… | |
… | |
9 | # exported functions, they croak on error |
9 | # exported functions, they croak on error |
10 | # and expect/generate UTF-8 |
10 | # and expect/generate UTF-8 |
11 | |
11 | |
12 | $utf8_encoded_json_text = to_json $perl_hash_or_arrayref; |
12 | $utf8_encoded_json_text = to_json $perl_hash_or_arrayref; |
13 | $perl_hash_or_arrayref = from_json $utf8_encoded_json_text; |
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 | |
14 | |
19 | # OO-interface |
15 | # OO-interface |
20 | |
16 | |
21 | $coder = JSON::XS->new->ascii->pretty->allow_nonref; |
17 | $coder = JSON::XS->new->ascii->pretty->allow_nonref; |
22 | $pretty_printed_unencoded = $coder->encode ($perl_scalar); |
18 | $pretty_printed_unencoded = $coder->encode ($perl_scalar); |
… | |
… | |
85 | |
81 | |
86 | package JSON::XS; |
82 | package JSON::XS; |
87 | |
83 | |
88 | use strict; |
84 | use strict; |
89 | |
85 | |
90 | our $VERSION = '1.4'; |
86 | our $VERSION = '1.5'; |
91 | our @ISA = qw(Exporter); |
87 | our @ISA = qw(Exporter); |
92 | |
88 | |
93 | our @EXPORT = qw(to_json from_json objToJson jsonToObj); |
89 | our @EXPORT = qw(to_json from_json); |
94 | |
90 | |
95 | use Exporter; |
91 | use Exporter; |
96 | use XSLoader; |
92 | use XSLoader; |
97 | |
93 | |
98 | =head1 FUNCTIONAL INTERFACE |
94 | =head1 FUNCTIONAL INTERFACE |
… | |
… | |
282 | |
278 | |
283 | Example, space_before and indent disabled, space_after enabled: |
279 | Example, space_before and indent disabled, space_after enabled: |
284 | |
280 | |
285 | {"key": "value"} |
281 | {"key": "value"} |
286 | |
282 | |
|
|
283 | =item $json = $json->relaxed ([$enable]) |
|
|
284 | |
|
|
285 | If C<$enable> is true (or missing), then C<decode> will accept some |
|
|
286 | extensions to normal JSON syntax (see below). C<encode> will not be |
|
|
287 | affected in anyway. I<Be aware that this option makes you accept invalid |
|
|
288 | JSON texts as if they were valid!>. I suggest only to use this option to |
|
|
289 | parse application-specific files written by humans (configuration files, |
|
|
290 | resource files etc.) |
|
|
291 | |
|
|
292 | If C<$enable> is false (the default), then C<decode> will only accept |
|
|
293 | valid JSON texts. |
|
|
294 | |
|
|
295 | Currently accepted extensions are: |
|
|
296 | |
|
|
297 | =over 4 |
|
|
298 | |
|
|
299 | =item * list items can have an end-comma |
|
|
300 | |
|
|
301 | JSON I<separates> array elements and key-value pairs with commas. This |
|
|
302 | can be annoying if you write JSON texts manually and want to be able to |
|
|
303 | quickly append elements, so this extension accepts comma at the end of |
|
|
304 | such items not just between them: |
|
|
305 | |
|
|
306 | [ |
|
|
307 | 1, |
|
|
308 | 2, <- this comma not normally allowed |
|
|
309 | ] |
|
|
310 | { |
|
|
311 | "k1": "v1", |
|
|
312 | "k2": "v2", <- this comma not normally allowed |
|
|
313 | } |
|
|
314 | |
|
|
315 | =item * shell-style '#'-comments |
|
|
316 | |
|
|
317 | Whenever JSON allows whitespace, shell-style comments are additionally |
|
|
318 | allowed. They are terminated by the first carriage-return or line-feed |
|
|
319 | character, after which more white-space and comments are allowed. |
|
|
320 | |
|
|
321 | [ |
|
|
322 | 1, # this comment not allowed in JSON |
|
|
323 | # neither this one... |
|
|
324 | ] |
|
|
325 | |
|
|
326 | =back |
|
|
327 | |
287 | =item $json = $json->canonical ([$enable]) |
328 | =item $json = $json->canonical ([$enable]) |
288 | |
329 | |
289 | If C<$enable> is true (or missing), then the C<encode> method will output JSON objects |
330 | If C<$enable> is true (or missing), then the C<encode> method will output JSON objects |
290 | by sorting their keys. This is adding a comparatively high overhead. |
331 | by sorting their keys. This is adding a comparatively high overhead. |
291 | |
332 | |
… | |
… | |
341 | |
382 | |
342 | The C<TO_JSON> method may safely call die if it wants. If C<TO_JSON> |
383 | The C<TO_JSON> method may safely call die if it wants. If C<TO_JSON> |
343 | returns other blessed objects, those will be handled in the same |
384 | returns other blessed objects, those will be handled in the same |
344 | way. C<TO_JSON> must take care of not causing an endless recursion cycle |
385 | way. C<TO_JSON> must take care of not causing an endless recursion cycle |
345 | (== crash) in this case. The name of C<TO_JSON> was chosen because other |
386 | (== crash) in this case. The name of C<TO_JSON> was chosen because other |
346 | methods called by the Perl core (== not the user of the object) are |
387 | methods called by the Perl core (== not by the user of the object) are |
347 | usually in upper case letters and to avoid collisions with the C<to_json> |
388 | usually in upper case letters and to avoid collisions with the C<to_json> |
348 | function. |
389 | function. |
349 | |
390 | |
350 | This setting does not yet influence C<decode> in any way, but in the |
391 | This setting does not yet influence C<decode> in any way, but in the |
351 | future, global hooks might get installed that influence C<decode> and are |
392 | future, global hooks might get installed that influence C<decode> and are |
352 | enabled by this setting. |
393 | enabled by this setting. |
353 | |
394 | |
354 | If C<$enable> is false, then the C<allow_blessed> setting will decide what |
395 | If C<$enable> is false, then the C<allow_blessed> setting will decide what |
355 | to do when a blessed object is found. |
396 | to do when a blessed object is found. |
|
|
397 | |
|
|
398 | =item $json = $json->filter_json_object ([$coderef->($hashref)]) |
|
|
399 | |
|
|
400 | When C<$coderef> is specified, it will be called from C<decode> each |
|
|
401 | time it decodes a JSON object. The only argument is a reference to the |
|
|
402 | newly-created hash. If the code references returns a single scalar (which |
|
|
403 | need not be a reference), this value (i.e. a copy of that scalar to avoid |
|
|
404 | aliasing) is inserted into the deserialised data structure. If it returns |
|
|
405 | an empty list (NOTE: I<not> C<undef>, which is a valid scalar), the |
|
|
406 | original deserialised hash will be inserted. This setting can slow down |
|
|
407 | decoding considerably. |
|
|
408 | |
|
|
409 | When C<$coderef> is omitted or undefined, any existing callback will |
|
|
410 | be removed and C<decode> will not change the deserialised hash in any |
|
|
411 | way. |
|
|
412 | |
|
|
413 | Example, convert all JSON objects into the integer 5: |
|
|
414 | |
|
|
415 | my $js = JSON::XS->new->filter_json_object (sub { 5 }); |
|
|
416 | # returns [5] |
|
|
417 | $js->decode ('[{}]') |
|
|
418 | # throw an exception because allow_nonref is not enabled |
|
|
419 | # so a lone 5 is not allowed. |
|
|
420 | $js->decode ('{"a":1, "b":2}'); |
|
|
421 | |
|
|
422 | =item $json = $json->filter_json_single_key_object ($key [=> $coderef->($value)]) |
|
|
423 | |
|
|
424 | Works remotely similar to C<filter_json_object>, but is only called for |
|
|
425 | JSON objects having a single key named C<$key>. |
|
|
426 | |
|
|
427 | This C<$coderef> is called before the one specified via |
|
|
428 | C<filter_json_object>, if any. It gets passed the single value in the JSON |
|
|
429 | object. If it returns a single value, it will be inserted into the data |
|
|
430 | structure. If it returns nothing (not even C<undef> but the empty list), |
|
|
431 | the callback from C<filter_json_object> will be called next, as if no |
|
|
432 | single-key callback were specified. |
|
|
433 | |
|
|
434 | If C<$coderef> is omitted or undefined, the corresponding callback will be |
|
|
435 | disabled. There can only ever be one callback for a given key. |
|
|
436 | |
|
|
437 | As this callback gets called less often then the C<filter_json_object> |
|
|
438 | one, decoding speed will not usually suffer as much. Therefore, single-key |
|
|
439 | objects make excellent targets to serialise Perl objects into, especially |
|
|
440 | as single-key JSON objects are as close to the type-tagged value concept |
|
|
441 | as JSON gets (its basically an ID/VALUE tuple). Of course, JSON does not |
|
|
442 | support this in any way, so you need to make sure your data never looks |
|
|
443 | like a serialised Perl hash. |
|
|
444 | |
|
|
445 | Typical names for the single object key are C<__class_whatever__>, or |
|
|
446 | C<$__dollars_are_rarely_used__$> or C<}ugly_brace_placement>, or even |
|
|
447 | things like C<__class_md5sum(classname)__>, to reduce the risk of clashing |
|
|
448 | with real hashes. |
|
|
449 | |
|
|
450 | Example, decode JSON objects of the form C<< { "__widget__" => <id> } >> |
|
|
451 | into the corresponding C<< $WIDGET{<id>} >> object: |
|
|
452 | |
|
|
453 | # return whatever is in $WIDGET{5}: |
|
|
454 | JSON::XS |
|
|
455 | ->new |
|
|
456 | ->filter_json_single_key_object (__widget__ => sub { |
|
|
457 | $WIDGET{ $_[0] } |
|
|
458 | }) |
|
|
459 | ->decode ('{"__widget__": 5') |
|
|
460 | |
|
|
461 | # this can be used with a TO_JSON method in some "widget" class |
|
|
462 | # for serialisation to json: |
|
|
463 | sub WidgetBase::TO_JSON { |
|
|
464 | my ($self) = @_; |
|
|
465 | |
|
|
466 | unless ($self->{id}) { |
|
|
467 | $self->{id} = ..get..some..id..; |
|
|
468 | $WIDGET{$self->{id}} = $self; |
|
|
469 | } |
|
|
470 | |
|
|
471 | { __widget__ => $self->{id} } |
|
|
472 | } |
356 | |
473 | |
357 | =item $json = $json->shrink ([$enable]) |
474 | =item $json = $json->shrink ([$enable]) |
358 | |
475 | |
359 | Perl usually over-allocates memory a bit when allocating space for |
476 | Perl usually over-allocates memory a bit when allocating space for |
360 | strings. This flag optionally resizes strings generated by either |
477 | strings. This flag optionally resizes strings generated by either |
… | |
… | |
393 | given character in a string. |
510 | given character in a string. |
394 | |
511 | |
395 | Setting the maximum depth to one disallows any nesting, so that ensures |
512 | Setting the maximum depth to one disallows any nesting, so that ensures |
396 | that the object is only a single hash/object or array. |
513 | that the object is only a single hash/object or array. |
397 | |
514 | |
398 | The argument to C<max_depth> will be rounded up to the next nearest power |
515 | The argument to C<max_depth> will be rounded up to the next highest power |
399 | of two. |
516 | of two. If no argument is given, the highest possible setting will be |
|
|
517 | used, which is rarely useful. |
|
|
518 | |
|
|
519 | See SECURITY CONSIDERATIONS, below, for more info on why this is useful. |
|
|
520 | |
|
|
521 | =item $json = $json->max_size ([$maximum_string_size]) |
|
|
522 | |
|
|
523 | Set the maximum length a JSON text may have (in bytes) where decoding is |
|
|
524 | being attempted. The default is C<0>, meaning no limit. When C<decode> |
|
|
525 | is called on a string longer then this number of characters it will not |
|
|
526 | attempt to decode the string but throw an exception. This setting has no |
|
|
527 | effect on C<encode> (yet). |
|
|
528 | |
|
|
529 | The argument to C<max_size> will be rounded up to the next B<highest> |
|
|
530 | power of two (so may be more than requested). If no argument is given, the |
|
|
531 | limit check will be deactivated (same as when C<0> is specified). |
400 | |
532 | |
401 | See SECURITY CONSIDERATIONS, below, for more info on why this is useful. |
533 | See SECURITY CONSIDERATIONS, below, for more info on why this is useful. |
402 | |
534 | |
403 | =item $json_text = $json->encode ($perl_scalar) |
535 | =item $json_text = $json->encode ($perl_scalar) |
404 | |
536 | |
… | |
… | |
466 | are represented by the same codepoints in the Perl string, so no manual |
598 | are represented by the same codepoints in the Perl string, so no manual |
467 | decoding is necessary. |
599 | decoding is necessary. |
468 | |
600 | |
469 | =item number |
601 | =item number |
470 | |
602 | |
471 | A JSON number becomes either an integer or numeric (floating point) |
603 | A JSON number becomes either an integer, numeric (floating point) or |
472 | scalar in perl, depending on its range and any fractional parts. On the |
604 | string scalar in perl, depending on its range and any fractional parts. On |
473 | Perl level, there is no difference between those as Perl handles all the |
605 | the Perl level, there is no difference between those as Perl handles all |
474 | conversion details, but an integer may take slightly less memory and might |
606 | the conversion details, but an integer may take slightly less memory and |
475 | represent more values exactly than (floating point) numbers. |
607 | might represent more values exactly than (floating point) numbers. |
|
|
608 | |
|
|
609 | If the number consists of digits only, JSON::XS will try to represent |
|
|
610 | it as an integer value. If that fails, it will try to represent it as |
|
|
611 | a numeric (floating point) value if that is possible without loss of |
|
|
612 | precision. Otherwise it will preserve the number as a string value. |
|
|
613 | |
|
|
614 | Numbers containing a fractional or exponential part will always be |
|
|
615 | represented as numeric (floating point) values, possibly at a loss of |
|
|
616 | precision. |
|
|
617 | |
|
|
618 | This might create round-tripping problems as numbers might become strings, |
|
|
619 | but as Perl is typeless there is no other way to do it. |
476 | |
620 | |
477 | =item true, false |
621 | =item true, false |
478 | |
622 | |
479 | These JSON atoms become C<JSON::XS::true> and C<JSON::XS::false>, |
623 | These JSON atoms become C<JSON::XS::true> and C<JSON::XS::false>, |
480 | respectively. They are overloaded to act almost exactly like the numbers |
624 | respectively. They are overloaded to act almost exactly like the numbers |
… | |
… | |
696 | It shows the number of encodes/decodes per second (JSON::XS uses |
840 | It shows the number of encodes/decodes per second (JSON::XS uses |
697 | the functional interface, while JSON::XS/2 uses the OO interface |
841 | the functional interface, while JSON::XS/2 uses the OO interface |
698 | with pretty-printing and hashkey sorting enabled, JSON::XS/3 enables |
842 | with pretty-printing and hashkey sorting enabled, JSON::XS/3 enables |
699 | shrink). Higher is better: |
843 | shrink). Higher is better: |
700 | |
844 | |
|
|
845 | Storable | 15779.925 | 14169.946 | |
|
|
846 | -----------+------------+------------+ |
701 | module | encode | decode | |
847 | module | encode | decode | |
702 | -----------|------------|------------| |
848 | -----------|------------|------------| |
703 | JSON | 7645.468 | 4208.613 | |
849 | JSON | 4990.842 | 4088.813 | |
704 | JSON::DWIW | 40721.398 | 77101.176 | |
850 | JSON::DWIW | 51653.990 | 71575.154 | |
705 | JSON::PC | 65948.176 | 78251.940 | |
851 | JSON::PC | 65948.176 | 74631.744 | |
706 | JSON::Syck | 22844.793 | 26479.192 | |
852 | JSON::PP | 8931.652 | 3817.168 | |
|
|
853 | JSON::Syck | 24877.248 | 27776.848 | |
707 | JSON::XS | 388361.481 | 199728.762 | |
854 | JSON::XS | 388361.481 | 227951.304 | |
708 | JSON::XS/2 | 218453.333 | 192399.266 | |
855 | JSON::XS/2 | 227951.304 | 218453.333 | |
709 | JSON::XS/3 | 338250.323 | 192399.266 | |
856 | JSON::XS/3 | 338250.323 | 218453.333 | |
710 | Storable | 15779.925 | 14169.946 | |
857 | Storable | 16500.016 | 135300.129 | |
711 | -----------+------------+------------+ |
858 | -----------+------------+------------+ |
712 | |
859 | |
713 | That is, JSON::XS is about five times faster than JSON::DWIW on encoding, |
860 | That is, JSON::XS is about five times faster than JSON::DWIW on encoding, |
714 | about three times faster on decoding, and over fourty times faster |
861 | about three times faster on decoding, and over fourty times faster |
715 | than JSON, even with pretty-printing and key sorting. It also compares |
862 | than JSON, even with pretty-printing and key sorting. It also compares |
… | |
… | |
718 | Using a longer test string (roughly 18KB, generated from Yahoo! Locals |
865 | Using a longer test string (roughly 18KB, generated from Yahoo! Locals |
719 | search API (http://nanoref.com/yahooapis/mgPdGg): |
866 | search API (http://nanoref.com/yahooapis/mgPdGg): |
720 | |
867 | |
721 | module | encode | decode | |
868 | module | encode | decode | |
722 | -----------|------------|------------| |
869 | -----------|------------|------------| |
723 | JSON | 254.685 | 37.665 | |
870 | JSON | 55.260 | 34.971 | |
724 | JSON::DWIW | 843.343 | 1049.731 | |
871 | JSON::DWIW | 825.228 | 1082.513 | |
725 | JSON::PC | 3602.116 | 2307.352 | |
872 | JSON::PC | 3571.444 | 2394.829 | |
|
|
873 | JSON::PP | 210.987 | 32.574 | |
726 | JSON::Syck | 505.107 | 787.899 | |
874 | JSON::Syck | 552.551 | 787.544 | |
727 | JSON::XS | 5747.196 | 3690.220 | |
875 | JSON::XS | 5780.463 | 4854.519 | |
728 | JSON::XS/2 | 3968.121 | 3676.634 | |
876 | JSON::XS/2 | 3869.998 | 4798.975 | |
729 | JSON::XS/3 | 6105.246 | 3662.508 | |
877 | JSON::XS/3 | 5862.880 | 4798.975 | |
730 | Storable | 4417.337 | 5285.161 | |
878 | Storable | 4445.002 | 5235.027 | |
731 | -----------+------------+------------+ |
879 | -----------+------------+------------+ |
732 | |
880 | |
733 | Again, JSON::XS leads by far (except for Storable which non-surprisingly |
881 | Again, JSON::XS leads by far (except for Storable which non-surprisingly |
734 | decodes faster). |
882 | decodes faster). |
735 | |
883 | |
… | |
… | |
752 | Second, you need to avoid resource-starving attacks. That means you should |
900 | Second, you need to avoid resource-starving attacks. That means you should |
753 | limit the size of JSON texts you accept, or make sure then when your |
901 | limit the size of JSON texts you accept, or make sure then when your |
754 | resources run out, thats just fine (e.g. by using a separate process that |
902 | resources run out, thats just fine (e.g. by using a separate process that |
755 | can crash safely). The size of a JSON text in octets or characters is |
903 | can crash safely). The size of a JSON text in octets or characters is |
756 | usually a good indication of the size of the resources required to decode |
904 | usually a good indication of the size of the resources required to decode |
757 | it into a Perl structure. |
905 | it into a Perl structure. While JSON::XS can check the size of the JSON |
|
|
906 | text, it might be too late when you already have it in memory, so you |
|
|
907 | might want to check the size before you accept the string. |
758 | |
908 | |
759 | Third, JSON::XS recurses using the C stack when decoding objects and |
909 | Third, JSON::XS recurses using the C stack when decoding objects and |
760 | arrays. The C stack is a limited resource: for instance, on my amd64 |
910 | arrays. The C stack is a limited resource: for instance, on my amd64 |
761 | machine with 8MB of stack size I can decode around 180k nested arrays but |
911 | machine with 8MB of stack size I can decode around 180k nested arrays but |
762 | only 14k nested JSON objects (due to perl itself recursing deeply on croak |
912 | only 14k nested JSON objects (due to perl itself recursing deeply on croak |