Skip to content
This repository


Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
Fetching contributors…


Cannot retrieve contributors at this time

file 1459 lines (1139 sloc) 66.219 kb
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458
    Cpanel::JSON::XS - JSON::XS for Cpanel, fast and correct serialising,
    also for 5.6.2

     use Cpanel::JSON::XS;

     # exported functions, they croak on error
     # and expect/generate UTF-8

     $utf8_encoded_json_text = encode_json $perl_hash_or_arrayref;
     $perl_hash_or_arrayref = decode_json $utf8_encoded_json_text;

     # OO-interface

     $coder = Cpanel::JSON::XS->new->ascii->pretty->allow_nonref;
     $pretty_printed_unencoded = $coder->encode ($perl_scalar);
     $perl_scalar = $coder->decode ($unicode_json_text);

     # Note that 5.6 misses most smart utf8 and encoding functionalities
     # of newer releases.

    This module converts Perl data structures to JSON and vice versa. Its
    primary goal is to be *correct* and its secondary goal is to be *fast*.
    To reach the latter goal it was written in C.

    As this is the n-th-something JSON module on CPAN, what was the reason
    to write yet another JSON module? While it seems there are many JSON
    modules, none of them correctly handle all corner cases, and in most
    cases their maintainers are unresponsive, gone missing, or not listening
    to bug reports for other reasons.

    See below for the Cpanel fork.

    See MAPPING, below, on how Cpanel::JSON::XS maps perl values to JSON
    values and vice versa.

    * smart Unicode handling

        This module knows how to handle Unicode with Perl version higher
        than 5.8.5, documents how and when it does so, and even documents
        what "correct" means.

    * round-trip integrity

        When you serialise a perl data structure using only data types
        supported by JSON and Perl, the deserialised data structure is
        identical on the Perl level. (e.g. the string "2.0" doesn't suddenly
        become "2" just because it looks like a number). There *are* minor
        exceptions to this, read the MAPPING section below to learn about

    * strict checking of JSON correctness

        There is no guessing, no generating of illegal JSON texts by
        default, and only JSON is accepted as input by default (the latter
        is a security feature).

    * fast

        Compared to other JSON modules and other serialisers such as
        Storable, this module usually compares favourably in terms of speed,

    * simple to use

        This module has both a simple functional interface as well as an
        object oriented interface.

    * reasonably versatile output formats

        You can choose between the most compact guaranteed-single-line
        format possible (nice for simple line-based protocols), a pure-ASCII
        format (for when your transport is not 8-bit clean, still supports
        the whole Unicode range), or a pretty-printed format (for when you
        want to read that stuff). Or you can combine those features in
        whatever way you like.

  cPanel fork
    Since the original author MLEHMANN has no public bugtracker, this cPanel
    fork sits now on github.

    src repo: <> original:

    RT: <> or

    Changes to JSON::XS

    - added "binary" extension, non-JSON and non JSON parsable, allows
    "\xNN" and "\NNN" sequences.

    - 5.6.2 support; sacrificing some utf8 features (assuming bytes
    all-over), no multi-byte unicode characters.

    - interop for true/false overloading. JSON::XS and JSON::PP
    representations are accepted and JSON::XS accepts Cpanel::JSON::XS
    booleans [#13]

    - ithread support. Cpanel::JSON::XS is thread-safe, JSON::XS not

    - performance optimizations for threaded Perls

    - additional fixes for:

      - [cpan #88061] AIX atof without USE_LONG_DOUBLE

      - #10 unshare_hek crash

      - #7 avoid re-blessing where possible (e.g. SvREADONLY restricted hashes)

    - public maintenance and bugtracker

    - use ppport.h, sanify XS.xs comment styles, harness C coding style

    - common::sense is optional. When available it is not used in the
    published production module, just during development and testing.

    - extended testsuite

    The following convenience methods are provided by this module. They are
    exported by default:

    $json_text = encode_json $perl_scalar
        Converts the given Perl data structure to a UTF-8 encoded, binary
        string (that is, the string contains octets only). Croaks on error.

        This function call is functionally identical to:

           $json_text = Cpanel::JSON::XS->new->utf8->encode ($perl_scalar)

        Except being faster.

    $perl_scalar = decode_json $json_text
        The opposite of "encode_json": expects an UTF-8 (binary) string and
        tries to parse that as an UTF-8 encoded JSON text, returning the
        resulting reference. Croaks on error.

        This function call is functionally identical to:

           $perl_scalar = Cpanel::JSON::XS->new->utf8->decode ($json_text)

        Except being faster.

    $is_boolean = Cpanel::JSON::XS::is_bool $scalar
        Returns true if the passed scalar represents either
        Cpanel::JSON::XS::true or Cpanel::JSON::XS::false, two constants
        that act like 1 and 0, respectively and are used to represent JSON
        "true" and "false" values in Perl.

        See MAPPING, below, for more information on how JSON values are
        mapped to Perl.

        from_json has been renamed to encode_json

        to_json has been renamed to encode_json

    Since this often leads to confusion, here are a few very clear words on
    how Unicode works in Perl, modulo bugs.

    1. Perl strings can store characters with ordinal values > 255.
        This enables you to store Unicode characters as single characters in
        a Perl string - very natural.

    2. Perl does *not* associate an encoding with your strings.
        ... until you force it to, e.g. when matching it against a regex, or
        printing the scalar to a file, in which case Perl either interprets
        your string as locale-encoded text, octets/binary, or as Unicode,
        depending on various settings. In no case is an encoding stored
        together with your data, it is *use* that decides encoding, not any
        magical meta data.

    3. The internal utf-8 flag has no meaning with regards to the encoding
    of your string.
    4. A "Unicode String" is simply a string where each character can be
    validly interpreted as a Unicode code point.
        If you have UTF-8 encoded data, it is no longer a Unicode string,
        but a Unicode string encoded in UTF-8, giving you a binary string.

    5. A string containing "high" (> 255) character values is *not* a UTF-8

    I hope this helps :)

    The object oriented interface lets you configure your own encoding or
    decoding style, within the limits of supported formats.

    $json = new Cpanel::JSON::XS
        Creates a new JSON object that can be used to de/encode JSON
        strings. All boolean flags described below are by default

        The mutators for flags all return the JSON object again and thus
        calls can be chained:

           my $json = Cpanel::JSON::XS->new->utf8->space_after->encode ({a => [1,2]})
           => {"a": [1, 2]}

    $json = $json->ascii ([$enable])
    $enabled = $json->get_ascii
        If $enable is true (or missing), then the "encode" method will not
        generate characters outside the code range 0..127 (which is ASCII).
        Any Unicode characters outside that range will be escaped using
        either a single \uXXXX (BMP characters) or a double \uHHHH\uLLLLL
        escape sequence, as per RFC4627. The resulting encoded JSON text can
        be treated as a native Unicode string, an ascii-encoded,
        latin1-encoded or UTF-8 encoded string, or any other superset of

        If $enable is false, then the "encode" method will not escape
        Unicode characters unless required by the JSON syntax or other
        flags. This results in a faster and more compact format.

        See also the section *ENCODING/CODESET FLAG NOTES* later in this

        The main use for this flag is to produce JSON texts that can be
        transmitted over a 7-bit channel, as the encoded JSON texts will not
        contain any 8 bit characters.

          Cpanel::JSON::XS->new->ascii (1)->encode ([chr 0x10401])
          => ["\ud801\udc01"]

    $json = $json->latin1 ([$enable])
    $enabled = $json->get_latin1
        If $enable is true (or missing), then the "encode" method will
        encode the resulting JSON text as latin1 (or iso-8859-1), escaping
        any characters outside the code range 0..255. The resulting string
        can be treated as a latin1-encoded JSON text or a native Unicode
        string. The "decode" method will not be affected in any way by this
        flag, as "decode" by default expects Unicode, which is a strict
        superset of latin1.

        If $enable is false, then the "encode" method will not escape
        Unicode characters unless required by the JSON syntax or other

        See also the section *ENCODING/CODESET FLAG NOTES* later in this

        The main use for this flag is efficiently encoding binary data as
        JSON text, as most octets will not be escaped, resulting in a
        smaller encoded size. The disadvantage is that the resulting JSON
        text is encoded in latin1 (and must correctly be treated as such
        when storing and transferring), a rare encoding for JSON. It is
        therefore most useful when you want to store data structures known
        to contain binary data efficiently in files or databases, not when
        talking to other JSON encoders/decoders.

          Cpanel::JSON::XS->new->latin1->encode (["\x{89}\x{abc}"]
          => ["\x{89}\\u0abc"] # (perl syntax, U+abc escaped, U+89 not)

    $json = $json->binary ([$enable])
    $enabled = $json = $json->get_binary
        If the $enable argument is true (or missing), then the "encode"
        method will not try to detect an UTF-8 encoding in any JSON string,
        it will strictly interpret it as byte sequence. The result might
        contain new "\xNN" sequences, which is unparsable JSON. The "decode"
        method forbids "\uNNNN" sequences and accepts "\xNN" and octal
        "\NNN" sequences.

        There is also a special logic for perl 5.6 and utf8. 5.6 encodes any
        string to utf-8 automatically when seeing a codepoint >= 0x80 and <
        0x100. With the binary flag enabled decode the perl utf8 encoded
        string to the original byte encoding and encode this with "\xNN"
        escapes. This will result to the same encodings as with newer perls.
        But note that binary multi-byte codepoints with 5.6 will result in
        "illegal unicode character in binary string" errors, unlike with
        newer perls.

        If $enable is false, then the "encode" method will smartly try to
        detect Unicode characters unless required by the JSON syntax or
        other flags and hex and octal sequences are forbidden.

        See also the section *ENCODING/CODESET FLAG NOTES* later in this

        The main use for this flag is to avoid the smart unicode detection
        and possible double encoding. The disadvantage is that the resulting
        JSON text is encoded in new "\xNN" and in latin1 characters and must
        correctly be treated as such when storing and transferring, a rare
        encoding for JSON. It will produce non-readable JSON strings in the
        browser. It is therefore most useful when you want to store data
        structures known to contain binary data efficiently in files or
        databases, not when talking to other JSON encoders/decoders. The
        binary decoding method can also be used when an encoder produced a
        non-JSON conformant hex or octal encoding "\xNN" or "\NNN".

          Cpanel::JSON::XS->new->binary->encode (["\x{89}\x{abc}"])
          5.6: Error: malformed or illegal unicode character in binary string
          >=5.8: ['\x89\xe0\xaa\xbc']

          Cpanel::JSON::XS->new->binary->encode (["\x{89}\x{bc}"])
          => ["\x89\xbc"]

          Cpanel::JSON::XS->new->binary->decode (["\x89\ua001"])
          Error: malformed or illegal unicode character in binary string

          Cpanel::JSON::XS->new->decode (["\x89"])
          Error: illegal hex character in non-binary string

    $json = $json->utf8 ([$enable])
    $enabled = $json->get_utf8
        If $enable is true (or missing), then the "encode" method will
        encode the JSON result into UTF-8, as required by many protocols,
        while the "decode" method expects to be handled an UTF-8-encoded
        string. Please note that UTF-8-encoded strings do not contain any
        characters outside the range 0..255, they are thus useful for
        bytewise/binary I/O. In future versions, enabling this option might
        enable autodetection of the UTF-16 and UTF-32 encoding families, as
        described in RFC4627.

        If $enable is false, then the "encode" method will return the JSON
        string as a (non-encoded) Unicode string, while "decode" expects
        thus a Unicode string. Any decoding or encoding (e.g. to UTF-8 or
        UTF-16) needs to be done yourself, e.g. using the Encode module.

        See also the section *ENCODING/CODESET FLAG NOTES* later in this

        Example, output UTF-16BE-encoded JSON:

          use Encode;
          $jsontext = encode "UTF-16BE", Cpanel::JSON::XS->new->encode ($object);

        Example, decode UTF-32LE-encoded JSON:

          use Encode;
          $object = Cpanel::JSON::XS->new->decode (decode "UTF-32LE", $jsontext);

    $json = $json->pretty ([$enable])
        This enables (or disables) all of the "indent", "space_before" and
        "space_after" (and in the future possibly more) flags in one call to
        generate the most readable (or most compact) form possible.

        Example, pretty-print some simple structure:

           my $json = Cpanel::JSON::XS->new->pretty(1)->encode ({a => [1,2]})
              "a" : [

    $json = $json->indent ([$enable])
    $enabled = $json->get_indent
        If $enable is true (or missing), then the "encode" method will use a
        multiline format as output, putting every array member or
        object/hash key-value pair into its own line, indenting them

        If $enable is false, no newlines or indenting will be produced, and
        the resulting JSON text is guaranteed not to contain any "newlines".

        This setting has no effect when decoding JSON texts.

    $json = $json->space_before ([$enable])
    $enabled = $json->get_space_before
        If $enable is true (or missing), then the "encode" method will add
        an extra optional space before the ":" separating keys from values
        in JSON objects.

        If $enable is false, then the "encode" method will not add any extra
        space at those places.

        This setting has no effect when decoding JSON texts. You will also
        most likely combine this setting with "space_after".

        Example, space_before enabled, space_after and indent disabled:

           {"key" :"value"}

    $json = $json->space_after ([$enable])
    $enabled = $json->get_space_after
        If $enable is true (or missing), then the "encode" method will add
        an extra optional space after the ":" separating keys from values in
        JSON objects and extra whitespace after the "," separating key-value
        pairs and array members.

        If $enable is false, then the "encode" method will not add any extra
        space at those places.

        This setting has no effect when decoding JSON texts.

        Example, space_before and indent disabled, space_after enabled:

           {"key": "value"}

    $json = $json->relaxed ([$enable])
    $enabled = $json->get_relaxed
        If $enable is true (or missing), then "decode" will accept some
        extensions to normal JSON syntax (see below). "encode" will not be
        affected in anyway. *Be aware that this option makes you accept
        invalid JSON texts as if they were valid!*. I suggest only to use
        this option to parse application-specific files written by humans
        (configuration files, resource files etc.)

        If $enable is false (the default), then "decode" will only accept
        valid JSON texts.

        Currently accepted extensions are:

        * list items can have an end-comma

            JSON *separates* array elements and key-value pairs with commas.
            This can be annoying if you write JSON texts manually and want
            to be able to quickly append elements, so this extension accepts
            comma at the end of such items not just between them:

                  2, <- this comma not normally allowed
                  "k1": "v1",
                  "k2": "v2", <- this comma not normally allowed

        * shell-style '#'-comments

            Whenever JSON allows whitespace, shell-style comments are
            additionally allowed. They are terminated by the first
            carriage-return or line-feed character, after which more
            white-space and comments are allowed.

                 1, # this comment not allowed in JSON
                    # neither this one...

    $json = $json->canonical ([$enable])
    $enabled = $json->get_canonical
        If $enable is true (or missing), then the "encode" method will
        output JSON objects by sorting their keys. This is adding a
        comparatively high overhead.

        If $enable is false, then the "encode" method will output key-value
        pairs in the order Perl stores them (which will likely change
        between runs of the same script).

        This option is useful if you want the same data structure to be
        encoded as the same JSON text (given the same overall settings). If
        it is disabled, the same hash might be encoded differently even if
        contains the same data, as key-value pairs have no inherent ordering
        in Perl.

        This setting has no effect when decoding JSON texts.

        This setting has currently no effect on tied hashes.

    $json = $json->allow_nonref ([$enable])
    $enabled = $json->get_allow_nonref
        If $enable is true (or missing), then the "encode" method can
        convert a non-reference into its corresponding string, number or
        null JSON value, which is an extension to RFC4627. Likewise,
        "decode" will accept those JSON values instead of croaking.

        If $enable is false, then the "encode" method will croak if it isn't
        passed an arrayref or hashref, as JSON texts must either be an
        object or array. Likewise, "decode" will croak if given something
        that is not a JSON object or array.

        Example, encode a Perl scalar as JSON value with enabled
        "allow_nonref", resulting in an invalid JSON text:

           Cpanel::JSON::XS->new->allow_nonref->encode ("Hello, World!")
           => "Hello, World!"

    $json = $json->allow_unknown ([$enable])
    $enabled = $json->get_allow_unknown
        If $enable is true (or missing), then "encode" will *not* throw an
        exception when it encounters values it cannot represent in JSON (for
        example, filehandles) but instead will encode a JSON "null" value.
        Note that blessed objects are not included here and are handled
        separately by c<allow_nonref>.

        If $enable is false (the default), then "encode" will throw an
        exception when it encounters anything it cannot encode as JSON.

        This option does not affect "decode" in any way, and it is
        recommended to leave it off unless you know your communications

    $json = $json->allow_blessed ([$enable])
    $enabled = $json->get_allow_blessed
        If $enable is true (or missing), then the "encode" method will not
        barf when it encounters a blessed reference. Instead, the value of
        the convert_blessed option will decide whether "null"
        ("convert_blessed" disabled or no "TO_JSON" method found) or a
        representation of the object ("convert_blessed" enabled and
        "TO_JSON" method found) is being encoded. Has no effect on "decode".

        If $enable is false (the default), then "encode" will throw an
        exception when it encounters a blessed object.

    $json = $json->convert_blessed ([$enable])
    $enabled = $json->get_convert_blessed
        If $enable is true (or missing), then "encode", upon encountering a
        blessed object, will check for the availability of the "TO_JSON"
        method on the object's class. If found, it will be called in scalar
        context and the resulting scalar will be encoded instead of the
        object. If no "TO_JSON" method is found, the value of
        "allow_blessed" will decide what to do.

        The "TO_JSON" method may safely call die if it wants. If "TO_JSON"
        returns other blessed objects, those will be handled in the same
        way. "TO_JSON" must take care of not causing an endless recursion
        cycle (== crash) in this case. The name of "TO_JSON" was chosen
        because other methods called by the Perl core (== not by the user of
        the object) are usually in upper case letters and to avoid
        collisions with any "to_json" function or method.

        This setting does not yet influence "decode" in any way, but in the
        future, global hooks might get installed that influence "decode" and
        are enabled by this setting.

        If $enable is false, then the "allow_blessed" setting will decide
        what to do when a blessed object is found.

    $json = $json->filter_json_object ([$coderef->($hashref)])
        When $coderef is specified, it will be called from "decode" each
        time it decodes a JSON object. The only argument is a reference to
        the newly-created hash. If the code references returns a single
        scalar (which need not be a reference), this value (i.e. a copy of
        that scalar to avoid aliasing) is inserted into the deserialised
        data structure. If it returns an empty list (NOTE: *not* "undef",
        which is a valid scalar), the original deserialised hash will be
        inserted. This setting can slow down decoding considerably.

        When $coderef is omitted or undefined, any existing callback will be
        removed and "decode" will not change the deserialised hash in any

        Example, convert all JSON objects into the integer 5:

           my $js = Cpanel::JSON::XS->new->filter_json_object (sub { 5 });
           # returns [5]
           $js->decode ('[{}]')
           # throw an exception because allow_nonref is not enabled
           # so a lone 5 is not allowed.
           $js->decode ('{"a":1, "b":2}');

    $json = $json->filter_json_single_key_object ($key [=>
        Works remotely similar to "filter_json_object", but is only called
        for JSON objects having a single key named $key.

        This $coderef is called before the one specified via
        "filter_json_object", if any. It gets passed the single value in the
        JSON object. If it returns a single value, it will be inserted into
        the data structure. If it returns nothing (not even "undef" but the
        empty list), the callback from "filter_json_object" will be called
        next, as if no single-key callback were specified.

        If $coderef is omitted or undefined, the corresponding callback will
        be disabled. There can only ever be one callback for a given key.

        As this callback gets called less often then the
        "filter_json_object" one, decoding speed will not usually suffer as
        much. Therefore, single-key objects make excellent targets to
        serialise Perl objects into, especially as single-key JSON objects
        are as close to the type-tagged value concept as JSON gets (it's
        basically an ID/VALUE tuple). Of course, JSON does not support this
        in any way, so you need to make sure your data never looks like a
        serialised Perl hash.

        Typical names for the single object key are "__class_whatever__", or
        "$__dollars_are_rarely_used__$" or "}ugly_brace_placement", or even
        things like "__class_md5sum(classname)__", to reduce the risk of
        clashing with real hashes.

        Example, decode JSON objects of the form "{ "__widget__" => <id> }"
        into the corresponding $WIDGET{<id>} object:

           # return whatever is in $WIDGET{5}:
              ->filter_json_single_key_object (__widget__ => sub {
                    $WIDGET{ $_[0] }
              ->decode ('{"__widget__": 5')

           # this can be used with a TO_JSON method in some "widget" class
           # for serialisation to json:
           sub WidgetBase::TO_JSON {
              my ($self) = @_;

              unless ($self->{id}) {
                 $self->{id} =;
                 $WIDGET{$self->{id}} = $self;

              { __widget__ => $self->{id} }

    $json = $json->shrink ([$enable])
    $enabled = $json->get_shrink
        Perl usually over-allocates memory a bit when allocating space for
        strings. This flag optionally resizes strings generated by either
        "encode" or "decode" to their minimum size possible. This can save
        memory when your JSON texts are either very very long or you have
        many short strings. It will also try to downgrade any strings to
        octet-form if possible: perl stores strings internally either in an
        encoding called UTF-X or in octet-form. The latter cannot store
        everything but uses less space in general (and some buggy Perl or C
        code might even rely on that internal representation being used).

        The actual definition of what shrink does might change in future
        versions, but it will always try to save space at the expense of

        If $enable is true (or missing), the string returned by "encode"
        will be shrunk-to-fit, while all strings generated by "decode" will
        also be shrunk-to-fit.

        If $enable is false, then the normal perl allocation algorithms are
        used. If you work with your data, then this is likely to be faster.

        In the future, this setting might control other things, such as
        converting strings that look like integers or floats into integers
        or floats internally (there is no difference on the Perl level),
        saving space.

    $json = $json->max_depth ([$maximum_nesting_depth])
    $max_depth = $json->get_max_depth
        Sets the maximum nesting level (default 512) accepted while encoding
        or decoding. If a higher nesting level is detected in JSON text or a
        Perl data structure, then the encoder and decoder will stop and
        croak at that point.

        Nesting level is defined by number of hash- or arrayrefs that the
        encoder needs to traverse to reach a given point or the number of
        "{" or "[" characters without their matching closing parenthesis
        crossed to reach a given character in a string.

        Setting the maximum depth to one disallows any nesting, so that
        ensures that the object is only a single hash/object or array.

        If no argument is given, the highest possible setting will be used,
        which is rarely useful.

        Note that nesting is implemented by recursion in C. The default
        value has been chosen to be as large as typical operating systems
        allow without crashing.

        See SECURITY CONSIDERATIONS, below, for more info on why this is

    $json = $json->max_size ([$maximum_string_size])
    $max_size = $json->get_max_size
        Set the maximum length a JSON text may have (in bytes) where
        decoding is being attempted. The default is 0, meaning no limit.
        When "decode" is called on a string that is longer then this many
        bytes, it will not attempt to decode the string but throw an
        exception. This setting has no effect on "encode" (yet).

        If no argument is given, the limit check will be deactivated (same
        as when 0 is specified).

        See SECURITY CONSIDERATIONS, below, for more info on why this is

    $json_text = $json->encode ($perl_scalar)
        Converts the given Perl data structure (a simple scalar or a
        reference to a hash or array) to its JSON representation. Simple
        scalars will be converted into JSON string or number sequences,
        while references to arrays become JSON arrays and references to
        hashes become JSON objects. Undefined Perl values (e.g. "undef")
        become JSON "null" values. Neither "true" nor "false" values will be

    $perl_scalar = $json->decode ($json_text)
        The opposite of "encode": expects a JSON text and tries to parse it,
        returning the resulting simple scalar or reference. Croaks on error.

        JSON numbers and strings become simple Perl scalars. JSON arrays
        become Perl arrayrefs and JSON objects become Perl hashrefs. "true"
        becomes 1, "false" becomes 0 and "null" becomes "undef".

    ($perl_scalar, $characters) = $json->decode_prefix ($json_text)
        This works like the "decode" method, but instead of raising an
        exception when there is trailing garbage after the first JSON
        object, it will silently stop parsing there and return the number of
        characters consumed so far.

        This is useful if your JSON texts are not delimited by an outer
        protocol (which is not the brightest thing to do in the first place)
        and you need to know where the JSON text ends.

           Cpanel::JSON::XS->new->decode_prefix ("[1] the tail")
           => ([], 3)

    $json->to_json ($perl_hash_or_arrayref)
        Deprecated method for perl 5.8 and newer. Use encode_json instead.

    $json->from_json ($utf8_encoded_json_text)
        Deprecated method for perl 5.8 and newer. Use decode_json instead.

    In some cases, there is the need for incremental parsing of JSON texts.
    While this module always has to keep both JSON text and resulting Perl
    data structure in memory at one time, it does allow you to parse a JSON
    stream incrementally. It does so by accumulating text until it has a
    full JSON object, which it then can decode. This process is similar to
    using "decode_prefix" to see if a full JSON object is available, but is
    much more efficient (and can be implemented with a minimum of method

    Cpanel::JSON::XS will only attempt to parse the JSON text once it is
    sure it has enough text to get a decisive result, using a very simple
    but truly incremental parser. This means that it sometimes won't stop as
    early as the full parser, for example, it doesn't detect mismatched
    parentheses. The only thing it guarantees is that it starts decoding as
    soon as a syntactically valid JSON text has been seen. This means you
    need to set resource limits (e.g. "max_size") to ensure the parser will
    stop parsing in the presence if syntax errors.

    The following methods implement this incremental parser.

    [void, scalar or list context] = $json->incr_parse ([$string])
        This is the central parsing function. It can both append new text
        and extract objects from the stream accumulated so far (both of
        these functions are optional).

        If $string is given, then this string is appended to the already
        existing JSON fragment stored in the $json object.

        After that, if the function is called in void context, it will
        simply return without doing anything further. This can be used to
        add more text in as many chunks as you want.

        If the method is called in scalar context, then it will try to
        extract exactly *one* JSON object. If that is successful, it will
        return this object, otherwise it will return "undef". If there is a
        parse error, this method will croak just as "decode" would do (one
        can then use "incr_skip" to skip the erroneous part). This is the
        most common way of using the method.

        And finally, in list context, it will try to extract as many objects
        from the stream as it can find and return them, or the empty list
        otherwise. For this to work, there must be no separators between the
        JSON objects or arrays, instead they must be concatenated
        back-to-back. If an error occurs, an exception will be raised as in
        the scalar context case. Note that in this case, any
        previously-parsed JSON texts will be lost.

        Example: Parse some JSON arrays/objects in a given string and return

           my @objs = Cpanel::JSON::XS->new->incr_parse ("[5][7][1,2]");

    $lvalue_string = $json->incr_text (>5.8 only)
        This method returns the currently stored JSON fragment as an lvalue,
        that is, you can manipulate it. This *only* works when a preceding
        call to "incr_parse" in *scalar context* successfully returned an
        object, and 2. only with Perl >= 5.8

        Under all other circumstances you must not call this function (I
        mean it. although in simple tests it might actually work, it *will*
        fail under real world conditions). As a special exception, you can
        also call this method before having parsed anything.

        This function is useful in two cases: a) finding the trailing text
        after a JSON object or b) parsing multiple JSON objects separated by
        non-JSON text (such as commas).

        This will reset the state of the incremental parser and will remove
        the parsed text from the input buffer so far. This is useful after
        "incr_parse" died, in which case the input buffer and incremental
        parser state is left unchanged, to skip the text parsed so far and
        to reset the parse state.

        The difference to "incr_reset" is that only text until the parse
        error occurred is removed.

        This completely resets the incremental parser, that is, after this
        call, it will be as if the parser had never parsed anything.

        This is useful if you want to repeatedly parse JSON objects and want
        to ignore any trailing data, which means you have to reset the
        parser after each successful decode.

    All options that affect decoding are supported, except "allow_nonref".
    The reason for this is that it cannot be made to work sensibly: JSON
    objects and arrays are self-delimited, i.e. you can concatenate them
    back to back and still decode them perfectly. This does not hold true
    for JSON numbers, however.

    For example, is the string 1 a single JSON number, or is it simply the
    start of 12? Or is 12 a single JSON number, or the concatenation of 1
    and 2? In neither case you can tell, and this is why Cpanel::JSON::XS
    takes the conservative route and disallows this case.

    Some examples will make all this clearer. First, a simple example that
    works similarly to "decode_prefix": We want to decode the JSON object at
    the start of a string and identify the portion after the JSON object:

       my $text = "[1,2,3] hello";

       my $json = new Cpanel::JSON::XS;

       my $obj = $json->incr_parse ($text)
          or die "expected JSON object or array at beginning of string";

       my $tail = $json->incr_text;
       # $tail now contains " hello"

    Easy, isn't it?

    Now for a more complicated example: Imagine a hypothetical protocol
    where you read some requests from a TCP stream, and each request is a
    JSON array, without any separation between them (in fact, it is often
    useful to use newlines as "separators", as these get interpreted as
    whitespace at the start of the JSON text, which makes it possible to
    test said protocol with "telnet"...).

    Here is how you'd do it (it is trivial to write this in an event-based

       my $json = new Cpanel::JSON::XS;

       # read some data from the socket
       while (sysread $socket, my $buf, 4096) {

          # split and decode as many requests as possible
          for my $request ($json->incr_parse ($buf)) {
             # act on the $request

    Another complicated example: Assume you have a string with JSON objects
    or arrays, all separated by (optional) comma characters (e.g. "[1],[2],
    [3]"). To parse them, we have to skip the commas between the JSON texts,
    and here is where the lvalue-ness of "incr_text" comes in useful:

       my $text = "[1],[2], [3]";
       my $json = new Cpanel::JSON::XS;

       # void context, so no parsing done
       $json->incr_parse ($text);

       # now extract as many objects as possible. note the
       # use of scalar context so incr_text can be called.
       while (my $obj = $json->incr_parse) {
          # do something with $obj

          # now skip the optional comma
          $json->incr_text =~ s/^ \s* , //x;

    Now lets go for a very complex example: Assume that you have a gigantic
    JSON array-of-objects, many gigabytes in size, and you want to parse it,
    but you cannot load it into memory fully (this has actually happened in
    the real world :).

    Well, you lost, you have to implement your own JSON parser. But
    Cpanel::JSON::XS can still help you: You implement a (very simple) array
    parser and let JSON decode the array elements, which are all full JSON
    objects on their own (this wouldn't work if the array elements could be
    JSON numbers, for example):

       my $json = new Cpanel::JSON::XS;

       # open the monster
       open my $fh, "<bigfile.json"
          or die "bigfile: $!";

       # first parse the initial "["
       for (;;) {
          sysread $fh, my $buf, 65536
             or die "read error: $!";
          $json->incr_parse ($buf); # void context, so no parsing

          # Exit the loop once we found and removed(!) the initial "[".
          # In essence, we are (ab-)using the $json object as a simple scalar
          # we append data to.
          last if $json->incr_text =~ s/^ \s* \[ //x;

       # now we have the skipped the initial "[", so continue
       # parsing all the elements.
       for (;;) {
          # in this loop we read data until we got a single JSON object
          for (;;) {
             if (my $obj = $json->incr_parse) {
                # do something with $obj

             # add more data
             sysread $fh, my $buf, 65536
                or die "read error: $!";
             $json->incr_parse ($buf); # void context, so no parsing

          # in this loop we read data until we either found and parsed the
          # separating "," between elements, or the final "]"
          for (;;) {
             # first skip whitespace
             $json->incr_text =~ s/^\s*//;

             # if we find "]", we are done
             if ($json->incr_text =~ s/^\]//) {
                print "finished.\n";

             # if we find ",", we can continue with the next element
             if ($json->incr_text =~ s/^,//) {

             # if we find anything else, we have a parse error!
             if (length $json->incr_text) {
                die "parse error near ", $json->incr_text;

             # else add more data
             sysread $fh, my $buf, 65536
                or die "read error: $!";
             $json->incr_parse ($buf); # void context, so no parsing

    This is a complex example, but most of the complexity comes from the
    fact that we are trying to be correct (bear with me if I am wrong, I
    never ran the above example :).

    This section describes how Cpanel::JSON::XS maps Perl values to JSON
    values and vice versa. These mappings are designed to "do the right
    thing" in most circumstances automatically, preserving round-tripping
    characteristics (what you put in comes out as something equivalent).

    For the more enlightened: note that in the following descriptions,
    lowercase *perl* refers to the Perl interpreter, while uppercase *Perl*
    refers to the abstract Perl language itself.

        A JSON object becomes a reference to a hash in Perl. No ordering of
        object keys is preserved (JSON does not preserve object key ordering

        A JSON array becomes a reference to an array in Perl.

        A JSON string becomes a string scalar in Perl - Unicode codepoints
        in JSON are represented by the same codepoints in the Perl string,
        so no manual decoding is necessary.

        A JSON number becomes either an integer, numeric (floating point) or
        string scalar in perl, depending on its range and any fractional
        parts. On the Perl level, there is no difference between those as
        Perl handles all the conversion details, but an integer may take
        slightly less memory and might represent more values exactly than
        floating point numbers.

        If the number consists of digits only, Cpanel::JSON::XS will try to
        represent it as an integer value. If that fails, it will try to
        represent it as a numeric (floating point) value if that is possible
        without loss of precision. Otherwise it will preserve the number as
        a string value (in which case you lose roundtripping ability, as the
        JSON number will be re-encoded to a JSON string).

        Numbers containing a fractional or exponential part will always be
        represented as numeric (floating point) values, possibly at a loss
        of precision (in which case you might lose perfect roundtripping
        ability, but the JSON number will still be re-encoded as a JSON

        Note that precision is not accuracy - binary floating point values
        cannot represent most decimal fractions exactly, and when converting
        from and to floating point, Cpanel::JSON::XS only guarantees
        precision up to but not including the least significant bit.

    true, false
        These JSON atoms become "JSON::XS::true" and "JSON::XS::false",
        respectively. They are overloaded to act almost exactly like the
        numbers 1 and 0. You can check whether a scalar is a JSON boolean by
        using the "Cpanel::JSON::XS::is_bool" function.

        A JSON null atom becomes "undef" in Perl.

    The mapping from Perl to JSON is slightly more difficult, as Perl is a
    truly typeless language, so we can only guess which JSON type is meant
    by a Perl value.

    hash references
        Perl hash references become JSON objects. As there is no inherent
        ordering in hash keys (or JSON objects), they will usually be
        encoded in a pseudo-random order that can change between runs of the
        same program but stays generally the same within a single run of a
        program. Cpanel::JSON::XS can optionally sort the hash keys
        (determined by the *canonical* flag), so the same datastructure will
        serialise to the same JSON text (given same settings and version of
        Cpanel::JSON::XS), but this incurs a runtime overhead and is only
        rarely useful, e.g. when you want to compare some JSON text against
        another for equality.

    array references
        Perl array references become JSON arrays.

    other references
        Other unblessed references are generally not allowed and will cause
        an exception to be thrown, except for references to the integers 0
        and 1, which get turned into "false" and "true" atoms in JSON. You
        can also use "JSON::XS::false" and "JSON::XS::true" to improve

           encode_json [\0, Cpanel::JSON::XS::true] # yields [false,true]

    Cpanel::JSON::XS::true, Cpanel::JSON::XS::false
        These special values become JSON true and JSON false values,
        respectively. You can also use "\1" and "\0" directly if you want.

    blessed objects
        Blessed objects are not directly representable in JSON. See the
        "allow_blessed" and "convert_blessed" methods on various options on
        how to deal with this: basically, you can choose between throwing an
        exception, encoding the reference as if it weren't blessed, or
        provide your own serialiser method.

    simple scalars
        Simple Perl scalars (any scalar that is not a reference) are the
        most difficult objects to encode: Cpanel::JSON::XS will encode
        undefined scalars as JSON "null" values, scalars that have last been
        used in a string context before encoding as JSON strings, and
        anything else as number value:

           # dump as number
           encode_json [2] # yields [2]
           encode_json [-3.0e17] # yields [-3e+17]
           my $value = 5; encode_json [$value] # yields [5]

           # used as string, so dump as string
           print $value;
           encode_json [$value] # yields ["5"]

           # undef becomes null
           encode_json [undef] # yields [null]

        You can force the type to be a JSON string by stringifying it:

           my $x = 3.1; # some variable containing a number
           "$x"; # stringified
           $x .= ""; # another, more awkward way to stringify
           print $x; # perl does it for you, too, quite often

        You can force the type to be a JSON number by numifying it:

           my $x = "3"; # some variable containing a string
           $x += 0; # numify it, ensuring it will be dumped as a number
           $x *= 1; # same thing, the choice is yours.

        You can not currently force the type in other, less obscure, ways.
        Tell me if you need this capability (but don't forget to explain why
        it's needed :).

        Note that numerical precision has the same meaning as under Perl (so
        binary to decimal conversion follows the same rules as in Perl,
        which can differ to other languages). Also, your perl interpreter
        might expose extensions to the floating point numbers of your
        platform, such as infinities or NaN's - these cannot be represented
        in JSON, and it is an error to pass those in.

    The interested reader might have seen a number of flags that signify
    encodings or codesets - "utf8", "latin1", "binary" and "ascii". There
    seems to be some confusion on what these do, so here is a short

    "utf8" controls whether the JSON text created by "encode" (and expected
    by "decode") is UTF-8 encoded or not, while "latin1" and "ascii" only
    control whether "encode" escapes character values outside their
    respective codeset range. Neither of these flags conflict with each
    other, although some combinations make less sense than others.

    Care has been taken to make all flags symmetrical with respect to
    "encode" and "decode", that is, texts encoded with any combination of
    these flag values will be correctly decoded when the same flags are used
    - in general, if you use different flag settings while encoding vs. when
    decoding you likely have a bug somewhere.

    Below comes a verbose discussion of these flags. Note that a "codeset"
    is simply an abstract set of character-codepoint pairs, while an
    encoding takes those codepoint numbers and *encodes* them, in our case
    into octets. Unicode is (among other things) a codeset, UTF-8 is an
    encoding, and ISO-8859-1 (= latin 1) and ASCII are both codesets *and*
    encodings at the same time, which can be confusing.

    "utf8" flag disabled
        When "utf8" is disabled (the default), then "encode"/"decode"
        generate and expect Unicode strings, that is, characters with high
        ordinal Unicode values (> 255) will be encoded as such characters,
        and likewise such characters are decoded as-is, no changes to them
        will be done, except "(re-)interpreting" them as Unicode codepoints
        or Unicode characters, respectively (to Perl, these are the same
        thing in strings unless you do funny/weird/dumb stuff).

        This is useful when you want to do the encoding yourself (e.g. when
        you want to have UTF-16 encoded JSON texts) or when some other layer
        does the encoding for you (for example, when printing to a terminal
        using a filehandle that transparently encodes to UTF-8 you certainly
        do NOT want to UTF-8 encode your data first and have Perl encode it
        another time).

    "utf8" flag enabled
        If the "utf8"-flag is enabled, "encode"/"decode" will encode all
        characters using the corresponding UTF-8 multi-byte sequence, and
        will expect your input strings to be encoded as UTF-8, that is, no
        "character" of the input string must have any value > 255, as UTF-8
        does not allow that.

        The "utf8" flag therefore switches between two modes: disabled means
        you will get a Unicode string in Perl, enabled means you get an
        UTF-8 encoded octet/binary string in Perl.

    "latin1", "binary" or "ascii" flags enabled
        With "latin1" (or "ascii") enabled, "encode" will escape characters
        with ordinal values > 255 (> 127 with "ascii") and encode the
        remaining characters as specified by the "utf8" flag. With "binary"
        enabled, ordinal values > 255 are illegal.

        If "utf8" is disabled, then the result is also correctly encoded in
        those character sets (as both are proper subsets of Unicode, meaning
        that a Unicode string with all character values < 256 is the same
        thing as a ISO-8859-1 string, and a Unicode string with all
        character values < 128 is the same thing as an ASCII string in

        If "utf8" is enabled, you still get a correct UTF-8-encoded string,
        regardless of these flags, just some more characters will be escaped
        using "\uXXXX" then before.

        Note that ISO-8859-1-*encoded* strings are not compatible with UTF-8
        encoding, while ASCII-encoded strings are. That is because the
        ISO-8859-1 encoding is NOT a subset of UTF-8 (despite the ISO-8859-1
        *codeset* being a subset of Unicode), while ASCII is.

        Surprisingly, "decode" will ignore these flags and so treat all
        input values as governed by the "utf8" flag. If it is disabled, this
        allows you to decode ISO-8859-1- and ASCII-encoded strings, as both
        strict subsets of Unicode. If it is enabled, you can correctly
        decode UTF-8 encoded strings.

        So neither "latin1", "binary" nor "ascii" are incompatible with the
        "utf8" flag - they only govern when the JSON output engine escapes a
        character or not.

        The main use for "latin1" or "binary" is to relatively efficiently
        store binary data as JSON, at the expense of breaking compatibility
        with most JSON decoders.

        The main use for "ascii" is to force the output to not contain
        characters with values > 127, which means you can interpret the
        resulting string as UTF-8, ISO-8859-1, ASCII, KOI8-R or most about
        any character set and 8-bit-encoding, and still get the same data
        structure back. This is useful when your channel for JSON transfer
        is not 8-bit clean or the encoding might be mangled in between (e.g.
        in mail), and works because ASCII is a proper subset of most 8-bit
        and multibyte encodings in use in the world.

  JSON and ECMAscript
    JSON syntax is based on how literals are represented in javascript (the
    not-standardised predecessor of ECMAscript) which is presumably why it
    is called "JavaScript Object Notation".

    However, JSON is not a subset (and also not a superset of course) of
    ECMAscript (the standard) or javascript (whatever browsers actually

    If you want to use javascript's "eval" function to "parse" JSON, you
    might run into parse errors for valid JSON texts, or the resulting data
    structure might not be queryable:

    One of the problems is that U+2028 and U+2029 are valid characters
    inside JSON strings, but are not allowed in ECMAscript string literals,
    so the following Perl fragment will not output something that can be
    guaranteed to be parsable by javascript's "eval":

       use Cpanel::JSON::XS;

       print encode_json [chr 0x2028];

    The right fix for this is to use a proper JSON parser in your javascript
    programs, and not rely on "eval" (see for example Douglas Crockford's
    json2.js parser).

    If this is not an option, you can, as a stop-gap measure, simply encode
    to ASCII-only JSON:

       use Cpanel::JSON::XS;

       print Cpanel::JSON::XS->new->ascii->encode ([chr 0x2028]);

    Note that this will enlarge the resulting JSON text quite a bit if you
    have many non-ASCII characters. You might be tempted to run some regexes
    to only escape U+2028 and U+2029, e.g.:

       # DO NOT USE THIS!
       my $json = Cpanel::JSON::XS->new->utf8->encode ([chr 0x2028]);
       $json =~ s/\xe2\x80\xa8/\\u2028/g; # escape U+2028
       $json =~ s/\xe2\x80\xa9/\\u2029/g; # escape U+2029
       print $json;

    Note that *this is a bad idea*: the above only works for U+2028 and
    U+2029 and thus only for fully ECMAscript-compliant parsers. Many
    existing javascript implementations, however, have issues with other
    characters as well - using "eval" naively simply *will* cause problems.

    Another problem is that some javascript implementations reserve some
    property names for their own purposes (which probably makes them
    non-ECMAscript-compliant). For example, Iceweasel reserves the
    "__proto__" property name for its own purposes.

    If that is a problem, you could parse try to filter the resulting JSON
    output for these property strings, e.g.:

       $json =~ s/"__proto__"\s*:/"__proto__renamed":/g;

    This works because "__proto__" is not valid outside of strings, so every
    occurrence of ""__proto__"\s*:" must be a string used as property name.

    If you know of other incompatibilities, please let me know.

    You often hear that JSON is a subset of YAML. This is, however, a mass
    hysteria(*) and very far from the truth (as of the time of this
    writing), so let me state it clearly: *in general, there is no way to
    configure JSON::XS to output a data structure as valid YAML* that works
    in all cases.

    If you really must use Cpanel::JSON::XS to generate YAML, you should use
    this algorithm (subject to change in future versions):

       my $to_yaml = Cpanel::JSON::XS->new->utf8->space_after (1);
       my $yaml = $to_yaml->encode ($ref) . "\n";

    This will *usually* generate JSON texts that also parse as valid YAML.
    Please note that YAML has hardcoded limits on (simple) object key
    lengths that JSON doesn't have and also has different and incompatible
    unicode character escape syntax, so you should make sure that your hash
    keys are noticeably shorter than the 1024 "stream characters" YAML
    allows and that you do not have characters with codepoint values outside
    the Unicode BMP (basic multilingual page). YAML also does not allow "\/"
    sequences in strings (which Cpanel::JSON::XS does not *currently*
    generate, but other JSON generators might).

    There might be other incompatibilities that I am not aware of (or the
    YAML specification has been changed yet again - it does so quite often).
    In general you should not try to generate YAML with a JSON generator or
    vice versa, or try to parse JSON with a YAML parser or vice versa:
    chances are high that you will run into severe interoperability problems
    when you least expect it.

    (*) I have been pressured multiple times by Brian Ingerson (one of the
        authors of the YAML specification) to remove this paragraph, despite
        him acknowledging that the actual incompatibilities exist. As I was
        personally bitten by this "JSON is YAML" lie, I refused and said I
        will continue to educate people about these issues, so others do not
        run into the same problem again and again. After this, Brian called
        me a (quote)*complete and worthless idiot*(unquote).

        In my opinion, instead of pressuring and insulting people who
        actually clarify issues with YAML and the wrong statements of some
        of its proponents, I would kindly suggest reading the JSON spec
        (which is not that difficult or long) and finally make YAML
        compatible to it, and educating users about the changes, instead of
        spreading lies about the real compatibility for many *years* and
        trying to silence people who point out that it isn't true.

        Addendum/2009: the YAML 1.2 spec is still incompatible with JSON,
        even though the incompatibilities have been documented (and are
        known to Brian) for many years and the spec makes explicit claims
        that YAML is a superset of JSON. It would be so easy to fix, but
        apparently, bullying people and corrupting userdata is so much

    It seems that JSON::XS is surprisingly fast, as shown in the following
    tables. They have been generated with the help of the "eg/bench" program
    in the JSON::XS distribution, to make it easy to compare on your own

    JSON::XS is with Data::MessagePack one of the fastest serializers,
    because JSON and JSON::XS do not support backrefs (no graph structures),
    only trees. Storable supports backrefs, i.e. graphs. Data::MessagePack
    encodes its data binary (as Storable) and supports only very simple
    subset of JSON.

    First comes a comparison between various modules using a very short
    single-line JSON string (also available at

       {"method": "handleMessage", "params": ["user1",
       "we were just talking"], "id": null, "array":[1,11,234,-5,1e5,1e7,
       1, 0]}

    It shows the number of encodes/decodes per second (JSON::XS uses the
    functional interface, while Cpanel::JSON::XS/2 uses the OO interface
    with pretty-printing and hashkey sorting enabled, Cpanel::JSON::XS/3
    enables shrink. JSON::DWIW/DS uses the deserialise function, while
    JSON::DWIW::FJ uses the from_json method). Higher is better:

       module | encode | decode |
       JSON::DWIW/DS | 86302.551 | 102300.098 |
       JSON::DWIW/FJ | 86302.551 | 75983.768 |
       JSON::PP | 15827.562 | 6638.658 |
       JSON::Syck | 63358.066 | 47662.545 |
       JSON::XS | 511500.488 | 511500.488 |
       JSON::XS/2 | 291271.111 | 388361.481 |
       JSON::XS/3 | 361577.931 | 361577.931 |
       Storable | 66788.280 | 265462.278 |

    That is, JSON::XS is almost six times faster than JSON::DWIW on
    encoding, about five times faster on decoding, and over thirty to
    seventy times faster than JSON's pure perl implementation. It also
    compares favourably to Storable for small amounts of data.

    Using a longer test string (roughly 18KB, generated from Yahoo! Locals
    search API (<>).

       module | encode | decode |
       JSON::DWIW/DS | 1647.927 | 2673.916 |
       JSON::DWIW/FJ | 1630.249 | 2596.128 |
       JSON::PP | 400.640 | 62.311 |
       JSON::Syck | 1481.040 | 1524.869 |
       JSON::XS | 20661.596 | 9541.183 |
       JSON::XS/2 | 10683.403 | 9416.938 |
       JSON::XS/3 | 20661.596 | 9400.054 |
       Storable | 19765.806 | 10000.725 |

    Again, JSON::XS leads by far (except for Storable which non-surprisingly
    decodes a bit faster).

    On large strings containing lots of high Unicode characters, some
    modules (such as JSON::PC) seem to decode faster than JSON::XS, but the
    result will be broken due to missing (or wrong) Unicode handling. Others
    refuse to decode or encode properly, so it was impossible to prepare a
    fair comparison table for that case.

    JSON-XS-3.01 broke interoperability with JSON-2.90 with booleans. See

    Cpanel::JSON::XS needs to know the JSON and JSON::XS versions to be able
    work with those objects, especially when encoding a booleans like
    "{"is_true":true}". So you need to load these modules before.

    true/false overloading is supported.

    JSON::XS and JSON::PP representations are accepted and older JSON::XS
    accepts Cpanel::JSON::XS booleans.

    I cannot think of any reason to still use JSON::XS anymore.

    When you are using JSON in a protocol, talking to untrusted potentially
    hostile creatures requires relatively few measures.

    First of all, your JSON decoder should be secure, that is, should not
    have any buffer overflows. Obviously, this module should ensure that and
    I am trying hard on making that true, but you never know.

    Second, you need to avoid resource-starving attacks. That means you
    should limit the size of JSON texts you accept, or make sure then when
    your resources run out, that's just fine (e.g. by using a separate
    process that can crash safely). The size of a JSON text in octets or
    characters is usually a good indication of the size of the resources
    required to decode it into a Perl structure. While JSON::XS can check
    the size of the JSON text, it might be too late when you already have it
    in memory, so you might want to check the size before you accept the

    Third, JSON::XS recurses using the C stack when decoding objects and
    arrays. The C stack is a limited resource: for instance, on my amd64
    machine with 8MB of stack size I can decode around 180k nested arrays
    but only 14k nested JSON objects (due to perl itself recursing deeply on
    croak to free the temporary). If that is exceeded, the program crashes.
    To be conservative, the default nesting limit is set to 512. If your
    process has a smaller stack, you should adjust this setting accordingly
    with the "max_depth" method.

    Something else could bomb you, too, that I forgot to think of. In that
    case, you get to keep the pieces. I am always open for hints, though...

    Also keep in mind that JSON::XS might leak contents of your Perl data
    structures in its error messages, so when you serialise sensitive
    information you might want to make sure that exceptions thrown by
    JSON::XS will not end up in front of untrusted eyes.

    If you are using JSON::XS to return packets to consumption by JavaScript
    scripts in a browser you should have a look at
    to see whether you are vulnerable to some common attack vectors (which
    really are browser design bugs, but it is still you who will have to
    deal with it, as major browser developers care only for features, not
    about getting security right).

    Cpanel::JSON::XS has experimental ithreads support, unlike JSON::XS. If
    you encouter any bugs with thread support please report them.

    While the goal of the JSON::XS module is to be correct, that
    unfortunately does not mean it's bug-free, only that the author thinks
    its design is bug-free. If you keep reporting bugs they will be fixed
    swiftly, though.

    Since the JSON::XS author refuses to use a public bugtracker and prefers
    private emails, we've setup a tracker at RT, so you might want to report
    any issues twice. Once in private to MLEHMANN to be fixed in JSON::XS
    for the masses and one to our the public tracker. Issues fixed by
    JSON::XS with a new release will also be backported to Cpanel::JSON::XS
    and 5.6.2, as long as Cpanel relies on 5.6.2 and JSON::XS as our
    serializer of choice.


    This module is available under the same licences as perl, the Artistic
    license and the GPL.

    The cpanel_json_xs command line utility for quick experiments.

      Marc Lehmann <>

      cPanel Inc. <>

      cPanel Inc. <>

Something went wrong with that request. Please try again.