… | |
… | |
86 | package JSON::XS; |
86 | package JSON::XS; |
87 | |
87 | |
88 | use strict; |
88 | use strict; |
89 | |
89 | |
90 | BEGIN { |
90 | BEGIN { |
91 | our $VERSION = '1.22'; |
91 | our $VERSION = '1.24'; |
92 | our @ISA = qw(Exporter); |
92 | our @ISA = qw(Exporter); |
93 | |
93 | |
94 | our @EXPORT = qw(to_json from_json objToJson jsonToObj); |
94 | our @EXPORT = qw(to_json from_json objToJson jsonToObj); |
95 | require Exporter; |
95 | require Exporter; |
96 | |
96 | |
… | |
… | |
402 | |
402 | |
403 | For the more enlightened: note that in the following descriptions, |
403 | For the more enlightened: note that in the following descriptions, |
404 | lowercase I<perl> refers to the Perl interpreter, while uppcercase I<Perl> |
404 | lowercase I<perl> refers to the Perl interpreter, while uppcercase I<Perl> |
405 | refers to the abstract Perl language itself. |
405 | refers to the abstract Perl language itself. |
406 | |
406 | |
|
|
407 | |
407 | =head2 JSON -> PERL |
408 | =head2 JSON -> PERL |
408 | |
409 | |
409 | =over 4 |
410 | =over 4 |
410 | |
411 | |
411 | =item object |
412 | =item object |
… | |
… | |
441 | =item null |
442 | =item null |
442 | |
443 | |
443 | A JSON null atom becomes C<undef> in Perl. |
444 | A JSON null atom becomes C<undef> in Perl. |
444 | |
445 | |
445 | =back |
446 | =back |
|
|
447 | |
446 | |
448 | |
447 | =head2 PERL -> JSON |
449 | =head2 PERL -> JSON |
448 | |
450 | |
449 | The mapping from Perl to JSON is slightly more difficult, as Perl is a |
451 | 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 |
452 | truly typeless language, so we can only guess which JSON type is meant by |
… | |
… | |
605 | |
607 | |
606 | Does not check input for validity. |
608 | Does not check input for validity. |
607 | |
609 | |
608 | =back |
610 | =back |
609 | |
611 | |
|
|
612 | |
|
|
613 | =head2 JSON and YAML |
|
|
614 | |
|
|
615 | You often hear that JSON is a subset (or a close subset) of YAML. This is, |
|
|
616 | however, a mass hysteria and very far from the truth. In general, there is |
|
|
617 | no way to configure JSON::XS to output a data structure as valid YAML. |
|
|
618 | |
|
|
619 | If you really must use JSON::XS to generate YAML, you should use this |
|
|
620 | algorithm (subject to change in future versions): |
|
|
621 | |
|
|
622 | my $to_yaml = JSON::XS->new->utf8->space_after (1); |
|
|
623 | my $yaml = $to_yaml->encode ($ref) . "\n"; |
|
|
624 | |
|
|
625 | This will usually generate JSON texts that also parse as valid |
|
|
626 | YAML. Please note that YAML has hardcoded limits on (simple) object key |
|
|
627 | lengths that JSON doesn't have, so you should make sure that your hash |
|
|
628 | keys are noticably shorter than the 1024 characters YAML allows. |
|
|
629 | |
|
|
630 | There might be other incompatibilities that I am not aware of. In general |
|
|
631 | you should not try to generate YAML with a JSON generator or vice versa, |
|
|
632 | or try to parse JSON with a YAML parser or vice versa: chances are high |
|
|
633 | that you will run into severe interoperability problems. |
|
|
634 | |
|
|
635 | |
610 | =head2 SPEED |
636 | =head2 SPEED |
611 | |
637 | |
612 | It seems that JSON::XS is surprisingly fast, as shown in the following |
638 | 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 |
639 | 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 |
640 | in the JSON::XS distribution, to make it easy to compare on your own |
… | |
… | |
616 | |
642 | |
617 | First comes a comparison between various modules using a very short |
643 | First comes a comparison between various modules using a very short |
618 | single-line JSON string: |
644 | single-line JSON string: |
619 | |
645 | |
620 | {"method": "handleMessage", "params": ["user1", "we were just talking"], \ |
646 | {"method": "handleMessage", "params": ["user1", "we were just talking"], \ |
621 | "id": null, [1,11,234,-5,1e5,1e7, true, false]} |
647 | "id": null, "array":[1,11,234,-5,1e5,1e7, true, false]} |
622 | |
648 | |
623 | It shows the number of encodes/decodes per second (JSON::XS uses the |
649 | It shows the number of encodes/decodes per second (JSON::XS uses |
624 | functional interface, while JSON::XS/2 uses the OO interface with |
650 | the functional interface, while JSON::XS/2 uses the OO interface |
625 | pretty-printing and hashkey sorting enabled). Higher is better: |
651 | with pretty-printing and hashkey sorting enabled, JSON::XS/3 enables |
|
|
652 | shrink). Higher is better: |
626 | |
653 | |
627 | -----------+------------+------------+ |
|
|
628 | module | encode | decode | |
654 | module | encode | decode | |
629 | -----------|------------|------------| |
655 | -----------|------------|------------| |
630 | JSON | 10597.029 | 5740.903 | |
656 | JSON | 7645.468 | 4208.613 | |
631 | JSON::DWIW | 78251.940 | 98457.840 | |
657 | JSON::DWIW | 40721.398 | 77101.176 | |
632 | JSON::PC | 70611.178 | 92794.336 | |
658 | JSON::PC | 65948.176 | 78251.940 | |
633 | JSON::Syck | 28767.517 | 38199.490 | |
659 | JSON::Syck | 22844.793 | 26479.192 | |
634 | JSON::XS | 419430.400 | 265462.278 | |
|
|
635 | JSON::XS/2 | 279620.267 | 265462.278 | |
|
|
636 | JSON::XS/3 | 388361.481 | 265462.278 | |
660 | JSON::XS | 388361.481 | 199728.762 | |
637 | Storable | 16294.887 | 16844.594 | |
661 | JSON::XS/2 | 218453.333 | 192399.266 | |
|
|
662 | JSON::XS/3 | 338250.323 | 192399.266 | |
|
|
663 | Storable | 15779.925 | 14169.946 | |
638 | -----------+------------+------------+ |
664 | -----------+------------+------------+ |
639 | |
665 | |
640 | That is, JSON::XS is about five times faster than JSON::DWIW on encoding, |
666 | That is, JSON::XS is about five times faster than JSON::DWIW on encoding, |
641 | about three times faster on decoding, and about fourty times faster |
667 | about three times faster on decoding, and over fourty times faster |
642 | than JSON, even with pretty-printing and key sorting. It also compares |
668 | than JSON, even with pretty-printing and key sorting. It also compares |
643 | favourably to Storable for small amounts of data. |
669 | favourably to Storable for small amounts of data. |
644 | |
670 | |
645 | Using a longer test string (roughly 18KB, generated from Yahoo! Locals |
671 | Using a longer test string (roughly 18KB, generated from Yahoo! Locals |
646 | search API (http://nanoref.com/yahooapis/mgPdGg): |
672 | search API (http://nanoref.com/yahooapis/mgPdGg): |
647 | |
673 | |
648 | module | encode | decode | |
674 | module | encode | decode | |
649 | -----------|------------|------------| |
675 | -----------|------------|------------| |
650 | JSON | 254.685 | 37.665 | |
676 | JSON | 254.685 | 37.665 | |
651 | JSON::DWIW | 1014.244 | 1087.678 | |
677 | JSON::DWIW | 843.343 | 1049.731 | |
652 | JSON::PC | 3602.116 | 2307.352 | |
678 | JSON::PC | 3602.116 | 2307.352 | |
653 | JSON::Syck | 558.035 | 776.263 | |
679 | JSON::Syck | 505.107 | 787.899 | |
654 | JSON::XS | 5747.196 | 3543.684 | |
680 | JSON::XS | 5747.196 | 3690.220 | |
655 | JSON::XS/2 | 3968.121 | 3589.170 | |
681 | JSON::XS/2 | 3968.121 | 3676.634 | |
656 | JSON::XS/3 | 6105.246 | 3561.134 | |
682 | JSON::XS/3 | 6105.246 | 3662.508 | |
657 | Storable | 4456.337 | 5320.020 | |
683 | Storable | 4417.337 | 5285.161 | |
658 | -----------+------------+------------+ |
684 | -----------+------------+------------+ |
659 | |
685 | |
660 | Again, JSON::XS leads by far. |
686 | Again, JSON::XS leads by far (except for Storable which non-surprisingly |
|
|
687 | decodes faster). |
661 | |
688 | |
662 | On large strings containing lots of high unicode characters, some modules |
689 | On large strings containing lots of high unicode characters, some modules |
663 | (such as JSON::PC) seem to decode faster than JSON::XS, but the result |
690 | (such as JSON::PC) seem to decode faster than JSON::XS, but the result |
664 | will be broken due to missing (or wrong) unicode handling. Others refuse |
691 | will be broken due to missing (or wrong) unicode handling. Others refuse |
665 | to decode or encode properly, so it was impossible to prepare a fair |
692 | to decode or encode properly, so it was impossible to prepare a fair |