ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/JSON-XS/XS.pm

Comparing JSON-XS/XS.pm (file contents):
Revision 1.23 by root, Sun Mar 25 21:19:13 2007 UTC vs.
Revision 1.104 by root, Thu May 8 15:33:06 2008 UTC

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
		5	=encoding utf-8
		6
		7	JSON::XS - 正しくて高速な JSON シリアライザ/デシリアライザ
		8	(http://fleur.hio.jp/perldoc/mix/lib/JSON/XS.html)
4		9
5	=head1 SYNOPSIS	10	=head1 SYNOPSIS
6		11
7	use JSON::XS;	12	use JSON::XS;
8		13
9	# exported functions, they croak on error	14	# exported functions, they croak on error
10	# and expect/generate UTF-8	15	# and expect/generate UTF-8
11		16
12	$utf8_encoded_json_text = to_json $perl_hash_or_arrayref;	17	$utf8_encoded_json_text = encode_json $perl_hash_or_arrayref;
13	$perl_hash_or_arrayref = from_json $utf8_encoded_json_text;	18	$perl_hash_or_arrayref = decode_json $utf8_encoded_json_text;
14
15	# objToJson and jsonToObj aliases to to_json and from_json
16	# are exported for compatibility to the JSON module,
17	# but should not be used in new code.
18		19
19	# OO-interface	20	# OO-interface
20		21
21	$coder = JSON::XS->new->ascii->pretty->allow_nonref;	22	$coder = JSON::XS->new->ascii->pretty->allow_nonref;
22	$pretty_printed_unencoded = $coder->encode ($perl_scalar);	23	$pretty_printed_unencoded = $coder->encode ($perl_scalar);
23	$perl_scalar = $coder->decode ($unicode_json_text);	24	$perl_scalar = $coder->decode ($unicode_json_text);
24		25
		26	# Note that JSON version 2.0 and above will automatically use JSON::XS
		27	# if available, at virtually no speed overhead either, so you should
		28	# be able to just:
		29
		30	use JSON;
		31
		32	# and do the same things, except that you have a pure-perl fallback now.
		33
25	=head1 DESCRIPTION	34	=head1 DESCRIPTION
26		35
27	This module converts Perl data structures to JSON and vice versa. Its	36	This module converts Perl data structures to JSON and vice versa. Its
28	primary goal is to be I<correct> and its secondary goal is to be	37	primary goal is to be I<correct> and its secondary goal is to be
29	I<fast>. To reach the latter goal it was written in C.	38	I<fast>. To reach the latter goal it was written in C.
		39
		40	Beginning with version 2.0 of the JSON module, when both JSON and
		41	JSON::XS are installed, then JSON will fall back on JSON::XS (this can be
		42	overriden) with no overhead due to emulation (by inheritign constructor
		43	and methods). If JSON::XS is not available, it will fall back to the
		44	compatible JSON::PP module as backend, so using JSON instead of JSON::XS
		45	gives you a portable JSON API that can be fast when you need and doesn't
		46	require a C compiler when that is a problem.
30		47
31	As this is the n-th-something JSON module on CPAN, what was the reason	48	As this is the n-th-something JSON module on CPAN, what was the reason
32	to write yet another JSON module? While it seems there are many JSON	49	to write yet another JSON module? While it seems there are many JSON
33	modules, none of them correctly handle all corner cases, and in most cases	50	modules, none of them correctly handle all corner cases, and in most cases
34	their maintainers are unresponsive, gone missing, or not listening to bug	51	their maintainers are unresponsive, gone missing, or not listening to bug
…		…
41		58
42	=head2 FEATURES	59	=head2 FEATURES
43		60
44	=over 4	61	=over 4
45		62
46	=item * correct unicode handling	63	=item * correct Unicode handling
47		64
48	This module knows how to handle Unicode, and even documents how and when	65	This module knows how to handle Unicode, documents how and when it does
49	it does so.	66	so, and even documents what "correct" means.
50		67
51	=item * round-trip integrity	68	=item * round-trip integrity
52		69
53	When you serialise a perl data structure using only datatypes supported	70	When you serialise a perl data structure using only datatypes supported
54	by JSON, the deserialised data structure is identical on the Perl level.	71	by JSON, the deserialised data structure is identical on the Perl level.
55	(e.g. the string "2.0" doesn't suddenly become "2" just because it looks	72	(e.g. the string "2.0" doesn't suddenly become "2" just because it looks
56	like a number).	73	like a number). There minor I<are> exceptions to this, read the MAPPING
		74	section below to learn about those.
57		75
58	=item * strict checking of JSON correctness	76	=item * strict checking of JSON correctness
59		77
60	There is no guessing, no generating of illegal JSON texts by default,	78	There is no guessing, no generating of illegal JSON texts by default,
61	and only JSON is accepted as input by default (the latter is a security	79	and only JSON is accepted as input by default (the latter is a security
62	feature).	80	feature).
63		81
64	=item * fast	82	=item * fast
65		83
66	Compared to other JSON modules, this module compares favourably in terms	84	Compared to other JSON modules and other serialisers such as Storable,
67	of speed, too.	85	this module usually compares favourably in terms of speed, too.
68		86
69	=item * simple to use	87	=item * simple to use
70		88
71	This module has both a simple functional interface as well as an OO	89	This module has both a simple functional interface as well as an objetc
72	interface.	90	oriented interface interface.
73		91
74	=item * reasonably versatile output formats	92	=item * reasonably versatile output formats
75		93
76	You can choose between the most compact guarenteed single-line format	94	You can choose between the most compact guaranteed-single-line format
77	possible (nice for simple line-based protocols), a pure-ascii format	95	possible (nice for simple line-based protocols), a pure-ascii format
78	(for when your transport is not 8-bit clean, still supports the whole	96	(for when your transport is not 8-bit clean, still supports the whole
79	unicode range), or a pretty-printed format (for when you want to read that	97	Unicode range), or a pretty-printed format (for when you want to read that
80	stuff). Or you can combine those features in whatever way you like.	98	stuff). Or you can combine those features in whatever way you like.
81		99
82	=back	100	=back
83		101
84	=cut	102	=cut
85		103
86	package JSON::XS;	104	package JSON::XS;
87		105
88	use strict;	106	use strict;
89		107
90	BEGIN {
91	our $VERSION = '0.8';	108	our $VERSION = '2.2';
92	our @ISA = qw(Exporter);	109	our @ISA = qw(Exporter);
93		110
94	our @EXPORT = qw(to_json from_json objToJson jsonToObj);	111	our @EXPORT = qw(encode_json decode_json to_json from_json);
95	require Exporter;
96		112
		113	sub to_json($) {
97	require XSLoader;	114	require Carp;
98	XSLoader::load JSON::XS::, $VERSION;	115	Carp::croak ("JSON::XS::to_json has been renamed to encode_json, either downgrade to pre-2.0 versions of JSON::XS or rename the call");
99	}	116	}
100		117
		118	sub from_json($) {
		119	require Carp;
		120	Carp::croak ("JSON::XS::from_json has been renamed to decode_json, either downgrade to pre-2.0 versions of JSON::XS or rename the call");
		121	}
		122
		123	use Exporter;
		124	use XSLoader;
		125
101	=head1 FUNCTIONAL INTERFACE	126	=head1 FUNCTIONAL INTERFACE
102		127
103	The following convinience methods are provided by this module. They are	128	The following convenience methods are provided by this module. They are
104	exported by default:	129	exported by default:
105		130
106	=over 4	131	=over 4
107		132
108	=item $json_text = to_json $perl_scalar	133	=item $json_text = encode_json $perl_scalar
109		134
110	Converts the given Perl data structure (a simple scalar or a reference to	135	Converts the given Perl data structure to a UTF-8 encoded, binary string
111	a hash or array) to a UTF-8 encoded, binary string (that is, the string contains	136	(that is, the string contains octets only). Croaks on error.
112	octets only). Croaks on error.
113		137
114	This function call is functionally identical to:	138	This function call is functionally identical to:
115		139
116	$json_text = JSON::XS->new->utf8->encode ($perl_scalar)	140	$json_text = JSON::XS->new->utf8->encode ($perl_scalar)
117		141
118	except being faster.	142	except being faster.
119		143
120	=item $perl_scalar = from_json $json_text	144	=item $perl_scalar = decode_json $json_text
121		145
122	The opposite of C<to_json>: expects an UTF-8 (binary) string and tries to	146	The opposite of C<encode_json>: expects an UTF-8 (binary) string and tries
123	parse that as an UTF-8 encoded JSON text, returning the resulting simple	147	to parse that as an UTF-8 encoded JSON text, returning the resulting
124	scalar or reference. Croaks on error.	148	reference. Croaks on error.
125		149
126	This function call is functionally identical to:	150	This function call is functionally identical to:
127		151
128	$perl_scalar = JSON::XS->new->utf8->decode ($json_text)	152	$perl_scalar = JSON::XS->new->utf8->decode ($json_text)
129		153
130	except being faster.	154	except being faster.
131		155
		156	=item $is_boolean = JSON::XS::is_bool $scalar
		157
		158	Returns true if the passed scalar represents either JSON::XS::true or
		159	JSON::XS::false, two constants that act like C<1> and C<0>, respectively
		160	and are used to represent JSON C<true> and C<false> values in Perl.
		161
		162	See MAPPING, below, for more information on how JSON values are mapped to
		163	Perl.
		164
132	=back	165	=back
		166
		167
		168	=head1 A FEW NOTES ON UNICODE AND PERL
		169
		170	Since this often leads to confusion, here are a few very clear words on
		171	how Unicode works in Perl, modulo bugs.
		172
		173	=over 4
		174
		175	=item 1. Perl strings can store characters with ordinal values > 255.
		176
		177	This enables you to store Unicode characters as single characters in a
		178	Perl string - very natural.
		179
		180	=item 2. Perl does I<not> associate an encoding with your strings.
		181
		182	... until you force it to, e.g. when matching it against a regex, or
		183	printing the scalar to a file, in which case Perl either interprets your
		184	string as locale-encoded text, octets/binary, or as Unicode, depending
		185	on various settings. In no case is an encoding stored together with your
		186	data, it is I<use> that decides encoding, not any magical meta data.
		187
		188	=item 3. The internal utf-8 flag has no meaning with regards to the
		189	encoding of your string.
		190
		191	Just ignore that flag unless you debug a Perl bug, a module written in
		192	XS or want to dive into the internals of perl. Otherwise it will only
		193	confuse you, as, despite the name, it says nothing about how your string
		194	is encoded. You can have Unicode strings with that flag set, with that
		195	flag clear, and you can have binary data with that flag set and that flag
		196	clear. Other possibilities exist, too.
		197
		198	If you didn't know about that flag, just the better, pretend it doesn't
		199	exist.
		200
		201	=item 4. A "Unicode String" is simply a string where each character can be
		202	validly interpreted as a Unicode codepoint.
		203
		204	If you have UTF-8 encoded data, it is no longer a Unicode string, but a
		205	Unicode string encoded in UTF-8, giving you a binary string.
		206
		207	=item 5. A string containing "high" (> 255) character values is I<not> a UTF-8 string.
		208
		209	It's a fact. Learn to live with it.
		210
		211	=back
		212
		213	I hope this helps :)
133		214
134		215
135	=head1 OBJECT-ORIENTED INTERFACE	216	=head1 OBJECT-ORIENTED INTERFACE
136		217
137	The object oriented interface lets you configure your own encoding or	218	The object oriented interface lets you configure your own encoding or
…		…
150	my $json = JSON::XS->new->utf8->space_after->encode ({a => [1,2]})	231	my $json = JSON::XS->new->utf8->space_after->encode ({a => [1,2]})
151	=> {"a": [1, 2]}	232	=> {"a": [1, 2]}
152		233
153	=item $json = $json->ascii ([$enable])	234	=item $json = $json->ascii ([$enable])
154		235
		236	=item $enabled = $json->get_ascii
		237
155	If C<$enable> is true (or missing), then the C<encode> method will not	238	If C<$enable> is true (or missing), then the C<encode> method will not
156	generate characters outside the code range C<0..127> (which is ASCII). Any	239	generate characters outside the code range C<0..127> (which is ASCII). Any
157	unicode characters outside that range will be escaped using either a	240	Unicode characters outside that range will be escaped using either a
158	single \uXXXX (BMP characters) or a double \uHHHH\uLLLLL escape sequence,	241	single \uXXXX (BMP characters) or a double \uHHHH\uLLLLL escape sequence,
159	as per RFC4627.	242	as per RFC4627. The resulting encoded JSON text can be treated as a native
		243	Unicode string, an ascii-encoded, latin1-encoded or UTF-8 encoded string,
		244	or any other superset of ASCII.
160		245
161	If C<$enable> is false, then the C<encode> method will not escape Unicode	246	If C<$enable> is false, then the C<encode> method will not escape Unicode
162	characters unless required by the JSON syntax. This results in a faster	247	characters unless required by the JSON syntax or other flags. This results
163	and more compact format.	248	in a faster and more compact format.
		249
		250	See also the section I<ENCODING/CODESET FLAG NOTES> later in this
		251	document.
		252
		253	The main use for this flag is to produce JSON texts that can be
		254	transmitted over a 7-bit channel, as the encoded JSON texts will not
		255	contain any 8 bit characters.
164		256
165	JSON::XS->new->ascii (1)->encode ([chr 0x10401])	257	JSON::XS->new->ascii (1)->encode ([chr 0x10401])
166	=> ["\ud801\udc01"]	258	=> ["\ud801\udc01"]
167		259
		260	=item $json = $json->latin1 ([$enable])
		261
		262	=item $enabled = $json->get_latin1
		263
		264	If C<$enable> is true (or missing), then the C<encode> method will encode
		265	the resulting JSON text as latin1 (or iso-8859-1), escaping any characters
		266	outside the code range C<0..255>. The resulting string can be treated as a
		267	latin1-encoded JSON text or a native Unicode string. The C<decode> method
		268	will not be affected in any way by this flag, as C<decode> by default
		269	expects Unicode, which is a strict superset of latin1.
		270
		271	If C<$enable> is false, then the C<encode> method will not escape Unicode
		272	characters unless required by the JSON syntax or other flags.
		273
		274	See also the section I<ENCODING/CODESET FLAG NOTES> later in this
		275	document.
		276
		277	The main use for this flag is efficiently encoding binary data as JSON
		278	text, as most octets will not be escaped, resulting in a smaller encoded
		279	size. The disadvantage is that the resulting JSON text is encoded
		280	in latin1 (and must correctly be treated as such when storing and
		281	transferring), a rare encoding for JSON. It is therefore most useful when
		282	you want to store data structures known to contain binary data efficiently
		283	in files or databases, not when talking to other JSON encoders/decoders.
		284
		285	JSON::XS->new->latin1->encode (["\x{89}\x{abc}"]
		286	=> ["\x{89}\\u0abc"] # (perl syntax, U+abc escaped, U+89 not)
		287
168	=item $json = $json->utf8 ([$enable])	288	=item $json = $json->utf8 ([$enable])
		289
		290	=item $enabled = $json->get_utf8
169		291
170	If C<$enable> is true (or missing), then the C<encode> method will encode	292	If C<$enable> is true (or missing), then the C<encode> method will encode
171	the JSON result into UTF-8, as required by many protocols, while the	293	the JSON result into UTF-8, as required by many protocols, while the
172	C<decode> method expects to be handled an UTF-8-encoded string. Please	294	C<decode> method expects to be handled an UTF-8-encoded string. Please
173	note that UTF-8-encoded strings do not contain any characters outside the	295	note that UTF-8-encoded strings do not contain any characters outside the
174	range C<0..255>, they are thus useful for bytewise/binary I/O. In future	296	range C<0..255>, they are thus useful for bytewise/binary I/O. In future
175	versions, enabling this option might enable autodetection of the UTF-16	297	versions, enabling this option might enable autodetection of the UTF-16
176	and UTF-32 encoding families, as described in RFC4627.	298	and UTF-32 encoding families, as described in RFC4627.
177		299
178	If C<$enable> is false, then the C<encode> method will return the JSON	300	If C<$enable> is false, then the C<encode> method will return the JSON
179	string as a (non-encoded) unicode string, while C<decode> expects thus a	301	string as a (non-encoded) Unicode string, while C<decode> expects thus a
180	unicode string. Any decoding or encoding (e.g. to UTF-8 or UTF-16) needs	302	Unicode string. Any decoding or encoding (e.g. to UTF-8 or UTF-16) needs
181	to be done yourself, e.g. using the Encode module.	303	to be done yourself, e.g. using the Encode module.
		304
		305	See also the section I<ENCODING/CODESET FLAG NOTES> later in this
		306	document.
182		307
183	Example, output UTF-16BE-encoded JSON:	308	Example, output UTF-16BE-encoded JSON:
184		309
185	use Encode;	310	use Encode;
186	$jsontext = encode "UTF-16BE", JSON::XS->new->encode ($object);	311	$jsontext = encode "UTF-16BE", JSON::XS->new->encode ($object);
…		…
207	]	332	]
208	}	333	}
209		334
210	=item $json = $json->indent ([$enable])	335	=item $json = $json->indent ([$enable])
211		336
		337	=item $enabled = $json->get_indent
		338
212	If C<$enable> is true (or missing), then the C<encode> method will use a multiline	339	If C<$enable> is true (or missing), then the C<encode> method will use a multiline
213	format as output, putting every array member or object/hash key-value pair	340	format as output, putting every array member or object/hash key-value pair
214	into its own line, identing them properly.	341	into its own line, indenting them properly.
215		342
216	If C<$enable> is false, no newlines or indenting will be produced, and the	343	If C<$enable> is false, no newlines or indenting will be produced, and the
217	resulting JSON text is guarenteed not to contain any C<newlines>.	344	resulting JSON text is guaranteed not to contain any C<newlines>.
218		345
219	This setting has no effect when decoding JSON texts.	346	This setting has no effect when decoding JSON texts.
220		347
221	=item $json = $json->space_before ([$enable])	348	=item $json = $json->space_before ([$enable])
		349
		350	=item $enabled = $json->get_space_before
222		351
223	If C<$enable> is true (or missing), then the C<encode> method will add an extra	352	If C<$enable> is true (or missing), then the C<encode> method will add an extra
224	optional space before the C<:> separating keys from values in JSON objects.	353	optional space before the C<:> separating keys from values in JSON objects.
225		354
226	If C<$enable> is false, then the C<encode> method will not add any extra	355	If C<$enable> is false, then the C<encode> method will not add any extra
…		…
232	Example, space_before enabled, space_after and indent disabled:	361	Example, space_before enabled, space_after and indent disabled:
233		362
234	{"key" :"value"}	363	{"key" :"value"}
235		364
236	=item $json = $json->space_after ([$enable])	365	=item $json = $json->space_after ([$enable])
		366
		367	=item $enabled = $json->get_space_after
237		368
238	If C<$enable> is true (or missing), then the C<encode> method will add an extra	369	If C<$enable> is true (or missing), then the C<encode> method will add an extra
239	optional space after the C<:> separating keys from values in JSON objects	370	optional space after the C<:> separating keys from values in JSON objects
240	and extra whitespace after the C<,> separating key-value pairs and array	371	and extra whitespace after the C<,> separating key-value pairs and array
241	members.	372	members.
…		…
247		378
248	Example, space_before and indent disabled, space_after enabled:	379	Example, space_before and indent disabled, space_after enabled:
249		380
250	{"key": "value"}	381	{"key": "value"}
251		382
		383	=item $json = $json->relaxed ([$enable])
		384
		385	=item $enabled = $json->get_relaxed
		386
		387	If C<$enable> is true (or missing), then C<decode> will accept some
		388	extensions to normal JSON syntax (see below). C<encode> will not be
		389	affected in anyway. I<Be aware that this option makes you accept invalid
		390	JSON texts as if they were valid!>. I suggest only to use this option to
		391	parse application-specific files written by humans (configuration files,
		392	resource files etc.)
		393
		394	If C<$enable> is false (the default), then C<decode> will only accept
		395	valid JSON texts.
		396
		397	Currently accepted extensions are:
		398
		399	=over 4
		400
		401	=item * list items can have an end-comma
		402
		403	JSON I<separates> array elements and key-value pairs with commas. This
		404	can be annoying if you write JSON texts manually and want to be able to
		405	quickly append elements, so this extension accepts comma at the end of
		406	such items not just between them:
		407
		408	[
		409	1,
		410	2, <- this comma not normally allowed
		411	]
		412	{
		413	"k1": "v1",
		414	"k2": "v2", <- this comma not normally allowed
		415	}
		416
		417	=item * shell-style '#'-comments
		418
		419	Whenever JSON allows whitespace, shell-style comments are additionally
		420	allowed. They are terminated by the first carriage-return or line-feed
		421	character, after which more white-space and comments are allowed.
		422
		423	[
		424	1, # this comment not allowed in JSON
		425	# neither this one...
		426	]
		427
		428	=back
		429
252	=item $json = $json->canonical ([$enable])	430	=item $json = $json->canonical ([$enable])
		431
		432	=item $enabled = $json->get_canonical
253		433
254	If C<$enable> is true (or missing), then the C<encode> method will output JSON objects	434	If C<$enable> is true (or missing), then the C<encode> method will output JSON objects
255	by sorting their keys. This is adding a comparatively high overhead.	435	by sorting their keys. This is adding a comparatively high overhead.
256		436
257	If C<$enable> is false, then the C<encode> method will output key-value	437	If C<$enable> is false, then the C<encode> method will output key-value
258	pairs in the order Perl stores them (which will likely change between runs	438	pairs in the order Perl stores them (which will likely change between runs
259	of the same script).	439	of the same script).
260		440
261	This option is useful if you want the same data structure to be encoded as	441	This option is useful if you want the same data structure to be encoded as
262	the same JSON text (given the same overall settings). If it is disabled,	442	the same JSON text (given the same overall settings). If it is disabled,
263	the same hash migh be encoded differently even if contains the same data,	443	the same hash might be encoded differently even if contains the same data,
264	as key-value pairs have no inherent ordering in Perl.	444	as key-value pairs have no inherent ordering in Perl.
265		445
266	This setting has no effect when decoding JSON texts.	446	This setting has no effect when decoding JSON texts.
267		447
268	=item $json = $json->allow_nonref ([$enable])	448	=item $json = $json->allow_nonref ([$enable])
		449
		450	=item $enabled = $json->get_allow_nonref
269		451
270	If C<$enable> is true (or missing), then the C<encode> method can convert a	452	If C<$enable> is true (or missing), then the C<encode> method can convert a
271	non-reference into its corresponding string, number or null JSON value,	453	non-reference into its corresponding string, number or null JSON value,
272	which is an extension to RFC4627. Likewise, C<decode> will accept those JSON	454	which is an extension to RFC4627. Likewise, C<decode> will accept those JSON
273	values instead of croaking.	455	values instead of croaking.
…		…
281	resulting in an invalid JSON text:	463	resulting in an invalid JSON text:
282		464
283	JSON::XS->new->allow_nonref->encode ("Hello, World!")	465	JSON::XS->new->allow_nonref->encode ("Hello, World!")
284	=> "Hello, World!"	466	=> "Hello, World!"
285		467
		468	=item $json = $json->allow_unknown ([$enable])
		469
		470	=item $enabled = $json->get_allow_unknown
		471
		472	If C<$enable> is true (or missing), then C<encode> will I<not> throw an
		473	exception when it encounters values it cannot represent in JSON (for
		474	example, filehandles) but instead will encode a JSON C<null> value. Note
		475	that blessed objects are not included here and are handled separately by
		476	c<allow_nonref>.
		477
		478	If C<$enable> is false (the default), then C<encode> will throw an
		479	exception when it encounters anything it cannot encode as JSON.
		480
		481	This option does not affect C<decode> in any way, and it is recommended to
		482	leave it off unless you know your communications partner.
		483
		484	=item $json = $json->allow_blessed ([$enable])
		485
		486	=item $enabled = $json->get_allow_blessed
		487
		488	If C<$enable> is true (or missing), then the C<encode> method will not
		489	barf when it encounters a blessed reference. Instead, the value of the
		490	B<convert_blessed> option will decide whether C<null> (C<convert_blessed>
		491	disabled or no C<TO_JSON> method found) or a representation of the
		492	object (C<convert_blessed> enabled and C<TO_JSON> method found) is being
		493	encoded. Has no effect on C<decode>.
		494
		495	If C<$enable> is false (the default), then C<encode> will throw an
		496	exception when it encounters a blessed object.
		497
		498	=item $json = $json->convert_blessed ([$enable])
		499
		500	=item $enabled = $json->get_convert_blessed
		501
		502	If C<$enable> is true (or missing), then C<encode>, upon encountering a
		503	blessed object, will check for the availability of the C<TO_JSON> method
		504	on the object's class. If found, it will be called in scalar context
		505	and the resulting scalar will be encoded instead of the object. If no
		506	C<TO_JSON> method is found, the value of C<allow_blessed> will decide what
		507	to do.
		508
		509	The C<TO_JSON> method may safely call die if it wants. If C<TO_JSON>
		510	returns other blessed objects, those will be handled in the same
		511	way. C<TO_JSON> must take care of not causing an endless recursion cycle
		512	(== crash) in this case. The name of C<TO_JSON> was chosen because other
		513	methods called by the Perl core (== not by the user of the object) are
		514	usually in upper case letters and to avoid collisions with any C<to_json>
		515	function or method.
		516
		517	This setting does not yet influence C<decode> in any way, but in the
		518	future, global hooks might get installed that influence C<decode> and are
		519	enabled by this setting.
		520
		521	If C<$enable> is false, then the C<allow_blessed> setting will decide what
		522	to do when a blessed object is found.
		523
		524	=item $json = $json->filter_json_object ([$coderef->($hashref)])
		525
		526	When C<$coderef> is specified, it will be called from C<decode> each
		527	time it decodes a JSON object. The only argument is a reference to the
		528	newly-created hash. If the code references returns a single scalar (which
		529	need not be a reference), this value (i.e. a copy of that scalar to avoid
		530	aliasing) is inserted into the deserialised data structure. If it returns
		531	an empty list (NOTE: I<not> C<undef>, which is a valid scalar), the
		532	original deserialised hash will be inserted. This setting can slow down
		533	decoding considerably.
		534
		535	When C<$coderef> is omitted or undefined, any existing callback will
		536	be removed and C<decode> will not change the deserialised hash in any
		537	way.
		538
		539	Example, convert all JSON objects into the integer 5:
		540
		541	my $js = JSON::XS->new->filter_json_object (sub { 5 });
		542	# returns [5]
		543	$js->decode ('[{}]')
		544	# throw an exception because allow_nonref is not enabled
		545	# so a lone 5 is not allowed.
		546	$js->decode ('{"a":1, "b":2}');
		547
		548	=item $json = $json->filter_json_single_key_object ($key [=> $coderef->($value)])
		549
		550	Works remotely similar to C<filter_json_object>, but is only called for
		551	JSON objects having a single key named C<$key>.
		552
		553	This C<$coderef> is called before the one specified via
		554	C<filter_json_object>, if any. It gets passed the single value in the JSON
		555	object. If it returns a single value, it will be inserted into the data
		556	structure. If it returns nothing (not even C<undef> but the empty list),
		557	the callback from C<filter_json_object> will be called next, as if no
		558	single-key callback were specified.
		559
		560	If C<$coderef> is omitted or undefined, the corresponding callback will be
		561	disabled. There can only ever be one callback for a given key.
		562
		563	As this callback gets called less often then the C<filter_json_object>
		564	one, decoding speed will not usually suffer as much. Therefore, single-key
		565	objects make excellent targets to serialise Perl objects into, especially
		566	as single-key JSON objects are as close to the type-tagged value concept
		567	as JSON gets (it's basically an ID/VALUE tuple). Of course, JSON does not
		568	support this in any way, so you need to make sure your data never looks
		569	like a serialised Perl hash.
		570
		571	Typical names for the single object key are C<__class_whatever__>, or
		572	C<$__dollars_are_rarely_used__$> or C<}ugly_brace_placement>, or even
		573	things like C<__class_md5sum(classname)__>, to reduce the risk of clashing
		574	with real hashes.
		575
		576	Example, decode JSON objects of the form C<< { "__widget__" => <id> } >>
		577	into the corresponding C<< $WIDGET{<id>} >> object:
		578
		579	# return whatever is in $WIDGET{5}:
		580	JSON::XS
		581	->new
		582	->filter_json_single_key_object (__widget__ => sub {
		583	$WIDGET{ $_[0] }
		584	})
		585	->decode ('{"__widget__": 5')
		586
		587	# this can be used with a TO_JSON method in some "widget" class
		588	# for serialisation to json:
		589	sub WidgetBase::TO_JSON {
		590	my ($self) = @_;
		591
		592	unless ($self->{id}) {
		593	$self->{id} = ..get..some..id..;
		594	$WIDGET{$self->{id}} = $self;
		595	}
		596
		597	{ __widget__ => $self->{id} }
		598	}
		599
286	=item $json = $json->shrink ([$enable])	600	=item $json = $json->shrink ([$enable])
287		601
		602	=item $enabled = $json->get_shrink
		603
288	Perl usually over-allocates memory a bit when allocating space for	604	Perl usually over-allocates memory a bit when allocating space for
289	strings. This flag optionally resizes strings generated by either	605	strings. This flag optionally resizes strings generated by either
290	C<encode> or C<decode> to their minimum size possible. This can save	606	C<encode> or C<decode> to their minimum size possible. This can save
291	memory when your JSON texts are either very very long or you have many	607	memory when your JSON texts are either very very long or you have many
292	short strings. It will also try to downgrade any strings to octet-form	608	short strings. It will also try to downgrade any strings to octet-form
293	if possible: perl stores strings internally either in an encoding called	609	if possible: perl stores strings internally either in an encoding called
294	UTF-X or in octet-form. The latter cannot store everything but uses less	610	UTF-X or in octet-form. The latter cannot store everything but uses less
295	space in general.	611	space in general (and some buggy Perl or C code might even rely on that
		612	internal representation being used).
296		613
		614	The actual definition of what shrink does might change in future versions,
		615	but it will always try to save space at the expense of time.
		616
297	If C<$enable> is true (or missing), the string returned by C<encode> will be shrunk-to-fit,	617	If C<$enable> is true (or missing), the string returned by C<encode> will
298	while all strings generated by C<decode> will also be shrunk-to-fit.	618	be shrunk-to-fit, while all strings generated by C<decode> will also be
		619	shrunk-to-fit.
299		620
300	If C<$enable> is false, then the normal perl allocation algorithms are used.	621	If C<$enable> is false, then the normal perl allocation algorithms are used.
301	If you work with your data, then this is likely to be faster.	622	If you work with your data, then this is likely to be faster.
302		623
303	In the future, this setting might control other things, such as converting	624	In the future, this setting might control other things, such as converting
304	strings that look like integers or floats into integers or floats	625	strings that look like integers or floats into integers or floats
305	internally (there is no difference on the Perl level), saving space.	626	internally (there is no difference on the Perl level), saving space.
306		627
307	=item $json = $json->max_depth ([$maximum_nesting_depth])	628	=item $json = $json->max_depth ([$maximum_nesting_depth])
308		629
		630	=item $max_depth = $json->get_max_depth
		631
309	Sets the maximum nesting level (default C<8192>) accepted while encoding	632	Sets the maximum nesting level (default C<512>) accepted while encoding
310	or decoding. If the JSON text or Perl data structure has an equal or	633	or decoding. If a higher nesting level is detected in JSON text or a Perl
311	higher nesting level then this limit, then the encoder and decoder will	634	data structure, then the encoder and decoder will stop and croak at that
312	stop and croak at that point.	635	point.
313		636
314	Nesting level is defined by number of hash- or arrayrefs that the encoder	637	Nesting level is defined by number of hash- or arrayrefs that the encoder
315	needs to traverse to reach a given point or the number of C<{> or C<[>	638	needs to traverse to reach a given point or the number of C<{> or C<[>
316	characters without their matching closing parenthesis crossed to reach a	639	characters without their matching closing parenthesis crossed to reach a
317	given character in a string.	640	given character in a string.
318		641
319	Setting the maximum depth to one disallows any nesting, so that ensures	642	Setting the maximum depth to one disallows any nesting, so that ensures
320	that the object is only a single hash/object or array.	643	that the object is only a single hash/object or array.
321		644
322	The argument to C<max_depth> will be rounded up to the next nearest power	645	If no argument is given, the highest possible setting will be used, which
323	of two.	646	is rarely useful.
		647
		648	Note that nesting is implemented by recursion in C. The default value has
		649	been chosen to be as large as typical operating systems allow without
		650	crashing.
		651
		652	See SECURITY CONSIDERATIONS, below, for more info on why this is useful.
		653
		654	=item $json = $json->max_size ([$maximum_string_size])
		655
		656	=item $max_size = $json->get_max_size
		657
		658	Set the maximum length a JSON text may have (in bytes) where decoding is
		659	being attempted. The default is C<0>, meaning no limit. When C<decode>
		660	is called on a string that is longer then this many bytes, it will not
		661	attempt to decode the string but throw an exception. This setting has no
		662	effect on C<encode> (yet).
		663
		664	If no argument is given, the limit check will be deactivated (same as when
		665	C<0> is specified).
324		666
325	See SECURITY CONSIDERATIONS, below, for more info on why this is useful.	667	See SECURITY CONSIDERATIONS, below, for more info on why this is useful.
326		668
327	=item $json_text = $json->encode ($perl_scalar)	669	=item $json_text = $json->encode ($perl_scalar)
328		670
…		…
340		682
341	JSON numbers and strings become simple Perl scalars. JSON arrays become	683	JSON numbers and strings become simple Perl scalars. JSON arrays become
342	Perl arrayrefs and JSON objects become Perl hashrefs. C<true> becomes	684	Perl arrayrefs and JSON objects become Perl hashrefs. C<true> becomes
343	C<1>, C<false> becomes C<0> and C<null> becomes C<undef>.	685	C<1>, C<false> becomes C<0> and C<null> becomes C<undef>.
344		686
		687	=item ($perl_scalar, $characters) = $json->decode_prefix ($json_text)
		688
		689	This works like the C<decode> method, but instead of raising an exception
		690	when there is trailing garbage after the first JSON object, it will
		691	silently stop parsing there and return the number of characters consumed
		692	so far.
		693
		694	This is useful if your JSON texts are not delimited by an outer protocol
		695	(which is not the brightest thing to do in the first place) and you need
		696	to know where the JSON text ends.
		697
		698	JSON::XS->new->decode_prefix ("[1] the tail")
		699	=> ([], 3)
		700
345	=back	701	=back
		702
		703
		704	=head1 INCREMENTAL PARSING
		705
		706	[This section and the API it details is still EXPERIMENTAL]
		707
		708	In some cases, there is the need for incremental parsing of JSON
		709	texts. While this module always has to keep both JSON text and resulting
		710	Perl data structure in memory at one time, it does allow you to parse a
		711	JSON stream incrementally. It does so by accumulating text until it has
		712	a full JSON object, which it then can decode. This process is similar to
		713	using C<decode_prefix> to see if a full JSON object is available, but is
		714	much more efficient (JSON::XS will only attempt to parse the JSON text
		715	once it is sure it has enough text to get a decisive result, using a very
		716	simple but truly incremental parser).
		717
		718	The following two methods deal with this.
		719
		720	=over 4
		721
		722	=item [void, scalar or list context] = $json->incr_parse ([$string])
		723
		724	This is the central parsing function. It can both append new text and
		725	extract objects from the stream accumulated so far (both of these
		726	functions are optional).
		727
		728	If C<$string> is given, then this string is appended to the already
		729	existing JSON fragment stored in the C<$json> object.
		730
		731	After that, if the function is called in void context, it will simply
		732	return without doing anything further. This can be used to add more text
		733	in as many chunks as you want.
		734
		735	If the method is called in scalar context, then it will try to extract
		736	exactly I<one> JSON object. If that is successful, it will return this
		737	object, otherwise it will return C<undef>. If there is a parse error,
		738	this method will croak just as C<decode> would do (one can then use
		739	C<incr_skip> to skip the errornous part). This is the most common way of
		740	using the method.
		741
		742	And finally, in list context, it will try to extract as many objects
		743	from the stream as it can find and return them, or the empty list
		744	otherwise. For this to work, there must be no separators between the JSON
		745	objects or arrays, instead they must be concatenated back-to-back. If
		746	an error occurs, an exception will be raised as in the scalar context
		747	case. Note that in this case, any previously-parsed JSON texts will be
		748	lost.
		749
		750	=item $lvalue_string = $json->incr_text
		751
		752	This method returns the currently stored JSON fragment as an lvalue, that
		753	is, you can manipulate it. This I<only> works when a preceding call to
		754	C<incr_parse> in I<scalar context> successfully returned an object. Under
		755	all other circumstances you must not call this function (I mean it.
		756	although in simple tests it might actually work, it I<will> fail under
		757	real world conditions). As a special exception, you can also call this
		758	method before having parsed anything.
		759
		760	This function is useful in two cases: a) finding the trailing text after a
		761	JSON object or b) parsing multiple JSON objects separated by non-JSON text
		762	(such as commas).
		763
		764	=item $json->incr_skip
		765
		766	This will reset the state of the incremental parser and will remove the
		767	parsed text from the input buffer. This is useful after C<incr_parse>
		768	died, in which case the input buffer and incremental parser state is left
		769	unchanged, to skip the text parsed so far and to reset the parse state.
		770
		771	=back
		772
		773	=head2 LIMITATIONS
		774
		775	All options that affect decoding are supported, except
		776	C<allow_nonref>. The reason for this is that it cannot be made to
		777	work sensibly: JSON objects and arrays are self-delimited, i.e. you can concatenate
		778	them back to back and still decode them perfectly. This does not hold true
		779	for JSON numbers, however.
		780
		781	For example, is the string C<1> a single JSON number, or is it simply the
		782	start of C<12>? Or is C<12> a single JSON number, or the concatenation
		783	of C<1> and C<2>? In neither case you can tell, and this is why JSON::XS
		784	takes the conservative route and disallows this case.
		785
		786	=head2 EXAMPLES
		787
		788	Some examples will make all this clearer. First, a simple example that
		789	works similarly to C<decode_prefix>: We want to decode the JSON object at
		790	the start of a string and identify the portion after the JSON object:
		791
		792	my $text = "[1,2,3] hello";
		793
		794	my $json = new JSON::XS;
		795
		796	my $obj = $json->incr_parse ($text)
		797	or die "expected JSON object or array at beginning of string";
		798
		799	my $tail = $json->incr_text;
		800	# $tail now contains " hello"
		801
		802	Easy, isn't it?
		803
		804	Now for a more complicated example: Imagine a hypothetical protocol where
		805	you read some requests from a TCP stream, and each request is a JSON
		806	array, without any separation between them (in fact, it is often useful to
		807	use newlines as "separators", as these get interpreted as whitespace at
		808	the start of the JSON text, which makes it possible to test said protocol
		809	with C<telnet>...).
		810
		811	Here is how you'd do it (it is trivial to write this in an event-based
		812	manner):
		813
		814	my $json = new JSON::XS;
		815
		816	# read some data from the socket
		817	while (sysread $socket, my $buf, 4096) {
		818
		819	# split and decode as many requests as possible
		820	for my $request ($json->incr_parse ($buf)) {
		821	# act on the $request
		822	}
		823	}
		824
		825	Another complicated example: Assume you have a string with JSON objects
		826	or arrays, all separated by (optional) comma characters (e.g. C<[1],[2],
		827	[3]>). To parse them, we have to skip the commas between the JSON texts,
		828	and here is where the lvalue-ness of C<incr_text> comes in useful:
		829
		830	my $text = "[1],[2], [3]";
		831	my $json = new JSON::XS;
		832
		833	# void context, so no parsing done
		834	$json->incr_parse ($text);
		835
		836	# now extract as many objects as possible. note the
		837	# use of scalar context so incr_text can be called.
		838	while (my $obj = $json->incr_parse) {
		839	# do something with $obj
		840
		841	# now skip the optional comma
		842	$json->incr_text =~ s/^ \s* , //x;
		843	}
		844
		845	Now lets go for a very complex example: Assume that you have a gigantic
		846	JSON array-of-objects, many gigabytes in size, and you want to parse it,
		847	but you cannot load it into memory fully (this has actually happened in
		848	the real world :).
		849
		850	Well, you lost, you have to implement your own JSON parser. But JSON::XS
		851	can still help you: You implement a (very simple) array parser and let
		852	JSON decode the array elements, which are all full JSON objects on their
		853	own (this wouldn't work if the array elements could be JSON numbers, for
		854	example):
		855
		856	my $json = new JSON::XS;
		857
		858	# open the monster
		859	open my $fh, "<bigfile.json"
		860	or die "bigfile: $!";
		861
		862	# first parse the initial "["
		863	for (;;) {
		864	sysread $fh, my $buf, 65536
		865	or die "read error: $!";
		866	$json->incr_parse ($buf); # void context, so no parsing
		867
		868	# Exit the loop once we found and removed(!) the initial "[".
		869	# In essence, we are (ab-)using the $json object as a simple scalar
		870	# we append data to.
		871	last if $json->incr_text =~ s/^ \s* \[ //x;
		872	}
		873
		874	# now we have the skipped the initial "[", so continue
		875	# parsing all the elements.
		876	for (;;) {
		877	# in this loop we read data until we got a single JSON object
		878	for (;;) {
		879	if (my $obj = $json->incr_parse) {
		880	# do something with $obj
		881	last;
		882	}
		883
		884	# add more data
		885	sysread $fh, my $buf, 65536
		886	or die "read error: $!";
		887	$json->incr_parse ($buf); # void context, so no parsing
		888	}
		889
		890	# in this loop we read data until we either found and parsed the
		891	# separating "," between elements, or the final "]"
		892	for (;;) {
		893	# first skip whitespace
		894	$json->incr_text =~ s/^\s*//;
		895
		896	# if we find "]", we are done
		897	if ($json->incr_text =~ s/^\]//) {
		898	print "finished.\n";
		899	exit;
		900	}
		901
		902	# if we find ",", we can continue with the next element
		903	if ($json->incr_text =~ s/^,//) {
		904	last;
		905	}
		906
		907	# if we find anything else, we have a parse error!
		908	if (length $json->incr_text) {
		909	die "parse error near ", $json->incr_text;
		910	}
		911
		912	# else add more data
		913	sysread $fh, my $buf, 65536
		914	or die "read error: $!";
		915	$json->incr_parse ($buf); # void context, so no parsing
		916	}
		917
		918	This is a complex example, but most of the complexity comes from the fact
		919	that we are trying to be correct (bear with me if I am wrong, I never ran
		920	the above example :).
		921
346		922
347		923
348	=head1 MAPPING	924	=head1 MAPPING
349		925
350	This section describes how JSON::XS maps Perl values to JSON values and	926	This section describes how JSON::XS maps Perl values to JSON values and
351	vice versa. These mappings are designed to "do the right thing" in most	927	vice versa. These mappings are designed to "do the right thing" in most
352	circumstances automatically, preserving round-tripping characteristics	928	circumstances automatically, preserving round-tripping characteristics
353	(what you put in comes out as something equivalent).	929	(what you put in comes out as something equivalent).
354		930
355	For the more enlightened: note that in the following descriptions,	931	For the more enlightened: note that in the following descriptions,
356	lowercase I<perl> refers to the Perl interpreter, while uppcercase I<Perl>	932	lowercase I<perl> refers to the Perl interpreter, while uppercase I<Perl>
357	refers to the abstract Perl language itself.	933	refers to the abstract Perl language itself.
358		934
		935
359	=head2 JSON -> PERL	936	=head2 JSON -> PERL
360		937
361	=over 4	938	=over 4
362		939
363	=item object	940	=item object
364		941
365	A JSON object becomes a reference to a hash in Perl. No ordering of object	942	A JSON object becomes a reference to a hash in Perl. No ordering of object
366	keys is preserved (JSON does not preserver object key ordering itself).	943	keys is preserved (JSON does not preserve object key ordering itself).
367		944
368	=item array	945	=item array
369		946
370	A JSON array becomes a reference to an array in Perl.	947	A JSON array becomes a reference to an array in Perl.
371		948
…		…
375	are represented by the same codepoints in the Perl string, so no manual	952	are represented by the same codepoints in the Perl string, so no manual
376	decoding is necessary.	953	decoding is necessary.
377		954
378	=item number	955	=item number
379		956
380	A JSON number becomes either an integer or numeric (floating point)	957	A JSON number becomes either an integer, numeric (floating point) or
381	scalar in perl, depending on its range and any fractional parts. On the	958	string scalar in perl, depending on its range and any fractional parts. On
382	Perl level, there is no difference between those as Perl handles all the	959	the Perl level, there is no difference between those as Perl handles all
383	conversion details, but an integer may take slightly less memory and might	960	the conversion details, but an integer may take slightly less memory and
384	represent more values exactly than (floating point) numbers.	961	might represent more values exactly than floating point numbers.
		962
		963	If the number consists of digits only, JSON::XS will try to represent
		964	it as an integer value. If that fails, it will try to represent it as
		965	a numeric (floating point) value if that is possible without loss of
		966	precision. Otherwise it will preserve the number as a string value (in
		967	which case you lose roundtripping ability, as the JSON number will be
		968	re-encoded toa JSON string).
		969
		970	Numbers containing a fractional or exponential part will always be
		971	represented as numeric (floating point) values, possibly at a loss of
		972	precision (in which case you might lose perfect roundtripping ability, but
		973	the JSON number will still be re-encoded as a JSON number).
385		974
386	=item true, false	975	=item true, false
387		976
388	These JSON atoms become C<0>, C<1>, respectively. Information is lost in	977	These JSON atoms become C<JSON::XS::true> and C<JSON::XS::false>,
389	this process. Future versions might represent those values differently,	978	respectively. They are overloaded to act almost exactly like the numbers
390	but they will be guarenteed to act like these integers would normally in	979	C<1> and C<0>. You can check whether a scalar is a JSON boolean by using
391	Perl.	980	the C<JSON::XS::is_bool> function.
392		981
393	=item null	982	=item null
394		983
395	A JSON null atom becomes C<undef> in Perl.	984	A JSON null atom becomes C<undef> in Perl.
396		985
397	=back	986	=back
		987
398		988
399	=head2 PERL -> JSON	989	=head2 PERL -> JSON
400		990
401	The mapping from Perl to JSON is slightly more difficult, as Perl is a	991	The mapping from Perl to JSON is slightly more difficult, as Perl is a
402	truly typeless language, so we can only guess which JSON type is meant by	992	truly typeless language, so we can only guess which JSON type is meant by
…		…
405	=over 4	995	=over 4
406		996
407	=item hash references	997	=item hash references
408		998
409	Perl hash references become JSON objects. As there is no inherent ordering	999	Perl hash references become JSON objects. As there is no inherent ordering
410	in hash keys, they will usually be encoded in a pseudo-random order that	1000	in hash keys (or JSON objects), they will usually be encoded in a
411	can change between runs of the same program but stays generally the same	1001	pseudo-random order that can change between runs of the same program but
412	within a single run of a program. JSON::XS can optionally sort the hash	1002	stays generally the same within a single run of a program. JSON::XS can
413	keys (determined by the I<canonical> flag), so the same datastructure	1003	optionally sort the hash keys (determined by the I<canonical> flag), so
414	will serialise to the same JSON text (given same settings and version of	1004	the same datastructure will serialise to the same JSON text (given same
415	JSON::XS), but this incurs a runtime overhead.	1005	settings and version of JSON::XS), but this incurs a runtime overhead
		1006	and is only rarely useful, e.g. when you want to compare some JSON text
		1007	against another for equality.
416		1008
417	=item array references	1009	=item array references
418		1010
419	Perl array references become JSON arrays.	1011	Perl array references become JSON arrays.
420		1012
		1013	=item other references
		1014
		1015	Other unblessed references are generally not allowed and will cause an
		1016	exception to be thrown, except for references to the integers C<0> and
		1017	C<1>, which get turned into C<false> and C<true> atoms in JSON. You can
		1018	also use C<JSON::XS::false> and C<JSON::XS::true> to improve readability.
		1019
		1020	encode_json [\0, JSON::XS::true] # yields [false,true]
		1021
		1022	=item JSON::XS::true, JSON::XS::false
		1023
		1024	These special values become JSON true and JSON false values,
		1025	respectively. You can also use C<\1> and C<\0> directly if you want.
		1026
421	=item blessed objects	1027	=item blessed objects
422		1028
423	Blessed objects are not allowed. JSON::XS currently tries to encode their	1029	Blessed objects are not directly representable in JSON. See the
424	underlying representation (hash- or arrayref), but this behaviour might	1030	C<allow_blessed> and C<convert_blessed> methods on various options on
425	change in future versions.	1031	how to deal with this: basically, you can choose between throwing an
		1032	exception, encoding the reference as if it weren't blessed, or provide
		1033	your own serialiser method.
426		1034
427	=item simple scalars	1035	=item simple scalars
428		1036
429	Simple Perl scalars (any scalar that is not a reference) are the most	1037	Simple Perl scalars (any scalar that is not a reference) are the most
430	difficult objects to encode: JSON::XS will encode undefined scalars as	1038	difficult objects to encode: JSON::XS will encode undefined scalars as
431	JSON null value, scalars that have last been used in a string context	1039	JSON C<null> values, scalars that have last been used in a string context
432	before encoding as JSON strings and anything else as number value:	1040	before encoding as JSON strings, and anything else as number value:
433		1041
434	# dump as number	1042	# dump as number
435	to_json [2] # yields [2]	1043	encode_json [2] # yields [2]
436	to_json [-3.0e17] # yields [-3e+17]	1044	encode_json [-3.0e17] # yields [-3e+17]
437	my $value = 5; to_json [$value] # yields [5]	1045	my $value = 5; encode_json [$value] # yields [5]
438		1046
439	# used as string, so dump as string	1047	# used as string, so dump as string
440	print $value;	1048	print $value;
441	to_json [$value] # yields ["5"]	1049	encode_json [$value] # yields ["5"]
442		1050
443	# undef becomes null	1051	# undef becomes null
444	to_json [undef] # yields [null]	1052	encode_json [undef] # yields [null]
445		1053
446	You can force the type to be a string by stringifying it:	1054	You can force the type to be a JSON string by stringifying it:
447		1055
448	my $x = 3.1; # some variable containing a number	1056	my $x = 3.1; # some variable containing a number
449	"$x"; # stringified	1057	"$x"; # stringified
450	$x .= ""; # another, more awkward way to stringify	1058	$x .= ""; # another, more awkward way to stringify
451	print $x; # perl does it for you, too, quite often	1059	print $x; # perl does it for you, too, quite often
452		1060
453	You can force the type to be a number by numifying it:	1061	You can force the type to be a JSON number by numifying it:
454		1062
455	my $x = "3"; # some variable containing a string	1063	my $x = "3"; # some variable containing a string
456	$x += 0; # numify it, ensuring it will be dumped as a number	1064	$x += 0; # numify it, ensuring it will be dumped as a number
457	$x *= 1; # same thing, the choise is yours.	1065	$x *= 1; # same thing, the choice is yours.
458		1066
459	You can not currently output JSON booleans or force the type in other,	1067	You can not currently force the type in other, less obscure, ways. Tell me
460	less obscure, ways. Tell me if you need this capability.	1068	if you need this capability (but don't forget to explain why it's needed
461		1069	:).
462	=item circular data structures
463
464	Those will be encoded until memory or stackspace runs out.
465		1070
466	=back	1071	=back
467		1072
468		1073
469	=head1 COMPARISON	1074	=head1 ENCODING/CODESET FLAG NOTES
470		1075
471	As already mentioned, this module was created because none of the existing	1076	The interested reader might have seen a number of flags that signify
472	JSON modules could be made to work correctly. First I will describe the	1077	encodings or codesets - C<utf8>, C<latin1> and C<ascii>. There seems to be
473	problems (or pleasures) I encountered with various existing JSON modules,	1078	some confusion on what these do, so here is a short comparison:
474	followed by some benchmark values. JSON::XS was designed not to suffer	1079
475	from any of these problems or limitations.	1080	C<utf8> controls whether the JSON text created by C<encode> (and expected
		1081	by C<decode>) is UTF-8 encoded or not, while C<latin1> and C<ascii> only
		1082	control whether C<encode> escapes character values outside their respective
		1083	codeset range. Neither of these flags conflict with each other, although
		1084	some combinations make less sense than others.
		1085
		1086	Care has been taken to make all flags symmetrical with respect to
		1087	C<encode> and C<decode>, that is, texts encoded with any combination of
		1088	these flag values will be correctly decoded when the same flags are used
		1089	- in general, if you use different flag settings while encoding vs. when
		1090	decoding you likely have a bug somewhere.
		1091
		1092	Below comes a verbose discussion of these flags. Note that a "codeset" is
		1093	simply an abstract set of character-codepoint pairs, while an encoding
		1094	takes those codepoint numbers and I<encodes> them, in our case into
		1095	octets. Unicode is (among other things) a codeset, UTF-8 is an encoding,
		1096	and ISO-8859-1 (= latin 1) and ASCII are both codesets I<and> encodings at
		1097	the same time, which can be confusing.
476		1098
477	=over 4	1099	=over 4
478		1100
479	=item JSON 1.07	1101	=item C<utf8> flag disabled
480		1102
481	Slow (but very portable, as it is written in pure Perl).	1103	When C<utf8> is disabled (the default), then C<encode>/C<decode> generate
		1104	and expect Unicode strings, that is, characters with high ordinal Unicode
		1105	values (> 255) will be encoded as such characters, and likewise such
		1106	characters are decoded as-is, no canges to them will be done, except
		1107	"(re-)interpreting" them as Unicode codepoints or Unicode characters,
		1108	respectively (to Perl, these are the same thing in strings unless you do
		1109	funny/weird/dumb stuff).
482		1110
483	Undocumented/buggy Unicode handling (how JSON handles unicode values is	1111	This is useful when you want to do the encoding yourself (e.g. when you
484	undocumented. One can get far by feeding it unicode strings and doing	1112	want to have UTF-16 encoded JSON texts) or when some other layer does
485	en-/decoding oneself, but unicode escapes are not working properly).	1113	the encoding for you (for example, when printing to a terminal using a
		1114	filehandle that transparently encodes to UTF-8 you certainly do NOT want
		1115	to UTF-8 encode your data first and have Perl encode it another time).
486		1116
487	No roundtripping (strings get clobbered if they look like numbers, e.g.	1117	=item C<utf8> flag enabled
488	the string C<2.0> will encode to C<2.0> instead of C<"2.0">, and that will
489	decode into the number 2.
490		1118
491	=item JSON::PC 0.01	1119	If the C<utf8>-flag is enabled, C<encode>/C<decode> will encode all
		1120	characters using the corresponding UTF-8 multi-byte sequence, and will
		1121	expect your input strings to be encoded as UTF-8, that is, no "character"
		1122	of the input string must have any value > 255, as UTF-8 does not allow
		1123	that.
492		1124
493	Very fast.	1125	The C<utf8> flag therefore switches between two modes: disabled means you
		1126	will get a Unicode string in Perl, enabled means you get an UTF-8 encoded
		1127	octet/binary string in Perl.
494		1128
495	Undocumented/buggy Unicode handling.	1129	=item C<latin1> or C<ascii> flags enabled
496		1130
497	No roundtripping.	1131	With C<latin1> (or C<ascii>) enabled, C<encode> will escape characters
		1132	with ordinal values > 255 (> 127 with C<ascii>) and encode the remaining
		1133	characters as specified by the C<utf8> flag.
498		1134
499	Has problems handling many Perl values (e.g. regex results and other magic	1135	If C<utf8> is disabled, then the result is also correctly encoded in those
500	values will make it croak).	1136	character sets (as both are proper subsets of Unicode, meaning that a
		1137	Unicode string with all character values < 256 is the same thing as a
		1138	ISO-8859-1 string, and a Unicode string with all character values < 128 is
		1139	the same thing as an ASCII string in Perl).
501		1140
502	Does not even generate valid JSON (C<{1,2}> gets converted to C<{1:2}>	1141	If C<utf8> is enabled, you still get a correct UTF-8-encoded string,
503	which is not a valid JSON text.	1142	regardless of these flags, just some more characters will be escaped using
		1143	C<\uXXXX> then before.
504		1144
505	Unmaintained (maintainer unresponsive for many months, bugs are not	1145	Note that ISO-8859-1-I<encoded> strings are not compatible with UTF-8
506	getting fixed).	1146	encoding, while ASCII-encoded strings are. That is because the ISO-8859-1
		1147	encoding is NOT a subset of UTF-8 (despite the ISO-8859-1 I<codeset> being
		1148	a subset of Unicode), while ASCII is.
507		1149
508	=item JSON::Syck 0.21	1150	Surprisingly, C<decode> will ignore these flags and so treat all input
		1151	values as governed by the C<utf8> flag. If it is disabled, this allows you
		1152	to decode ISO-8859-1- and ASCII-encoded strings, as both strict subsets of
		1153	Unicode. If it is enabled, you can correctly decode UTF-8 encoded strings.
509		1154
510	Very buggy (often crashes).	1155	So neither C<latin1> nor C<ascii> are incompatible with the C<utf8> flag -
		1156	they only govern when the JSON output engine escapes a character or not.
511		1157
512	Very inflexible (no human-readable format supported, format pretty much	1158	The main use for C<latin1> is to relatively efficiently store binary data
513	undocumented. I need at least a format for easy reading by humans and a	1159	as JSON, at the expense of breaking compatibility with most JSON decoders.
514	single-line compact format for use in a protocol, and preferably a way to
515	generate ASCII-only JSON texts).
516		1160
517	Completely broken (and confusingly documented) Unicode handling (unicode	1161	The main use for C<ascii> is to force the output to not contain characters
518	escapes are not working properly, you need to set ImplicitUnicode to	1162	with values > 127, which means you can interpret the resulting string
519	I<different> values on en- and decoding to get symmetric behaviour).	1163	as UTF-8, ISO-8859-1, ASCII, KOI8-R or most about any character set and
520		1164	8-bit-encoding, and still get the same data structure back. This is useful
521	No roundtripping (simple cases work, but this depends on wether the scalar	1165	when your channel for JSON transfer is not 8-bit clean or the encoding
522	value was used in a numeric context or not).	1166	might be mangled in between (e.g. in mail), and works because ASCII is a
523		1167	proper subset of most 8-bit and multibyte encodings in use in the world.
524	Dumping hashes may skip hash values depending on iterator state.
525
526	Unmaintained (maintainer unresponsive for many months, bugs are not
527	getting fixed).
528
529	Does not check input for validity (i.e. will accept non-JSON input and
530	return "something" instead of raising an exception. This is a security
531	issue: imagine two banks transfering money between each other using
532	JSON. One bank might parse a given non-JSON request and deduct money,
533	while the other might reject the transaction with a syntax error. While a
534	good protocol will at least recover, that is extra unnecessary work and
535	the transaction will still not succeed).
536
537	=item JSON::DWIW 0.04
538
539	Very fast. Very natural. Very nice.
540
541	Undocumented unicode handling (but the best of the pack. Unicode escapes
542	still don't get parsed properly).
543
544	Very inflexible.
545
546	No roundtripping.
547
548	Does not generate valid JSON texts (key strings are often unquoted, empty keys
549	result in nothing being output)
550
551	Does not check input for validity.
552		1168
553	=back	1169	=back
		1170
		1171
		1172	=head2 JSON and YAML
		1173
		1174	You often hear that JSON is a subset of YAML. This is, however, a mass
		1175	hysteria(*) and very far from the truth (as of the time of this writing),
		1176	so let me state it clearly: I<in general, there is no way to configure
		1177	JSON::XS to output a data structure as valid YAML> that works in all
		1178	cases.
		1179
		1180	If you really must use JSON::XS to generate YAML, you should use this
		1181	algorithm (subject to change in future versions):
		1182
		1183	my $to_yaml = JSON::XS->new->utf8->space_after (1);
		1184	my $yaml = $to_yaml->encode ($ref) . "\n";
		1185
		1186	This will I<usually> generate JSON texts that also parse as valid
		1187	YAML. Please note that YAML has hardcoded limits on (simple) object key
		1188	lengths that JSON doesn't have and also has different and incompatible
		1189	unicode handling, so you should make sure that your hash keys are
		1190	noticeably shorter than the 1024 "stream characters" YAML allows and that
		1191	you do not have characters with codepoint values outside the Unicode BMP
		1192	(basic multilingual page). YAML also does not allow C<\/> sequences in
		1193	strings (which JSON::XS does not I<currently> generate, but other JSON
		1194	generators might).
		1195
		1196	There might be other incompatibilities that I am not aware of (or the YAML
		1197	specification has been changed yet again - it does so quite often). In
		1198	general you should not try to generate YAML with a JSON generator or vice
		1199	versa, or try to parse JSON with a YAML parser or vice versa: chances are
		1200	high that you will run into severe interoperability problems when you
		1201	least expect it.
		1202
		1203	=over 4
		1204
		1205	=item (*)
		1206
		1207	I have been pressured multiple times by Brian Ingerson (one of the
		1208	authors of the YAML specification) to remove this paragraph, despite him
		1209	acknowledging that the actual incompatibilities exist. As I was personally
		1210	bitten by this "JSON is YAML" lie, I refused and said I will continue to
		1211	educate people about these issues, so others do not run into the same
		1212	problem again and again. After this, Brian called me a (quote)I<complete
		1213	and worthless idiot>(unquote).
		1214
		1215	In my opinion, instead of pressuring and insulting people who actually
		1216	clarify issues with YAML and the wrong statements of some of its
		1217	proponents, I would kindly suggest reading the JSON spec (which is not
		1218	that difficult or long) and finally make YAML compatible to it, and
		1219	educating users about the changes, instead of spreading lies about the
		1220	real compatibility for many I<years> and trying to silence people who
		1221	point out that it isn't true.
		1222
		1223	=back
		1224
554		1225
555	=head2 SPEED	1226	=head2 SPEED
556		1227
557	It seems that JSON::XS is surprisingly fast, as shown in the following	1228	It seems that JSON::XS is surprisingly fast, as shown in the following
558	tables. They have been generated with the help of the C<eg/bench> program	1229	tables. They have been generated with the help of the C<eg/bench> program
559	in the JSON::XS distribution, to make it easy to compare on your own	1230	in the JSON::XS distribution, to make it easy to compare on your own
560	system.	1231	system.
561		1232
562	First comes a comparison between various modules using a very short JSON	1233	First comes a comparison between various modules using
563	string:	1234	a very short single-line JSON string (also available at
		1235	L<http://dist.schmorp.de/misc/json/short.json>).
564		1236
565	{"method": "handleMessage", "params": ["user1", "we were just talking"], "id": null}	1237	{"method": "handleMessage", "params": ["user1",
		1238	"we were just talking"], "id": null, "array":[1,11,234,-5,1e5,1e7,
		1239	true, false]}
566		1240
567	It shows the number of encodes/decodes per second (JSON::XS uses the	1241	It shows the number of encodes/decodes per second (JSON::XS uses
568	functional interface, while JSON::XS/2 uses the OO interface with	1242	the functional interface, while JSON::XS/2 uses the OO interface
569	pretty-printing and hashkey sorting enabled). Higher is better:	1243	with pretty-printing and hashkey sorting enabled, JSON::XS/3 enables
		1244	shrink). Higher is better:
570		1245
571	module | encode | decode |	1246	module | encode | decode |
572	-----------|------------|------------|	1247	-----------|------------|------------|
573	JSON | 11488.516 | 7823.035 |	1248	JSON 1.x | 4990.842 | 4088.813 |
574	JSON::DWIW | 94708.054 | 129094.260 |	1249	JSON::DWIW | 51653.990 | 71575.154 |
575	JSON::PC | 63884.157 | 128528.212 |	1250	JSON::PC | 65948.176 | 74631.744 |
576	JSON::Syck | 34898.677 | 42096.911 |	1251	JSON::PP | 8931.652 | 3817.168 |
577	JSON::XS | 654027.064 | 396423.669 |	1252	JSON::Syck | 24877.248 | 27776.848 |
578	JSON::XS/2 | 371564.190 | 371725.613 |	1253	JSON::XS | 388361.481 | 227951.304 |
		1254	JSON::XS/2 | 227951.304 | 218453.333 |
		1255	JSON::XS/3 | 338250.323 | 218453.333 |
		1256	Storable | 16500.016 | 135300.129 |
579	-----------+------------+------------+	1257	-----------+------------+------------+
580		1258
581	That is, JSON::XS is more than six times faster than JSON::DWIW on	1259	That is, JSON::XS is about five times faster than JSON::DWIW on encoding,
582	encoding, more than three times faster on decoding, and about thirty times	1260	about three times faster on decoding, and over forty times faster
583	faster than JSON, even with pretty-printing and key sorting.	1261	than JSON, even with pretty-printing and key sorting. It also compares
		1262	favourably to Storable for small amounts of data.
584		1263
585	Using a longer test string (roughly 18KB, generated from Yahoo! Locals	1264	Using a longer test string (roughly 18KB, generated from Yahoo! Locals
586	search API (http://nanoref.com/yahooapis/mgPdGg):	1265	search API (L<http://dist.schmorp.de/misc/json/long.json>).
587		1266
588	module | encode | decode |	1267	module | encode | decode |
589	-----------|------------|------------|	1268	-----------|------------|------------|
590	JSON | 273.023 | 44.674 |	1269	JSON 1.x | 55.260 | 34.971 |
591	JSON::DWIW | 1089.383 | 1145.704 |	1270	JSON::DWIW | 825.228 | 1082.513 |
592	JSON::PC | 3097.419 | 2393.921 |	1271	JSON::PC | 3571.444 | 2394.829 |
593	JSON::Syck | 514.060 | 843.053 |	1272	JSON::PP | 210.987 | 32.574 |
594	JSON::XS | 6479.668 | 3636.364 |	1273	JSON::Syck | 552.551 | 787.544 |
595	JSON::XS/2 | 3774.221 | 3599.124 |	1274	JSON::XS | 5780.463 | 4854.519 |
		1275	JSON::XS/2 | 3869.998 | 4798.975 |
		1276	JSON::XS/3 | 5862.880 | 4798.975 |
		1277	Storable | 4445.002 | 5235.027 |
596	-----------+------------+------------+	1278	-----------+------------+------------+
597		1279
598	Again, JSON::XS leads by far.	1280	Again, JSON::XS leads by far (except for Storable which non-surprisingly
		1281	decodes faster).
599		1282
600	On large strings containing lots of high unicode characters, some modules	1283	On large strings containing lots of high Unicode characters, some modules
601	(such as JSON::PC) seem to decode faster than JSON::XS, but the result	1284	(such as JSON::PC) seem to decode faster than JSON::XS, but the result
602	will be broken due to missing (or wrong) unicode handling. Others refuse	1285	will be broken due to missing (or wrong) Unicode handling. Others refuse
603	to decode or encode properly, so it was impossible to prepare a fair	1286	to decode or encode properly, so it was impossible to prepare a fair
604	comparison table for that case.	1287	comparison table for that case.
605		1288
606		1289
607	=head1 SECURITY CONSIDERATIONS	1290	=head1 SECURITY CONSIDERATIONS
…		…
613	any buffer overflows. Obviously, this module should ensure that and I am	1296	any buffer overflows. Obviously, this module should ensure that and I am
614	trying hard on making that true, but you never know.	1297	trying hard on making that true, but you never know.
615		1298
616	Second, you need to avoid resource-starving attacks. That means you should	1299	Second, you need to avoid resource-starving attacks. That means you should
617	limit the size of JSON texts you accept, or make sure then when your	1300	limit the size of JSON texts you accept, or make sure then when your
618	resources run out, thats just fine (e.g. by using a separate process that	1301	resources run out, that's just fine (e.g. by using a separate process that
619	can crash safely). The size of a JSON text in octets or characters is	1302	can crash safely). The size of a JSON text in octets or characters is
620	usually a good indication of the size of the resources required to decode	1303	usually a good indication of the size of the resources required to decode
621	it into a Perl structure.	1304	it into a Perl structure. While JSON::XS can check the size of the JSON
		1305	text, it might be too late when you already have it in memory, so you
		1306	might want to check the size before you accept the string.
622		1307
623	Third, JSON::XS recurses using the C stack when decoding objects and	1308	Third, JSON::XS recurses using the C stack when decoding objects and
624	arrays. The C stack is a limited resource: for instance, on my amd64	1309	arrays. The C stack is a limited resource: for instance, on my amd64
625	machine with 8MB of stack size I can decode around 180k nested arrays	1310	machine with 8MB of stack size I can decode around 180k nested arrays but
626	but only 14k nested JSON objects. If that is exceeded, the program	1311	only 14k nested JSON objects (due to perl itself recursing deeply on croak
		1312	to free the temporary). If that is exceeded, the program crashes. To be
627	crashes. Thats why the default nesting limit is set to 8192. If your	1313	conservative, the default nesting limit is set to 512. If your process
628	process has a smaller stack, you should adjust this setting accordingly	1314	has a smaller stack, you should adjust this setting accordingly with the
629	with the C<max_depth> method.	1315	C<max_depth> method.
630		1316
631	And last but least, something else could bomb you that I forgot to think	1317	Something else could bomb you, too, that I forgot to think of. In that
632	of. In that case, you get to keep the pieces. I am alway sopen for hints,	1318	case, you get to keep the pieces. I am always open for hints, though...
633	though...	1319
		1320	Also keep in mind that JSON::XS might leak contents of your Perl data
		1321	structures in its error messages, so when you serialise sensitive
		1322	information you might want to make sure that exceptions thrown by JSON::XS
		1323	will not end up in front of untrusted eyes.
		1324
		1325	If you are using JSON::XS to return packets to consumption
		1326	by JavaScript scripts in a browser you should have a look at
		1327	L<http://jpsykes.com/47/practical-csrf-and-json-security> to see whether
		1328	you are vulnerable to some common attack vectors (which really are browser
		1329	design bugs, but it is still you who will have to deal with it, as major
		1330	browser developers care only for features, not about getting security
		1331	right).
		1332
		1333
		1334	=head1 THREADS
		1335
		1336	This module is I<not> guaranteed to be thread safe and there are no
		1337	plans to change this until Perl gets thread support (as opposed to the
		1338	horribly slow so-called "threads" which are simply slow and bloated
		1339	process simulations - use fork, it's I<much> faster, cheaper, better).
		1340
		1341	(It might actually work, but you have been warned).
634		1342
635		1343
636	=head1 BUGS	1344	=head1 BUGS
637		1345
638	While the goal of this module is to be correct, that unfortunately does	1346	While the goal of this module is to be correct, that unfortunately does
639	not mean its bug-free, only that I think its design is bug-free. It is	1347	not mean it's bug-free, only that I think its design is bug-free. If you
640	still relatively early in its development. If you keep reporting bugs they	1348	keep reporting bugs they will be fixed swiftly, though.
641	will be fixed swiftly, though.	1349
		1350	Please refrain from using rt.cpan.org or any other bug reporting
		1351	service. I put the contact address into my modules for a reason.
642		1352
643	=cut	1353	=cut
644		1354
		1355	our $true = do { bless \(my $dummy = 1), "JSON::XS::Boolean" };
		1356	our $false = do { bless \(my $dummy = 0), "JSON::XS::Boolean" };
		1357
		1358	sub true() { $true }
		1359	sub false() { $false }
		1360
		1361	sub is_bool($) {
		1362	UNIVERSAL::isa $_[0], "JSON::XS::Boolean"
		1363	# or UNIVERSAL::isa $_[0], "JSON::Literal"
		1364	}
		1365
		1366	XSLoader::load "JSON::XS", $VERSION;
		1367
		1368	package JSON::XS::Boolean;
		1369
		1370	use overload
		1371	"0+" => sub { ${$_[0]} },
		1372	"++" => sub { $_[0] = ${$_[0]} + 1 },
		1373	"--" => sub { $_[0] = ${$_[0]} - 1 },
		1374	fallback => 1;
		1375
645	1;	1376	1;
		1377
		1378	=head1 SEE ALSO
		1379
		1380	The F<json_xs> command line utility for quick experiments.
646		1381
647	=head1 AUTHOR	1382	=head1 AUTHOR
648		1383
649	Marc Lehmann <schmorp@schmorp.de>	1384	Marc Lehmann <schmorp@schmorp.de>
650	http://home.schmorp.de/	1385	http://home.schmorp.de/

Diff Legend

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