $json->allow_blessed ([$enable])
419 $enabled = $json->get_allow_blessed
420 If $enable is true (or missing), then the "encode" method will not
421 barf when it encounters a blessed reference. Instead, the value of
422 the convert_blessed option will decide whether "null"
423 ("convert_blessed" disabled or no "TO_JSON" method found) or a
424 representation of the object ("convert_blessed" enabled and
425 "TO_JSON" method found) is being encoded. Has no effect on "decode".
426
427 If $enable is false (the default), then "encode" will throw an
428 exception when it encounters a blessed object.
429
430 $json = $json->convert_blessed ([$enable])
431 $enabled = $json->get_convert_blessed
432 If $enable is true (or missing), then "encode", upon encountering a
433 blessed object, will check for the availability of the "TO_JSON"
434 method on the object's class. If found, it will be called in scalar
435 context and the resulting scalar will be encoded instead of the
436 object. If no "TO_JSON" method is found, the value of
437 "allow_blessed" will decide what to do.
438
439 The "TO_JSON" method may safely call die if it wants. If "TO_JSON"
440 returns other blessed objects, those will be handled in the same
441 way. "TO_JSON" must take care of not causing an endless recursion
442 cycle (== crash) in this case. The name of "TO_JSON" was chosen
443 because other methods called by the Perl core (== not by the user of
444 the object) are usually in upper case letters and to avoid
445 collisions with any "to_json" function or method.
446
447 This setting does not yet influence "decode" in any way, but in the
448 future, global hooks might get installed that influence "decode" and
449 are enabled by this setting.
450
451 If $enable is false, then the "allow_blessed" setting will decide
452 what to do when a blessed object is found.
453
454 $json = $json->filter_json_object ([$coderef->($hashref)])
455 When $coderef is specified, it will be called from "decode" each
456 time it decodes a JSON object. The only argument is a reference to
457 the newly-created hash. If the code references returns a single
458 scalar (which need not be a reference), this value (i.e. a copy of
459 that scalar to avoid aliasing) is inserted into the deserialised
460 data structure. If it returns an empty list (NOTE: *not* "undef",
461 which is a valid scalar), the original deserialised hash will be
462 inserted. This setting can slow down decoding considerably.
463
464 When $coderef is omitted or undefined, any existing callback will be
465 removed and "decode" will not change the deserialised hash in any
466 way.
467
468 Example, convert all JSON objects into the integer 5:
469
470 my $js = JSON::XS->new->filter_json_object (sub { 5 });
471 # returns [5]
472 $js->decode ('[{}]')
473 # throw an exception because allow_nonref is not enabled
474 # so a lone 5 is not allowed.
475 $js->decode ('{"a":1, "b":2}');
476
477 $json = $json->filter_json_single_key_object ($key [=>
478 $coderef->($value)])
479 Works remotely similar to "filter_json_object", but is only called
480 for JSON objects having a single key named $key.
481
482 This $coderef is called before the one specified via
483 "filter_json_object", if any. It gets passed the single value in the
484 JSON object. If it returns a single value, it will be inserted into
485 the data structure. If it returns nothing (not even "undef" but the
486 empty list), the callback from "filter_json_object" will be called
487 next, as if no single-key callback were specified.
488
489 If $coderef is omitted or undefined, the corresponding callback will
490 be disabled. There can only ever be one callback for a given key.
491
492 As this callback gets called less often then the
493 "filter_json_object" one, decoding speed will not usually suffer as
494 much. Therefore, single-key objects make excellent targets to
495 serialise Perl objects into, especially as single-key JSON objects
496 are as close to the type-tagged value concept as JSON gets (it's
497 basically an ID/VALUE tuple). Of course, JSON does not support this
498 in any way, so you need to make sure your data never looks like a
499 serialised Perl hash.
500
501 Typical names for the single object key are "__class_whatever__", or
502 "$__dollars_are_rarely_used__$" or "}ugly_brace_placement", or even
503 things like "__class_md5sum(classname)__", to reduce the risk of
504 clashing with real hashes.
505
506 Example, decode JSON objects of the form "{ "__widget__" => <id> }"
507 into the corresponding $WIDGET{<id>} object:
508
509 # return whatever is in $WIDGET{5}:
510 JSON::XS
511 ->new
512 ->filter_json_single_key_object (__widget__ => sub {
513 $WIDGET{ $_[0] }
514 })
515 ->decode ('{"__widget__": 5')
516
517 # this can be used with a TO_JSON method in some "widget" class
518 # for serialisation to json:
519 sub WidgetBase::TO_JSON {
520 my ($self) = @_;
521
522 unless ($self->{id}) {
523 $self->{id} = ..get..some..id..;
524 $WIDGET{$self->{id}} = $self;
525 }
526
527 { __widget__ => $self->{id} }
528 }
529
219 $json = $json->shrink ([$enable]) 530 $json = $json->shrink ([$enable])
531 $enabled = $json->get_shrink
220 Perl usually over-allocates memory a bit when allocating space for 532 Perl usually over-allocates memory a bit when allocating space for
221 strings. This flag optionally resizes strings generated by either 533 strings. This flag optionally resizes strings generated by either
222 "encode" or "decode" to their minimum size possible. This can save 534 "encode" or "decode" to their minimum size possible. This can save
223 memory when your JSON strings are either very very long or you have 535 memory when your JSON texts are either very very long or you have
224 many short strings. It will also try to downgrade any strings to 536 many short strings. It will also try to downgrade any strings to
225 octet-form if possible: perl stores strings internally either in an 537 octet-form if possible: perl stores strings internally either in an
226 encoding called UTF-X or in octet-form. The latter cannot store 538 encoding called UTF-X or in octet-form. The latter cannot store
227 everything but uses less space in general. 539 everything but uses less space in general (and some buggy Perl or C
540 code might even rely on that internal representation being used).
541
542 The actual definition of what shrink does might change in future
543 versions, but it will always try to save space at the expense of
544 time.
228 545
229 If $enable is true (or missing), the string returned by "encode" 546 If $enable is true (or missing), the string returned by "encode"
230 will be shrunk-to-fit, while all strings generated by "decode" will 547 will be shrunk-to-fit, while all strings generated by "decode" will
231 also be shrunk-to-fit. 548 also be shrunk-to-fit.
232 549
… …
236 In the future, this setting might control other things, such as 553 In the future, this setting might control other things, such as
237 converting strings that look like integers or floats into integers 554 converting strings that look like integers or floats into integers
238 or floats internally (there is no difference on the Perl level), 555 or floats internally (there is no difference on the Perl level),
239 saving space. 556 saving space.
240 557
558 $json = $json->max_depth ([$maximum_nesting_depth])
559 $max_depth = $json->get_max_depth
560 Sets the maximum nesting level (default 512) accepted while encoding
561 or decoding. If a higher nesting level is detected in JSON text or a
562 Perl data structure, then the encoder and decoder will stop and
563 croak at that point.
564
565 Nesting level is defined by number of hash- or arrayrefs that the
566 encoder needs to traverse to reach a given point or the number of
567 "{" or "[" characters without their matching closing parenthesis
568 crossed to reach a given character in a string.
569
570 Setting the maximum depth to one disallows any nesting, so that
571 ensures that the object is only a single hash/object or array.
572
573 If no argument is given, the highest possible setting will be used,
574 which is rarely useful.
575
576 Note that nesting is implemented by recursion in C. The default
577 value has been chosen to be as large as typical operating systems
578 allow without crashing.
579
580 See SECURITY CONSIDERATIONS, below, for more info on why this is
581 useful.
582
583 $json = $json->max_size ([$maximum_string_size])
584 $max_size = $json->get_max_size
585 Set the maximum length a JSON text may have (in bytes) where
586 decoding is being attempted. The default is 0, meaning no limit.
587 When "decode" is called on a string that is longer then this many
588 bytes, it will not attempt to decode the string but throw an
589 exception. This setting has no effect on "encode" (yet).
590
591 If no argument is given, the limit check will be deactivated (same
592 as when 0 is specified).
593
594 See SECURITY CONSIDERATIONS, below, for more info on why this is
595 useful.
596
241 $json_string = $json->encode ($perl_scalar) 597 $json_text = $json->encode ($perl_scalar)
242 Converts the given Perl data structure (a simple scalar or a 598 Converts the given Perl data structure (a simple scalar or a
243 reference to a hash or array) to its JSON representation. Simple 599 reference to a hash or array) to its JSON representation. Simple
244 scalars will be converted into JSON string or number sequences, 600 scalars will be converted into JSON string or number sequences,
245 while references to arrays become JSON arrays and references to 601 while references to arrays become JSON arrays and references to
246 hashes become JSON objects. Undefined Perl values (e.g. "undef") 602 hashes become JSON objects. Undefined Perl values (e.g. "undef")
247 become JSON "null" values. Neither "true" nor "false" values will be 603 become JSON "null" values. Neither "true" nor "false" values will be
248 generated. 604 generated.
249 605
250 $perl_scalar = $json->decode ($json_string) 606 $perl_scalar = $json->decode ($json_text)
251 The opposite of "encode": expects a JSON string and tries to parse 607 The opposite of "encode": expects a JSON text and tries to parse it,
252 it, returning the resulting simple scalar or reference. Croaks on 608 returning the resulting simple scalar or reference. Croaks on error.
253 error.
254 609
255 JSON numbers and strings become simple Perl scalars. JSON arrays 610 JSON numbers and strings become simple Perl scalars. JSON arrays
256 become Perl arrayrefs and JSON objects become Perl hashrefs. "true" 611 become Perl arrayrefs and JSON objects become Perl hashrefs. "true"
257 becomes 1, "false" becomes 0 and "null" becomes "undef". 612 becomes 1, "false" becomes 0 and "null" becomes "undef".
613
614 ($perl_scalar, $characters) = $json->decode_prefix ($json_text)
615 This works like the "decode" method, but instead of raising an
616 exception when there is trailing garbage after the first JSON
617 object, it will silently stop parsing there and return the number of
618 characters consumed so far.
619
620 This is useful if your JSON texts are not delimited by an outer
621 protocol (which is not the brightest thing to do in the first place)
622 and you need to know where the JSON text ends.
623
624 JSON::XS->new->decode_prefix ("[1] the tail")
625 => ([], 3)
626
627INCREMENTAL PARSING
628 In some cases, there is the need for incremental parsing of JSON texts.
629 While this module always has to keep both JSON text and resulting Perl
630 data structure in memory at one time, it does allow you to parse a JSON
631 stream incrementally. It does so by accumulating text until it has a
632 full JSON object, which it then can decode. This process is similar to
633 using "decode_prefix" to see if a full JSON object is available, but is
634 much more efficient (JSON::XS will only attempt to parse the JSON text
635 once it is sure it has enough text to get a decisive result, using a
636 very simple but truly incremental parser).
637
638 The following two methods deal with this.
639
640 [void, scalar or list context] = $json->incr_parse ([$string])
641 This is the central parsing function. It can both append new text
642 and extract objects from the stream accumulated so far (both of
643 these functions are optional).
644
645 If $string is given, then this string is appended to the already
646 existing JSON fragment stored in the $json object.
647
648 After that, if the function is called in void context, it will
649 simply return without doing anything further. This can be used to
650 add more text in as many chunks as you want.
651
652 If the method is called in scalar context, then it will try to
653 extract exactly *one* JSON object. If that is successful, it will
654 return this object, otherwise it will return "undef". If there is a
655 parse error, this method will croak just as "decode" would do (one
656 can then use "incr_skip" to skip the errornous part). This is the
657 most common way of using the method.
658
659 And finally, in list context, it will try to extract as many objects
660 from the stream as it can find and return them, or the empty list
661 otherwise. For this to work, there must be no separators between the
662 JSON objects or arrays, instead they must be concatenated
663 back-to-back. If an error occurs, an exception will be raised as in
664 the scalar context case. Note that in this case, any
665 previously-parsed JSON texts will be lost.
666
667 $lvalue_string = $json->incr_text
668 This method returns the currently stored JSON fragment as an lvalue,
669 that is, you can manipulate it. This *only* works when a preceding
670 call to "incr_parse" in *scalar context* successfully returned an
671 object. Under all other circumstances you must not call this
672 function (I mean it. although in simple tests it might actually
673 work, it *will* fail under real world conditions). As a special
674 exception, you can also call this method before having parsed
675 anything.
676
677 This function is useful in two cases: a) finding the trailing text
678 after a JSON object or b) parsing multiple JSON objects separated by
679 non-JSON text (such as commas).
680
681 $json->incr_skip
682 This will reset the state of the incremental parser and will remove
683 the parsed text from the input buffer. This is useful after
684 "incr_parse" died, in which case the input buffer and incremental
685 parser state is left unchanged, to skip the text parsed so far and
686 to reset the parse state.
687
688 $json->incr_reset
689 This completely resets the incremental parser, that is, after this
690 call, it will be as if the parser had never parsed anything.
691
692 This is useful if you want ot repeatedly parse JSON objects and want
693 to ignore any trailing data, which means you have to reset the
694 parser after each successful decode.
695
696 LIMITATIONS
697 All options that affect decoding are supported, except "allow_nonref".
698 The reason for this is that it cannot be made to work sensibly: JSON
699 objects and arrays are self-delimited, i.e. you can concatenate them
700 back to back and still decode them perfectly. This does not hold true
701 for JSON numbers, however.
702
703 For example, is the string 1 a single JSON number, or is it simply the
704 start of 12? Or is 12 a single JSON number, or the concatenation of 1
705 and 2? In neither case you can tell, and this is why JSON::XS takes the
706 conservative route and disallows this case.
707
708 EXAMPLES
709 Some examples will make all this clearer. First, a simple example that
710 works similarly to "decode_prefix": We want to decode the JSON object at
711 the start of a string and identify the portion after the JSON object:
712
713 my $text = "[1,2,3] hello";
714
715 my $json = new JSON::XS;
716
717 my $obj = $json->incr_parse ($text)
718 or die "expected JSON object or array at beginning of string";
719
720 my $tail = $json->incr_text;
721 # $tail now contains " hello"
722
723 Easy, isn't it?
724
725 Now for a more complicated example: Imagine a hypothetical protocol
726 where you read some requests from a TCP stream, and each request is a
727 JSON array, without any separation between them (in fact, it is often
728 useful to use newlines as "separators", as these get interpreted as
729 whitespace at the start of the JSON text, which makes it possible to
730 test said protocol with "telnet"...).
731
732 Here is how you'd do it (it is trivial to write this in an event-based
733 manner):
734
735 my $json = new JSON::XS;
736
737 # read some data from the socket
738 while (sysread $socket, my $buf, 4096) {
739
740 # split and decode as many requests as possible
741 for my $request ($json->incr_parse ($buf)) {
742 # act on the $request
743 }
744 }
745
746 Another complicated example: Assume you have a string with JSON objects
747 or arrays, all separated by (optional) comma characters (e.g. "[1],[2],
748 [3]"). To parse them, we have to skip the commas between the JSON texts,
749 and here is where the lvalue-ness of "incr_text" comes in useful:
750
751 my $text = "[1],[2], [3]";
752 my $json = new JSON::XS;
753
754 # void context, so no parsing done
755 $json->incr_parse ($text);
756
757 # now extract as many objects as possible. note the
758 # use of scalar context so incr_text can be called.
759 while (my $obj = $json->incr_parse) {
760 # do something with $obj
761
762 # now skip the optional comma
763 $json->incr_text =~ s/^ \s* , //x;
764 }
765
766 Now lets go for a very complex example: Assume that you have a gigantic
767 JSON array-of-objects, many gigabytes in size, and you want to parse it,
768 but you cannot load it into memory fully (this has actually happened in
769 the real world :).
770
771 Well, you lost, you have to implement your own JSON parser. But JSON::XS
772 can still help you: You implement a (very simple) array parser and let
773 JSON decode the array elements, which are all full JSON objects on their
774 own (this wouldn't work if the array elements could be JSON numbers, for
775 example):
776
777 my $json = new JSON::XS;
778
779 # open the monster
780 open my $fh, "<bigfile.json"
781 or die "bigfile: $!";
782
783 # first parse the initial "["
784 for (;;) {
785 sysread $fh, my $buf, 65536
786 or die "read error: $!";
787 $json->incr_parse ($buf); # void context, so no parsing
788
789 # Exit the loop once we found and removed(!) the initial "[".
790 # In essence, we are (ab-)using the $json object as a simple scalar
791 # we append data to.
792 last if $json->incr_text =~ s/^ \s* \[ //x;
793 }
794
795 # now we have the skipped the initial "[", so continue
796 # parsing all the elements.
797 for (;;) {
798 # in this loop we read data until we got a single JSON object
799 for (;;) {
800 if (my $obj = $json->incr_parse) {
801 # do something with $obj
802 last;
803 }
804
805 # add more data
806 sysread $fh, my $buf, 65536
807 or die "read error: $!";
808 $json->incr_parse ($buf); # void context, so no parsing
809 }
810
811 # in this loop we read data until we either found and parsed the
812 # separating "," between elements, or the final "]"
813 for (;;) {
814 # first skip whitespace
815 $json->incr_text =~ s/^\s*//;
816
817 # if we find "]", we are done
818 if ($json->incr_text =~ s/^\]//) {
819 print "finished.\n";
820 exit;
821 }
822
823 # if we find ",", we can continue with the next element
824 if ($json->incr_text =~ s/^,//) {
825 last;
826 }
827
828 # if we find anything else, we have a parse error!
829 if (length $json->incr_text) {
830 die "parse error near ", $json->incr_text;
831 }
832
833 # else add more data
834 sysread $fh, my $buf, 65536
835 or die "read error: $!";
836 $json->incr_parse ($buf); # void context, so no parsing
837 }
838
839 This is a complex example, but most of the complexity comes from the
840 fact that we are trying to be correct (bear with me if I am wrong, I
841 never ran the above example :).
258 842
259MAPPING 843MAPPING
260 This section describes how JSON::XS maps Perl values to JSON values and 844 This section describes how JSON::XS maps Perl values to JSON values and
261 vice versa. These mappings are designed to "do the right thing" in most 845 vice versa. These mappings are designed to "do the right thing" in most
262 circumstances automatically, preserving round-tripping characteristics 846 circumstances automatically, preserving round-tripping characteristics
263 (what you put in comes out as something equivalent). 847 (what you put in comes out as something equivalent).
264 848
265 For the more enlightened: note that in the following descriptions, 849 For the more enlightened: note that in the following descriptions,
266 lowercase *perl* refers to the Perl interpreter, while uppcercase *Perl* 850 lowercase *perl* refers to the Perl interpreter, while uppercase *Perl*
267 refers to the abstract Perl language itself. 851 refers to the abstract Perl language itself.
268 852
269 JSON -> PERL 853 JSON -> PERL
270 object 854 object
271 A JSON object becomes a reference to a hash in Perl. No ordering of 855 A JSON object becomes a reference to a hash in Perl. No ordering of
272 object keys is preserved (JSON does not preserver object key 856 object keys is preserved (JSON does not preserve object key ordering
273 ordering itself). 857 itself).
274 858
275 array 859 array
276 A JSON array becomes a reference to an array in Perl. 860 A JSON array becomes a reference to an array in Perl.
277 861
278 string 862 string
279 A JSON string becomes a string scalar in Perl - Unicode codepoints 863 A JSON string becomes a string scalar in Perl - Unicode codepoints
280 in JSON are represented by the same codepoints in the Perl string, 864 in JSON are represented by the same codepoints in the Perl string,
281 so no manual decoding is necessary. 865 so no manual decoding is necessary.
282 866
283 number 867 number
284 A JSON number becomes either an integer or numeric (floating point) 868 A JSON number becomes either an integer, numeric (floating point) or
285 scalar in perl, depending on its range and any fractional parts. On 869 string scalar in perl, depending on its range and any fractional
286 the Perl level, there is no difference between those as Perl handles 870 parts. On the Perl level, there is no difference between those as
287 all the conversion details, but an integer may take slightly less 871 Perl handles all the conversion details, but an integer may take
288 memory and might represent more values exactly than (floating point) 872 slightly less memory and might represent more values exactly than
873 floating point numbers.
874
875 If the number consists of digits only, JSON::XS will try to
876 represent it as an integer value. If that fails, it will try to
877 represent it as a numeric (floating point) value if that is possible
878 without loss of precision. Otherwise it will preserve the number as
879 a string value (in which case you lose roundtripping ability, as the
880 JSON number will be re-encoded toa JSON string).
881
882 Numbers containing a fractional or exponential part will always be
883 represented as numeric (floating point) values, possibly at a loss
884 of precision (in which case you might lose perfect roundtripping
885 ability, but the JSON number will still be re-encoded as a JSON
289 numbers. 886 number).
290 887
291 true, false 888 true, false
292 These JSON atoms become 0, 1, respectively. Information is lost in 889 These JSON atoms become "JSON::XS::true" and "JSON::XS::false",
293 this process. Future versions might represent those values 890 respectively. They are overloaded to act almost exactly like the
294 differently, but they will be guarenteed to act like these integers 891 numbers 1 and 0. You can check whether a scalar is a JSON boolean by
295 would normally in Perl. 892 using the "JSON::XS::is_bool" function.
296 893
297 null 894 null
298 A JSON null atom becomes "undef" in Perl. 895 A JSON null atom becomes "undef" in Perl.
299 896
300 PERL -> JSON 897 PERL -> JSON
… …
302 truly typeless language, so we can only guess which JSON type is meant 899 truly typeless language, so we can only guess which JSON type is meant
303 by a Perl value. 900 by a Perl value.
304 901
305 hash references 902 hash references
306 Perl hash references become JSON objects. As there is no inherent 903 Perl hash references become JSON objects. As there is no inherent
307 ordering in hash keys, they will usually be encoded in a 904 ordering in hash keys (or JSON objects), they will usually be
308 pseudo-random order that can change between runs of the same program 905 encoded in a pseudo-random order that can change between runs of the
309 but stays generally the same within a single run of a program. 906 same program but stays generally the same within a single run of a
310 JSON::XS can optionally sort the hash keys (determined by the 907 program. JSON::XS can optionally sort the hash keys (determined by
311 *canonical* flag), so the same datastructure will serialise to the 908 the *canonical* flag), so the same datastructure will serialise to
312 same JSON text (given same settings and version of JSON::XS), but 909 the same JSON text (given same settings and version of JSON::XS),
313 this incurs a runtime overhead. 910 but this incurs a runtime overhead and is only rarely useful, e.g.
911 when you want to compare some JSON text against another for
912 equality.
314 913
315 array references 914 array references
316 Perl array references become JSON arrays. 915 Perl array references become JSON arrays.
317 916
917 other references
918 Other unblessed references are generally not allowed and will cause
919 an exception to be thrown, except for references to the integers 0
920 and 1, which get turned into "false" and "true" atoms in JSON. You
921 can also use "JSON::XS::false" and "JSON::XS::true" to improve
922 readability.
923
924 encode_json [\0, JSON::XS::true] # yields [false,true]
925
926 JSON::XS::true, JSON::XS::false
927 These special values become JSON true and JSON false values,
928 respectively. You can also use "\1" and "\0" directly if you want.
929
318 blessed objects 930 blessed objects
319 Blessed objects are not allowed. JSON::XS currently tries to encode 931 Blessed objects are not directly representable in JSON. See the
320 their underlying representation (hash- or arrayref), but this 932 "allow_blessed" and "convert_blessed" methods on various options on
321 behaviour might change in future versions. 933 how to deal with this: basically, you can choose between throwing an
934 exception, encoding the reference as if it weren't blessed, or
935 provide your own serialiser method.
322 936
323 simple scalars 937 simple scalars
324 Simple Perl scalars (any scalar that is not a reference) are the 938 Simple Perl scalars (any scalar that is not a reference) are the
325 most difficult objects to encode: JSON::XS will encode undefined 939 most difficult objects to encode: JSON::XS will encode undefined
326 scalars as JSON null value, scalars that have last been used in a 940 scalars as JSON "null" values, scalars that have last been used in a
327 string context before encoding as JSON strings and anything else as 941 string context before encoding as JSON strings, and anything else as
328 number value: 942 number value:
329 943
330 # dump as number 944 # dump as number
331 to_json [2] # yields [2] 945 encode_json [2] # yields [2]
332 to_json [-3.0e17] # yields [-3e+17] 946 encode_json [-3.0e17] # yields [-3e+17]
333 my $value = 5; to_json [$value] # yields [5] 947 my $value = 5; encode_json [$value] # yields [5]
334 948
335 # used as string, so dump as string 949 # used as string, so dump as string
336 print $value; 950 print $value;
337 to_json [$value] # yields ["5"] 951 encode_json [$value] # yields ["5"]
338 952
339 # undef becomes null 953 # undef becomes null
340 to_json [undef] # yields [null] 954 encode_json [undef] # yields [null]
341 955
342 You can force the type to be a string by stringifying it: 956 You can force the type to be a JSON string by stringifying it:
343 957
344 my $x = 3.1; # some variable containing a number 958 my $x = 3.1; # some variable containing a number
345 "$x"; # stringified 959 "$x"; # stringified
346 $x .= ""; # another, more awkward way to stringify 960 $x .= ""; # another, more awkward way to stringify
347 print $x; # perl does it for you, too, quite often 961 print $x; # perl does it for you, too, quite often
348 962
349 You can force the type to be a number by numifying it: 963 You can force the type to be a JSON number by numifying it:
350 964
351 my $x = "3"; # some variable containing a string 965 my $x = "3"; # some variable containing a string
352 $x += 0; # numify it, ensuring it will be dumped as a number 966 $x += 0; # numify it, ensuring it will be dumped as a number
353 $x *= 1; # same thing, the choise is yours. 967 $x *= 1; # same thing, the choice is yours.
354 968
355 You can not currently output JSON booleans or force the type in 969 You can not currently force the type in other, less obscure, ways.
356 other, less obscure, ways. Tell me if you need this capability. 970 Tell me if you need this capability (but don't forget to explain why
971 it's needed :).
357 972
358 circular data structures 973ENCODING/CODESET FLAG NOTES
359 Those will be encoded until memory or stackspace runs out. 974 The interested reader might have seen a number of flags that signify
975 encodings or codesets - "utf8", "latin1" and "ascii". There seems to be
976 some confusion on what these do, so here is a short comparison:
360 977
361COMPARISON 978 "utf8" controls whether the JSON text created by "encode" (and expected
362 As already mentioned, this module was created because none of the 979 by "decode") is UTF-8 encoded or not, while "latin1" and "ascii" only
363 existing JSON modules could be made to work correctly. First I will 980 control whether "encode" escapes character values outside their
364 describe the problems (or pleasures) I encountered with various existing 981 respective codeset range. Neither of these flags conflict with each
365 JSON modules, followed by some benchmark values. JSON::XS was designed 982 other, although some combinations make less sense than others.
366 not to suffer from any of these problems or limitations.
367 983
368 JSON 1.07 984 Care has been taken to make all flags symmetrical with respect to
369 Slow (but very portable, as it is written in pure Perl). 985 "encode" and "decode", that is, texts encoded with any combination of
986 these flag values will be correctly decoded when the same flags are used
987 - in general, if you use different flag settings while encoding vs. when
988 decoding you likely have a bug somewhere.
370 989
371 Undocumented/buggy Unicode handling (how JSON handles unicode values 990 Below comes a verbose discussion of these flags. Note that a "codeset"
372 is undocumented. One can get far by feeding it unicode strings and 991 is simply an abstract set of character-codepoint pairs, while an
373 doing en-/decoding oneself, but unicode escapes are not working 992 encoding takes those codepoint numbers and *encodes* them, in our case
993 into octets. Unicode is (among other things) a codeset, UTF-8 is an
994 encoding, and ISO-8859-1 (= latin 1) and ASCII are both codesets *and*
995 encodings at the same time, which can be confusing.
996
997 "utf8" flag disabled
998 When "utf8" is disabled (the default), then "encode"/"decode"
999 generate and expect Unicode strings, that is, characters with high
1000 ordinal Unicode values (> 255) will be encoded as such characters,
1001 and likewise such characters are decoded as-is, no canges to them
1002 will be done, except "(re-)interpreting" them as Unicode codepoints
1003 or Unicode characters, respectively (to Perl, these are the same
1004 thing in strings unless you do funny/weird/dumb stuff).
1005
1006 This is useful when you want to do the encoding yourself (e.g. when
1007 you want to have UTF-16 encoded JSON texts) or when some other layer
1008 does the encoding for you (for example, when printing to a terminal
1009 using a filehandle that transparently encodes to UTF-8 you certainly
1010 do NOT want to UTF-8 encode your data first and have Perl encode it
1011 another time).
1012
1013 "utf8" flag enabled
1014 If the "utf8"-flag is enabled, "encode"/"decode" will encode all
1015 characters using the corresponding UTF-8 multi-byte sequence, and
1016 will expect your input strings to be encoded as UTF-8, that is, no
1017 "character" of the input string must have any value > 255, as UTF-8
1018 does not allow that.
1019
1020 The "utf8" flag therefore switches between two modes: disabled means
1021 you will get a Unicode string in Perl, enabled means you get an
1022 UTF-8 encoded octet/binary string in Perl.
1023
1024 "latin1" or "ascii" flags enabled
1025 With "latin1" (or "ascii") enabled, "encode" will escape characters
1026 with ordinal values > 255 (> 127 with "ascii") and encode the
1027 remaining characters as specified by the "utf8" flag.
1028
1029 If "utf8" is disabled, then the result is also correctly encoded in
1030 those character sets (as both are proper subsets of Unicode, meaning
1031 that a Unicode string with all character values < 256 is the same
1032 thing as a ISO-8859-1 string, and a Unicode string with all
1033 character values < 128 is the same thing as an ASCII string in
374 properly). 1034 Perl).
375 1035
376 No roundtripping (strings get clobbered if they look like numbers, 1036 If "utf8" is enabled, you still get a correct UTF-8-encoded string,
377 e.g. the string 2.0 will encode to 2.0 instead of "2.0", and that 1037 regardless of these flags, just some more characters will be escaped
378 will decode into the number 2. 1038 using "\uXXXX" then before.
379 1039
380 JSON::PC 0.01 1040 Note that ISO-8859-1-*encoded* strings are not compatible with UTF-8
381 Very fast. 1041 encoding, while ASCII-encoded strings are. That is because the
1042 ISO-8859-1 encoding is NOT a subset of UTF-8 (despite the ISO-8859-1
1043 *codeset* being a subset of Unicode), while ASCII is.
382 1044
383 Undocumented/buggy Unicode handling. 1045 Surprisingly, "decode" will ignore these flags and so treat all
1046 input values as governed by the "utf8" flag. If it is disabled, this
1047 allows you to decode ISO-8859-1- and ASCII-encoded strings, as both
1048 strict subsets of Unicode. If it is enabled, you can correctly
1049 decode UTF-8 encoded strings.
384 1050
385 No roundtripping. 1051 So neither "latin1" nor "ascii" are incompatible with the "utf8"
1052 flag - they only govern when the JSON output engine escapes a
1053 character or not.
386 1054
387 Has problems handling many Perl values (e.g. regex results and other 1055 The main use for "latin1" is to relatively efficiently store binary
388 magic values will make it croak). 1056 data as JSON, at the expense of breaking compatibility with most
1057 JSON decoders.
389 1058
390 Does not even generate valid JSON ("{1,2}" gets converted to "{1:2}" 1059 The main use for "ascii" is to force the output to not contain
391 which is not a valid JSON string. 1060 characters with values > 127, which means you can interpret the
1061 resulting string as UTF-8, ISO-8859-1, ASCII, KOI8-R or most about
1062 any character set and 8-bit-encoding, and still get the same data
1063 structure back. This is useful when your channel for JSON transfer
1064 is not 8-bit clean or the encoding might be mangled in between (e.g.
1065 in mail), and works because ASCII is a proper subset of most 8-bit
1066 and multibyte encodings in use in the world.
392 1067
393 Unmaintained (maintainer unresponsive for many months, bugs are not 1068 JSON and YAML
394 getting fixed). 1069 You often hear that JSON is a subset of YAML. This is, however, a mass
1070 hysteria(*) and very far from the truth (as of the time of this
1071 writing), so let me state it clearly: *in general, there is no way to
1072 configure JSON::XS to output a data structure as valid YAML* that works
1073 in all cases.
395 1074
396 JSON::Syck 0.21 1075 If you really must use JSON::XS to generate YAML, you should use this
397 Very buggy (often crashes). 1076 algorithm (subject to change in future versions):
398 1077
399 Very inflexible (no human-readable format supported, format pretty 1078 my $to_yaml = JSON::XS->new->utf8->space_after (1);
400 much undocumented. I need at least a format for easy reading by 1079 my $yaml = $to_yaml->encode ($ref) . "\n";
401 humans and a single-line compact format for use in a protocol, and
402 preferably a way to generate ASCII-only JSON strings).
403 1080
404 Completely broken (and confusingly documented) Unicode handling 1081 This will *usually* generate JSON texts that also parse as valid YAML.
405 (unicode escapes are not working properly, you need to set 1082 Please note that YAML has hardcoded limits on (simple) object key
406 ImplicitUnicode to *different* values on en- and decoding to get 1083 lengths that JSON doesn't have and also has different and incompatible
407 symmetric behaviour). 1084 unicode handling, so you should make sure that your hash keys are
1085 noticeably shorter than the 1024 "stream characters" YAML allows and
1086 that you do not have characters with codepoint values outside the
1087 Unicode BMP (basic multilingual page). YAML also does not allow "\/"
1088 sequences in strings (which JSON::XS does not *currently* generate, but
1089 other JSON generators might).
408 1090
409 No roundtripping (simple cases work, but this depends on wether the 1091 There might be other incompatibilities that I am not aware of (or the
410 scalar value was used in a numeric context or not). 1092 YAML specification has been changed yet again - it does so quite often).
1093 In general you should not try to generate YAML with a JSON generator or
1094 vice versa, or try to parse JSON with a YAML parser or vice versa:
1095 chances are high that you will run into severe interoperability problems
1096 when you least expect it.
411 1097
412 Dumping hashes may skip hash values depending on iterator state. 1098 (*) I have been pressured multiple times by Brian Ingerson (one of the
1099 authors of the YAML specification) to remove this paragraph, despite
1100 him acknowledging that the actual incompatibilities exist. As I was
1101 personally bitten by this "JSON is YAML" lie, I refused and said I
1102 will continue to educate people about these issues, so others do not
1103 run into the same problem again and again. After this, Brian called
1104 me a (quote)*complete and worthless idiot*(unquote).
413 1105
414 Unmaintained (maintainer unresponsive for many months, bugs are not 1106 In my opinion, instead of pressuring and insulting people who
415 getting fixed). 1107 actually clarify issues with YAML and the wrong statements of some
416 1108 of its proponents, I would kindly suggest reading the JSON spec
417 Does not check input for validity (i.e. will accept non-JSON input 1109 (which is not that difficult or long) and finally make YAML
418 and return "something" instead of raising an exception. This is a 1110 compatible to it, and educating users about the changes, instead of
419 security issue: imagine two banks transfering money between each 1111 spreading lies about the real compatibility for many *years* and
420 other using JSON. One bank might parse a given non-JSON request and 1112 trying to silence people who point out that it isn't true.
421 deduct money, while the other might reject the transaction with a
422 syntax error. While a good protocol will at least recover, that is
423 extra unnecessary work and the transaction will still not succeed).
424
425 JSON::DWIW 0.04
426 Very fast. Very natural. Very nice.
427
428 Undocumented unicode handling (but the best of the pack. Unicode
429 escapes still don't get parsed properly).
430
431 Very inflexible.
432
433 No roundtripping.
434
435 Does not generate valid JSON (key strings are often unquoted, empty
436 keys result in nothing being output)
437
438 Does not check input for validity.
439 1113
440 SPEED 1114 SPEED
441 It seems that JSON::XS is surprisingly fast, as shown in the following 1115 It seems that JSON::XS is surprisingly fast, as shown in the following
442 tables. They have been generated with the help of the "eg/bench" program 1116 tables. They have been generated with the help of the "eg/bench" program
443 in the JSON::XS distribution, to make it easy to compare on your own 1117 in the JSON::XS distribution, to make it easy to compare on your own
444 system. 1118 system.
445 1119
446 First comes a comparison between various modules using a very short JSON 1120 First comes a comparison between various modules using a very short
447 string (83 bytes), showing the number of encodes/decodes per second 1121 single-line JSON string (also available at
448 (JSON::XS is the functional interface, while JSON::XS/2 is the OO 1122 <http://dist.schmorp.de/misc/json/short.json>).
449 interface with pretty-printing and hashkey sorting enabled). Higher is 1123
450 better: 1124 {"method": "handleMessage", "params": ["user1",
1125 "we were just talking"], "id": null, "array":[1,11,234,-5,1e5,1e7,
1126 true, false]}
1127
1128 It shows the number of encodes/decodes per second (JSON::XS uses the
1129 functional interface, while JSON::XS/2 uses the OO interface with
1130 pretty-printing and hashkey sorting enabled, JSON::XS/3 enables shrink).
1131 Higher is better:
451 1132
452 module | encode | decode | 1133 module | encode | decode |
453 -----------|------------|------------| 1134 -----------|------------|------------|
454 JSON | 14006 | 6820 | 1135 JSON 1.x | 4990.842 | 4088.813 |
455 JSON::DWIW | 200937 | 120386 | 1136 JSON::DWIW | 51653.990 | 71575.154 |
456 JSON::PC | 85065 | 129366 | 1137 JSON::PC | 65948.176 | 74631.744 |
457 JSON::Syck | 59898 | 44232 | 1138 JSON::PP | 8931.652 | 3817.168 |
458 JSON::XS | 1171478 | 342435 | 1139 JSON::Syck | 24877.248 | 27776.848 |
459 JSON::XS/2 | 730760 | 328714 | 1140 JSON::XS | 388361.481 | 227951.304 |
1141 JSON::XS/2 | 227951.304 | 218453.333 |
1142 JSON::XS/3 | 338250.323 | 218453.333 |
1143 Storable | 16500.016 | 135300.129 |
460 -----------+------------+------------+ 1144 -----------+------------+------------+
461 1145
462 That is, JSON::XS is 6 times faster than than JSON::DWIW and about 80 1146 That is, JSON::XS is about five times faster than JSON::DWIW on
1147 encoding, about three times faster on decoding, and over forty times
463 times faster than JSON, even with pretty-printing and key sorting. 1148 faster than JSON, even with pretty-printing and key sorting. It also
1149 compares favourably to Storable for small amounts of data.
464 1150
465 Using a longer test string (roughly 18KB, generated from Yahoo! Locals 1151 Using a longer test string (roughly 18KB, generated from Yahoo! Locals
466 search API (http://nanoref.com/yahooapis/mgPdGg): 1152 search API (<http://dist.schmorp.de/misc/json/long.json>).
467 1153
468 module | encode | decode | 1154 module | encode | decode |
469 -----------|------------|------------| 1155 -----------|------------|------------|
470 JSON | 673 | 38 | 1156 JSON 1.x | 55.260 | 34.971 |
471 JSON::DWIW | 5271 | 770 | 1157 JSON::DWIW | 825.228 | 1082.513 |
1158 JSON::PC | 3571.444 | 2394.829 |
472 JSON::PC | 9901 | 2491 | 1159 JSON::PP | 210.987 | 32.574 |
473 JSON::Syck | 2360 | 786 | 1160 JSON::Syck | 552.551 | 787.544 |
474 JSON::XS | 37398 | 3202 | 1161 JSON::XS | 5780.463 | 4854.519 |
475 JSON::XS/2 | 13765 | 3153 | 1162 JSON::XS/2 | 3869.998 | 4798.975 |
1163 JSON::XS/3 | 5862.880 | 4798.975 |
1164 Storable | 4445.002 | 5235.027 |
476 -----------+------------+------------+ 1165 -----------+------------+------------+
477 1166
478 Again, JSON::XS leads by far in the encoding case, while still beating 1167 Again, JSON::XS leads by far (except for Storable which non-surprisingly
479 every other module in the decoding case. 1168 decodes faster).
480 1169
481 On large strings containing lots of unicode characters, some modules 1170 On large strings containing lots of high Unicode characters, some
482 (such as JSON::PC) decode faster than JSON::XS, but the result will be 1171 modules (such as JSON::PC) seem to decode faster than JSON::XS, but the
483 broken due to missing unicode handling. Others refuse to decode or 1172 result will be broken due to missing (or wrong) Unicode handling. Others
484 encode properly, so it was impossible to prepare a fair comparison table 1173 refuse to decode or encode properly, so it was impossible to prepare a
485 for that case. 1174 fair comparison table for that case.
486 1175
487RESOURCE LIMITS 1176SECURITY CONSIDERATIONS
488 JSON::XS does not impose any limits on the size of JSON texts or Perl 1177 When you are using JSON in a protocol, talking to untrusted potentially
489 values they represent - if your machine can handle it, JSON::XS will 1178 hostile creatures requires relatively few measures.
490 encode or decode it. Future versions might optionally impose structure 1179
491 depth and memory use resource limits. 1180 First of all, your JSON decoder should be secure, that is, should not
1181 have any buffer overflows. Obviously, this module should ensure that and
1182 I am trying hard on making that true, but you never know.
1183
1184 Second, you need to avoid resource-starving attacks. That means you
1185 should limit the size of JSON texts you accept, or make sure then when
1186 your resources run out, that's just fine (e.g. by using a separate
1187 process that can crash safely). The size of a JSON text in octets or
1188 characters is usually a good indication of the size of the resources
1189 required to decode it into a Perl structure. While JSON::XS can check
1190 the size of the JSON text, it might be too late when you already have it
1191 in memory, so you might want to check the size before you accept the
1192 string.
1193
1194 Third, JSON::XS recurses using the C stack when decoding objects and
1195 arrays. The C stack is a limited resource: for instance, on my amd64
1196 machine with 8MB of stack size I can decode around 180k nested arrays
1197 but only 14k nested JSON objects (due to perl itself recursing deeply on
1198 croak to free the temporary). If that is exceeded, the program crashes.
1199 To be conservative, the default nesting limit is set to 512. If your
1200 process has a smaller stack, you should adjust this setting accordingly
1201 with the "max_depth" method.
1202
1203 Something else could bomb you, too, that I forgot to think of. In that
1204 case, you get to keep the pieces. I am always open for hints, though...
1205
1206 Also keep in mind that JSON::XS might leak contents of your Perl data
1207 structures in its error messages, so when you serialise sensitive
1208 information you might want to make sure that exceptions thrown by
1209 JSON::XS will not end up in front of untrusted eyes.
1210
1211 If you are using JSON::XS to return packets to consumption by JavaScript
1212 scripts in a browser you should have a look at
1213 <http://jpsykes.com/47/practical-csrf-and-json-security> to see whether
1214 you are vulnerable to some common attack vectors (which really are
1215 browser design bugs, but it is still you who will have to deal with it,
1216 as major browser developers care only for features, not about getting
1217 security right).
1218
1219THREADS
1220 This module is *not* guaranteed to be thread safe and there are no plans
1221 to change this until Perl gets thread support (as opposed to the
1222 horribly slow so-called "threads" which are simply slow and bloated
1223 process simulations - use fork, it's *much* faster, cheaper, better).
1224
1225 (It might actually work, but you have been warned).
492 1226
493BUGS 1227BUGS
494 While the goal of this module is to be correct, that unfortunately does 1228 While the goal of this module is to be correct, that unfortunately does
495 not mean its bug-free, only that I think its design is bug-free. It is 1229 not mean it's bug-free, only that I think its design is bug-free. If you
496 still very young and not well-tested. If you keep reporting bugs they 1230 keep reporting bugs they will be fixed swiftly, though.
497 will be fixed swiftly, though. 1231
1232 Please refrain from using rt.cpan.org or any other bug reporting
1233 service. I put the contact address into my modules for a reason.
1234
1235SEE ALSO
1236 The json_xs command line utility for quick experiments.
498 1237
499AUTHOR 1238AUTHOR
500 Marc Lehmann <schmorp@schmorp.de> 1239 Marc Lehmann <schmorp@schmorp.de>
501 http://home.schmorp.de/ 1240 http://home.schmorp.de/
502 1241

Diff Legend

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