[ViewVC] Diff of: cvs/JSON-XS/XS.pm

Comparing JSON-XS/XS.pm (file contents):
Revision 1.75 by root, Thu Nov 29 13:35:35 2007 UTC vs.
Revision 1.99 by root, Thu Mar 27 06:37:35 2008 UTC

…		…
1	=head1 NAME	1	=head1 NAME
		2
		3	=encoding utf-8
2		4
3	JSON::XS - JSON serialising/deserialising, done correctly and fast	5	JSON::XS - JSON serialising/deserialising, done correctly and fast
4		6
5	JSON::XS - 正しくて高速な JSON シリアライザ/デシリアライザ	7	JSON::XS - 正しくて高速な JSON シリアライザ/デシリアライザ
6	(http://fleur.hio.jp/perldoc/mix/lib/JSON/XS.html)	8	(http://fleur.hio.jp/perldoc/mix/lib/JSON/XS.html)
10	use JSON::XS;	12	use JSON::XS;
11		13
12	# exported functions, they croak on error	14	# exported functions, they croak on error
13	# and expect/generate UTF-8	15	# and expect/generate UTF-8
14		16
15	$utf8_encoded_json_text = to_json $perl_hash_or_arrayref;	17	$utf8_encoded_json_text = encode_json $perl_hash_or_arrayref;
16	$perl_hash_or_arrayref = from_json $utf8_encoded_json_text;	18	$perl_hash_or_arrayref = decode_json $utf8_encoded_json_text;
17		19
18	# OO-interface	20	# OO-interface
19		21
20	$coder = JSON::XS->new->ascii->pretty->allow_nonref;	22	$coder = JSON::XS->new->ascii->pretty->allow_nonref;
21	$pretty_printed_unencoded = $coder->encode ($perl_scalar);	23	$pretty_printed_unencoded = $coder->encode ($perl_scalar);
22	$perl_scalar = $coder->decode ($unicode_json_text);	24	$perl_scalar = $coder->decode ($unicode_json_text);
23		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
24	=head1 DESCRIPTION	34	=head1 DESCRIPTION
25		35
26	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
27	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
28	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.
29		47
30	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
31	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
32	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
33	their maintainers are unresponsive, gone missing, or not listening to bug	51	their maintainers are unresponsive, gone missing, or not listening to bug
…		…
42		60
43	=over 4	61	=over 4
44		62
45	=item * correct Unicode handling	63	=item * correct Unicode handling
46		64
47	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
48	it does so.	66	so, and even documents what "correct" means.
49		67
50	=item * round-trip integrity	68	=item * round-trip integrity
51		69
52	When you serialise a perl data structure using only datatypes supported	70	When you serialise a perl data structure using only datatypes supported
53	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.
54	(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
55	like a number).	73	like a number). There minor I<are> exceptions to this, read the MAPPING
		74	section below to learn about those.
56		75
57	=item * strict checking of JSON correctness	76	=item * strict checking of JSON correctness
58		77
59	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,
60	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
61	feature).	80	feature).
62		81
63	=item * fast	82	=item * fast
64		83
65	Compared to other JSON modules, this module compares favourably in terms	84	Compared to other JSON modules and other serialisers such as Storable,
66	of speed, too.	85	this module usually compares favourably in terms of speed, too.
67		86
68	=item * simple to use	87	=item * simple to use
69		88
70	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
71	interface.	90	oriented interface interface.
72		91
73	=item * reasonably versatile output formats	92	=item * reasonably versatile output formats
74		93
75	You can choose between the most compact guaranteed single-line format	94	You can choose between the most compact guaranteed-single-line format
76	possible (nice for simple line-based protocols), a pure-ascii format	95	possible (nice for simple line-based protocols), a pure-ascii format
77	(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
78	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
79	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.
80		99
…		…
84		103
85	package JSON::XS;	104	package JSON::XS;
86		105
87	use strict;	106	use strict;
88		107
89	our $VERSION = '2.0';	108	our $VERSION = '2.2';
90	our @ISA = qw(Exporter);	109	our @ISA = qw(Exporter);
91		110
92	our @EXPORT = qw(to_json from_json);	111	our @EXPORT = qw(encode_json decode_json to_json from_json);
		112
		113	sub to_json($) {
		114	require Carp;
		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");
		116	}
		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	}
93		122
94	use Exporter;	123	use Exporter;
95	use XSLoader;	124	use XSLoader;
96		125
97	=head1 FUNCTIONAL INTERFACE	126	=head1 FUNCTIONAL INTERFACE
…		…
99	The following convenience methods are provided by this module. They are	128	The following convenience methods are provided by this module. They are
100	exported by default:	129	exported by default:
101		130
102	=over 4	131	=over 4
103		132
104	=item $json_text = to_json $perl_scalar	133	=item $json_text = encode_json $perl_scalar
105		134
106	Converts the given Perl data structure to a UTF-8 encoded, binary string	135	Converts the given Perl data structure to a UTF-8 encoded, binary string
107	(that is, the string contains octets only). Croaks on error.	136	(that is, the string contains octets only). Croaks on error.
108		137
109	This function call is functionally identical to:	138	This function call is functionally identical to:
110		139
111	$json_text = JSON::XS->new->utf8->encode ($perl_scalar)	140	$json_text = JSON::XS->new->utf8->encode ($perl_scalar)
112		141
113	except being faster.	142	except being faster.
114		143
115	=item $perl_scalar = from_json $json_text	144	=item $perl_scalar = decode_json $json_text
116		145
117	The opposite of C<to_json>: expects an UTF-8 (binary) string and tries	146	The opposite of C<encode_json>: expects an UTF-8 (binary) string and tries
118	to parse that as an UTF-8 encoded JSON text, returning the resulting	147	to parse that as an UTF-8 encoded JSON text, returning the resulting
119	reference. Croaks on error.	148	reference. Croaks on error.
120		149
121	This function call is functionally identical to:	150	This function call is functionally identical to:
122		151
…		…
148	This enables you to store Unicode characters as single characters in a	177	This enables you to store Unicode characters as single characters in a
149	Perl string - very natural.	178	Perl string - very natural.
150		179
151	=item 2. Perl does I<not> associate an encoding with your strings.	180	=item 2. Perl does I<not> associate an encoding with your strings.
152		181
153	Unless you force it to, e.g. when matching it against a regex, or printing	182	... until you force it to, e.g. when matching it against a regex, or
154	the scalar to a file, in which case Perl either interprets your string as	183	printing the scalar to a file, in which case Perl either interprets your
155	locale-encoded text, octets/binary, or as Unicode, depending on various	184	string as locale-encoded text, octets/binary, or as Unicode, depending
156	settings. In no case is an encoding stored together with your data, it is	185	on various settings. In no case is an encoding stored together with your
157	I<use> that decides encoding, not any magical metadata.	186	data, it is I<use> that decides encoding, not any magical meta data.
158		187
159	=item 3. The internal utf-8 flag has no meaning with regards to the	188	=item 3. The internal utf-8 flag has no meaning with regards to the
160	encoding of your string.	189	encoding of your string.
161		190
162	Just ignore that flag unless you debug a Perl bug, a module written in	191	Just ignore that flag unless you debug a Perl bug, a module written in
…		…
216		245
217	If C<$enable> is false, then the C<encode> method will not escape Unicode	246	If C<$enable> is false, then the C<encode> method will not escape Unicode
218	characters unless required by the JSON syntax or other flags. This results	247	characters unless required by the JSON syntax or other flags. This results
219	in a faster and more compact format.	248	in a faster and more compact format.
220		249
		250	See also the section I<ENCODING/CODESET FLAG NOTES> later in this
		251	document.
		252
221	The main use for this flag is to produce JSON texts that can be	253	The main use for this flag is to produce JSON texts that can be
222	transmitted over a 7-bit channel, as the encoded JSON texts will not	254	transmitted over a 7-bit channel, as the encoded JSON texts will not
223	contain any 8 bit characters.	255	contain any 8 bit characters.
224		256
225	JSON::XS->new->ascii (1)->encode ([chr 0x10401])	257	JSON::XS->new->ascii (1)->encode ([chr 0x10401])
…		…
236	will not be affected in any way by this flag, as C<decode> by default	268	will not be affected in any way by this flag, as C<decode> by default
237	expects Unicode, which is a strict superset of latin1.	269	expects Unicode, which is a strict superset of latin1.
238		270
239	If C<$enable> is false, then the C<encode> method will not escape Unicode	271	If C<$enable> is false, then the C<encode> method will not escape Unicode
240	characters unless required by the JSON syntax or other flags.	272	characters unless required by the JSON syntax or other flags.
		273
		274	See also the section I<ENCODING/CODESET FLAG NOTES> later in this
		275	document.
241		276
242	The main use for this flag is efficiently encoding binary data as JSON	277	The main use for this flag is efficiently encoding binary data as JSON
243	text, as most octets will not be escaped, resulting in a smaller encoded	278	text, as most octets will not be escaped, resulting in a smaller encoded
244	size. The disadvantage is that the resulting JSON text is encoded	279	size. The disadvantage is that the resulting JSON text is encoded
245	in latin1 (and must correctly be treated as such when storing and	280	in latin1 (and must correctly be treated as such when storing and
…		…
264		299
265	If C<$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
266	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
267	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
268	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.
269		307
270	Example, output UTF-16BE-encoded JSON:	308	Example, output UTF-16BE-encoded JSON:
271		309
272	use Encode;	310	use Encode;
273	$jsontext = encode "UTF-16BE", JSON::XS->new->encode ($object);	311	$jsontext = encode "UTF-16BE", JSON::XS->new->encode ($object);
…		…
425	resulting in an invalid JSON text:	463	resulting in an invalid JSON text:
426		464
427	JSON::XS->new->allow_nonref->encode ("Hello, World!")	465	JSON::XS->new->allow_nonref->encode ("Hello, World!")
428	=> "Hello, World!"	466	=> "Hello, World!"
429		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
430	=item $json = $json->allow_blessed ([$enable])	484	=item $json = $json->allow_blessed ([$enable])
431		485
432	=item $enabled = $json->get_allow_blessed	486	=item $enabled = $json->get_allow_blessed
433		487
434	If C<$enable> is true (or missing), then the C<encode> method will not	488	If C<$enable> is true (or missing), then the C<encode> method will not
435	barf when it encounters a blessed reference. Instead, the value of the	489	barf when it encounters a blessed reference. Instead, the value of the
436	B<convert_blessed> option will decide whether C<null> (C<convert_blessed>	490	B<convert_blessed> option will decide whether C<null> (C<convert_blessed>
437	disabled or no C<to_json> method found) or a representation of the	491	disabled or no C<TO_JSON> method found) or a representation of the
438	object (C<convert_blessed> enabled and C<to_json> method found) is being	492	object (C<convert_blessed> enabled and C<TO_JSON> method found) is being
439	encoded. Has no effect on C<decode>.	493	encoded. Has no effect on C<decode>.
440		494
441	If C<$enable> is false (the default), then C<encode> will throw an	495	If C<$enable> is false (the default), then C<encode> will throw an
442	exception when it encounters a blessed object.	496	exception when it encounters a blessed object.
443		497
…		…
455	The C<TO_JSON> method may safely call die if it wants. If C<TO_JSON>	509	The C<TO_JSON> method may safely call die if it wants. If C<TO_JSON>
456	returns other blessed objects, those will be handled in the same	510	returns other blessed objects, those will be handled in the same
457	way. C<TO_JSON> must take care of not causing an endless recursion cycle	511	way. C<TO_JSON> must take care of not causing an endless recursion cycle
458	(== crash) in this case. The name of C<TO_JSON> was chosen because other	512	(== crash) in this case. The name of C<TO_JSON> was chosen because other
459	methods called by the Perl core (== not by the user of the object) are	513	methods called by the Perl core (== not by the user of the object) are
460	usually in upper case letters and to avoid collisions with the C<to_json>	514	usually in upper case letters and to avoid collisions with any C<to_json>
461	function.	515	function or method.
462		516
463	This setting does not yet influence C<decode> in any way, but in the	517	This setting does not yet influence C<decode> in any way, but in the
464	future, global hooks might get installed that influence C<decode> and are	518	future, global hooks might get installed that influence C<decode> and are
465	enabled by this setting.	519	enabled by this setting.
466		520
…		…
643	=> ([], 3)	697	=> ([], 3)
644		698
645	=back	699	=back
646		700
647		701
		702	=head1 INCREMENTAL PARSING
		703
		704	[This section and the API it details is still EXPERIMENTAL]
		705
		706	In some cases, there is the need for incremental parsing of JSON
		707	texts. While this module always has to keep both JSON text and resulting
		708	Perl data structure in memory at one time, it does allow you to parse a
		709	JSON stream incrementally. It does so by accumulating text until it has
		710	a full JSON object, which it then can decode. This process is similar to
		711	using C<decode_prefix> to see if a full JSON object is available, but is
		712	much more efficient (JSON::XS will only attempt to parse the JSON text
		713	once it is sure it has enough text to get a decisive result, using a very
		714	simple but truly incremental parser).
		715
		716	The following two methods deal with this.
		717
		718	=over 4
		719
		720	=item [void, scalar or list context] = $json->incr_parse ([$string])
		721
		722	This is the central parsing function. It can both append new text and
		723	extract objects from the stream accumulated so far (both of these
		724	functions are optional).
		725
		726	If C<$string> is given, then this string is appended to the already
		727	existing JSON fragment stored in the C<$json> object.
		728
		729	After that, if the function is called in void context, it will simply
		730	return without doing anything further. This can be used to add more text
		731	in as many chunks as you want.
		732
		733	If the method is called in scalar context, then it will try to extract
		734	exactly I<one> JSON object. If that is successful, it will return this
		735	object, otherwise it will return C<undef>. If there is a parse error,
		736	this method will croak just as C<decode> would do (one can then use
		737	C<incr_skip> to skip the errornous part). This is the most common way of
		738	using the method.
		739
		740	And finally, in list context, it will try to extract as many objects
		741	from the stream as it can find and return them, or the empty list
		742	otherwise. For this to work, there must be no separators between the JSON
		743	objects or arrays, instead they must be concatenated back-to-back. If
		744	an error occurs, an exception will be raised as in the scalar context
		745	case. Note that in this case, any previously-parsed JSON texts will be
		746	lost.
		747
		748	=item $lvalue_string = $json->incr_text
		749
		750	This method returns the currently stored JSON fragment as an lvalue, that
		751	is, you can manipulate it. This I<only> works when a preceding call to
		752	C<incr_parse> in I<scalar context> successfully returned an object. Under
		753	all other circumstances you must not call this function (I mean it.
		754	although in simple tests it might actually work, it I<will> fail under
		755	real world conditions). As a special exception, you can also call this
		756	method before having parsed anything.
		757
		758	This function is useful in two cases: a) finding the trailing text after a
		759	JSON object or b) parsing multiple JSON objects separated by non-JSON text
		760	(such as commas).
		761
		762	=item $json->incr_skip
		763
		764	This will reset the state of the incremental parser and will remove the
		765	parsed text from the input buffer. This is useful after C<incr_parse>
		766	died, in which case the input buffer and incremental parser state is left
		767	unchanged, to skip the text parsed so far and to reset the parse state.
		768
		769	=back
		770
		771	=head2 LIMITATIONS
		772
		773	All options that affect decoding are supported, except
		774	C<allow_nonref>. The reason for this is that it cannot be made to
		775	work sensibly: JSON objects and arrays are self-delimited, i.e. you can concatenate
		776	them back to back and still decode them perfectly. This does not hold true
		777	for JSON numbers, however.
		778
		779	For example, is the string C<1> a single JSON number, or is it simply the
		780	start of C<12>? Or is C<12> a single JSON number, or the concatenation
		781	of C<1> and C<2>? In neither case you can tell, and this is why JSON::XS
		782	takes the conservative route and disallows this case.
		783
		784	=head2 EXAMPLES
		785
		786	Some examples will make all this clearer. First, a simple example that
		787	works similarly to C<decode_prefix>: We want to decode the JSON object at
		788	the start of a string and identify the portion after the JSON object:
		789
		790	my $text = "[1,2,3] hello";
		791
		792	my $json = new JSON::XS;
		793
		794	my $obj = $json->incr_parse ($text)
		795	or die "expected JSON object or array at beginning of string";
		796
		797	my $tail = $json->incr_text;
		798	# $tail now contains " hello"
		799
		800	Easy, isn't it?
		801
		802	Now for a more complicated example: Imagine a hypothetical protocol where
		803	you read some requests from a TCP stream, and each request is a JSON
		804	array, without any separation between them (in fact, it is often useful to
		805	use newlines as "separators", as these get interpreted as whitespace at
		806	the start of the JSON text, which makes it possible to test said protocol
		807	with C<telnet>...).
		808
		809	Here is how you'd do it (it is trivial to write this in an event-based
		810	manner):
		811
		812	my $json = new JSON::XS;
		813
		814	# read some data from the socket
		815	while (sysread $socket, my $buf, 4096) {
		816
		817	# split and decode as many requests as possible
		818	for my $request ($json->incr_parse ($buf)) {
		819	# act on the $request
		820	}
		821	}
		822
		823	Another complicated example: Assume you have a string with JSON objects
		824	or arrays, all separated by (optional) comma characters (e.g. C<[1],[2],
		825	[3]>). To parse them, we have to skip the commas between the JSON texts,
		826	and here is where the lvalue-ness of C<incr_text> comes in useful:
		827
		828	my $text = "[1],[2], [3]";
		829	my $json = new JSON::XS;
		830
		831	# void context, so no parsing done
		832	$json->incr_parse ($text);
		833
		834	# now extract as many objects as possible. note the
		835	# use of scalar context so incr_text can be called.
		836	while (my $obj = $json->incr_parse) {
		837	# do something with $obj
		838
		839	# now skip the optional comma
		840	$json->incr_text =~ s/^ \s* , //x;
		841	}
		842
		843	Now lets go for a very complex example: Assume that you have a gigantic
		844	JSON array-of-objects, many gigabytes in size, and you want to parse it,
		845	but you cannot load it into memory fully (this has actually happened in
		846	the real world :).
		847
		848	Well, you lost, you have to implement your own JSON parser. But JSON::XS
		849	can still help you: You implement a (very simple) array parser and let
		850	JSON decode the array elements, which are all full JSON objects on their
		851	own (this wouldn't work if the array elements could be JSON numbers, for
		852	example):
		853
		854	my $json = new JSON::XS;
		855
		856	# open the monster
		857	open my $fh, "<bigfile.json"
		858	or die "bigfile: $!";
		859
		860	# first parse the initial "["
		861	for (;;) {
		862	sysread $fh, my $buf, 65536
		863	or die "read error: $!";
		864	$json->incr_parse ($buf); # void context, so no parsing
		865
		866	# Exit the loop once we found and removed(!) the initial "[".
		867	# In essence, we are (ab-)using the $json object as a simple scalar
		868	# we append data to.
		869	last if $json->incr_text =~ s/^ \s* \[ //x;
		870	}
		871
		872	# now we have the skipped the initial "[", so continue
		873	# parsing all the elements.
		874	for (;;) {
		875	# in this loop we read data until we got a single JSON object
		876	for (;;) {
		877	if (my $obj = $json->incr_parse) {
		878	# do something with $obj
		879	last;
		880	}
		881
		882	# add more data
		883	sysread $fh, my $buf, 65536
		884	or die "read error: $!";
		885	$json->incr_parse ($buf); # void context, so no parsing
		886	}
		887
		888	# in this loop we read data until we either found and parsed the
		889	# separating "," between elements, or the final "]"
		890	for (;;) {
		891	# first skip whitespace
		892	$json->incr_text =~ s/^\s*//;
		893
		894	# if we find "]", we are done
		895	if ($json->incr_text =~ s/^\]//) {
		896	print "finished.\n";
		897	exit;
		898	}
		899
		900	# if we find ",", we can continue with the next element
		901	if ($json->incr_text =~ s/^,//) {
		902	last;
		903	}
		904
		905	# if we find anything else, we have a parse error!
		906	if (length $json->incr_text) {
		907	die "parse error near ", $json->incr_text;
		908	}
		909
		910	# else add more data
		911	sysread $fh, my $buf, 65536
		912	or die "read error: $!";
		913	$json->incr_parse ($buf); # void context, so no parsing
		914	}
		915
		916	This is a complex example, but most of the complexity comes from the fact
		917	that we are trying to be correct (bear with me if I am wrong, I never ran
		918	the above example :).
		919
		920
		921
648	=head1 MAPPING	922	=head1 MAPPING
649		923
650	This section describes how JSON::XS maps Perl values to JSON values and	924	This section describes how JSON::XS maps Perl values to JSON values and
651	vice versa. These mappings are designed to "do the right thing" in most	925	vice versa. These mappings are designed to "do the right thing" in most
652	circumstances automatically, preserving round-tripping characteristics	926	circumstances automatically, preserving round-tripping characteristics
…		…
680		954
681	A JSON number becomes either an integer, numeric (floating point) or	955	A JSON number becomes either an integer, numeric (floating point) or
682	string scalar in perl, depending on its range and any fractional parts. On	956	string scalar in perl, depending on its range and any fractional parts. On
683	the Perl level, there is no difference between those as Perl handles all	957	the Perl level, there is no difference between those as Perl handles all
684	the conversion details, but an integer may take slightly less memory and	958	the conversion details, but an integer may take slightly less memory and
685	might represent more values exactly than (floating point) numbers.	959	might represent more values exactly than floating point numbers.
686		960
687	If the number consists of digits only, JSON::XS will try to represent	961	If the number consists of digits only, JSON::XS will try to represent
688	it as an integer value. If that fails, it will try to represent it as	962	it as an integer value. If that fails, it will try to represent it as
689	a numeric (floating point) value if that is possible without loss of	963	a numeric (floating point) value if that is possible without loss of
690	precision. Otherwise it will preserve the number as a string value.	964	precision. Otherwise it will preserve the number as a string value (in
		965	which case you lose roundtripping ability, as the JSON number will be
		966	re-encoded toa JSON string).
691		967
692	Numbers containing a fractional or exponential part will always be	968	Numbers containing a fractional or exponential part will always be
693	represented as numeric (floating point) values, possibly at a loss of	969	represented as numeric (floating point) values, possibly at a loss of
694	precision.	970	precision (in which case you might lose perfect roundtripping ability, but
695		971	the JSON number will still be re-encoded as a JSON number).
696	This might create round-tripping problems as numbers might become strings,
697	but as Perl is typeless there is no other way to do it.
698		972
699	=item true, false	973	=item true, false
700		974
701	These JSON atoms become C<JSON::XS::true> and C<JSON::XS::false>,	975	These JSON atoms become C<JSON::XS::true> and C<JSON::XS::false>,
702	respectively. They are overloaded to act almost exactly like the numbers	976	respectively. They are overloaded to act almost exactly like the numbers
…		…
739	Other unblessed references are generally not allowed and will cause an	1013	Other unblessed references are generally not allowed and will cause an
740	exception to be thrown, except for references to the integers C<0> and	1014	exception to be thrown, except for references to the integers C<0> and
741	C<1>, which get turned into C<false> and C<true> atoms in JSON. You can	1015	C<1>, which get turned into C<false> and C<true> atoms in JSON. You can
742	also use C<JSON::XS::false> and C<JSON::XS::true> to improve readability.	1016	also use C<JSON::XS::false> and C<JSON::XS::true> to improve readability.
743		1017
744	to_json [\0,JSON::XS::true] # yields [false,true]	1018	encode_json [\0,JSON::XS::true] # yields [false,true]
745		1019
746	=item JSON::XS::true, JSON::XS::false	1020	=item JSON::XS::true, JSON::XS::false
747		1021
748	These special values become JSON true and JSON false values,	1022	These special values become JSON true and JSON false values,
749	respectively. You can also use C<\1> and C<\0> directly if you want.	1023	respectively. You can also use C<\1> and C<\0> directly if you want.
750		1024
751	=item blessed objects	1025	=item blessed objects
752		1026
753	Blessed objects are not allowed. JSON::XS currently tries to encode their	1027	Blessed objects are not directly representable in JSON. See the
754	underlying representation (hash- or arrayref), but this behaviour might	1028	C<allow_blessed> and C<convert_blessed> methods on various options on
755	change in future versions.	1029	how to deal with this: basically, you can choose between throwing an
		1030	exception, encoding the reference as if it weren't blessed, or provide
		1031	your own serialiser method.
756		1032
757	=item simple scalars	1033	=item simple scalars
758		1034
759	Simple Perl scalars (any scalar that is not a reference) are the most	1035	Simple Perl scalars (any scalar that is not a reference) are the most
760	difficult objects to encode: JSON::XS will encode undefined scalars as	1036	difficult objects to encode: JSON::XS will encode undefined scalars as
761	JSON null value, scalars that have last been used in a string context	1037	JSON C<null> values, scalars that have last been used in a string context
762	before encoding as JSON strings and anything else as number value:	1038	before encoding as JSON strings, and anything else as number value:
763		1039
764	# dump as number	1040	# dump as number
765	to_json [2] # yields [2]	1041	encode_json [2] # yields [2]
766	to_json [-3.0e17] # yields [-3e+17]	1042	encode_json [-3.0e17] # yields [-3e+17]
767	my $value = 5; to_json [$value] # yields [5]	1043	my $value = 5; encode_json [$value] # yields [5]
768		1044
769	# used as string, so dump as string	1045	# used as string, so dump as string
770	print $value;	1046	print $value;
771	to_json [$value] # yields ["5"]	1047	encode_json [$value] # yields ["5"]
772		1048
773	# undef becomes null	1049	# undef becomes null
774	to_json [undef] # yields [null]	1050	encode_json [undef] # yields [null]
775		1051
776	You can force the type to be a JSON string by stringifying it:	1052	You can force the type to be a JSON string by stringifying it:
777		1053
778	my $x = 3.1; # some variable containing a number	1054	my $x = 3.1; # some variable containing a number
779	"$x"; # stringified	1055	"$x"; # stringified
…		…
785	my $x = "3"; # some variable containing a string	1061	my $x = "3"; # some variable containing a string
786	$x += 0; # numify it, ensuring it will be dumped as a number	1062	$x += 0; # numify it, ensuring it will be dumped as a number
787	$x *= 1; # same thing, the choice is yours.	1063	$x *= 1; # same thing, the choice is yours.
788		1064
789	You can not currently force the type in other, less obscure, ways. Tell me	1065	You can not currently force the type in other, less obscure, ways. Tell me
790	if you need this capability.	1066	if you need this capability (but don't forget to explain why it's needed
		1067	:).
791		1068
792	=back	1069	=back
793		1070
794		1071
795	=head1 COMPARISON	1072	=head1 ENCODING/CODESET FLAG NOTES
796		1073
797	As already mentioned, this module was created because none of the existing	1074	The interested reader might have seen a number of flags that signify
798	JSON modules could be made to work correctly. First I will describe the	1075	encodings or codesets - C<utf8>, C<latin1> and C<ascii>. There seems to be
799	problems (or pleasures) I encountered with various existing JSON modules,	1076	some confusion on what these do, so here is a short comparison:
800	followed by some benchmark values. JSON::XS was designed not to suffer	1077
801	from any of these problems or limitations.	1078	C<utf8> controls whether the JSON text created by C<encode> (and expected
		1079	by C<decode>) is UTF-8 encoded or not, while C<latin1> and C<ascii> only
		1080	control whether C<encode> escapes character values outside their respective
		1081	codeset range. Neither of these flags conflict with each other, although
		1082	some combinations make less sense than others.
		1083
		1084	Care has been taken to make all flags symmetrical with respect to
		1085	C<encode> and C<decode>, that is, texts encoded with any combination of
		1086	these flag values will be correctly decoded when the same flags are used
		1087	- in general, if you use different flag settings while encoding vs. when
		1088	decoding you likely have a bug somewhere.
		1089
		1090	Below comes a verbose discussion of these flags. Note that a "codeset" is
		1091	simply an abstract set of character-codepoint pairs, while an encoding
		1092	takes those codepoint numbers and I<encodes> them, in our case into
		1093	octets. Unicode is (among other things) a codeset, UTF-8 is an encoding,
		1094	and ISO-8859-1 (= latin 1) and ASCII are both codesets I<and> encodings at
		1095	the same time, which can be confusing.
802		1096
803	=over 4	1097	=over 4
804		1098
805	=item JSON 1.07	1099	=item C<utf8> flag disabled
806		1100
807	Slow (but very portable, as it is written in pure Perl).	1101	When C<utf8> is disabled (the default), then C<encode>/C<decode> generate
		1102	and expect Unicode strings, that is, characters with high ordinal Unicode
		1103	values (> 255) will be encoded as such characters, and likewise such
		1104	characters are decoded as-is, no canges to them will be done, except
		1105	"(re-)interpreting" them as Unicode codepoints or Unicode characters,
		1106	respectively (to Perl, these are the same thing in strings unless you do
		1107	funny/weird/dumb stuff).
808		1108
809	Undocumented/buggy Unicode handling (how JSON handles Unicode values is	1109	This is useful when you want to do the encoding yourself (e.g. when you
810	undocumented. One can get far by feeding it Unicode strings and doing	1110	want to have UTF-16 encoded JSON texts) or when some other layer does
811	en-/decoding oneself, but Unicode escapes are not working properly).	1111	the encoding for you (for example, when printing to a terminal using a
		1112	filehandle that transparently encodes to UTF-8 you certainly do NOT want
		1113	to UTF-8 encode your data first and have Perl encode it another time).
812		1114
813	No round-tripping (strings get clobbered if they look like numbers, e.g.	1115	=item C<utf8> flag enabled
814	the string C<2.0> will encode to C<2.0> instead of C<"2.0">, and that will
815	decode into the number 2.
816		1116
817	=item JSON::PC 0.01	1117	If the C<utf8>-flag is enabled, C<encode>/C<decode> will encode all
		1118	characters using the corresponding UTF-8 multi-byte sequence, and will
		1119	expect your input strings to be encoded as UTF-8, that is, no "character"
		1120	of the input string must have any value > 255, as UTF-8 does not allow
		1121	that.
818		1122
819	Very fast.	1123	The C<utf8> flag therefore switches between two modes: disabled means you
		1124	will get a Unicode string in Perl, enabled means you get an UTF-8 encoded
		1125	octet/binary string in Perl.
820		1126
821	Undocumented/buggy Unicode handling.	1127	=item C<latin1> or C<ascii> flags enabled
822		1128
823	No round-tripping.	1129	With C<latin1> (or C<ascii>) enabled, C<encode> will escape characters
		1130	with ordinal values > 255 (> 127 with C<ascii>) and encode the remaining
		1131	characters as specified by the C<utf8> flag.
824		1132
825	Has problems handling many Perl values (e.g. regex results and other magic	1133	If C<utf8> is disabled, then the result is also correctly encoded in those
826	values will make it croak).	1134	character sets (as both are proper subsets of Unicode, meaning that a
		1135	Unicode string with all character values < 256 is the same thing as a
		1136	ISO-8859-1 string, and a Unicode string with all character values < 128 is
		1137	the same thing as an ASCII string in Perl).
827		1138
828	Does not even generate valid JSON (C<{1,2}> gets converted to C<{1:2}>	1139	If C<utf8> is enabled, you still get a correct UTF-8-encoded string,
829	which is not a valid JSON text.	1140	regardless of these flags, just some more characters will be escaped using
		1141	C<\uXXXX> then before.
830		1142
831	Unmaintained (maintainer unresponsive for many months, bugs are not	1143	Note that ISO-8859-1-I<encoded> strings are not compatible with UTF-8
832	getting fixed).	1144	encoding, while ASCII-encoded strings are. That is because the ISO-8859-1
		1145	encoding is NOT a subset of UTF-8 (despite the ISO-8859-1 I<codeset> being
		1146	a subset of Unicode), while ASCII is.
833		1147
834	=item JSON::Syck 0.21	1148	Surprisingly, C<decode> will ignore these flags and so treat all input
		1149	values as governed by the C<utf8> flag. If it is disabled, this allows you
		1150	to decode ISO-8859-1- and ASCII-encoded strings, as both strict subsets of
		1151	Unicode. If it is enabled, you can correctly decode UTF-8 encoded strings.
835		1152
836	Very buggy (often crashes).	1153	So neither C<latin1> nor C<ascii> are incompatible with the C<utf8> flag -
		1154	they only govern when the JSON output engine escapes a character or not.
837		1155
838	Very inflexible (no human-readable format supported, format pretty much	1156	The main use for C<latin1> is to relatively efficiently store binary data
839	undocumented. I need at least a format for easy reading by humans and a	1157	as JSON, at the expense of breaking compatibility with most JSON decoders.
840	single-line compact format for use in a protocol, and preferably a way to
841	generate ASCII-only JSON texts).
842		1158
843	Completely broken (and confusingly documented) Unicode handling (Unicode	1159	The main use for C<ascii> is to force the output to not contain characters
844	escapes are not working properly, you need to set ImplicitUnicode to	1160	with values > 127, which means you can interpret the resulting string
845	I<different> values on en- and decoding to get symmetric behaviour).	1161	as UTF-8, ISO-8859-1, ASCII, KOI8-R or most about any character set and
846		1162	8-bit-encoding, and still get the same data structure back. This is useful
847	No round-tripping (simple cases work, but this depends on whether the scalar	1163	when your channel for JSON transfer is not 8-bit clean or the encoding
848	value was used in a numeric context or not).	1164	might be mangled in between (e.g. in mail), and works because ASCII is a
849		1165	proper subset of most 8-bit and multibyte encodings in use in the world.
850	Dumping hashes may skip hash values depending on iterator state.
851
852	Unmaintained (maintainer unresponsive for many months, bugs are not
853	getting fixed).
854
855	Does not check input for validity (i.e. will accept non-JSON input and
856	return "something" instead of raising an exception. This is a security
857	issue: imagine two banks transferring money between each other using
858	JSON. One bank might parse a given non-JSON request and deduct money,
859	while the other might reject the transaction with a syntax error. While a
860	good protocol will at least recover, that is extra unnecessary work and
861	the transaction will still not succeed).
862
863	=item JSON::DWIW 0.04
864
865	Very fast. Very natural. Very nice.
866
867	Undocumented Unicode handling (but the best of the pack. Unicode escapes
868	still don't get parsed properly).
869
870	Very inflexible.
871
872	No round-tripping.
873
874	Does not generate valid JSON texts (key strings are often unquoted, empty keys
875	result in nothing being output)
876
877	Does not check input for validity.
878		1166
879	=back	1167	=back
880		1168
881		1169
882	=head2 JSON and YAML	1170	=head2 JSON and YAML
883		1171
884	You often hear that JSON is a subset (or a close subset) of YAML. This is,	1172	You often hear that JSON is a subset of YAML. This is, however, a mass
885	however, a mass hysteria and very far from the truth. In general, there is	1173	hysteria(*) and very far from the truth (as of the time of this writing),
886	no way to configure JSON::XS to output a data structure as valid YAML.	1174	so let me state it clearly: I<in general, there is no way to configure
		1175	JSON::XS to output a data structure as valid YAML> that works in all
		1176	cases.
887		1177
888	If you really must use JSON::XS to generate YAML, you should use this	1178	If you really must use JSON::XS to generate YAML, you should use this
889	algorithm (subject to change in future versions):	1179	algorithm (subject to change in future versions):
890		1180
891	my $to_yaml = JSON::XS->new->utf8->space_after (1);	1181	my $to_yaml = JSON::XS->new->utf8->space_after (1);
892	my $yaml = $to_yaml->encode ($ref) . "\n";	1182	my $yaml = $to_yaml->encode ($ref) . "\n";
893		1183
894	This will usually generate JSON texts that also parse as valid	1184	This will I<usually> generate JSON texts that also parse as valid
895	YAML. Please note that YAML has hardcoded limits on (simple) object key	1185	YAML. Please note that YAML has hardcoded limits on (simple) object key
896	lengths that JSON doesn't have, so you should make sure that your hash	1186	lengths that JSON doesn't have and also has different and incompatible
		1187	unicode handling, so you should make sure that your hash keys are
897	keys are noticeably shorter than the 1024 characters YAML allows.	1188	noticeably shorter than the 1024 "stream characters" YAML allows and that
		1189	you do not have characters with codepoint values outside the Unicode BMP
		1190	(basic multilingual page). YAML also does not allow C<\/> sequences in
		1191	strings (which JSON::XS does not I<currently> generate, but other JSON
		1192	generators might).
898		1193
899	There might be other incompatibilities that I am not aware of. In general	1194	There might be other incompatibilities that I am not aware of (or the YAML
		1195	specification has been changed yet again - it does so quite often). In
900	you should not try to generate YAML with a JSON generator or vice versa,	1196	general you should not try to generate YAML with a JSON generator or vice
901	or try to parse JSON with a YAML parser or vice versa: chances are high	1197	versa, or try to parse JSON with a YAML parser or vice versa: chances are
902	that you will run into severe interoperability problems.	1198	high that you will run into severe interoperability problems when you
		1199	least expect it.
		1200
		1201	=over 4
		1202
		1203	=item (*)
		1204
		1205	I have been pressured multiple times by Brian Ingerson (one of the
		1206	authors of the YAML specification) to remove this paragraph, despite him
		1207	acknowledging that the actual incompatibilities exist. As I was personally
		1208	bitten by this "JSON is YAML" lie, I refused and said I will continue to
		1209	educate people about these issues, so others do not run into the same
		1210	problem again and again. After this, Brian called me a (quote)I<complete
		1211	and worthless idiot>(unquote).
		1212
		1213	In my opinion, instead of pressuring and insulting people who actually
		1214	clarify issues with YAML and the wrong statements of some of its
		1215	proponents, I would kindly suggest reading the JSON spec (which is not
		1216	that difficult or long) and finally make YAML compatible to it, and
		1217	educating users about the changes, instead of spreading lies about the
		1218	real compatibility for many I<years> and trying to silence people who
		1219	point out that it isn't true.
		1220
		1221	=back
903		1222
904		1223
905	=head2 SPEED	1224	=head2 SPEED
906		1225
907	It seems that JSON::XS is surprisingly fast, as shown in the following	1226	It seems that JSON::XS is surprisingly fast, as shown in the following
908	tables. They have been generated with the help of the C<eg/bench> program	1227	tables. They have been generated with the help of the C<eg/bench> program
909	in the JSON::XS distribution, to make it easy to compare on your own	1228	in the JSON::XS distribution, to make it easy to compare on your own
910	system.	1229	system.
911		1230
912	First comes a comparison between various modules using a very short	1231	First comes a comparison between various modules using
913	single-line JSON string:	1232	a very short single-line JSON string (also available at
		1233	L<http://dist.schmorp.de/misc/json/short.json>).
914		1234
915	{"method": "handleMessage", "params": ["user1", "we were just talking"], \	1235	{"method": "handleMessage", "params": ["user1", "we were just talking"], \
916	"id": null, "array":[1,11,234,-5,1e5,1e7, true, false]}	1236	"id": null, "array":[1,11,234,-5,1e5,1e7, true, false]}
917		1237
918	It shows the number of encodes/decodes per second (JSON::XS uses	1238	It shows the number of encodes/decodes per second (JSON::XS uses
…		…
937	about three times faster on decoding, and over forty times faster	1257	about three times faster on decoding, and over forty times faster
938	than JSON, even with pretty-printing and key sorting. It also compares	1258	than JSON, even with pretty-printing and key sorting. It also compares
939	favourably to Storable for small amounts of data.	1259	favourably to Storable for small amounts of data.
940		1260
941	Using a longer test string (roughly 18KB, generated from Yahoo! Locals	1261	Using a longer test string (roughly 18KB, generated from Yahoo! Locals
942	search API (http://nanoref.com/yahooapis/mgPdGg):	1262	search API (L<http://dist.schmorp.de/misc/json/long.json>).
943		1263
944	module \| encode \| decode \|	1264	module \| encode \| decode \|
945	-----------\|------------\|------------\|	1265	-----------\|------------\|------------\|
946	JSON 1.x \| 55.260 \| 34.971 \|	1266	JSON 1.x \| 55.260 \| 34.971 \|
947	JSON::DWIW \| 825.228 \| 1082.513 \|	1267	JSON::DWIW \| 825.228 \| 1082.513 \|
…		…
984		1304
985	Third, JSON::XS recurses using the C stack when decoding objects and	1305	Third, JSON::XS recurses using the C stack when decoding objects and
986	arrays. The C stack is a limited resource: for instance, on my amd64	1306	arrays. The C stack is a limited resource: for instance, on my amd64
987	machine with 8MB of stack size I can decode around 180k nested arrays but	1307	machine with 8MB of stack size I can decode around 180k nested arrays but
988	only 14k nested JSON objects (due to perl itself recursing deeply on croak	1308	only 14k nested JSON objects (due to perl itself recursing deeply on croak
989	to free the temporary). If that is exceeded, the program crashes. to be	1309	to free the temporary). If that is exceeded, the program crashes. To be
990	conservative, the default nesting limit is set to 512. If your process	1310	conservative, the default nesting limit is set to 512. If your process
991	has a smaller stack, you should adjust this setting accordingly with the	1311	has a smaller stack, you should adjust this setting accordingly with the
992	C<max_depth> method.	1312	C<max_depth> method.
993		1313
994	And last but least, something else could bomb you that I forgot to think	1314	Something else could bomb you, too, that I forgot to think of. In that
995	of. In that case, you get to keep the pieces. I am always open for hints,	1315	case, you get to keep the pieces. I am always open for hints, though...
996	though...	1316
		1317	Also keep in mind that JSON::XS might leak contents of your Perl data
		1318	structures in its error messages, so when you serialise sensitive
		1319	information you might want to make sure that exceptions thrown by JSON::XS
		1320	will not end up in front of untrusted eyes.
997		1321
998	If you are using JSON::XS to return packets to consumption	1322	If you are using JSON::XS to return packets to consumption
999	by JavaScript scripts in a browser you should have a look at	1323	by JavaScript scripts in a browser you should have a look at
1000	L<http://jpsykes.com/47/practical-csrf-and-json-security> to see whether	1324	L<http://jpsykes.com/47/practical-csrf-and-json-security> to see whether
1001	you are vulnerable to some common attack vectors (which really are browser	1325	you are vulnerable to some common attack vectors (which really are browser
1002	design bugs, but it is still you who will have to deal with it, as major	1326	design bugs, but it is still you who will have to deal with it, as major
1003	browser developers care only for features, not about doing security	1327	browser developers care only for features, not about getting security
1004	right).	1328	right).
1005		1329
1006		1330
1007	=head1 THREADS	1331	=head1 THREADS
1008		1332
1009	This module is I<not> guaranteed to be thread safe and there are no	1333	This module is I<not> guaranteed to be thread safe and there are no
1010	plans to change this until Perl gets thread support (as opposed to the	1334	plans to change this until Perl gets thread support (as opposed to the
1011	horribly slow so-called "threads" which are simply slow and bloated	1335	horribly slow so-called "threads" which are simply slow and bloated
1012	process simulations - use fork, its I<much> faster, cheaper, better).	1336	process simulations - use fork, it's I<much> faster, cheaper, better).
1013		1337
1014	(It might actually work, but you have been warned).	1338	(It might actually work, but you have been warned).
1015		1339
1016		1340
1017	=head1 BUGS	1341	=head1 BUGS
1018		1342
1019	While the goal of this module is to be correct, that unfortunately does	1343	While the goal of this module is to be correct, that unfortunately does
1020	not mean its bug-free, only that I think its design is bug-free. It is	1344	not mean it's bug-free, only that I think its design is bug-free. It is
1021	still relatively early in its development. If you keep reporting bugs they	1345	still relatively early in its development. If you keep reporting bugs they
1022	will be fixed swiftly, though.	1346	will be fixed swiftly, though.
1023		1347
1024	Please refrain from using rt.cpan.org or any other bug reporting	1348	Please refrain from using rt.cpan.org or any other bug reporting
1025	service. I put the contact address into my modules for a reason.	1349	service. I put the contact address into my modules for a reason.
…		…
1047	"--" => sub { $_[0] = ${$_[0]} - 1 },	1371	"--" => sub { $_[0] = ${$_[0]} - 1 },
1048	fallback => 1;	1372	fallback => 1;
1049		1373
1050	1;	1374	1;
1051		1375
		1376	=head1 SEE ALSO
		1377
		1378	The F<json_xs> command line utility for quick experiments.
		1379
1052	=head1 AUTHOR	1380	=head1 AUTHOR
1053		1381
1054	Marc Lehmann <schmorp@schmorp.de>	1382	Marc Lehmann <schmorp@schmorp.de>
1055	http://home.schmorp.de/	1383	http://home.schmorp.de/
1056		1384

Diff Legend

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

Comparing JSON-XS/XS.pm (file contents): Revision 1.75 by root, Thu Nov 29 13:35:35 2007 UTC vs. Revision 1.99 by root, Thu Mar 27 06:37:35 2008 UTC

Diff Legend

Comparing JSON-XS/XS.pm (file contents):
Revision 1.75 by root, Thu Nov 29 13:35:35 2007 UTC vs.
Revision 1.99 by root, Thu Mar 27 06:37:35 2008 UTC