1 |
=head1 NAME |
2 |
|
3 |
JSON::XS - JSON serialising/deserialising, done correctly and fast |
4 |
|
5 |
=head1 SYNOPSIS |
6 |
|
7 |
use JSON::XS; |
8 |
|
9 |
# exported functions, they croak on error |
10 |
# and expect/generate UTF-8 |
11 |
|
12 |
$utf8_encoded_json_text = to_json $perl_hash_or_arrayref; |
13 |
$perl_hash_or_arrayref = from_json $utf8_encoded_json_text; |
14 |
|
15 |
# objToJson and jsonToObj aliases to to_json and from_json |
16 |
# are exported for compatibility to the JSON module, |
17 |
# but should not be used in new code. |
18 |
|
19 |
# OO-interface |
20 |
|
21 |
$coder = JSON::XS->new->ascii->pretty->allow_nonref; |
22 |
$pretty_printed_unencoded = $coder->encode ($perl_scalar); |
23 |
$perl_scalar = $coder->decode ($unicode_json_text); |
24 |
|
25 |
=head1 DESCRIPTION |
26 |
|
27 |
This module converts Perl data structures to JSON and vice versa. Its |
28 |
primary goal is to be I<correct> and its secondary goal is to be |
29 |
I<fast>. To reach the latter goal it was written in C. |
30 |
|
31 |
As this is the n-th-something JSON module on CPAN, what was the reason |
32 |
to write yet another JSON module? While it seems there are many JSON |
33 |
modules, none of them correctly handle all corner cases, and in most cases |
34 |
their maintainers are unresponsive, gone missing, or not listening to bug |
35 |
reports for other reasons. |
36 |
|
37 |
See COMPARISON, below, for a comparison to some other JSON modules. |
38 |
|
39 |
See MAPPING, below, on how JSON::XS maps perl values to JSON values and |
40 |
vice versa. |
41 |
|
42 |
=head2 FEATURES |
43 |
|
44 |
=over 4 |
45 |
|
46 |
=item * correct unicode handling |
47 |
|
48 |
This module knows how to handle Unicode, and even documents how and when |
49 |
it does so. |
50 |
|
51 |
=item * round-trip integrity |
52 |
|
53 |
When you serialise a perl data structure using only datatypes supported |
54 |
by JSON, the deserialised data structure is identical on the Perl level. |
55 |
(e.g. the string "2.0" doesn't suddenly become "2" just because it looks |
56 |
like a number). |
57 |
|
58 |
=item * strict checking of JSON correctness |
59 |
|
60 |
There is no guessing, no generating of illegal JSON texts by default, |
61 |
and only JSON is accepted as input by default (the latter is a security |
62 |
feature). |
63 |
|
64 |
=item * fast |
65 |
|
66 |
Compared to other JSON modules, this module compares favourably in terms |
67 |
of speed, too. |
68 |
|
69 |
=item * simple to use |
70 |
|
71 |
This module has both a simple functional interface as well as an OO |
72 |
interface. |
73 |
|
74 |
=item * reasonably versatile output formats |
75 |
|
76 |
You can choose between the most compact guarenteed single-line format |
77 |
possible (nice for simple line-based protocols), a pure-ascii format |
78 |
(for when your transport is not 8-bit clean, still supports the whole |
79 |
unicode range), or a pretty-printed format (for when you want to read that |
80 |
stuff). Or you can combine those features in whatever way you like. |
81 |
|
82 |
=back |
83 |
|
84 |
=cut |
85 |
|
86 |
package JSON::XS; |
87 |
|
88 |
use strict; |
89 |
|
90 |
BEGIN { |
91 |
our $VERSION = '1.2'; |
92 |
our @ISA = qw(Exporter); |
93 |
|
94 |
our @EXPORT = qw(to_json from_json objToJson jsonToObj); |
95 |
require Exporter; |
96 |
|
97 |
require XSLoader; |
98 |
XSLoader::load JSON::XS::, $VERSION; |
99 |
} |
100 |
|
101 |
=head1 FUNCTIONAL INTERFACE |
102 |
|
103 |
The following convinience methods are provided by this module. They are |
104 |
exported by default: |
105 |
|
106 |
=over 4 |
107 |
|
108 |
=item $json_text = to_json $perl_scalar |
109 |
|
110 |
Converts the given Perl data structure (a simple scalar or a reference to |
111 |
a hash or array) to a UTF-8 encoded, binary string (that is, the string contains |
112 |
octets only). Croaks on error. |
113 |
|
114 |
This function call is functionally identical to: |
115 |
|
116 |
$json_text = JSON::XS->new->utf8->encode ($perl_scalar) |
117 |
|
118 |
except being faster. |
119 |
|
120 |
=item $perl_scalar = from_json $json_text |
121 |
|
122 |
The opposite of C<to_json>: expects an UTF-8 (binary) string and tries to |
123 |
parse that as an UTF-8 encoded JSON text, returning the resulting simple |
124 |
scalar or reference. Croaks on error. |
125 |
|
126 |
This function call is functionally identical to: |
127 |
|
128 |
$perl_scalar = JSON::XS->new->utf8->decode ($json_text) |
129 |
|
130 |
except being faster. |
131 |
|
132 |
=back |
133 |
|
134 |
|
135 |
=head1 OBJECT-ORIENTED INTERFACE |
136 |
|
137 |
The object oriented interface lets you configure your own encoding or |
138 |
decoding style, within the limits of supported formats. |
139 |
|
140 |
=over 4 |
141 |
|
142 |
=item $json = new JSON::XS |
143 |
|
144 |
Creates a new JSON::XS object that can be used to de/encode JSON |
145 |
strings. All boolean flags described below are by default I<disabled>. |
146 |
|
147 |
The mutators for flags all return the JSON object again and thus calls can |
148 |
be chained: |
149 |
|
150 |
my $json = JSON::XS->new->utf8->space_after->encode ({a => [1,2]}) |
151 |
=> {"a": [1, 2]} |
152 |
|
153 |
=item $json = $json->ascii ([$enable]) |
154 |
|
155 |
If C<$enable> is true (or missing), then the C<encode> method will not |
156 |
generate characters outside the code range C<0..127> (which is ASCII). Any |
157 |
unicode characters outside that range will be escaped using either a |
158 |
single \uXXXX (BMP characters) or a double \uHHHH\uLLLLL escape sequence, |
159 |
as per RFC4627. The resulting encoded JSON text can be treated as a native |
160 |
unicode string, an ascii-encoded, latin1-encoded or UTF-8 encoded string, |
161 |
or any other superset of ASCII. |
162 |
|
163 |
If C<$enable> is false, then the C<encode> method will not escape Unicode |
164 |
characters unless required by the JSON syntax or other flags. This results |
165 |
in a faster and more compact format. |
166 |
|
167 |
The main use for this flag is to produce JSON texts that can be |
168 |
transmitted over a 7-bit channel, as the encoded JSON texts will not |
169 |
contain any 8 bit characters. |
170 |
|
171 |
JSON::XS->new->ascii (1)->encode ([chr 0x10401]) |
172 |
=> ["\ud801\udc01"] |
173 |
|
174 |
=item $json = $json->latin1 ([$enable]) |
175 |
|
176 |
If C<$enable> is true (or missing), then the C<encode> method will encode |
177 |
the resulting JSON text as latin1 (or iso-8859-1), escaping any characters |
178 |
outside the code range C<0..255>. The resulting string can be treated as a |
179 |
latin1-encoded JSON text or a native unicode string. The C<decode> method |
180 |
will not be affected in any way by this flag, as C<decode> by default |
181 |
expects unicode, which is a strict superset of latin1. |
182 |
|
183 |
If C<$enable> is false, then the C<encode> method will not escape Unicode |
184 |
characters unless required by the JSON syntax or other flags. |
185 |
|
186 |
The main use for this flag is efficiently encoding binary data as JSON |
187 |
text, as most octets will not be escaped, resulting in a smaller encoded |
188 |
size. The disadvantage is that the resulting JSON text is encoded |
189 |
in latin1 (and must correctly be treated as such when storing and |
190 |
transfering), a rare encoding for JSON. It is therefore most useful when |
191 |
you want to store data structures known to contain binary data efficiently |
192 |
in files or databases, not when talking to other JSON encoders/decoders. |
193 |
|
194 |
JSON::XS->new->latin1->encode (["\x{89}\x{abc}"] |
195 |
=> ["\x{89}\\u0abc"] # (perl syntax, U+abc escaped, U+89 not) |
196 |
|
197 |
=item $json = $json->utf8 ([$enable]) |
198 |
|
199 |
If C<$enable> is true (or missing), then the C<encode> method will encode |
200 |
the JSON result into UTF-8, as required by many protocols, while the |
201 |
C<decode> method expects to be handled an UTF-8-encoded string. Please |
202 |
note that UTF-8-encoded strings do not contain any characters outside the |
203 |
range C<0..255>, they are thus useful for bytewise/binary I/O. In future |
204 |
versions, enabling this option might enable autodetection of the UTF-16 |
205 |
and UTF-32 encoding families, as described in RFC4627. |
206 |
|
207 |
If C<$enable> is false, then the C<encode> method will return the JSON |
208 |
string as a (non-encoded) unicode string, while C<decode> expects thus a |
209 |
unicode string. Any decoding or encoding (e.g. to UTF-8 or UTF-16) needs |
210 |
to be done yourself, e.g. using the Encode module. |
211 |
|
212 |
Example, output UTF-16BE-encoded JSON: |
213 |
|
214 |
use Encode; |
215 |
$jsontext = encode "UTF-16BE", JSON::XS->new->encode ($object); |
216 |
|
217 |
Example, decode UTF-32LE-encoded JSON: |
218 |
|
219 |
use Encode; |
220 |
$object = JSON::XS->new->decode (decode "UTF-32LE", $jsontext); |
221 |
|
222 |
=item $json = $json->pretty ([$enable]) |
223 |
|
224 |
This enables (or disables) all of the C<indent>, C<space_before> and |
225 |
C<space_after> (and in the future possibly more) flags in one call to |
226 |
generate the most readable (or most compact) form possible. |
227 |
|
228 |
Example, pretty-print some simple structure: |
229 |
|
230 |
my $json = JSON::XS->new->pretty(1)->encode ({a => [1,2]}) |
231 |
=> |
232 |
{ |
233 |
"a" : [ |
234 |
1, |
235 |
2 |
236 |
] |
237 |
} |
238 |
|
239 |
=item $json = $json->indent ([$enable]) |
240 |
|
241 |
If C<$enable> is true (or missing), then the C<encode> method will use a multiline |
242 |
format as output, putting every array member or object/hash key-value pair |
243 |
into its own line, identing them properly. |
244 |
|
245 |
If C<$enable> is false, no newlines or indenting will be produced, and the |
246 |
resulting JSON text is guarenteed not to contain any C<newlines>. |
247 |
|
248 |
This setting has no effect when decoding JSON texts. |
249 |
|
250 |
=item $json = $json->space_before ([$enable]) |
251 |
|
252 |
If C<$enable> is true (or missing), then the C<encode> method will add an extra |
253 |
optional space before the C<:> separating keys from values in JSON objects. |
254 |
|
255 |
If C<$enable> is false, then the C<encode> method will not add any extra |
256 |
space at those places. |
257 |
|
258 |
This setting has no effect when decoding JSON texts. You will also |
259 |
most likely combine this setting with C<space_after>. |
260 |
|
261 |
Example, space_before enabled, space_after and indent disabled: |
262 |
|
263 |
{"key" :"value"} |
264 |
|
265 |
=item $json = $json->space_after ([$enable]) |
266 |
|
267 |
If C<$enable> is true (or missing), then the C<encode> method will add an extra |
268 |
optional space after the C<:> separating keys from values in JSON objects |
269 |
and extra whitespace after the C<,> separating key-value pairs and array |
270 |
members. |
271 |
|
272 |
If C<$enable> is false, then the C<encode> method will not add any extra |
273 |
space at those places. |
274 |
|
275 |
This setting has no effect when decoding JSON texts. |
276 |
|
277 |
Example, space_before and indent disabled, space_after enabled: |
278 |
|
279 |
{"key": "value"} |
280 |
|
281 |
=item $json = $json->canonical ([$enable]) |
282 |
|
283 |
If C<$enable> is true (or missing), then the C<encode> method will output JSON objects |
284 |
by sorting their keys. This is adding a comparatively high overhead. |
285 |
|
286 |
If C<$enable> is false, then the C<encode> method will output key-value |
287 |
pairs in the order Perl stores them (which will likely change between runs |
288 |
of the same script). |
289 |
|
290 |
This option is useful if you want the same data structure to be encoded as |
291 |
the same JSON text (given the same overall settings). If it is disabled, |
292 |
the same hash migh be encoded differently even if contains the same data, |
293 |
as key-value pairs have no inherent ordering in Perl. |
294 |
|
295 |
This setting has no effect when decoding JSON texts. |
296 |
|
297 |
=item $json = $json->allow_nonref ([$enable]) |
298 |
|
299 |
If C<$enable> is true (or missing), then the C<encode> method can convert a |
300 |
non-reference into its corresponding string, number or null JSON value, |
301 |
which is an extension to RFC4627. Likewise, C<decode> will accept those JSON |
302 |
values instead of croaking. |
303 |
|
304 |
If C<$enable> is false, then the C<encode> method will croak if it isn't |
305 |
passed an arrayref or hashref, as JSON texts must either be an object |
306 |
or array. Likewise, C<decode> will croak if given something that is not a |
307 |
JSON object or array. |
308 |
|
309 |
Example, encode a Perl scalar as JSON value with enabled C<allow_nonref>, |
310 |
resulting in an invalid JSON text: |
311 |
|
312 |
JSON::XS->new->allow_nonref->encode ("Hello, World!") |
313 |
=> "Hello, World!" |
314 |
|
315 |
=item $json = $json->shrink ([$enable]) |
316 |
|
317 |
Perl usually over-allocates memory a bit when allocating space for |
318 |
strings. This flag optionally resizes strings generated by either |
319 |
C<encode> or C<decode> to their minimum size possible. This can save |
320 |
memory when your JSON texts are either very very long or you have many |
321 |
short strings. It will also try to downgrade any strings to octet-form |
322 |
if possible: perl stores strings internally either in an encoding called |
323 |
UTF-X or in octet-form. The latter cannot store everything but uses less |
324 |
space in general (and some buggy Perl or C code might even rely on that |
325 |
internal representation being used). |
326 |
|
327 |
The actual definition of what shrink does might change in future versions, |
328 |
but it will always try to save space at the expense of time. |
329 |
|
330 |
If C<$enable> is true (or missing), the string returned by C<encode> will |
331 |
be shrunk-to-fit, while all strings generated by C<decode> will also be |
332 |
shrunk-to-fit. |
333 |
|
334 |
If C<$enable> is false, then the normal perl allocation algorithms are used. |
335 |
If you work with your data, then this is likely to be faster. |
336 |
|
337 |
In the future, this setting might control other things, such as converting |
338 |
strings that look like integers or floats into integers or floats |
339 |
internally (there is no difference on the Perl level), saving space. |
340 |
|
341 |
=item $json = $json->max_depth ([$maximum_nesting_depth]) |
342 |
|
343 |
Sets the maximum nesting level (default C<512>) accepted while encoding |
344 |
or decoding. If the JSON text or Perl data structure has an equal or |
345 |
higher nesting level then this limit, then the encoder and decoder will |
346 |
stop and croak at that point. |
347 |
|
348 |
Nesting level is defined by number of hash- or arrayrefs that the encoder |
349 |
needs to traverse to reach a given point or the number of C<{> or C<[> |
350 |
characters without their matching closing parenthesis crossed to reach a |
351 |
given character in a string. |
352 |
|
353 |
Setting the maximum depth to one disallows any nesting, so that ensures |
354 |
that the object is only a single hash/object or array. |
355 |
|
356 |
The argument to C<max_depth> will be rounded up to the next nearest power |
357 |
of two. |
358 |
|
359 |
See SECURITY CONSIDERATIONS, below, for more info on why this is useful. |
360 |
|
361 |
=item $json_text = $json->encode ($perl_scalar) |
362 |
|
363 |
Converts the given Perl data structure (a simple scalar or a reference |
364 |
to a hash or array) to its JSON representation. Simple scalars will be |
365 |
converted into JSON string or number sequences, while references to arrays |
366 |
become JSON arrays and references to hashes become JSON objects. Undefined |
367 |
Perl values (e.g. C<undef>) become JSON C<null> values. Neither C<true> |
368 |
nor C<false> values will be generated. |
369 |
|
370 |
=item $perl_scalar = $json->decode ($json_text) |
371 |
|
372 |
The opposite of C<encode>: expects a JSON text and tries to parse it, |
373 |
returning the resulting simple scalar or reference. Croaks on error. |
374 |
|
375 |
JSON numbers and strings become simple Perl scalars. JSON arrays become |
376 |
Perl arrayrefs and JSON objects become Perl hashrefs. C<true> becomes |
377 |
C<1>, C<false> becomes C<0> and C<null> becomes C<undef>. |
378 |
|
379 |
=item ($perl_scalar, $characters) = $json->decode_prefix ($json_text) |
380 |
|
381 |
This works like the C<decode> method, but instead of raising an exception |
382 |
when there is trailing garbage after the first JSON object, it will |
383 |
silently stop parsing there and return the number of characters consumed |
384 |
so far. |
385 |
|
386 |
This is useful if your JSON texts are not delimited by an outer protocol |
387 |
(which is not the brightest thing to do in the first place) and you need |
388 |
to know where the JSON text ends. |
389 |
|
390 |
JSON::XS->new->decode_prefix ("[1] the tail") |
391 |
=> ([], 3) |
392 |
|
393 |
=back |
394 |
|
395 |
|
396 |
=head1 MAPPING |
397 |
|
398 |
This section describes how JSON::XS maps Perl values to JSON values and |
399 |
vice versa. These mappings are designed to "do the right thing" in most |
400 |
circumstances automatically, preserving round-tripping characteristics |
401 |
(what you put in comes out as something equivalent). |
402 |
|
403 |
For the more enlightened: note that in the following descriptions, |
404 |
lowercase I<perl> refers to the Perl interpreter, while uppcercase I<Perl> |
405 |
refers to the abstract Perl language itself. |
406 |
|
407 |
=head2 JSON -> PERL |
408 |
|
409 |
=over 4 |
410 |
|
411 |
=item object |
412 |
|
413 |
A JSON object becomes a reference to a hash in Perl. No ordering of object |
414 |
keys is preserved (JSON does not preserver object key ordering itself). |
415 |
|
416 |
=item array |
417 |
|
418 |
A JSON array becomes a reference to an array in Perl. |
419 |
|
420 |
=item string |
421 |
|
422 |
A JSON string becomes a string scalar in Perl - Unicode codepoints in JSON |
423 |
are represented by the same codepoints in the Perl string, so no manual |
424 |
decoding is necessary. |
425 |
|
426 |
=item number |
427 |
|
428 |
A JSON number becomes either an integer or numeric (floating point) |
429 |
scalar in perl, depending on its range and any fractional parts. On the |
430 |
Perl level, there is no difference between those as Perl handles all the |
431 |
conversion details, but an integer may take slightly less memory and might |
432 |
represent more values exactly than (floating point) numbers. |
433 |
|
434 |
=item true, false |
435 |
|
436 |
These JSON atoms become C<0>, C<1>, respectively. Information is lost in |
437 |
this process. Future versions might represent those values differently, |
438 |
but they will be guarenteed to act like these integers would normally in |
439 |
Perl. |
440 |
|
441 |
=item null |
442 |
|
443 |
A JSON null atom becomes C<undef> in Perl. |
444 |
|
445 |
=back |
446 |
|
447 |
=head2 PERL -> JSON |
448 |
|
449 |
The mapping from Perl to JSON is slightly more difficult, as Perl is a |
450 |
truly typeless language, so we can only guess which JSON type is meant by |
451 |
a Perl value. |
452 |
|
453 |
=over 4 |
454 |
|
455 |
=item hash references |
456 |
|
457 |
Perl hash references become JSON objects. As there is no inherent ordering |
458 |
in hash keys (or JSON objects), they will usually be encoded in a |
459 |
pseudo-random order that can change between runs of the same program but |
460 |
stays generally the same within a single run of a program. JSON::XS can |
461 |
optionally sort the hash keys (determined by the I<canonical> flag), so |
462 |
the same datastructure will serialise to the same JSON text (given same |
463 |
settings and version of JSON::XS), but this incurs a runtime overhead |
464 |
and is only rarely useful, e.g. when you want to compare some JSON text |
465 |
against another for equality. |
466 |
|
467 |
=item array references |
468 |
|
469 |
Perl array references become JSON arrays. |
470 |
|
471 |
=item other references |
472 |
|
473 |
Other unblessed references are generally not allowed and will cause an |
474 |
exception to be thrown, except for references to the integers C<0> and |
475 |
C<1>, which get turned into C<false> and C<true> atoms in JSON. You can |
476 |
also use C<JSON::XS::false> and C<JSON::XS::true> to improve readability. |
477 |
|
478 |
to_json [\0,JSON::XS::true] # yields [false,true] |
479 |
|
480 |
=item blessed objects |
481 |
|
482 |
Blessed objects are not allowed. JSON::XS currently tries to encode their |
483 |
underlying representation (hash- or arrayref), but this behaviour might |
484 |
change in future versions. |
485 |
|
486 |
=item simple scalars |
487 |
|
488 |
Simple Perl scalars (any scalar that is not a reference) are the most |
489 |
difficult objects to encode: JSON::XS will encode undefined scalars as |
490 |
JSON null value, scalars that have last been used in a string context |
491 |
before encoding as JSON strings and anything else as number value: |
492 |
|
493 |
# dump as number |
494 |
to_json [2] # yields [2] |
495 |
to_json [-3.0e17] # yields [-3e+17] |
496 |
my $value = 5; to_json [$value] # yields [5] |
497 |
|
498 |
# used as string, so dump as string |
499 |
print $value; |
500 |
to_json [$value] # yields ["5"] |
501 |
|
502 |
# undef becomes null |
503 |
to_json [undef] # yields [null] |
504 |
|
505 |
You can force the type to be a string by stringifying it: |
506 |
|
507 |
my $x = 3.1; # some variable containing a number |
508 |
"$x"; # stringified |
509 |
$x .= ""; # another, more awkward way to stringify |
510 |
print $x; # perl does it for you, too, quite often |
511 |
|
512 |
You can force the type to be a number by numifying it: |
513 |
|
514 |
my $x = "3"; # some variable containing a string |
515 |
$x += 0; # numify it, ensuring it will be dumped as a number |
516 |
$x *= 1; # same thing, the choise is yours. |
517 |
|
518 |
You can not currently output JSON booleans or force the type in other, |
519 |
less obscure, ways. Tell me if you need this capability. |
520 |
|
521 |
=back |
522 |
|
523 |
|
524 |
=head1 COMPARISON |
525 |
|
526 |
As already mentioned, this module was created because none of the existing |
527 |
JSON modules could be made to work correctly. First I will describe the |
528 |
problems (or pleasures) I encountered with various existing JSON modules, |
529 |
followed by some benchmark values. JSON::XS was designed not to suffer |
530 |
from any of these problems or limitations. |
531 |
|
532 |
=over 4 |
533 |
|
534 |
=item JSON 1.07 |
535 |
|
536 |
Slow (but very portable, as it is written in pure Perl). |
537 |
|
538 |
Undocumented/buggy Unicode handling (how JSON handles unicode values is |
539 |
undocumented. One can get far by feeding it unicode strings and doing |
540 |
en-/decoding oneself, but unicode escapes are not working properly). |
541 |
|
542 |
No roundtripping (strings get clobbered if they look like numbers, e.g. |
543 |
the string C<2.0> will encode to C<2.0> instead of C<"2.0">, and that will |
544 |
decode into the number 2. |
545 |
|
546 |
=item JSON::PC 0.01 |
547 |
|
548 |
Very fast. |
549 |
|
550 |
Undocumented/buggy Unicode handling. |
551 |
|
552 |
No roundtripping. |
553 |
|
554 |
Has problems handling many Perl values (e.g. regex results and other magic |
555 |
values will make it croak). |
556 |
|
557 |
Does not even generate valid JSON (C<{1,2}> gets converted to C<{1:2}> |
558 |
which is not a valid JSON text. |
559 |
|
560 |
Unmaintained (maintainer unresponsive for many months, bugs are not |
561 |
getting fixed). |
562 |
|
563 |
=item JSON::Syck 0.21 |
564 |
|
565 |
Very buggy (often crashes). |
566 |
|
567 |
Very inflexible (no human-readable format supported, format pretty much |
568 |
undocumented. I need at least a format for easy reading by humans and a |
569 |
single-line compact format for use in a protocol, and preferably a way to |
570 |
generate ASCII-only JSON texts). |
571 |
|
572 |
Completely broken (and confusingly documented) Unicode handling (unicode |
573 |
escapes are not working properly, you need to set ImplicitUnicode to |
574 |
I<different> values on en- and decoding to get symmetric behaviour). |
575 |
|
576 |
No roundtripping (simple cases work, but this depends on wether the scalar |
577 |
value was used in a numeric context or not). |
578 |
|
579 |
Dumping hashes may skip hash values depending on iterator state. |
580 |
|
581 |
Unmaintained (maintainer unresponsive for many months, bugs are not |
582 |
getting fixed). |
583 |
|
584 |
Does not check input for validity (i.e. will accept non-JSON input and |
585 |
return "something" instead of raising an exception. This is a security |
586 |
issue: imagine two banks transfering money between each other using |
587 |
JSON. One bank might parse a given non-JSON request and deduct money, |
588 |
while the other might reject the transaction with a syntax error. While a |
589 |
good protocol will at least recover, that is extra unnecessary work and |
590 |
the transaction will still not succeed). |
591 |
|
592 |
=item JSON::DWIW 0.04 |
593 |
|
594 |
Very fast. Very natural. Very nice. |
595 |
|
596 |
Undocumented unicode handling (but the best of the pack. Unicode escapes |
597 |
still don't get parsed properly). |
598 |
|
599 |
Very inflexible. |
600 |
|
601 |
No roundtripping. |
602 |
|
603 |
Does not generate valid JSON texts (key strings are often unquoted, empty keys |
604 |
result in nothing being output) |
605 |
|
606 |
Does not check input for validity. |
607 |
|
608 |
=back |
609 |
|
610 |
=head2 SPEED |
611 |
|
612 |
It seems that JSON::XS is surprisingly fast, as shown in the following |
613 |
tables. They have been generated with the help of the C<eg/bench> program |
614 |
in the JSON::XS distribution, to make it easy to compare on your own |
615 |
system. |
616 |
|
617 |
First comes a comparison between various modules using a very short JSON |
618 |
string: |
619 |
|
620 |
{"method": "handleMessage", "params": ["user1", "we were just talking"], "id": null} |
621 |
|
622 |
It shows the number of encodes/decodes per second (JSON::XS uses the |
623 |
functional interface, while JSON::XS/2 uses the OO interface with |
624 |
pretty-printing and hashkey sorting enabled). Higher is better: |
625 |
|
626 |
module | encode | decode | |
627 |
-----------|------------|------------| |
628 |
JSON | 11488.516 | 7823.035 | |
629 |
JSON::DWIW | 94708.054 | 129094.260 | |
630 |
JSON::PC | 63884.157 | 128528.212 | |
631 |
JSON::Syck | 34898.677 | 42096.911 | |
632 |
JSON::XS | 654027.064 | 396423.669 | |
633 |
JSON::XS/2 | 371564.190 | 371725.613 | |
634 |
-----------+------------+------------+ |
635 |
|
636 |
That is, JSON::XS is more than six times faster than JSON::DWIW on |
637 |
encoding, more than three times faster on decoding, and about thirty times |
638 |
faster than JSON, even with pretty-printing and key sorting. |
639 |
|
640 |
Using a longer test string (roughly 18KB, generated from Yahoo! Locals |
641 |
search API (http://nanoref.com/yahooapis/mgPdGg): |
642 |
|
643 |
module | encode | decode | |
644 |
-----------|------------|------------| |
645 |
JSON | 273.023 | 44.674 | |
646 |
JSON::DWIW | 1089.383 | 1145.704 | |
647 |
JSON::PC | 3097.419 | 2393.921 | |
648 |
JSON::Syck | 514.060 | 843.053 | |
649 |
JSON::XS | 6479.668 | 3636.364 | |
650 |
JSON::XS/2 | 3774.221 | 3599.124 | |
651 |
-----------+------------+------------+ |
652 |
|
653 |
Again, JSON::XS leads by far. |
654 |
|
655 |
On large strings containing lots of high unicode characters, some modules |
656 |
(such as JSON::PC) seem to decode faster than JSON::XS, but the result |
657 |
will be broken due to missing (or wrong) unicode handling. Others refuse |
658 |
to decode or encode properly, so it was impossible to prepare a fair |
659 |
comparison table for that case. |
660 |
|
661 |
|
662 |
=head1 SECURITY CONSIDERATIONS |
663 |
|
664 |
When you are using JSON in a protocol, talking to untrusted potentially |
665 |
hostile creatures requires relatively few measures. |
666 |
|
667 |
First of all, your JSON decoder should be secure, that is, should not have |
668 |
any buffer overflows. Obviously, this module should ensure that and I am |
669 |
trying hard on making that true, but you never know. |
670 |
|
671 |
Second, you need to avoid resource-starving attacks. That means you should |
672 |
limit the size of JSON texts you accept, or make sure then when your |
673 |
resources run out, thats just fine (e.g. by using a separate process that |
674 |
can crash safely). The size of a JSON text in octets or characters is |
675 |
usually a good indication of the size of the resources required to decode |
676 |
it into a Perl structure. |
677 |
|
678 |
Third, JSON::XS recurses using the C stack when decoding objects and |
679 |
arrays. The C stack is a limited resource: for instance, on my amd64 |
680 |
machine with 8MB of stack size I can decode around 180k nested arrays but |
681 |
only 14k nested JSON objects (due to perl itself recursing deeply on croak |
682 |
to free the temporary). If that is exceeded, the program crashes. to be |
683 |
conservative, the default nesting limit is set to 512. If your process |
684 |
has a smaller stack, you should adjust this setting accordingly with the |
685 |
C<max_depth> method. |
686 |
|
687 |
And last but least, something else could bomb you that I forgot to think |
688 |
of. In that case, you get to keep the pieces. I am always open for hints, |
689 |
though... |
690 |
|
691 |
|
692 |
=head1 BUGS |
693 |
|
694 |
While the goal of this module is to be correct, that unfortunately does |
695 |
not mean its bug-free, only that I think its design is bug-free. It is |
696 |
still relatively early in its development. If you keep reporting bugs they |
697 |
will be fixed swiftly, though. |
698 |
|
699 |
=cut |
700 |
|
701 |
sub true() { \1 } |
702 |
sub false() { \0 } |
703 |
|
704 |
1; |
705 |
|
706 |
=head1 AUTHOR |
707 |
|
708 |
Marc Lehmann <schmorp@schmorp.de> |
709 |
http://home.schmorp.de/ |
710 |
|
711 |
=cut |
712 |
|