ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/JSON-XS/XS.pm

Comparing JSON-XS/XS.pm (file contents):
Revision 1.1 by root, Thu Mar 22 16:40:16 2007 UTC vs.
Revision 1.143 by root, Fri Oct 25 20:02:54 2013 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		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)
		9
5	=head1 SYNOPSIS	10	=head1 SYNOPSIS
6		11
7	use JSON::XS;	12	use JSON::XS;
8		13
		14	# exported functions, they croak on error
		15	# and expect/generate UTF-8
		16
		17	$utf8_encoded_json_text = encode_json $perl_hash_or_arrayref;
		18	$perl_hash_or_arrayref = decode_json $utf8_encoded_json_text;
		19
		20	# OO-interface
		21
		22	$coder = JSON::XS->new->ascii->pretty->allow_nonref;
		23	$pretty_printed_unencoded = $coder->encode ($perl_scalar);
		24	$perl_scalar = $coder->decode ($unicode_json_text);
		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
9	=head1 DESCRIPTION	34	=head1 DESCRIPTION
10		35
		36	This module converts Perl data structures to JSON and vice versa. Its
		37	primary goal is to be I<correct> and its secondary goal is to be
		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	overridden) with no overhead due to emulation (by inheriting 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.
		47
		48	As this is the n-th-something JSON module on CPAN, what was the reason
		49	to write yet another JSON module? While it seems there are many JSON
		50	modules, none of them correctly handle all corner cases, and in most cases
		51	their maintainers are unresponsive, gone missing, or not listening to bug
		52	reports for other reasons.
		53
		54	See MAPPING, below, on how JSON::XS maps perl values to JSON values and
		55	vice versa.
		56
		57	=head2 FEATURES
		58
11	=over 4	59	=over 4
12		60
		61	=item * correct Unicode handling
		62
		63	This module knows how to handle Unicode, documents how and when it does
		64	so, and even documents what "correct" means.
		65
		66	=item * round-trip integrity
		67
		68	When you serialise a perl data structure using only data types supported
		69	by JSON and Perl, the deserialised data structure is identical on the Perl
		70	level. (e.g. the string "2.0" doesn't suddenly become "2" just because
		71	it looks like a number). There I<are> minor exceptions to this, read the
		72	MAPPING section below to learn about those.
		73
		74	=item * strict checking of JSON correctness
		75
		76	There is no guessing, no generating of illegal JSON texts by default,
		77	and only JSON is accepted as input by default (the latter is a security
		78	feature).
		79
		80	=item * fast
		81
		82	Compared to other JSON modules and other serialisers such as Storable,
		83	this module usually compares favourably in terms of speed, too.
		84
		85	=item * simple to use
		86
		87	This module has both a simple functional interface as well as an object
		88	oriented interface.
		89
		90	=item * reasonably versatile output formats
		91
		92	You can choose between the most compact guaranteed-single-line format
		93	possible (nice for simple line-based protocols), a pure-ASCII format
		94	(for when your transport is not 8-bit clean, still supports the whole
		95	Unicode range), or a pretty-printed format (for when you want to read that
		96	stuff). Or you can combine those features in whatever way you like.
		97
		98	=back
		99
13	=cut	100	=cut
14		101
15	package JSON::XS;	102	package JSON::XS;
16		103
17	BEGIN {	104	use common::sense;
18	$VERSION = '0.1';	105
		106	our $VERSION = 2.34;
19	@ISA = qw(Exporter);	107	our @ISA = qw(Exporter);
20		108
21	require Exporter;	109	our @EXPORT = qw(encode_json decode_json);
22		110
23	require XSLoader;	111	use Exporter;
24	XSLoader::load JSON::XS::, $VERSION;	112	use XSLoader;
		113
		114	=head1 FUNCTIONAL INTERFACE
		115
		116	The following convenience methods are provided by this module. They are
		117	exported by default:
		118
		119	=over 4
		120
		121	=item $json_text = encode_json $perl_scalar
		122
		123	Converts the given Perl data structure to a UTF-8 encoded, binary string
		124	(that is, the string contains octets only). Croaks on error.
		125
		126	This function call is functionally identical to:
		127
		128	$json_text = JSON::XS->new->utf8->encode ($perl_scalar)
		129
		130	Except being faster.
		131
		132	=item $perl_scalar = decode_json $json_text
		133
		134	The opposite of C<encode_json>: expects an UTF-8 (binary) string and tries
		135	to parse that as an UTF-8 encoded JSON text, returning the resulting
		136	reference. Croaks on error.
		137
		138	This function call is functionally identical to:
		139
		140	$perl_scalar = JSON::XS->new->utf8->decode ($json_text)
		141
		142	Except being faster.
		143
		144	=item $is_boolean = JSON::XS::is_bool $scalar
		145
		146	Returns true if the passed scalar represents either JSON::XS::true or
		147	JSON::XS::false, two constants that act like C<1> and C<0>, respectively
		148	and are used to represent JSON C<true> and C<false> values in Perl.
		149
		150	See MAPPING, below, for more information on how JSON values are mapped to
		151	Perl.
		152
		153	=back
		154
		155
		156	=head1 A FEW NOTES ON UNICODE AND PERL
		157
		158	Since this often leads to confusion, here are a few very clear words on
		159	how Unicode works in Perl, modulo bugs.
		160
		161	=over 4
		162
		163	=item 1. Perl strings can store characters with ordinal values > 255.
		164
		165	This enables you to store Unicode characters as single characters in a
		166	Perl string - very natural.
		167
		168	=item 2. Perl does I<not> associate an encoding with your strings.
		169
		170	... until you force it to, e.g. when matching it against a regex, or
		171	printing the scalar to a file, in which case Perl either interprets your
		172	string as locale-encoded text, octets/binary, or as Unicode, depending
		173	on various settings. In no case is an encoding stored together with your
		174	data, it is I<use> that decides encoding, not any magical meta data.
		175
		176	=item 3. The internal utf-8 flag has no meaning with regards to the
		177	encoding of your string.
		178
		179	Just ignore that flag unless you debug a Perl bug, a module written in
		180	XS or want to dive into the internals of perl. Otherwise it will only
		181	confuse you, as, despite the name, it says nothing about how your string
		182	is encoded. You can have Unicode strings with that flag set, with that
		183	flag clear, and you can have binary data with that flag set and that flag
		184	clear. Other possibilities exist, too.
		185
		186	If you didn't know about that flag, just the better, pretend it doesn't
		187	exist.
		188
		189	=item 4. A "Unicode String" is simply a string where each character can be
		190	validly interpreted as a Unicode code point.
		191
		192	If you have UTF-8 encoded data, it is no longer a Unicode string, but a
		193	Unicode string encoded in UTF-8, giving you a binary string.
		194
		195	=item 5. A string containing "high" (> 255) character values is I<not> a UTF-8 string.
		196
		197	It's a fact. Learn to live with it.
		198
		199	=back
		200
		201	I hope this helps :)
		202
		203
		204	=head1 OBJECT-ORIENTED INTERFACE
		205
		206	The object oriented interface lets you configure your own encoding or
		207	decoding style, within the limits of supported formats.
		208
		209	=over 4
		210
		211	=item $json = new JSON::XS
		212
		213	Creates a new JSON::XS object that can be used to de/encode JSON
		214	strings. All boolean flags described below are by default I<disabled>.
		215
		216	The mutators for flags all return the JSON object again and thus calls can
		217	be chained:
		218
		219	my $json = JSON::XS->new->utf8->space_after->encode ({a => [1,2]})
		220	=> {"a": [1, 2]}
		221
		222	=item $json = $json->ascii ([$enable])
		223
		224	=item $enabled = $json->get_ascii
		225
		226	If C<$enable> is true (or missing), then the C<encode> method will not
		227	generate characters outside the code range C<0..127> (which is ASCII). Any
		228	Unicode characters outside that range will be escaped using either a
		229	single \uXXXX (BMP characters) or a double \uHHHH\uLLLLL escape sequence,
		230	as per RFC4627. The resulting encoded JSON text can be treated as a native
		231	Unicode string, an ascii-encoded, latin1-encoded or UTF-8 encoded string,
		232	or any other superset of ASCII.
		233
		234	If C<$enable> is false, then the C<encode> method will not escape Unicode
		235	characters unless required by the JSON syntax or other flags. This results
		236	in a faster and more compact format.
		237
		238	See also the section I<ENCODING/CODESET FLAG NOTES> later in this
		239	document.
		240
		241	The main use for this flag is to produce JSON texts that can be
		242	transmitted over a 7-bit channel, as the encoded JSON texts will not
		243	contain any 8 bit characters.
		244
		245	JSON::XS->new->ascii (1)->encode ([chr 0x10401])
		246	=> ["\ud801\udc01"]
		247
		248	=item $json = $json->latin1 ([$enable])
		249
		250	=item $enabled = $json->get_latin1
		251
		252	If C<$enable> is true (or missing), then the C<encode> method will encode
		253	the resulting JSON text as latin1 (or iso-8859-1), escaping any characters
		254	outside the code range C<0..255>. The resulting string can be treated as a
		255	latin1-encoded JSON text or a native Unicode string. The C<decode> method
		256	will not be affected in any way by this flag, as C<decode> by default
		257	expects Unicode, which is a strict superset of latin1.
		258
		259	If C<$enable> is false, then the C<encode> method will not escape Unicode
		260	characters unless required by the JSON syntax or other flags.
		261
		262	See also the section I<ENCODING/CODESET FLAG NOTES> later in this
		263	document.
		264
		265	The main use for this flag is efficiently encoding binary data as JSON
		266	text, as most octets will not be escaped, resulting in a smaller encoded
		267	size. The disadvantage is that the resulting JSON text is encoded
		268	in latin1 (and must correctly be treated as such when storing and
		269	transferring), a rare encoding for JSON. It is therefore most useful when
		270	you want to store data structures known to contain binary data efficiently
		271	in files or databases, not when talking to other JSON encoders/decoders.
		272
		273	JSON::XS->new->latin1->encode (["\x{89}\x{abc}"]
		274	=> ["\x{89}\\u0abc"] # (perl syntax, U+abc escaped, U+89 not)
		275
		276	=item $json = $json->utf8 ([$enable])
		277
		278	=item $enabled = $json->get_utf8
		279
		280	If C<$enable> is true (or missing), then the C<encode> method will encode
		281	the JSON result into UTF-8, as required by many protocols, while the
		282	C<decode> method expects to be handled an UTF-8-encoded string. Please
		283	note that UTF-8-encoded strings do not contain any characters outside the
		284	range C<0..255>, they are thus useful for bytewise/binary I/O. In future
		285	versions, enabling this option might enable autodetection of the UTF-16
		286	and UTF-32 encoding families, as described in RFC4627.
		287
		288	If C<$enable> is false, then the C<encode> method will return the JSON
		289	string as a (non-encoded) Unicode string, while C<decode> expects thus a
		290	Unicode string. Any decoding or encoding (e.g. to UTF-8 or UTF-16) needs
		291	to be done yourself, e.g. using the Encode module.
		292
		293	See also the section I<ENCODING/CODESET FLAG NOTES> later in this
		294	document.
		295
		296	Example, output UTF-16BE-encoded JSON:
		297
		298	use Encode;
		299	$jsontext = encode "UTF-16BE", JSON::XS->new->encode ($object);
		300
		301	Example, decode UTF-32LE-encoded JSON:
		302
		303	use Encode;
		304	$object = JSON::XS->new->decode (decode "UTF-32LE", $jsontext);
		305
		306	=item $json = $json->pretty ([$enable])
		307
		308	This enables (or disables) all of the C<indent>, C<space_before> and
		309	C<space_after> (and in the future possibly more) flags in one call to
		310	generate the most readable (or most compact) form possible.
		311
		312	Example, pretty-print some simple structure:
		313
		314	my $json = JSON::XS->new->pretty(1)->encode ({a => [1,2]})
		315	=>
		316	{
		317	"a" : [
		318	1,
		319	2
		320	]
		321	}
		322
		323	=item $json = $json->indent ([$enable])
		324
		325	=item $enabled = $json->get_indent
		326
		327	If C<$enable> is true (or missing), then the C<encode> method will use a multiline
		328	format as output, putting every array member or object/hash key-value pair
		329	into its own line, indenting them properly.
		330
		331	If C<$enable> is false, no newlines or indenting will be produced, and the
		332	resulting JSON text is guaranteed not to contain any C<newlines>.
		333
		334	This setting has no effect when decoding JSON texts.
		335
		336	=item $json = $json->space_before ([$enable])
		337
		338	=item $enabled = $json->get_space_before
		339
		340	If C<$enable> is true (or missing), then the C<encode> method will add an extra
		341	optional space before the C<:> separating keys from values in JSON objects.
		342
		343	If C<$enable> is false, then the C<encode> method will not add any extra
		344	space at those places.
		345
		346	This setting has no effect when decoding JSON texts. You will also
		347	most likely combine this setting with C<space_after>.
		348
		349	Example, space_before enabled, space_after and indent disabled:
		350
		351	{"key" :"value"}
		352
		353	=item $json = $json->space_after ([$enable])
		354
		355	=item $enabled = $json->get_space_after
		356
		357	If C<$enable> is true (or missing), then the C<encode> method will add an extra
		358	optional space after the C<:> separating keys from values in JSON objects
		359	and extra whitespace after the C<,> separating key-value pairs and array
		360	members.
		361
		362	If C<$enable> is false, then the C<encode> method will not add any extra
		363	space at those places.
		364
		365	This setting has no effect when decoding JSON texts.
		366
		367	Example, space_before and indent disabled, space_after enabled:
		368
		369	{"key": "value"}
		370
		371	=item $json = $json->relaxed ([$enable])
		372
		373	=item $enabled = $json->get_relaxed
		374
		375	If C<$enable> is true (or missing), then C<decode> will accept some
		376	extensions to normal JSON syntax (see below). C<encode> will not be
		377	affected in anyway. I<Be aware that this option makes you accept invalid
		378	JSON texts as if they were valid!>. I suggest only to use this option to
		379	parse application-specific files written by humans (configuration files,
		380	resource files etc.)
		381
		382	If C<$enable> is false (the default), then C<decode> will only accept
		383	valid JSON texts.
		384
		385	Currently accepted extensions are:
		386
		387	=over 4
		388
		389	=item * list items can have an end-comma
		390
		391	JSON I<separates> array elements and key-value pairs with commas. This
		392	can be annoying if you write JSON texts manually and want to be able to
		393	quickly append elements, so this extension accepts comma at the end of
		394	such items not just between them:
		395
		396	[
		397	1,
		398	2, <- this comma not normally allowed
		399	]
		400	{
		401	"k1": "v1",
		402	"k2": "v2", <- this comma not normally allowed
		403	}
		404
		405	=item * shell-style '#'-comments
		406
		407	Whenever JSON allows whitespace, shell-style comments are additionally
		408	allowed. They are terminated by the first carriage-return or line-feed
		409	character, after which more white-space and comments are allowed.
		410
		411	[
		412	1, # this comment not allowed in JSON
		413	# neither this one...
		414	]
		415
		416	=back
		417
		418	=item $json = $json->canonical ([$enable])
		419
		420	=item $enabled = $json->get_canonical
		421
		422	If C<$enable> is true (or missing), then the C<encode> method will output JSON objects
		423	by sorting their keys. This is adding a comparatively high overhead.
		424
		425	If C<$enable> is false, then the C<encode> method will output key-value
		426	pairs in the order Perl stores them (which will likely change between runs
		427	of the same script, and can change even within the same run from 5.18
		428	onwards).
		429
		430	This option is useful if you want the same data structure to be encoded as
		431	the same JSON text (given the same overall settings). If it is disabled,
		432	the same hash might be encoded differently even if contains the same data,
		433	as key-value pairs have no inherent ordering in Perl.
		434
		435	This setting has no effect when decoding JSON texts.
		436
		437	This setting has currently no effect on tied hashes.
		438
		439	=item $json = $json->allow_nonref ([$enable])
		440
		441	=item $enabled = $json->get_allow_nonref
		442
		443	If C<$enable> is true (or missing), then the C<encode> method can convert a
		444	non-reference into its corresponding string, number or null JSON value,
		445	which is an extension to RFC4627. Likewise, C<decode> will accept those JSON
		446	values instead of croaking.
		447
		448	If C<$enable> is false, then the C<encode> method will croak if it isn't
		449	passed an arrayref or hashref, as JSON texts must either be an object
		450	or array. Likewise, C<decode> will croak if given something that is not a
		451	JSON object or array.
		452
		453	Example, encode a Perl scalar as JSON value with enabled C<allow_nonref>,
		454	resulting in an invalid JSON text:
		455
		456	JSON::XS->new->allow_nonref->encode ("Hello, World!")
		457	=> "Hello, World!"
		458
		459	=item $json = $json->allow_unknown ([$enable])
		460
		461	=item $enabled = $json->get_allow_unknown
		462
		463	If C<$enable> is true (or missing), then C<encode> will I<not> throw an
		464	exception when it encounters values it cannot represent in JSON (for
		465	example, filehandles) but instead will encode a JSON C<null> value. Note
		466	that blessed objects are not included here and are handled separately by
		467	c<allow_nonref>.
		468
		469	If C<$enable> is false (the default), then C<encode> will throw an
		470	exception when it encounters anything it cannot encode as JSON.
		471
		472	This option does not affect C<decode> in any way, and it is recommended to
		473	leave it off unless you know your communications partner.
		474
		475	=item $json = $json->allow_blessed ([$enable])
		476
		477	=item $enabled = $json->get_allow_blessed
		478
		479	If C<$enable> is true (or missing), then the C<encode> method will not
		480	barf when it encounters a blessed reference. Instead, the value of the
		481	B<convert_blessed> option will decide whether C<null> (C<convert_blessed>
		482	disabled or no C<TO_JSON> method found) or a representation of the
		483	object (C<convert_blessed> enabled and C<TO_JSON> method found) is being
		484	encoded. Has no effect on C<decode>.
		485
		486	If C<$enable> is false (the default), then C<encode> will throw an
		487	exception when it encounters a blessed object.
		488
		489	=item $json = $json->convert_blessed ([$enable])
		490
		491	=item $enabled = $json->get_convert_blessed
		492
		493	If C<$enable> is true (or missing), then C<encode>, upon encountering a
		494	blessed object, will check for the availability of the C<TO_JSON> method
		495	on the object's class. If found, it will be called in scalar context
		496	and the resulting scalar will be encoded instead of the object. If no
		497	C<TO_JSON> method is found, the value of C<allow_blessed> will decide what
		498	to do.
		499
		500	The C<TO_JSON> method may safely call die if it wants. If C<TO_JSON>
		501	returns other blessed objects, those will be handled in the same
		502	way. C<TO_JSON> must take care of not causing an endless recursion cycle
		503	(== crash) in this case. The name of C<TO_JSON> was chosen because other
		504	methods called by the Perl core (== not by the user of the object) are
		505	usually in upper case letters and to avoid collisions with any C<to_json>
		506	function or method.
		507
		508	This setting does not yet influence C<decode> in any way, but in the
		509	future, global hooks might get installed that influence C<decode> and are
		510	enabled by this setting.
		511
		512	If C<$enable> is false, then the C<allow_blessed> setting will decide what
		513	to do when a blessed object is found.
		514
		515	=item $json = $json->filter_json_object ([$coderef->($hashref)])
		516
		517	When C<$coderef> is specified, it will be called from C<decode> each
		518	time it decodes a JSON object. The only argument is a reference to the
		519	newly-created hash. If the code references returns a single scalar (which
		520	need not be a reference), this value (i.e. a copy of that scalar to avoid
		521	aliasing) is inserted into the deserialised data structure. If it returns
		522	an empty list (NOTE: I<not> C<undef>, which is a valid scalar), the
		523	original deserialised hash will be inserted. This setting can slow down
		524	decoding considerably.
		525
		526	When C<$coderef> is omitted or undefined, any existing callback will
		527	be removed and C<decode> will not change the deserialised hash in any
		528	way.
		529
		530	Example, convert all JSON objects into the integer 5:
		531
		532	my $js = JSON::XS->new->filter_json_object (sub { 5 });
		533	# returns [5]
		534	$js->decode ('[{}]')
		535	# throw an exception because allow_nonref is not enabled
		536	# so a lone 5 is not allowed.
		537	$js->decode ('{"a":1, "b":2}');
		538
		539	=item $json = $json->filter_json_single_key_object ($key [=> $coderef->($value)])
		540
		541	Works remotely similar to C<filter_json_object>, but is only called for
		542	JSON objects having a single key named C<$key>.
		543
		544	This C<$coderef> is called before the one specified via
		545	C<filter_json_object>, if any. It gets passed the single value in the JSON
		546	object. If it returns a single value, it will be inserted into the data
		547	structure. If it returns nothing (not even C<undef> but the empty list),
		548	the callback from C<filter_json_object> will be called next, as if no
		549	single-key callback were specified.
		550
		551	If C<$coderef> is omitted or undefined, the corresponding callback will be
		552	disabled. There can only ever be one callback for a given key.
		553
		554	As this callback gets called less often then the C<filter_json_object>
		555	one, decoding speed will not usually suffer as much. Therefore, single-key
		556	objects make excellent targets to serialise Perl objects into, especially
		557	as single-key JSON objects are as close to the type-tagged value concept
		558	as JSON gets (it's basically an ID/VALUE tuple). Of course, JSON does not
		559	support this in any way, so you need to make sure your data never looks
		560	like a serialised Perl hash.
		561
		562	Typical names for the single object key are C<__class_whatever__>, or
		563	C<$__dollars_are_rarely_used__$> or C<}ugly_brace_placement>, or even
		564	things like C<__class_md5sum(classname)__>, to reduce the risk of clashing
		565	with real hashes.
		566
		567	Example, decode JSON objects of the form C<< { "__widget__" => <id> } >>
		568	into the corresponding C<< $WIDGET{<id>} >> object:
		569
		570	# return whatever is in $WIDGET{5}:
		571	JSON::XS
		572	->new
		573	->filter_json_single_key_object (__widget__ => sub {
		574	$WIDGET{ $_[0] }
		575	})
		576	->decode ('{"__widget__": 5')
		577
		578	# this can be used with a TO_JSON method in some "widget" class
		579	# for serialisation to json:
		580	sub WidgetBase::TO_JSON {
		581	my ($self) = @_;
		582
		583	unless ($self->{id}) {
		584	$self->{id} = ..get..some..id..;
		585	$WIDGET{$self->{id}} = $self;
		586	}
		587
		588	{ __widget__ => $self->{id} }
		589	}
		590
		591	=item $json = $json->shrink ([$enable])
		592
		593	=item $enabled = $json->get_shrink
		594
		595	Perl usually over-allocates memory a bit when allocating space for
		596	strings. This flag optionally resizes strings generated by either
		597	C<encode> or C<decode> to their minimum size possible. This can save
		598	memory when your JSON texts are either very very long or you have many
		599	short strings. It will also try to downgrade any strings to octet-form
		600	if possible: perl stores strings internally either in an encoding called
		601	UTF-X or in octet-form. The latter cannot store everything but uses less
		602	space in general (and some buggy Perl or C code might even rely on that
		603	internal representation being used).
		604
		605	The actual definition of what shrink does might change in future versions,
		606	but it will always try to save space at the expense of time.
		607
		608	If C<$enable> is true (or missing), the string returned by C<encode> will
		609	be shrunk-to-fit, while all strings generated by C<decode> will also be
		610	shrunk-to-fit.
		611
		612	If C<$enable> is false, then the normal perl allocation algorithms are used.
		613	If you work with your data, then this is likely to be faster.
		614
		615	In the future, this setting might control other things, such as converting
		616	strings that look like integers or floats into integers or floats
		617	internally (there is no difference on the Perl level), saving space.
		618
		619	=item $json = $json->max_depth ([$maximum_nesting_depth])
		620
		621	=item $max_depth = $json->get_max_depth
		622
		623	Sets the maximum nesting level (default C<512>) accepted while encoding
		624	or decoding. If a higher nesting level is detected in JSON text or a Perl
		625	data structure, then the encoder and decoder will stop and croak at that
		626	point.
		627
		628	Nesting level is defined by number of hash- or arrayrefs that the encoder
		629	needs to traverse to reach a given point or the number of C<{> or C<[>
		630	characters without their matching closing parenthesis crossed to reach a
		631	given character in a string.
		632
		633	Setting the maximum depth to one disallows any nesting, so that ensures
		634	that the object is only a single hash/object or array.
		635
		636	If no argument is given, the highest possible setting will be used, which
		637	is rarely useful.
		638
		639	Note that nesting is implemented by recursion in C. The default value has
		640	been chosen to be as large as typical operating systems allow without
		641	crashing.
		642
		643	See SECURITY CONSIDERATIONS, below, for more info on why this is useful.
		644
		645	=item $json = $json->max_size ([$maximum_string_size])
		646
		647	=item $max_size = $json->get_max_size
		648
		649	Set the maximum length a JSON text may have (in bytes) where decoding is
		650	being attempted. The default is C<0>, meaning no limit. When C<decode>
		651	is called on a string that is longer then this many bytes, it will not
		652	attempt to decode the string but throw an exception. This setting has no
		653	effect on C<encode> (yet).
		654
		655	If no argument is given, the limit check will be deactivated (same as when
		656	C<0> is specified).
		657
		658	See SECURITY CONSIDERATIONS, below, for more info on why this is useful.
		659
		660	=item $json_text = $json->encode ($perl_scalar)
		661
		662	Converts the given Perl value or data structure to its JSON
		663	representation. Croaks on error.
		664
		665	=item $perl_scalar = $json->decode ($json_text)
		666
		667	The opposite of C<encode>: expects a JSON text and tries to parse it,
		668	returning the resulting simple scalar or reference. Croaks on error.
		669
		670	=item ($perl_scalar, $characters) = $json->decode_prefix ($json_text)
		671
		672	This works like the C<decode> method, but instead of raising an exception
		673	when there is trailing garbage after the first JSON object, it will
		674	silently stop parsing there and return the number of characters consumed
		675	so far.
		676
		677	This is useful if your JSON texts are not delimited by an outer protocol
		678	and you need to know where the JSON text ends.
		679
		680	JSON::XS->new->decode_prefix ("[1] the tail")
		681	=> ([], 3)
		682
		683	=back
		684
		685
		686	=head1 INCREMENTAL PARSING
		687
		688	In some cases, there is the need for incremental parsing of JSON
		689	texts. While this module always has to keep both JSON text and resulting
		690	Perl data structure in memory at one time, it does allow you to parse a
		691	JSON stream incrementally. It does so by accumulating text until it has
		692	a full JSON object, which it then can decode. This process is similar to
		693	using C<decode_prefix> to see if a full JSON object is available, but
		694	is much more efficient (and can be implemented with a minimum of method
		695	calls).
		696
		697	JSON::XS will only attempt to parse the JSON text once it is sure it
		698	has enough text to get a decisive result, using a very simple but
		699	truly incremental parser. This means that it sometimes won't stop as
		700	early as the full parser, for example, it doesn't detect mismatched
		701	parentheses. The only thing it guarantees is that it starts decoding as
		702	soon as a syntactically valid JSON text has been seen. This means you need
		703	to set resource limits (e.g. C<max_size>) to ensure the parser will stop
		704	parsing in the presence if syntax errors.
		705
		706	The following methods implement this incremental parser.
		707
		708	=over 4
		709
		710	=item [void, scalar or list context] = $json->incr_parse ([$string])
		711
		712	This is the central parsing function. It can both append new text and
		713	extract objects from the stream accumulated so far (both of these
		714	functions are optional).
		715
		716	If C<$string> is given, then this string is appended to the already
		717	existing JSON fragment stored in the C<$json> object.
		718
		719	After that, if the function is called in void context, it will simply
		720	return without doing anything further. This can be used to add more text
		721	in as many chunks as you want.
		722
		723	If the method is called in scalar context, then it will try to extract
		724	exactly I<one> JSON object. If that is successful, it will return this
		725	object, otherwise it will return C<undef>. If there is a parse error,
		726	this method will croak just as C<decode> would do (one can then use
		727	C<incr_skip> to skip the erroneous part). This is the most common way of
		728	using the method.
		729
		730	And finally, in list context, it will try to extract as many objects
		731	from the stream as it can find and return them, or the empty list
		732	otherwise. For this to work, there must be no separators between the JSON
		733	objects or arrays, instead they must be concatenated back-to-back. If
		734	an error occurs, an exception will be raised as in the scalar context
		735	case. Note that in this case, any previously-parsed JSON texts will be
		736	lost.
		737
		738	Example: Parse some JSON arrays/objects in a given string and return
		739	them.
		740
		741	my @objs = JSON::XS->new->incr_parse ("[5][7][1,2]");
		742
		743	=item $lvalue_string = $json->incr_text
		744
		745	This method returns the currently stored JSON fragment as an lvalue, that
		746	is, you can manipulate it. This I<only> works when a preceding call to
		747	C<incr_parse> in I<scalar context> successfully returned an object. Under
		748	all other circumstances you must not call this function (I mean it.
		749	although in simple tests it might actually work, it I<will> fail under
		750	real world conditions). As a special exception, you can also call this
		751	method before having parsed anything.
		752
		753	This function is useful in two cases: a) finding the trailing text after a
		754	JSON object or b) parsing multiple JSON objects separated by non-JSON text
		755	(such as commas).
		756
		757	=item $json->incr_skip
		758
		759	This will reset the state of the incremental parser and will remove
		760	the parsed text from the input buffer so far. This is useful after
		761	C<incr_parse> died, in which case the input buffer and incremental parser
		762	state is left unchanged, to skip the text parsed so far and to reset the
		763	parse state.
		764
		765	The difference to C<incr_reset> is that only text until the parse error
		766	occurred is removed.
		767
		768	=item $json->incr_reset
		769
		770	This completely resets the incremental parser, that is, after this call,
		771	it will be as if the parser had never parsed anything.
		772
		773	This is useful if you want to repeatedly parse JSON objects and want to
		774	ignore any trailing data, which means you have to reset the parser after
		775	each successful decode.
		776
		777	=back
		778
		779	=head2 LIMITATIONS
		780
		781	All options that affect decoding are supported, except
		782	C<allow_nonref>. The reason for this is that it cannot be made to work
		783	sensibly: JSON objects and arrays are self-delimited, i.e. you can
		784	concatenate them back to back and still decode them perfectly. This does
		785	not hold true for JSON numbers, however.
		786
		787	For example, is the string C<1> a single JSON number, or is it simply the
		788	start of C<12>? Or is C<12> a single JSON number, or the concatenation
		789	of C<1> and C<2>? In neither case you can tell, and this is why JSON::XS
		790	takes the conservative route and disallows this case.
		791
		792	=head2 EXAMPLES
		793
		794	Some examples will make all this clearer. First, a simple example that
		795	works similarly to C<decode_prefix>: We want to decode the JSON object at
		796	the start of a string and identify the portion after the JSON object:
		797
		798	my $text = "[1,2,3] hello";
		799
		800	my $json = new JSON::XS;
		801
		802	my $obj = $json->incr_parse ($text)
		803	or die "expected JSON object or array at beginning of string";
		804
		805	my $tail = $json->incr_text;
		806	# $tail now contains " hello"
		807
		808	Easy, isn't it?
		809
		810	Now for a more complicated example: Imagine a hypothetical protocol where
		811	you read some requests from a TCP stream, and each request is a JSON
		812	array, without any separation between them (in fact, it is often useful to
		813	use newlines as "separators", as these get interpreted as whitespace at
		814	the start of the JSON text, which makes it possible to test said protocol
		815	with C<telnet>...).
		816
		817	Here is how you'd do it (it is trivial to write this in an event-based
		818	manner):
		819
		820	my $json = new JSON::XS;
		821
		822	# read some data from the socket
		823	while (sysread $socket, my $buf, 4096) {
		824
		825	# split and decode as many requests as possible
		826	for my $request ($json->incr_parse ($buf)) {
		827	# act on the $request
		828	}
		829	}
		830
		831	Another complicated example: Assume you have a string with JSON objects
		832	or arrays, all separated by (optional) comma characters (e.g. C<[1],[2],
		833	[3]>). To parse them, we have to skip the commas between the JSON texts,
		834	and here is where the lvalue-ness of C<incr_text> comes in useful:
		835
		836	my $text = "[1],[2], [3]";
		837	my $json = new JSON::XS;
		838
		839	# void context, so no parsing done
		840	$json->incr_parse ($text);
		841
		842	# now extract as many objects as possible. note the
		843	# use of scalar context so incr_text can be called.
		844	while (my $obj = $json->incr_parse) {
		845	# do something with $obj
		846
		847	# now skip the optional comma
		848	$json->incr_text =~ s/^ \s* , //x;
		849	}
		850
		851	Now lets go for a very complex example: Assume that you have a gigantic
		852	JSON array-of-objects, many gigabytes in size, and you want to parse it,
		853	but you cannot load it into memory fully (this has actually happened in
		854	the real world :).
		855
		856	Well, you lost, you have to implement your own JSON parser. But JSON::XS
		857	can still help you: You implement a (very simple) array parser and let
		858	JSON decode the array elements, which are all full JSON objects on their
		859	own (this wouldn't work if the array elements could be JSON numbers, for
		860	example):
		861
		862	my $json = new JSON::XS;
		863
		864	# open the monster
		865	open my $fh, "<bigfile.json"
		866	or die "bigfile: $!";
		867
		868	# first parse the initial "["
		869	for (;;) {
		870	sysread $fh, my $buf, 65536
		871	or die "read error: $!";
		872	$json->incr_parse ($buf); # void context, so no parsing
		873
		874	# Exit the loop once we found and removed(!) the initial "[".
		875	# In essence, we are (ab-)using the $json object as a simple scalar
		876	# we append data to.
		877	last if $json->incr_text =~ s/^ \s* \[ //x;
		878	}
		879
		880	# now we have the skipped the initial "[", so continue
		881	# parsing all the elements.
		882	for (;;) {
		883	# in this loop we read data until we got a single JSON object
		884	for (;;) {
		885	if (my $obj = $json->incr_parse) {
		886	# do something with $obj
		887	last;
		888	}
		889
		890	# add more data
		891	sysread $fh, my $buf, 65536
		892	or die "read error: $!";
		893	$json->incr_parse ($buf); # void context, so no parsing
		894	}
		895
		896	# in this loop we read data until we either found and parsed the
		897	# separating "," between elements, or the final "]"
		898	for (;;) {
		899	# first skip whitespace
		900	$json->incr_text =~ s/^\s*//;
		901
		902	# if we find "]", we are done
		903	if ($json->incr_text =~ s/^\]//) {
		904	print "finished.\n";
		905	exit;
		906	}
		907
		908	# if we find ",", we can continue with the next element
		909	if ($json->incr_text =~ s/^,//) {
		910	last;
		911	}
		912
		913	# if we find anything else, we have a parse error!
		914	if (length $json->incr_text) {
		915	die "parse error near ", $json->incr_text;
		916	}
		917
		918	# else add more data
		919	sysread $fh, my $buf, 65536
		920	or die "read error: $!";
		921	$json->incr_parse ($buf); # void context, so no parsing
		922	}
		923
		924	This is a complex example, but most of the complexity comes from the fact
		925	that we are trying to be correct (bear with me if I am wrong, I never ran
		926	the above example :).
		927
		928
		929
		930	=head1 MAPPING
		931
		932	This section describes how JSON::XS maps Perl values to JSON values and
		933	vice versa. These mappings are designed to "do the right thing" in most
		934	circumstances automatically, preserving round-tripping characteristics
		935	(what you put in comes out as something equivalent).
		936
		937	For the more enlightened: note that in the following descriptions,
		938	lowercase I<perl> refers to the Perl interpreter, while uppercase I<Perl>
		939	refers to the abstract Perl language itself.
		940
		941
		942	=head2 JSON -> PERL
		943
		944	=over 4
		945
		946	=item object
		947
		948	A JSON object becomes a reference to a hash in Perl. No ordering of object
		949	keys is preserved (JSON does not preserve object key ordering itself).
		950
		951	=item array
		952
		953	A JSON array becomes a reference to an array in Perl.
		954
		955	=item string
		956
		957	A JSON string becomes a string scalar in Perl - Unicode codepoints in JSON
		958	are represented by the same codepoints in the Perl string, so no manual
		959	decoding is necessary.
		960
		961	=item number
		962
		963	A JSON number becomes either an integer, numeric (floating point) or
		964	string scalar in perl, depending on its range and any fractional parts. On
		965	the Perl level, there is no difference between those as Perl handles all
		966	the conversion details, but an integer may take slightly less memory and
		967	might represent more values exactly than floating point numbers.
		968
		969	If the number consists of digits only, JSON::XS will try to represent
		970	it as an integer value. If that fails, it will try to represent it as
		971	a numeric (floating point) value if that is possible without loss of
		972	precision. Otherwise it will preserve the number as a string value (in
		973	which case you lose roundtripping ability, as the JSON number will be
		974	re-encoded to a JSON string).
		975
		976	Numbers containing a fractional or exponential part will always be
		977	represented as numeric (floating point) values, possibly at a loss of
		978	precision (in which case you might lose perfect roundtripping ability, but
		979	the JSON number will still be re-encoded as a JSON number).
		980
		981	Note that precision is not accuracy - binary floating point values cannot
		982	represent most decimal fractions exactly, and when converting from and to
		983	floating point, JSON::XS only guarantees precision up to but not including
		984	the least significant bit.
		985
		986	=item true, false
		987
		988	These JSON atoms become C<JSON::XS::true> and C<JSON::XS::false>,
		989	respectively. They are overloaded to act almost exactly like the numbers
		990	C<1> and C<0>. You can check whether a scalar is a JSON boolean by using
		991	the C<JSON::XS::is_bool> function.
		992
		993	=item null
		994
		995	A JSON null atom becomes C<undef> in Perl.
		996
		997	=back
		998
		999
		1000	=head2 PERL -> JSON
		1001
		1002	The mapping from Perl to JSON is slightly more difficult, as Perl is a
		1003	truly typeless language, so we can only guess which JSON type is meant by
		1004	a Perl value.
		1005
		1006	=over 4
		1007
		1008	=item hash references
		1009
		1010	Perl hash references become JSON objects. As there is no inherent
		1011	ordering in hash keys (or JSON objects), they will usually be encoded
		1012	in a pseudo-random order. JSON::XS can optionally sort the hash keys
		1013	(determined by the I<canonical> flag), so the same datastructure will
		1014	serialise to the same JSON text (given same settings and version of
		1015	JSON::XS), but this incurs a runtime overhead and is only rarely useful,
		1016	e.g. when you want to compare some JSON text against another for equality.
		1017
		1018	=item array references
		1019
		1020	Perl array references become JSON arrays.
		1021
		1022	=item other references
		1023
		1024	Other unblessed references are generally not allowed and will cause an
		1025	exception to be thrown, except for references to the integers C<0> and
		1026	C<1>, which get turned into C<false> and C<true> atoms in JSON. You can
		1027	also use C<JSON::XS::false> and C<JSON::XS::true> to improve readability.
		1028
		1029	encode_json [\0, JSON::XS::true] # yields [false,true]
		1030
		1031	=item JSON::XS::true, JSON::XS::false
		1032
		1033	These special values become JSON true and JSON false values,
		1034	respectively. You can also use C<\1> and C<\0> directly if you want.
		1035
		1036	=item blessed objects
		1037
		1038	Blessed objects are not directly representable in JSON. See the
		1039	C<allow_blessed> and C<convert_blessed> methods on various options on
		1040	how to deal with this: basically, you can choose between throwing an
		1041	exception, encoding the reference as if it weren't blessed, or provide
		1042	your own serialiser method.
		1043
		1044	=item simple scalars
		1045
		1046	Simple Perl scalars (any scalar that is not a reference) are the most
		1047	difficult objects to encode: JSON::XS will encode undefined scalars as
		1048	JSON C<null> values, scalars that have last been used in a string context
		1049	before encoding as JSON strings, and anything else as number value:
		1050
		1051	# dump as number
		1052	encode_json [2] # yields [2]
		1053	encode_json [-3.0e17] # yields [-3e+17]
		1054	my $value = 5; encode_json [$value] # yields [5]
		1055
		1056	# used as string, so dump as string
		1057	print $value;
		1058	encode_json [$value] # yields ["5"]
		1059
		1060	# undef becomes null
		1061	encode_json [undef] # yields [null]
		1062
		1063	You can force the type to be a JSON string by stringifying it:
		1064
		1065	my $x = 3.1; # some variable containing a number
		1066	"$x"; # stringified
		1067	$x .= ""; # another, more awkward way to stringify
		1068	print $x; # perl does it for you, too, quite often
		1069
		1070	You can force the type to be a JSON number by numifying it:
		1071
		1072	my $x = "3"; # some variable containing a string
		1073	$x += 0; # numify it, ensuring it will be dumped as a number
		1074	$x *= 1; # same thing, the choice is yours.
		1075
		1076	You can not currently force the type in other, less obscure, ways. Tell me
		1077	if you need this capability (but don't forget to explain why it's needed
		1078	:).
		1079
		1080	Note that numerical precision has the same meaning as under Perl (so
		1081	binary to decimal conversion follows the same rules as in Perl, which
		1082	can differ to other languages). Also, your perl interpreter might expose
		1083	extensions to the floating point numbers of your platform, such as
		1084	infinities or NaN's - these cannot be represented in JSON, and it is an
		1085	error to pass those in.
		1086
		1087	=back
		1088
		1089
		1090	=head1 ENCODING/CODESET FLAG NOTES
		1091
		1092	The interested reader might have seen a number of flags that signify
		1093	encodings or codesets - C<utf8>, C<latin1> and C<ascii>. There seems to be
		1094	some confusion on what these do, so here is a short comparison:
		1095
		1096	C<utf8> controls whether the JSON text created by C<encode> (and expected
		1097	by C<decode>) is UTF-8 encoded or not, while C<latin1> and C<ascii> only
		1098	control whether C<encode> escapes character values outside their respective
		1099	codeset range. Neither of these flags conflict with each other, although
		1100	some combinations make less sense than others.
		1101
		1102	Care has been taken to make all flags symmetrical with respect to
		1103	C<encode> and C<decode>, that is, texts encoded with any combination of
		1104	these flag values will be correctly decoded when the same flags are used
		1105	- in general, if you use different flag settings while encoding vs. when
		1106	decoding you likely have a bug somewhere.
		1107
		1108	Below comes a verbose discussion of these flags. Note that a "codeset" is
		1109	simply an abstract set of character-codepoint pairs, while an encoding
		1110	takes those codepoint numbers and I<encodes> them, in our case into
		1111	octets. Unicode is (among other things) a codeset, UTF-8 is an encoding,
		1112	and ISO-8859-1 (= latin 1) and ASCII are both codesets I<and> encodings at
		1113	the same time, which can be confusing.
		1114
		1115	=over 4
		1116
		1117	=item C<utf8> flag disabled
		1118
		1119	When C<utf8> is disabled (the default), then C<encode>/C<decode> generate
		1120	and expect Unicode strings, that is, characters with high ordinal Unicode
		1121	values (> 255) will be encoded as such characters, and likewise such
		1122	characters are decoded as-is, no changes to them will be done, except
		1123	"(re-)interpreting" them as Unicode codepoints or Unicode characters,
		1124	respectively (to Perl, these are the same thing in strings unless you do
		1125	funny/weird/dumb stuff).
		1126
		1127	This is useful when you want to do the encoding yourself (e.g. when you
		1128	want to have UTF-16 encoded JSON texts) or when some other layer does
		1129	the encoding for you (for example, when printing to a terminal using a
		1130	filehandle that transparently encodes to UTF-8 you certainly do NOT want
		1131	to UTF-8 encode your data first and have Perl encode it another time).
		1132
		1133	=item C<utf8> flag enabled
		1134
		1135	If the C<utf8>-flag is enabled, C<encode>/C<decode> will encode all
		1136	characters using the corresponding UTF-8 multi-byte sequence, and will
		1137	expect your input strings to be encoded as UTF-8, that is, no "character"
		1138	of the input string must have any value > 255, as UTF-8 does not allow
		1139	that.
		1140
		1141	The C<utf8> flag therefore switches between two modes: disabled means you
		1142	will get a Unicode string in Perl, enabled means you get an UTF-8 encoded
		1143	octet/binary string in Perl.
		1144
		1145	=item C<latin1> or C<ascii> flags enabled
		1146
		1147	With C<latin1> (or C<ascii>) enabled, C<encode> will escape characters
		1148	with ordinal values > 255 (> 127 with C<ascii>) and encode the remaining
		1149	characters as specified by the C<utf8> flag.
		1150
		1151	If C<utf8> is disabled, then the result is also correctly encoded in those
		1152	character sets (as both are proper subsets of Unicode, meaning that a
		1153	Unicode string with all character values < 256 is the same thing as a
		1154	ISO-8859-1 string, and a Unicode string with all character values < 128 is
		1155	the same thing as an ASCII string in Perl).
		1156
		1157	If C<utf8> is enabled, you still get a correct UTF-8-encoded string,
		1158	regardless of these flags, just some more characters will be escaped using
		1159	C<\uXXXX> then before.
		1160
		1161	Note that ISO-8859-1-I<encoded> strings are not compatible with UTF-8
		1162	encoding, while ASCII-encoded strings are. That is because the ISO-8859-1
		1163	encoding is NOT a subset of UTF-8 (despite the ISO-8859-1 I<codeset> being
		1164	a subset of Unicode), while ASCII is.
		1165
		1166	Surprisingly, C<decode> will ignore these flags and so treat all input
		1167	values as governed by the C<utf8> flag. If it is disabled, this allows you
		1168	to decode ISO-8859-1- and ASCII-encoded strings, as both strict subsets of
		1169	Unicode. If it is enabled, you can correctly decode UTF-8 encoded strings.
		1170
		1171	So neither C<latin1> nor C<ascii> are incompatible with the C<utf8> flag -
		1172	they only govern when the JSON output engine escapes a character or not.
		1173
		1174	The main use for C<latin1> is to relatively efficiently store binary data
		1175	as JSON, at the expense of breaking compatibility with most JSON decoders.
		1176
		1177	The main use for C<ascii> is to force the output to not contain characters
		1178	with values > 127, which means you can interpret the resulting string
		1179	as UTF-8, ISO-8859-1, ASCII, KOI8-R or most about any character set and
		1180	8-bit-encoding, and still get the same data structure back. This is useful
		1181	when your channel for JSON transfer is not 8-bit clean or the encoding
		1182	might be mangled in between (e.g. in mail), and works because ASCII is a
		1183	proper subset of most 8-bit and multibyte encodings in use in the world.
		1184
		1185	=back
		1186
		1187
		1188	=head2 JSON and ECMAscript
		1189
		1190	JSON syntax is based on how literals are represented in javascript (the
		1191	not-standardised predecessor of ECMAscript) which is presumably why it is
		1192	called "JavaScript Object Notation".
		1193
		1194	However, JSON is not a subset (and also not a superset of course) of
		1195	ECMAscript (the standard) or javascript (whatever browsers actually
		1196	implement).
		1197
		1198	If you want to use javascript's C<eval> function to "parse" JSON, you
		1199	might run into parse errors for valid JSON texts, or the resulting data
		1200	structure might not be queryable:
		1201
		1202	One of the problems is that U+2028 and U+2029 are valid characters inside
		1203	JSON strings, but are not allowed in ECMAscript string literals, so the
		1204	following Perl fragment will not output something that can be guaranteed
		1205	to be parsable by javascript's C<eval>:
		1206
		1207	use JSON::XS;
		1208
		1209	print encode_json [chr 0x2028];
		1210
		1211	The right fix for this is to use a proper JSON parser in your javascript
		1212	programs, and not rely on C<eval> (see for example Douglas Crockford's
		1213	F<json2.js> parser).
		1214
		1215	If this is not an option, you can, as a stop-gap measure, simply encode to
		1216	ASCII-only JSON:
		1217
		1218	use JSON::XS;
		1219
		1220	print JSON::XS->new->ascii->encode ([chr 0x2028]);
		1221
		1222	Note that this will enlarge the resulting JSON text quite a bit if you
		1223	have many non-ASCII characters. You might be tempted to run some regexes
		1224	to only escape U+2028 and U+2029, e.g.:
		1225
		1226	# DO NOT USE THIS!
		1227	my $json = JSON::XS->new->utf8->encode ([chr 0x2028]);
		1228	$json =~ s/\xe2\x80\xa8/\\u2028/g; # escape U+2028
		1229	$json =~ s/\xe2\x80\xa9/\\u2029/g; # escape U+2029
		1230	print $json;
		1231
		1232	Note that I<this is a bad idea>: the above only works for U+2028 and
		1233	U+2029 and thus only for fully ECMAscript-compliant parsers. Many existing
		1234	javascript implementations, however, have issues with other characters as
		1235	well - using C<eval> naively simply I<will> cause problems.
		1236
		1237	Another problem is that some javascript implementations reserve
		1238	some property names for their own purposes (which probably makes
		1239	them non-ECMAscript-compliant). For example, Iceweasel reserves the
		1240	C<__proto__> property name for its own purposes.
		1241
		1242	If that is a problem, you could parse try to filter the resulting JSON
		1243	output for these property strings, e.g.:
		1244
		1245	$json =~ s/"__proto__"\s*:/"__proto__renamed":/g;
		1246
		1247	This works because C<__proto__> is not valid outside of strings, so every
		1248	occurrence of C<"__proto__"\s*:> must be a string used as property name.
		1249
		1250	If you know of other incompatibilities, please let me know.
		1251
		1252
		1253	=head2 JSON and YAML
		1254
		1255	You often hear that JSON is a subset of YAML. This is, however, a mass
		1256	hysteria(*) and very far from the truth (as of the time of this writing),
		1257	so let me state it clearly: I<in general, there is no way to configure
		1258	JSON::XS to output a data structure as valid YAML> that works in all
		1259	cases.
		1260
		1261	If you really must use JSON::XS to generate YAML, you should use this
		1262	algorithm (subject to change in future versions):
		1263
		1264	my $to_yaml = JSON::XS->new->utf8->space_after (1);
		1265	my $yaml = $to_yaml->encode ($ref) . "\n";
		1266
		1267	This will I<usually> generate JSON texts that also parse as valid
		1268	YAML. Please note that YAML has hardcoded limits on (simple) object key
		1269	lengths that JSON doesn't have and also has different and incompatible
		1270	unicode character escape syntax, so you should make sure that your hash
		1271	keys are noticeably shorter than the 1024 "stream characters" YAML allows
		1272	and that you do not have characters with codepoint values outside the
		1273	Unicode BMP (basic multilingual page). YAML also does not allow C<\/>
		1274	sequences in strings (which JSON::XS does not I<currently> generate, but
		1275	other JSON generators might).
		1276
		1277	There might be other incompatibilities that I am not aware of (or the YAML
		1278	specification has been changed yet again - it does so quite often). In
		1279	general you should not try to generate YAML with a JSON generator or vice
		1280	versa, or try to parse JSON with a YAML parser or vice versa: chances are
		1281	high that you will run into severe interoperability problems when you
		1282	least expect it.
		1283
		1284	=over 4
		1285
		1286	=item (*)
		1287
		1288	I have been pressured multiple times by Brian Ingerson (one of the
		1289	authors of the YAML specification) to remove this paragraph, despite him
		1290	acknowledging that the actual incompatibilities exist. As I was personally
		1291	bitten by this "JSON is YAML" lie, I refused and said I will continue to
		1292	educate people about these issues, so others do not run into the same
		1293	problem again and again. After this, Brian called me a (quote)I<complete
		1294	and worthless idiot>(unquote).
		1295
		1296	In my opinion, instead of pressuring and insulting people who actually
		1297	clarify issues with YAML and the wrong statements of some of its
		1298	proponents, I would kindly suggest reading the JSON spec (which is not
		1299	that difficult or long) and finally make YAML compatible to it, and
		1300	educating users about the changes, instead of spreading lies about the
		1301	real compatibility for many I<years> and trying to silence people who
		1302	point out that it isn't true.
		1303
		1304	Addendum/2009: the YAML 1.2 spec is still incompatible with JSON, even
		1305	though the incompatibilities have been documented (and are known to Brian)
		1306	for many years and the spec makes explicit claims that YAML is a superset
		1307	of JSON. It would be so easy to fix, but apparently, bullying people and
		1308	corrupting userdata is so much easier.
		1309
		1310	=back
		1311
		1312
		1313	=head2 SPEED
		1314
		1315	It seems that JSON::XS is surprisingly fast, as shown in the following
		1316	tables. They have been generated with the help of the C<eg/bench> program
		1317	in the JSON::XS distribution, to make it easy to compare on your own
		1318	system.
		1319
		1320	First comes a comparison between various modules using
		1321	a very short single-line JSON string (also available at
		1322	L<http://dist.schmorp.de/misc/json/short.json>).
		1323
		1324	{"method": "handleMessage", "params": ["user1",
		1325	"we were just talking"], "id": null, "array":[1,11,234,-5,1e5,1e7,
		1326	1, 0]}
		1327
		1328	It shows the number of encodes/decodes per second (JSON::XS uses
		1329	the functional interface, while JSON::XS/2 uses the OO interface
		1330	with pretty-printing and hashkey sorting enabled, JSON::XS/3 enables
		1331	shrink. JSON::DWIW/DS uses the deserialise function, while JSON::DWIW::FJ
		1332	uses the from_json method). Higher is better:
		1333
		1334	module | encode | decode |
		1335	--------------|------------|------------|
		1336	JSON::DWIW/DS | 86302.551 | 102300.098 |
		1337	JSON::DWIW/FJ | 86302.551 | 75983.768 |
		1338	JSON::PP | 15827.562 | 6638.658 |
		1339	JSON::Syck | 63358.066 | 47662.545 |
		1340	JSON::XS | 511500.488 | 511500.488 |
		1341	JSON::XS/2 | 291271.111 | 388361.481 |
		1342	JSON::XS/3 | 361577.931 | 361577.931 |
		1343	Storable | 66788.280 | 265462.278 |
		1344	--------------+------------+------------+
		1345
		1346	That is, JSON::XS is almost six times faster than JSON::DWIW on encoding,
		1347	about five times faster on decoding, and over thirty to seventy times
		1348	faster than JSON's pure perl implementation. It also compares favourably
		1349	to Storable for small amounts of data.
		1350
		1351	Using a longer test string (roughly 18KB, generated from Yahoo! Locals
		1352	search API (L<http://dist.schmorp.de/misc/json/long.json>).
		1353
		1354	module | encode | decode |
		1355	--------------|------------|------------|
		1356	JSON::DWIW/DS | 1647.927 | 2673.916 |
		1357	JSON::DWIW/FJ | 1630.249 | 2596.128 |
		1358	JSON::PP | 400.640 | 62.311 |
		1359	JSON::Syck | 1481.040 | 1524.869 |
		1360	JSON::XS | 20661.596 | 9541.183 |
		1361	JSON::XS/2 | 10683.403 | 9416.938 |
		1362	JSON::XS/3 | 20661.596 | 9400.054 |
		1363	Storable | 19765.806 | 10000.725 |
		1364	--------------+------------+------------+
		1365
		1366	Again, JSON::XS leads by far (except for Storable which non-surprisingly
		1367	decodes a bit faster).
		1368
		1369	On large strings containing lots of high Unicode characters, some modules
		1370	(such as JSON::PC) seem to decode faster than JSON::XS, but the result
		1371	will be broken due to missing (or wrong) Unicode handling. Others refuse
		1372	to decode or encode properly, so it was impossible to prepare a fair
		1373	comparison table for that case.
		1374
		1375
		1376	=head1 SECURITY CONSIDERATIONS
		1377
		1378	When you are using JSON in a protocol, talking to untrusted potentially
		1379	hostile creatures requires relatively few measures.
		1380
		1381	First of all, your JSON decoder should be secure, that is, should not have
		1382	any buffer overflows. Obviously, this module should ensure that and I am
		1383	trying hard on making that true, but you never know.
		1384
		1385	Second, you need to avoid resource-starving attacks. That means you should
		1386	limit the size of JSON texts you accept, or make sure then when your
		1387	resources run out, that's just fine (e.g. by using a separate process that
		1388	can crash safely). The size of a JSON text in octets or characters is
		1389	usually a good indication of the size of the resources required to decode
		1390	it into a Perl structure. While JSON::XS can check the size of the JSON
		1391	text, it might be too late when you already have it in memory, so you
		1392	might want to check the size before you accept the string.
		1393
		1394	Third, JSON::XS recurses using the C stack when decoding objects and
		1395	arrays. The C stack is a limited resource: for instance, on my amd64
		1396	machine with 8MB of stack size I can decode around 180k nested arrays but
		1397	only 14k nested JSON objects (due to perl itself recursing deeply on croak
		1398	to free the temporary). If that is exceeded, the program crashes. To be
		1399	conservative, the default nesting limit is set to 512. If your process
		1400	has a smaller stack, you should adjust this setting accordingly with the
		1401	C<max_depth> method.
		1402
		1403	Something else could bomb you, too, that I forgot to think of. In that
		1404	case, you get to keep the pieces. I am always open for hints, though...
		1405
		1406	Also keep in mind that JSON::XS might leak contents of your Perl data
		1407	structures in its error messages, so when you serialise sensitive
		1408	information you might want to make sure that exceptions thrown by JSON::XS
		1409	will not end up in front of untrusted eyes.
		1410
		1411	If you are using JSON::XS to return packets to consumption
		1412	by JavaScript scripts in a browser you should have a look at
		1413	L<http://blog.archive.jpsykes.com/47/practical-csrf-and-json-security/> to
		1414	see whether you are vulnerable to some common attack vectors (which really
		1415	are browser design bugs, but it is still you who will have to deal with
		1416	it, as major browser developers care only for features, not about getting
		1417	security right).
		1418
		1419
		1420	=head1 THREADS
		1421
		1422	This module is I<not> guaranteed to be thread safe and there are no
		1423	plans to change this until Perl gets thread support (as opposed to the
		1424	horribly slow so-called "threads" which are simply slow and bloated
		1425	process simulations - use fork, it's I<much> faster, cheaper, better).
		1426
		1427	(It might actually work, but you have been warned).
		1428
		1429
		1430	=head1 THE PERILS OF SETLOCALE
		1431
		1432	Sometimes people avoid the Perl locale support and directly call the
		1433	system's setlocale function with C<LC_ALL>.
		1434
		1435	This breaks both perl and modules such as JSON::XS, as stringification of
		1436	numbers no longer works correctly (e.g. C<$x = 0.1; print "$x"+1> might
		1437	print C<1>, and JSON::XS might output illegal JSON as JSON::XS relies on
		1438	perl to stringify numbers).
		1439
		1440	The solution is simple: don't call C<setlocale>, or use it for only those
		1441	categories you need, such as C<LC_MESSAGES> or C<LC_CTYPE>.
		1442
		1443	If you need C<LC_NUMERIC>, you should enable it only around the code that
		1444	actually needs it (avoiding stringification of numbers), and restore it
		1445	afterwards.
		1446
		1447
		1448	=head1 BUGS
		1449
		1450	While the goal of this module is to be correct, that unfortunately does
		1451	not mean it's bug-free, only that I think its design is bug-free. If you
		1452	keep reporting bugs they will be fixed swiftly, though.
		1453
		1454	Please refrain from using rt.cpan.org or any other bug reporting
		1455	service. I put the contact address into my modules for a reason.
		1456
		1457	=cut
		1458
		1459	our $true = do { bless \(my $dummy = 1), "JSON::XS::Boolean" };
		1460	our $false = do { bless \(my $dummy = 0), "JSON::XS::Boolean" };
		1461
		1462	sub true() { $true }
		1463	sub false() { $false }
		1464
		1465	sub is_bool($) {
		1466	UNIVERSAL::isa $_[0], "JSON::XS::Boolean"
		1467	# or UNIVERSAL::isa $_[0], "JSON::Literal"
25	}	1468	}
26		1469
27	=item	1470	XSLoader::load "JSON::XS", $VERSION;
28		1471
29	=cut	1472	package JSON::XS::Boolean;
30		1473
31	use JSON::DWIW;	1474	use overload
32	use Benchmark;	1475	"0+" => sub { ${$_[0]} },
33		1476	"++" => sub { $_[0] = ${$_[0]} + 1 },
34	use utf8;	1477	"--" => sub { $_[0] = ${$_[0]} - 1 },
35	#my $json = '{"ü":1,"a":[1,{"3":4},2],"b":5,"üü":2}';	1478	fallback => 1;
36	my $json = '{"test":9555555555555555555,"hu" : -1e+5, "arr" : [ 1,2,3,4,5]}';
37
38	my $js = JSON::XS->new;
39	warn $js->indent (0);
40	warn $js->canonical (0);
41	warn $js->ascii (0);
42	warn $js->space_after (0);
43	use Data::Dumper;
44	warn Dumper $js->decode ($json);
45	warn Dumper $js->encode ($js->decode ($json));
46	#my $x = {"üü" => 2, "ü" => 1, "a" => [1,{3,4},2], b => 5};
47
48	#my $js2 = JSON::DWIW->new;
49	#
50	#timethese 200000, {
51	# a => sub { $js->encode ($x) },
52	# b => sub { $js2->to_json ($x) },
53	#};
54		1479
55	1;	1480	1;
56		1481
57	=back	1482	=head1 SEE ALSO
		1483
		1484	The F<json_xs> command line utility for quick experiments.
58		1485
59	=head1 AUTHOR	1486	=head1 AUTHOR
60		1487
61	Marc Lehmann <schmorp@schmorp.de>	1488	Marc Lehmann <schmorp@schmorp.de>
62	http://home.schmorp.de/	1489	http://home.schmorp.de/

Diff Legend

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