| … | |
… | |
| 64 | |
64 | |
| 65 | package CBOR::XS; |
65 | package CBOR::XS; |
| 66 | |
66 | |
| 67 | use common::sense; |
67 | use common::sense; |
| 68 | |
68 | |
| 69 | our $VERSION = 1.11; |
69 | our $VERSION = 1.25; |
| 70 | our @ISA = qw(Exporter); |
70 | our @ISA = qw(Exporter); |
| 71 | |
71 | |
| 72 | our @EXPORT = qw(encode_cbor decode_cbor); |
72 | our @EXPORT = qw(encode_cbor decode_cbor); |
| 73 | |
73 | |
| 74 | use Exporter; |
74 | use Exporter; |
| … | |
… | |
| 218 | isn't prepared for this will not leak memory. |
218 | isn't prepared for this will not leak memory. |
| 219 | |
219 | |
| 220 | If C<$enable> is false (the default), then C<decode> will throw an error |
220 | If C<$enable> is false (the default), then C<decode> will throw an error |
| 221 | when it encounters a self-referential/cyclic data structure. |
221 | when it encounters a self-referential/cyclic data structure. |
| 222 | |
222 | |
|
|
223 | FUTURE DIRECTION: the motivation behind this option is to avoid I<real> |
|
|
224 | cycles - future versions of this module might chose to decode cyclic data |
|
|
225 | structures using weak references when this option is off, instead of |
|
|
226 | throwing an error. |
|
|
227 | |
| 223 | This option does not affect C<encode> in any way - shared values and |
228 | This option does not affect C<encode> in any way - shared values and |
| 224 | references will always be decoded properly if present. |
229 | references will always be decoded properly if present. |
| 225 | |
230 | |
| 226 | =item $cbor = $cbor->pack_strings ([$enable]) |
231 | =item $cbor = $cbor->pack_strings ([$enable]) |
| 227 | |
232 | |
| … | |
… | |
| 330 | and you need to know where the first CBOR string ends amd the next one |
335 | and you need to know where the first CBOR string ends amd the next one |
| 331 | starts. |
336 | starts. |
| 332 | |
337 | |
| 333 | CBOR::XS->new->decode_prefix ("......") |
338 | CBOR::XS->new->decode_prefix ("......") |
| 334 | => ("...", 3) |
339 | => ("...", 3) |
|
|
340 | |
|
|
341 | =back |
|
|
342 | |
|
|
343 | =head2 INCREMENTAL PARSING |
|
|
344 | |
|
|
345 | In some cases, there is the need for incremental parsing of JSON |
|
|
346 | texts. While this module always has to keep both CBOR text and resulting |
|
|
347 | Perl data structure in memory at one time, it does allow you to parse a |
|
|
348 | CBOR stream incrementally, using a similar to using "decode_prefix" to see |
|
|
349 | if a full CBOR object is available, but is much more efficient. |
|
|
350 | |
|
|
351 | It basically works by parsing as much of a CBOR string as possible - if |
|
|
352 | the CBOR data is not complete yet, the pasrer will remember where it was, |
|
|
353 | to be able to restart when more data has been accumulated. Once enough |
|
|
354 | data is available to either decode a complete CBOR value or raise an |
|
|
355 | error, a real decode will be attempted. |
|
|
356 | |
|
|
357 | A typical use case would be a network protocol that consists of sending |
|
|
358 | and receiving CBOR-encoded messages. The solution that works with CBOR and |
|
|
359 | about anything else is by prepending a length to every CBOR value, so the |
|
|
360 | receiver knows how many octets to read. More compact (and slightly slower) |
|
|
361 | would be to just send CBOR values back-to-back, as C<CBOR::XS> knows where |
|
|
362 | a CBOR value ends, and doesn't need an explicit length. |
|
|
363 | |
|
|
364 | The following methods help with this: |
|
|
365 | |
|
|
366 | =over 4 |
|
|
367 | |
|
|
368 | =item @decoded = $cbor->incr_parse ($buffer) |
|
|
369 | |
|
|
370 | This method attempts to decode exactly one CBOR value from the beginning |
|
|
371 | of the given C<$buffer>. The value is removed from the C<$buffer> on |
|
|
372 | success. When C<$buffer> doesn't contain a complete value yet, it returns |
|
|
373 | nothing. Finally, when the C<$buffer> doesn't start with something |
|
|
374 | that could ever be a valid CBOR value, it raises an exception, just as |
|
|
375 | C<decode> would. In the latter case the decoder state is undefined and |
|
|
376 | must be reset before being able to parse further. |
|
|
377 | |
|
|
378 | This method modifies the C<$buffer> in place. When no CBOR value can be |
|
|
379 | decoded, the decoder stores the current string offset. On the next call, |
|
|
380 | continues decoding at the place where it stopped before. For this to make |
|
|
381 | sense, the C<$buffer> must begin with the same octets as on previous |
|
|
382 | unsuccessful calls. |
|
|
383 | |
|
|
384 | You can call this method in scalar context, in which case it either |
|
|
385 | returns a decoded value or C<undef>. This makes it impossible to |
|
|
386 | distinguish between CBOR null values (which decode to C<undef>) and an |
|
|
387 | unsuccessful decode, which is often acceptable. |
|
|
388 | |
|
|
389 | =item @decoded = $cbor->incr_parse_multiple ($buffer) |
|
|
390 | |
|
|
391 | Same as C<incr_parse>, but attempts to decode as many CBOR values as |
|
|
392 | possible in one go, instead of at most one. Calls to C<incr_parse> and |
|
|
393 | C<incr_parse_multiple> can be interleaved. |
|
|
394 | |
|
|
395 | =item $cbor->incr_reset |
|
|
396 | |
|
|
397 | Resets the incremental decoder. This throws away any saved state, so that |
|
|
398 | subsequent calls to C<incr_parse> or C<incr_parse_multiple> start to parse |
|
|
399 | a new CBOR value from the beginning of the C<$buffer> again. |
|
|
400 | |
|
|
401 | This method can be caled at any time, but it I<must> be called if you want |
|
|
402 | to change your C<$buffer> or there was a decoding error and you want to |
|
|
403 | reuse the C<$cbor> object for future incremental parsings. |
| 335 | |
404 | |
| 336 | =back |
405 | =back |
| 337 | |
406 | |
| 338 | |
407 | |
| 339 | =head1 MAPPING |
408 | =head1 MAPPING |
| … | |
… | |
| 798 | perl core distribution (e.g. L<URI>), it is (currently) up to the user to |
867 | perl core distribution (e.g. L<URI>), it is (currently) up to the user to |
| 799 | provide these modules. The decoding usually fails with an exception if the |
868 | provide these modules. The decoding usually fails with an exception if the |
| 800 | required module cannot be loaded. |
869 | required module cannot be loaded. |
| 801 | |
870 | |
| 802 | =over 4 |
871 | =over 4 |
|
|
872 | |
|
|
873 | =item 0, 1 (date/time string, seconds since the epoch) |
|
|
874 | |
|
|
875 | These tags are decoded into L<Time::Piece> objects. The corresponding |
|
|
876 | C<Time::Piece::TO_CBOR> method always encodes into tag 1 values currently. |
|
|
877 | |
|
|
878 | The L<Time::Piece> API is generally surprisingly bad, and fractional |
|
|
879 | seconds are only accidentally kept intact, so watch out. On the plus side, |
|
|
880 | the module comes with perl since 5.10, which has to count for something. |
| 803 | |
881 | |
| 804 | =item 2, 3 (positive/negative bignum) |
882 | =item 2, 3 (positive/negative bignum) |
| 805 | |
883 | |
| 806 | These tags are decoded into L<Math::BigInt> objects. The corresponding |
884 | These tags are decoded into L<Math::BigInt> objects. The corresponding |
| 807 | C<Math::BigInt::TO_CBOR> method encodes "small" bigints into normal CBOR |
885 | C<Math::BigInt::TO_CBOR> method encodes "small" bigints into normal CBOR |
| … | |
… | |
| 972 | service. I put the contact address into my modules for a reason. |
1050 | service. I put the contact address into my modules for a reason. |
| 973 | |
1051 | |
| 974 | =cut |
1052 | =cut |
| 975 | |
1053 | |
| 976 | our %FILTER = ( |
1054 | our %FILTER = ( |
| 977 | # 0 # rfc4287 datetime, utf-8 |
1055 | 0 => sub { # rfc4287 datetime, utf-8 |
| 978 | # 1 # unix timestamp, any |
1056 | require Time::Piece; |
|
|
1057 | # Time::Piece::Strptime uses the "incredibly flexible date parsing routine" |
|
|
1058 | # from FreeBSD, which can't parse ISO 8601, RFC3339, RFC4287 or much of anything |
|
|
1059 | # else either. Whats incredibe over standard strptime totally escapes me. |
|
|
1060 | # doesn't do fractional times, either. sigh. |
|
|
1061 | # In fact, it's all a lie, it uses whatever strptime it wants, and of course, |
|
|
1062 | # they are all incomptible. The openbsd one simply ignores %z (but according to the |
|
|
1063 | # docs, it would be much more incredibly flexible indeed. If it worked, that is.). |
|
|
1064 | scalar eval { |
|
|
1065 | my $s = $_[1]; |
|
|
1066 | |
|
|
1067 | $s =~ s/Z$/+00:00/; |
|
|
1068 | $s =~ s/(\.[0-9]+)?([+-][0-9][0-9]):([0-9][0-9])$// |
|
|
1069 | or die; |
|
|
1070 | |
|
|
1071 | my $b = $1 - ($2 * 60 + $3) * 60; # fractional part + offset. hopefully |
|
|
1072 | my $d = Time::Piece->strptime ($s, "%Y-%m-%dT%H:%M:%S"); |
|
|
1073 | |
|
|
1074 | Time::Piece::gmtime ($d->epoch + $b) |
|
|
1075 | } || die "corrupted CBOR date/time string ($_[0])"; |
|
|
1076 | }, |
|
|
1077 | |
|
|
1078 | 1 => sub { # seconds since the epoch, possibly fractional |
|
|
1079 | require Time::Piece; |
|
|
1080 | scalar Time::Piece::gmtime (pop) |
|
|
1081 | }, |
| 979 | |
1082 | |
| 980 | 2 => sub { # pos bigint |
1083 | 2 => sub { # pos bigint |
| 981 | require Math::BigInt; |
1084 | require Math::BigInt; |
| 982 | Math::BigInt->new ("0x" . unpack "H*", pop) |
1085 | Math::BigInt->new ("0x" . unpack "H*", pop) |
| 983 | }, |
1086 | }, |
| … | |
… | |
| 1019 | } |
1122 | } |
| 1020 | |
1123 | |
| 1021 | sub URI::TO_CBOR { |
1124 | sub URI::TO_CBOR { |
| 1022 | my $uri = $_[0]->as_string; |
1125 | my $uri = $_[0]->as_string; |
| 1023 | utf8::upgrade $uri; |
1126 | utf8::upgrade $uri; |
| 1024 | CBOR::XS::tag 32, $uri |
1127 | tag 32, $uri |
| 1025 | } |
1128 | } |
| 1026 | |
1129 | |
| 1027 | sub Math::BigInt::TO_CBOR { |
1130 | sub Math::BigInt::TO_CBOR { |
| 1028 | if ($_[0] >= -2147483648 && $_[0] <= 2147483647) { |
1131 | if ($_[0] >= -2147483648 && $_[0] <= 2147483647) { |
| 1029 | $_[0]->numify |
1132 | $_[0]->numify |
| 1030 | } else { |
1133 | } else { |
| 1031 | my $hex = substr $_[0]->as_hex, 2; |
1134 | my $hex = substr $_[0]->as_hex, 2; |
| 1032 | $hex = "0$hex" if 1 & length $hex; # sigh |
1135 | $hex = "0$hex" if 1 & length $hex; # sigh |
| 1033 | CBOR::XS::tag $_[0] >= 0 ? 2 : 3, pack "H*", $hex |
1136 | tag $_[0] >= 0 ? 2 : 3, pack "H*", $hex |
| 1034 | } |
1137 | } |
| 1035 | } |
1138 | } |
| 1036 | |
1139 | |
| 1037 | sub Math::BigFloat::TO_CBOR { |
1140 | sub Math::BigFloat::TO_CBOR { |
| 1038 | my ($m, $e) = $_[0]->parts; |
1141 | my ($m, $e) = $_[0]->parts; |
| 1039 | CBOR::XS::tag 4, [$e->numify, $m] |
1142 | tag 4, [$e->numify, $m] |
|
|
1143 | } |
|
|
1144 | |
|
|
1145 | sub Time::Piece::TO_CBOR { |
|
|
1146 | tag 1, 0 + $_[0]->epoch |
| 1040 | } |
1147 | } |
| 1041 | |
1148 | |
| 1042 | XSLoader::load "CBOR::XS", $VERSION; |
1149 | XSLoader::load "CBOR::XS", $VERSION; |
| 1043 | |
1150 | |
| 1044 | =head1 SEE ALSO |
1151 | =head1 SEE ALSO |
